Enveloped viruses initiate infection by merging their membrane with the target cell membrane. Our research is aimed at understanding the molecular mechanisms of enveloped virus entry into cells. We investigate entry/fusion of diverse viruses, such as HIV, influenza virus and Avian Sarcoma and Leukosis Virus (ASLV), using functional assays, including single virus tracking. Virus co-labeling with fluorescent membrane, content and/or core markers enables the detection of lipid mixing (hemifusion) and viral content and core release (fusion pore formation) during viral entry. Additional labeling and imaging strategies enable the visualization of transport, fusion and uncoating of HIV cores en route to the nucleus. We are interested in delineating the mechanisms of antiviral activity of host restriction factors that target the viral fusion and in identifying novel small molecule inhibitors of virus fusion.
HIV-1 entry and fusion
Our laboratory has characterized the mechanism of HIV Env glycoprotein-mediated membrane fusion. Through the usage of various interventions and fusion inhibitors, we identified key intermediate steps of this process, such as a target membrane-inserted pre-hairpin intermediate, and showed that the requisite number of gp41 completes refolding into the final 6-helix bundle structure after the formation of a fusion pore (Figure 1).

Figure 1
Single particle tracking revealed that HIV, which has long been thought to infect host cells by fusing directly with the plasma membrane, enters several target cells via endocytosis and pH-independent fusion with endosomes(Figure 2 and Movie 1). Our imaging assay showed that HIV fusion initiated at the cell surface did not progress beyond a lipid mixing stage. These surprising findings offered a new paradigm for HIV entry and implicated yet unknown endocytic trafficking factors in facilitating HIV fusion. Through a targeted shRNA screen, we have since identified several endocytic trafficking proteins as essential factors for HIV fusion and are investigating their roles in the entry process.

Figure 2
In spite of the apparent lack of HIV fusion at the plasma membrane, viruses “wedged” between two adjacent cells can mediate cell-cell fusion. Unlike HIV-cell fusion, HIV-mediated cell-cell fusion (referred to as fusion-from-without, FFWO) is very inefficient and highly dependent on actin dynamics. Actin-dependence suggests a role for an external cell-generated force in HIV-1 fusion at the cell surface. Particles adhered to the plasma membranes of both neighboring cells – a condition that is satisfied in FFWO, but not for viruses bound to a single cell – could be subjected to forces that likely originate from Env-mediated signaling and actin remodeling. We thus hypothesize that HIV relies on mechanical tension in a cell membrane to dilate nascent fusion pores and release its genome into the cytoplasm.

We have also implemented a novel virus labeling strategy that biases towards detection of virus fusion at the plasma membrane. Virus particles are co-labeled with a releasable marker (mCherry, red) and an extra-viral pH sensor, ecliptic pHluorin (EcpH, green), which is fully quenched at mildly acidic pH (Figure 3). ). Whereas we were not able to reliably detect HIV fusion at the cell surface, the pH-sensor revealed that these particles occasionally shuttle between neutral and acidic compartments in target cells expressing CD4 (Figure 3 and Movie 2). It thus appears that a small fraction of viral particles is recycled to the plasma membrane and re-internalized.

Figure 3
In collaboration with the Emory Chemical Biology Discovery Center (Drs. Yuhong Du and Haian Fu), we have carried out high-throughput screening for small molecule HIV fusion inhibitors. The hits from a 100,000 compound library are being analyzed and validated. Among the hits were several purinergic receptor antagonists as inhibitors of HIV fusion. Although published work suggested a role for purinergic receptors in HIV entry and fusion, we have shown that NF279, a P2X1 receptor antagonist, blocks the binding of both CCR5 and CXCR4 coreceptors by the Env-CD4 complexes (Figure 4). Consistent with its coreceptor-blocking function, NF279 also antagonizes the signaling function of CCR5 and CXCR4, as evidenced by the suppression of calcium responses elicited by their specific ligands RANTES and SDF1-α (Movie 3).

Figure 4
Real-time imaging of HIV core uncoating in living cells
Disassembly of the cone-shaped HIV capsid (uncoating) after virus-cell fusion is a prerequisite for establishing a life-long infection. We have developed a novel strategy to visualize HIV uncoating that is based on a fluorescently tagged oligomeric form of a capsid-binding host protein cyclophilin A (CypA-DsRed). CypA-DsRed, which is specifically packaged into virions through the high-avidity binding to the HIV-1 capsid, does not compromise the infectivity. This probe remains associated with cores after virus-cell fusion and is released upon uncoating or by treatment with cyclosporine A CsA,(Figure 5. Importantly, the rate of CypA-DsRed loss from individual post-fusion cores is accelerated by reverse transcription and is modulated by the capsid (CA) mutations that alter the core stability (Figure 5). The CypA-DsRed based imaging assay showed biphasic HIV-1 uncoating, with a large number of cores shedding the CA marker shortly after fusion (Movies 4 and a small fraction of cores undergoing gradual uncoating at late times post-infection Movie 5). We are currently delineating CypA-DsRed interactions with the HIV core, using imaging and structural approaches, and assessing the effect of CA-interacting host factors on the rate of uncoating. We are also interested in visualizing HIV uncoating in the context of transport of pre-integration complexes through the nuclear pore.

Figure 5
Host factors restricting virus entry/fusion
Interferon-induced transmembrane proteins (IFITMs, Figure 6) inhibit infection of diverse enveloped viruses, including the influenza A virus (IAV), West Nile, Ebola and other viruses. We have examined the mechanism of IAV restriction by IFITM3 protein using direct virus-cell fusion assay and single virus imaging in live cells. IFITM3 did not inhibit lipid mixing, but abrogated the release of viral content into the cytoplasm (Figure 6 and Movies 6 and Movie 7). IFITM3’s ability to block fusion pore formation at a post-hemifusion stage shows that this protein stabilizes the cytoplasmic leaflet of endosomal membranes without adversely affecting the lumenal leaflet. Alternatively, IFITM3 may redirect IAV fusion to a non-productive pathway by promoting fusion with intralumenal vesicles within multivesicular bodies/late endosomes.

We have also examined the mechanisms of action of broad antiviral effects of human alpha-defensins. We found that the human neutrophil peptide 1 potently blocks HIV-1 fusion by interfering with virtually all major steps of viral fusion, including HIV-1 endocytosis. We also found that even sub-inhibitory doses of this defensin in the presence of serum dramatically potentiate the neutralizing activity of antibodies against the HIV-1 gp41 N-terminal heptad repeat domain.

Figure 6
Correlated light-electron microscopy (CLEM) approach to elucidate fusion intermediates of virus entry
In collaboration with Dr. Elizabeth Wright at Emory University, we are developing novel approaches to combine live cell single virus fluorescence microscopy with cryo-electron tomography of the viral fusion sites. Proof-of-concept correlated images of single retrovirus engulfed by a clathrin-coated vesicle are shown in Figure 7. Together with Drs. Wright and Spearman, we have recently acquired a state-of-the-art Leica cryo-CLEM system, which is equipped with a cryo-stage and a cryo-transfer system that allow direct fluorescence imaging of frozen samples on grids, using a 50x/0.90NA long working distance objective.

Figure 7
ASLV as a model system for studies of endosomal entry
The receptor- and low pH-driven ASLV fusion is amenable for studies of virus entry via endocytosis. Owing to the extreme stability of Env-receptor complexes at neutral pH, ASLV fusion can be reversibly arrested by raising the pH and then synchronously released by removing the fusion block. Synchronized ASLV fusion is initiated within seconds after acidification and often culminates in release of the viral core from an endosome (Figure 8 and Movie 8).

Figure 8
Endosomal pH is one of the critical determinants of the location and timing of envelope protein refolding, which in turn drives viral fusion. Through incorporating pH-sensitive probes into the viral membrane, we were able to monitor the pH in virus-carrying endosomes and subsequent virus-endosome fusion (Figure 9 and Movie 9).

Figure 9
Reconstitution of virus fusion in a minimal model system
We have reconstituted fusion of single retroviral particles pseudotyped with the Vesicular Stomatitis Virus (VSV) G protein with dextran-supported lipid bilayers (Figure 10). Incorporation of diffusible fluorescent labels into the viral membrane and the viral interior enabled detection of the lipid mixing (hemifusion) and content transfer (full fusion) steps of VSV G-mediated fusion at low pH. We found that phosphatidylserine (PS) and bis(monoacylglycero)phosphate (BMP) greatly enhance the efficiency of hemifusion and permitted full virus fusion (Figure 10 and Movie 10). The strong fusion-enhancing effect of BMP, a late endosome-resident lipid, is consistent with the model that VSV initiates fusion in early endosomes, but releases its core into the cytosol after reaching late endosomal compartments.

Figure 10