Yellow fever virus (YFV) infects cells through a multi-step sequence: it latches onto sugar-coated molecules on the cell surface, rides into the cell inside a bubble of membrane, and then uses the acidic environment deep within that bubble to fuse its own outer shell with the bubble wall, dumping its genetic material into the cell’s interior. Each step depends on a single workhorse molecule, the envelope (E) protein, which dominates the virus’s outer surface and undergoes dramatic shape changes to drive the process forward. The details of how those steps unfold reveal why the virus targets certain tissues, how the live vaccine differs from wild strains, and where researchers see opportunities to block infection.
The Virus Particle and Its Envelope Armor
A mature yellow fever virion is a small sphere, roughly 50 nanometers across, with an outer shell built from 180 copies of the E protein arranged in a herringbone pattern around an icosahedral frame.1International Journal of Biomedicine. Abstract OR-8: Cryo-EM Structure of Mature Yellow Fever Virus The E protein is the virus’s Swiss Army knife: it handles attachment to the host cell, responds to pH changes, and physically merges the viral and cellular membranes. It folds into three distinct domains, labeled I, II, and III. Domain III, positioned at the outermost tip, contains the major site where neutralizing antibodies bind and is thought to interact most directly with host cell receptors.2PubMed Central. Structure of yellow fever virus envelope protein domain III Domain II harbors a “fusion loop,” a short stretch of hydrophobic amino acids that will eventually plunge into the host membrane to initiate fusion.
Before the virus is fully mature, a small chaperone protein called pr sits on top of the E protein, capping the fusion loop to prevent the virus from fusing prematurely with membranes inside the cell that made it. As the newly assembled virus travels through the secretory pathway and encounters the neutral pH of the extracellular environment, pr is knocked off through a localized shape change, arming the virus for its next encounter with a host cell.3PubMed Central. New insight into flavivirus maturation from structure/function studies of the yellow fever virus envelope protein complex This activation step is essential: a virus particle that still carries pr on its surface cannot fuse and is effectively dead on arrival.
Subtle differences in E protein structure matter enormously. A single amino acid at position 380 (an arginine in the 17D vaccine strain) stabilizes the way neighboring E proteins pack together on the viral surface by forming bonds across the gaps between protein clusters called rafts.4Nature Communications. A single residue in the yellow fever virus envelope protein modulates virion architecture and antigenicity That same residue also happens to be a dominant determinant for binding to certain sugars on the host cell, which influences how quickly the virus is cleared from the bloodstream.5PubMed Central. E protein domain III determinants of yellow fever virus 17D vaccine strain enhance binding to glycosaminoglycans, impede virus spread, and attenuate virulence One residue, in other words, simultaneously shapes the virus’s physical stability, its ability to grab onto cells, and its behavior in a living host.
Grabbing Hold of the Cell Surface
The first thing YFV needs is a handshake with something on the outside of a target cell. The best-studied attachment factor is heparan sulfate, a long chain of sugars that decorates proteins embedded in the membranes of many human cell types. When researchers stripped heparan sulfate from cells using enzymes, or flooded the system with the closely related molecule heparin, yellow fever virus binding and infection dropped dramatically, with heparin alone blocking infection by about 97%.6PubMed. Heparan sulfate-mediated binding of infectious dengue virus type 2 and yellow fever virus Heparan sulfate is not a true receptor in the classical sense; it acts more like molecular Velcro, concentrating virus particles near the cell surface so they can find a genuine entry receptor.
What that definitive entry receptor is remains one of the bigger open questions in YFV biology. Research on related flaviviruses has flagged several candidate molecules, including C-type lectin receptors such as DC-SIGN, phosphatidylserine receptors like TIM-1, and TAM-family receptors such as AXL and Tyro3.7Scientific Reports. Human Schwann cells are susceptible to infection with Zika and yellow fever viruses, but not dengue virus Different cell types express different combinations of these molecules, which likely contributes to the virus’s preference for certain tissues. But no single molecule has been definitively proven to be the obligatory entry receptor for YFV in the way that ACE2 is for SARS-CoV-2. The search continues, and it is complicated by the fact that the virus can probably use more than one receptor depending on the cell type it encounters.
Riding Into the Cell
Once attached, the virus does not bore through the membrane or fuse at the surface. Instead, it hijacks the cell’s own import machinery. The cell wraps a patch of membrane around the virus particle, pinches it off, and carries the resulting bubble, called an endosome, inward. For YFV and most other mosquito-borne flaviviruses, this process depends on clathrin, a protein that forms a cage-like lattice under the membrane to shape and pull in the budding vesicle.8PubMed Central. Beyond the Surface: Endocytosis of Mosquito-Borne Flaviviruses The process also requires dynamin, a protein that acts like molecular scissors to snip the vesicle free from the cell surface.9PubMed Central. A Sensitive Yellow Fever Virus Entry Reporter Identifies Valosin-Containing Protein (VCP/p97) as an Essential Host Factor for Flavivirus Uncoating
Some evidence suggests that a subset of flaviviruses use a specialized version of clathrin-mediated entry that depends on additional cellular factors not required by the textbook pathway.8PubMed Central. Beyond the Surface: Endocytosis of Mosquito-Borne Flaviviruses Researchers have also identified host proteins like LY6E and RPLP1 as essential for productive YFV entry, and a protein called VCP/p97, a cellular machine normally involved in protein quality control, has been shown to be required for uncoating the virus once it reaches the cytoplasm.9PubMed Central. A Sensitive Yellow Fever Virus Entry Reporter Identifies Valosin-Containing Protein (VCP/p97) as an Essential Host Factor for Flavivirus Uncoating These host-dependency factors represent potential drug targets because they are cellular proteins the virus cannot easily mutate away from needing.
The Acid Bath That Triggers Fusion
The endosome is not a static holding tank. As it matures, proton pumps in its membrane actively lower the pH inside, creating an increasingly acidic environment. This acidification is the trigger the virus has been waiting for. At low pH, the E protein undergoes a sweeping conformational change: the flat-lying dimers on the viral surface dissociate, and the individual E proteins rear up, exposing their fusion loops to plunge into the endosomal membrane.10PubMed. Molecular mechanisms of flavivirus membrane fusion If you block endosomal acidification with drugs, YFV entry stalls completely.9PubMed Central. A Sensitive Yellow Fever Virus Entry Reporter Identifies Valosin-Containing Protein (VCP/p97) as an Essential Host Factor for Flavivirus Uncoating
The pr protein that capped the fusion loop during maturation is also pH-sensitive. At the low pH found inside endosomes, pr binds the E protein tightly, with an affinity in the low-nanomolar range. In experiments using artificial membranes, the presence of pr at a one-to-one ratio with the E protein completely blocked the E protein from inserting into the membrane at acidic pH.11bioRxiv. Molecular mechanisms regulating the pH-dependent pr/E interaction in yellow fever virus This might seem contradictory, since acidic pH is supposed to trigger fusion. The resolution is timing: pr is shed at neutral pH during virus maturation and is no longer present on the mature virion. The acid sensitivity of the pr–E interaction is a safety mechanism for the cell that produced the virus, not the cell the virus is about to invade.
Researchers have been able to freeze the fusion process midway by carefully controlling pH. When flavivirus particles are exposed to mildly acidic conditions and then quickly returned to alkaline pH, the E proteins get stuck in a “prehairpin” intermediate: they have inserted their fusion loops into the target membrane but have not yet folded back on themselves to pull the two membranes together.12PLoS Pathogens. Characterization of a Structural Intermediate of Flavivirus Membrane Fusion Completing the fold requires sustained low pH and results in domain III swinging around to pack against the other end of the molecule, forming a stable trimer of E proteins that drives the two membranes into full merger. A pore opens, and the viral RNA genome spills into the cytoplasm.
Setting Up Shop Inside the Cell
Once free in the cytoplasm, the viral RNA is immediately readable by the cell’s protein-making machinery because it is a positive-sense strand, essentially a ready-made messenger RNA. Ribosomes translate it into a single long polyprotein that is then cleaved by viral and host enzymes into three structural proteins and seven nonstructural (NS) proteins. Several of those NS proteins have a talent for reshaping the cell’s internal membranes. They commandeer the endoplasmic reticulum, the sprawling membrane network where the cell normally manufactures and folds its own proteins, and sculpt it into specialized compartments: small invaginated vesicles called vesicle packets and larger tangles of membrane known as convoluted membranes.13PubMed Central. Compartmentalized replication organelle of flavivirus at the ER and the factors involved These remodeled structures serve as the virus’s replication factories, sheltering the copying machinery from the cell’s innate immune sensors while concentrating everything needed to churn out new RNA genomes.
The NS5 protein is the largest and arguably most important nonstructural protein. It contains the RNA-dependent RNA polymerase, the enzyme that copies the viral genome, and it also interacts with host factors to support replication. One such interaction involves the translation initiation factor eIF3L, a host protein that helps ribosomes start reading messenger RNAs. NS5 binds eIF3L through a region that is conserved across multiple flaviviruses, suggesting the virus may redirect or fine-tune the host translation machinery to favor its own output.14PubMed Central. The eukaryotic translation initiation factor 3 subunit L protein interacts with Flavivirus NS5 and may modulate yellow fever virus replication
Why the Liver Takes the Hardest Hit
Yellow fever is named for the jaundice that marks severe cases, and the liver is its principal battlefield. Viral antigens concentrate most heavily in hepatocytes in the midzonal region of the liver lobule, a zone that sits between the oxygen-rich periportal area and the oxygen-poor central vein.15PubMed. Revisiting the liver in human yellow fever: virus-induced apoptosis in hepatocytes associated with TGF-beta, TNF-alpha and NK cells activity Infected hepatocytes die predominantly through apoptosis rather than necrosis, a pattern that helps explain the relatively mild inflammatory infiltrate seen in fatal cases despite massive cell loss.
At a molecular level, YFV infection of human hepatocytes tips the balance between damaging reactive oxygen species and the cell’s antioxidant defenses. Within two to four days of infection, reactive oxygen species levels climb significantly, while enzymes that normally neutralize them lose activity. The result is oxidative damage to lipids, proteins, and DNA, compounding the injury from the virus itself.16PubMed Central. Yellow fever virus infection in human hepatocyte cells triggers an imbalance in redox homeostasis with increased reactive oxygen species production, oxidative stress, and decreased antioxidant enzymes This oxidative stress appears to be a driver of pathology rather than just a bystander, offering a potential explanation for why liver failure develops so rapidly in severe yellow fever.
The liver is not the only tissue the virus reaches. In the early phase of infection, YFV infects dendritic cells and macrophages, the immune sentinels stationed in the skin near the site of a mosquito bite. Both wild-type and vaccine strains replicate in these cells, but the vaccine strain (17D) infects them more efficiently and delivers more RNA into the cytoplasm, which actually helps by triggering a stronger antiviral immune response.17PLOS Neglected Tropical Diseases. Characterization of Yellow Fever Virus Infection of Human and Non-human Primate Antigen Presenting Cells and Their Interaction with CD4+ T Cells The virus’s NS1 protein also contributes to tissue-specific damage. NS1 from YFV triggers the strongest vascular leakage in liver endothelial cells, mirroring the clinical disease, while NS1 from encephalitic flaviviruses preferentially disrupts brain endothelial cells.18Cell Reports. Flavivirus NS1 Proteins Direct Tissue-Specific Endothelial Dysfunction and Vascular Leakage
How the Virus Dodges the Immune Alarm
Cells have an early warning system against viruses: interferons, signaling molecules that tell neighboring cells to ramp up their antiviral defenses. YFV has evolved a targeted countermeasure. The NS5 protein binds to STAT2, a key transcription factor in the interferon signaling chain, but it does so through an unusual mechanism: the interaction only occurs in cells that have already been stimulated by type I interferon. The virus essentially hijacks the cell’s own response to interferon as a trigger for shutting that response down. This process requires a specific modification of NS5 (a type of polyubiquitination carried out by the host enzyme TRIM23) and depends on STAT1 being activated first.19Cell Host & Microbe. Yellow Fever Virus NS5 Inhibits Type I Interferon Signaling through Binding to STAT2
The importance of interferon signaling is underscored by experiments in mice. Animals lacking the receptor for type I interferon develop robust infection with the vaccine strain but survive. When those animals also lack the receptor for type III interferon (a related but distinct pathway), they succumb to the infection, with the virus invading the brain and the blood-brain barrier breaking down.20PubMed Central. Type III Interferon-Mediated Signaling Is Critical for Controlling Live Attenuated Yellow Fever Virus Infection In Vivo Type III interferon, in other words, serves as a critical backup system, particularly in protecting the nervous system from viral invasion. This layered defense explains why most healthy people clear even wild-type YFV infection without neurological involvement, and why the rare severe cases tend to occur in individuals with compromised immune signaling.
Vaccine Versus Wild-Type Differences in Entry
The 17D vaccine strain, developed in the 1930s through serial passage, differs from its wild-type ancestor (the Asibi strain) by only a handful of amino acid changes, yet it behaves quite differently during entry. The vaccine strain binds to and infects host cells more efficiently than Asibi, delivering more viral RNA into the cytoplasm.21PubMed Central. Vaccine and Wild-Type Strains of Yellow Fever Virus Engage Distinct Entry Mechanisms and Differentially Stimulate Antiviral Immune Responses This might seem backwards: shouldn’t more efficient entry make a virus more dangerous? The answer lies in what happens next. Greater RNA delivery triggers a stronger cytokine-mediated antiviral response, essentially setting off the alarm louder and faster. The virus replicates briefly but is quickly contained.
In mosquitoes, the story is reversed. When Aedes aegypti mosquitoes feed on blood containing the 17D vaccine strain, very few midgut cells become infected, suggesting an entry defect at the insect gut barrier. Researchers observed only scattered foci of infection by immunofluorescence, consistent with the vaccine E protein having lower affinity for whatever attachment factors mosquito midgut cells use.22PLOS Neglected Tropical Diseases. Midgut barriers prevent the replication and dissemination of the yellow fever vaccine in Aedes aegypti This means the vaccine strain is biologically unable to complete the mosquito-to-human transmission cycle, an important safety feature that has held up across decades of mass vaccination.
Blocking Entry With Antibodies
Understanding the entry mechanism has direct implications for therapy and vaccine design. Neutralizing antibodies work by gumming up one or more of the steps described above. One well-characterized monoclonal antibody, called 5A, targets the E protein dimer and blocks both attachment and fusion. In experiments, 5A reduced virus binding to cells in a dose-dependent manner and also inhibited the pH-triggered fusion of virus particles with artificial membranes.23Cell Reports. Crystal Structures of the Yellow Fever Virus Envelope Protein and Its Complex with a Potent Neutralizing Monoclonal Antibody Other researchers have mapped a “neutralizing supersite” on the virus surface where the most potent antibodies converge, with the best of them neutralizing infection at concentrations below 0.01 micrograms per milliliter.24The Innovation. A neutralizing-protective supersite of human monoclonal antibodies for yellow fever virus
These ultra-potent antibodies could matter for more than academic interest. Yellow fever has no approved antiviral drug, and severe cases are treated with supportive care only. If monoclonal antibodies targeting the entry pathway can be produced at scale, they could offer a therapeutic option for patients who present after infection is established but before organ failure sets in. The same structural insights also inform next-generation vaccine design, since the goal of any vaccine is to teach the immune system to produce exactly these kinds of entry-blocking antibodies.
The Mosquito Entry Problem Nobody Has Solved
Nearly everything described above applies to mammalian cells. In the Aedes aegypti mosquitoes that carry YFV between humans, the entry picture is far murkier. No attachment factor and no entry receptor have been identified for flavivirus infection of mosquito cells.22PLOS Neglected Tropical Diseases. Midgut barriers prevent the replication and dissemination of the yellow fever vaccine in Aedes aegypti Domain III of the E protein is involved, just as it is in mammalian cells, but the specific molecular partners on the mosquito side remain unknown. This gap is not trivial. Interrupting the virus’s ability to infect mosquito midgut cells would break the transmission cycle at its source, potentially offering a vector-control strategy that complements vaccines. Until the mosquito receptors are found, that approach remains out of reach.