Click Image for Figure and Legend

Virus Mediated Cell-to-Cell Fusion
Paramyxoviruses enter cells by fusion of the viral envelope with the cellular plasma membrane and the viral fusion protein (F) mediates this process. This protein is synthesized as a precursor that has to be cleaved for the protein to be biologically active. The initial folding of the F protein causes it to be trapped energetically in a metastable form. Cleavage causes a significant conformational change in the protein but it remains in a metastable form. Activation of the F protein for fusion activity requires coexpression of the viral receptor-binding protein, hemagglutinin-neuraminidase (HN). When HN binds its receptor, sialic acid, HN undergoes a conformational change, which in turn causes a further conformational change in the F protein. Finally, refolding of the F protein and the formation of a six-helix bundle (6HB) (core trimer) cause the merger of the cell target membrane with the viral membrane. Recently, we have found that activation of the paramyxovirus fusion protein is modulated by inside-out signaling from the cytoplasmic tail. We are using a combination of cellular assays, biophysical measurements and microscopic assays as well as protein structural determination at the atomic level to approach this problem.

We have hypothesized that the F protein is a biological nanomachine that undergoes a series of conformational changes and it is controlled such that fusion only occurs at the right time and in the right place.

Recently we determined the crystal structure of both the pre-fusion and post-fusion forms of the paramyxovirus F protein in collaboration with Theodore Jardetzky (Northwestern University). The pre-fusion F structure was solved for PIV5 and the post-fusion F structure was solved for hPIV3. In the large part the structures are thought to be representative of the F structures of all paramyxoviruses. To produce a soluble form of the pre-fusion F protein we appended an engineered, trimeric coiled coil domain (GCNt) to heptad repeat B (HRB) to mimic the transmembrane domains. The resulting F protein (F-GCNt) assembled into well-defined trimers. The pre-fusion F stucture contains a globular head attached to a trimeric coiled-coil stalk formed by the C-terminal heptad repeat B region (HRB). The globular head contains three domains (DI-DIII). The fusion peptides at the N-terminus of heptad repeat A (HRA) region are sequestered between adjacent subunits, with cleavage/activation sites exposed at the protein surface.

The post-fusion F forms a trimer, which reveals a globular, predominantly beta-sheet containing head domain, a neck region formed by both b-sheet and a-helices and a stalk region that is predominantly a-helical. The structure contained the 6HB expected of the post-fusion conformation of the protein. The only part of the structure lacking electron density are amino acids (95-135) begin 5 residues N-terminal to the internal cleavage site, extending through the fusion peptide (starting at residue 110) and the N-terminal part of HRA , but these residues would be draped flexibly on the exterior of the stalk region. It had been widely anticipated that cleavage of F at the cleavage site was a requirement for conversion to the post-fusion form. Nonetheless, many lines of evidence suggested that the observed F conformation represented the post-fusion form, although the polypeptide chains were intact in the crystal and the fusion peptide was not located at the appropriate end of the 6HB.

The PIV5 pre-fusion F and hPIV3 post-fusion F structures are in strikingly different conformations, consistent with a transition from pre- to post-fusion forms. We have observed related forms of the F protein in electron micrographs of F. None of the intersubunit contacts are conserved in the pre- and post-fusion forms. The two F structures are related by flipping the stalk and TM domains relative to the F head. Substantial compacting of the head is observed in hPIV3 post-fusion F compared to PIV5 pre-fusion F. DI domains pivot slightly inwards, shearing intersubunit contacts, and DII domains swing across, contacting neighboring subunits.   Individual DI and DII domains in the two structures remain similar. DIII undergoes major refolding between the two structures, projecting a new coiled coil (HRA) upwards and away from DI, the pre-fusion stalk and the viral membrane. The fusion peptide, located at the top of the HRA coiled coil, moves ~115Å from its initial position between subunits in the pre-fusion conformation, allowing DII domains to reposition. None of the post-fusion HRA intersubunit coiled-coil contacts are observed in F-GCNt. Instead they are replaced by two sets of 6-helix rings at the DIII interfaces. For the HRA coiled coil to form, DIII must rotate and collapse inwards, further compacting the head.

The conformational change also requires the opening and translocation of the HRB stalk. In the pre-fusion form, HRB is located at the base of the head region. During the conversion to the post-fusion conformation, HRB segments must separate and swing around the base of the head, to pack against the HRA coiled coil. In the prefusion conformation, HRA is broken up into 4 helices, 2 b-strands and 5 loop, kink or turn segments. Thus, the conformational changes in HRA involve the refolding of 11 distinct segments into a single, extended a-helical conformation.

The pre-fusion F structure provides a model for the stepwise induction of membrane fusion by paramyxoviruses and reveals how multiple sequence elements play distinct structural roles in the pre- and post-fusion conformations.

We also solved the atomic structure of the tetrameric paramyxovirus HN protein to a resolution of 2. 8 Å, in collaboration with Theodore Jardetzky. Based on the HN structure we propose a model for HN involvement in membrane fusion that is consistent with the available data and that involves ligand-dependent changes in the HN oligomer that are driven by surface:surface interactions. In this model, the HN dimer/tetramer forms in the absence of ligands and can interact with the F protein, potentially through lateral interactions on two sides of the tetramer. Engagement of cell surface receptors could trigger the partial disassembly of the HN tetramer, assuming that the energy of binding of the individual HN sites to distinct sialic receptors is sufficient to perturb the weak NA domain interactions. Opening of the tetrameric head, driven by the energy of receptor engagement, could lead to changes in both the HN stalk region (the HN stalk domain is thought to interact with F) and the interaction with F, thus activating F for membrane fusion.