The HIV Virus Structure: Key Components and Architecture

HIV is a roughly spherical particle about 120 nanometers across, built from a stolen patch of human cell membrane wrapped around a cone-shaped protein shell that protects two copies of its RNA genome. That description sounds simple, but each layer is an engineered solution to a specific problem the virus faces: getting inside cells, hiding from the immune system, and converting its genetic blueprint into DNA once it arrives. The architecture is surprisingly ordered and has been mapped in increasing detail over the past two decades using imaging techniques that can resolve individual protein subunits.

The Lipid Envelope

The outermost layer of HIV is not something the virus makes from scratch. It is a piece of the host cell’s own plasma membrane, grabbed during budding. This lipid bilayer is not a random slice of the cell surface, though. HIV buds preferentially from membrane regions enriched in raft-type lipids, and the resulting viral envelope reflects that composition. The virus also displays certain lipids on its outer surface that are normally kept on the inner leaflet of healthy cell membranes, a sign that the usual lipid organization has been disrupted during budding.¹ This matters because the chemical makeup of the envelope is critical for the virus’s ability to fuse with and enter new target cells. A disrupted or chemically altered envelope can cripple the virus before it even reaches a receptor.

Because the envelope is host-derived, it also carries host proteins. This is not entirely accidental. While you might expect the virus to passively sweep up whatever membrane proteins happen to be nearby, evidence suggests the incorporation of host proteins can be selective and conserved across viral strains.² Some of these borrowed host molecules may help the virus evade immune detection or improve its ability to attach to new cells.

Envelope Glycoproteins and the Glycan Shield

Studding the lipid envelope are the virus’s own surface proteins, known collectively as Env. Each functional unit is a trimer: three copies of a complex made from two linked subunits called gp120 and gp41. The gp120 portion sits on the outside and is responsible for recognizing and binding to receptors on target cells. The gp41 portion anchors the complex in the membrane and drives the fusion machinery that lets the virus merge with a host cell. Structures of the prefusion-closed Env trimer have been solved, and antibodies that recognize all major exposed surfaces of this trimer have now been cataloged into roughly two dozen classes spanning six broad epitope categories.³ The interface between gp120 and gp41 has emerged as a particularly important target for broadly neutralizing antibodies.⁴

What makes the Env trimer so difficult for the immune system to neutralize is the glycan shield. Each gp120 molecule is coated with sugar chains, roughly half the molecular weight of gp120 is carbohydrate. These glycans are not just decoration. Simulations and structural studies show that the sugars form an extensive network of interactions with one another, creating a higher-order structure that defines which parts of the protein surface are buried and which remain exposed to antibodies.⁵ The shield is conformationally heterogeneous, meaning the sugars are constantly shifting position, and the degree of shielding they provide changes dynamically.⁶

The overall extent of glycan coverage is conserved across HIV strains, but the precise number, location, and chemical processing of individual glycans varies from one strain to another.⁷ This strain-dependent variation is one reason why designing a vaccine that works against all circulating HIV variants has been so challenging. The glycan shield essentially uses the host’s own sugar-modification machinery to camouflage the virus. Antibodies that manage to target glycan-dependent epitopes are among the most broadly neutralizing ones discovered so far, and understanding how the shield moves and rearranges has become central to vaccine design efforts.

The Matrix Shell

Immediately beneath the lipid envelope sits a layer of matrix protein, known as MA or p17. About 2,000 copies of this protein line the inner surface of the membrane, forming a protective shell. MA is not just structural scaffolding. During virus assembly, the MA domain of the precursor Gag polyprotein is what steers the whole assembly machinery to the correct location on the host cell’s plasma membrane.⁸ It does this partly through a fat molecule (myristate) attached to one end, which acts as a membrane anchor, and partly by specifically recognizing a lipid called PI(4,5)P2 that is concentrated at plasma membrane assembly sites enriched in cholesterol.⁹

In the mature virus, the matrix layer maintains its association with the inner face of the membrane. Electron tomography images consistently show the capsid core positioned about 11 nanometers away from the envelope/MA layer, suggesting a defined spacing between these structural shells.¹⁰ That gap is filled with various proteins and enzymes that travel along with the viral core.

The Conical Capsid

The most visually distinctive feature of a mature HIV particle is its capsid core, and its shape is genuinely unusual among viruses. Rather than a sphere or an icosahedron, HIV builds a cone. Structurally, this cone follows the geometry of a fullerene: a closed shell assembled from about 250 hexameric rings and exactly 12 pentameric rings of the capsid protein CA.¹¹ The pentamers introduce curvature, much like the pentagons in a soccer ball allow a flat surface to close into a sphere. In HIV’s case, the asymmetric placement of pentamers produces the tapered cone shape, with a wider base and a narrower tip. Individual particles vary somewhat in their exact cone angle and dimensions, but the overall architecture is remarkably consistent.

The capsid is not a passive container. Each hexameric ring has a central pore lined by a ring of six positively charged arginine residues, creating a channel that functions like a molecular gate. This pore selectively imports nucleotides from the host cell’s cytoplasm into the capsid interior, fueling the reverse transcription of viral RNA into DNA while the core is still intact.¹² Removing arginine residues from the pore causes a proportional drop in nucleotide uptake, reverse transcription, and infectivity. Molecular simulations have shown that a small cofactor molecule called IP6 enhances nucleotide transport through these pores, while an experimental inhibitor called hexacarboxybenzene blocks it.¹³ The capsid pore is now being explored as a potential drug target distinct from the enzymes traditionally targeted by antiretroviral drugs.

The Nucleocapsid and Genomic RNA

Inside the conical capsid sit two copies of the viral RNA genome, each about 9,700 nucleotides long. The two strands are physically linked near their starting ends through a kissing-loop interaction at a region called the dimerization initiation site. This pairing involves specific structural motifs in the RNA’s untranslated leader region. In the dimeric form, key functional domains like stem-loop 1, the polyadenylation signal, and the primer binding site fold as independent units. When the RNA is monomeric, those same domains reconfigure into alternative long-range and short-range pairings, essentially acting as a structural switch that helps determine whether the RNA gets packaged into new virus particles or stays behind to be translated into proteins.¹⁴

The RNA is not floating freely inside the capsid. It is bound along much of its length by about 1,500 to 2,000 copies of the nucleocapsid protein, NC or p7. NC is small but critical. It contains two zinc-finger motifs that grip the RNA through a combination of electrostatic contacts (charged amino acids binding the phosphate backbone) and stacking interactions (an aromatic amino acid in the second zinc finger wedging between RNA bases).¹⁵ NC also acts as a chaperone during reverse transcription, helping the viral RNA refold and facilitating strand transfers that are essential for copying the genome into DNA.

The Viral Enzymes

Packed inside the capsid alongside the RNA-NC complex are three enzymes the virus needs for its replication cycle. Reverse transcriptase (RT) copies the single-stranded RNA genome into double-stranded DNA. It works as a two-subunit complex and is loaded into the capsid core during assembly. Interestingly, RT incorporation into the core appears to be independent of several other viral components: studies using viruses with deletions in integrase, nucleocapsid, or genomic RNA showed that RT heterodimers were present in the cores at normal levels regardless of those deletions.¹⁶

Integrase is the enzyme that splices the newly synthesized viral DNA into the host cell’s chromosomes, establishing the permanent infection that makes HIV so difficult to eradicate. Protease, the third enzyme, is the one responsible for triggering the virus’s transition from an immature, non-infectious particle into its mature, infectious form. All three enzymes are encoded as part of the Gag-Pol polyprotein, and about 5% of the polyproteins in any given particle are Gag-Pol rather than plain Gag.¹⁷ That ratio ensures enough enzyme molecules are present for the virus to function, while keeping the structural lattice predominantly composed of Gag.

How the Virus Matures

A freshly budded HIV particle looks nothing like the cone-containing mature virus. It starts as an immature virion, in which uncleaved Gag polyproteins form a roughly spherical lattice of hexameric rings lining the inside of the membrane. This immature lattice has gaps and defects, including a large opening at the site where the particle pinched off from the cell. At lattice edges, partial hexamers exist in a dynamic equilibrium, with their helical bundles flickering between ordered and disordered states.¹⁸

Maturation begins when the viral protease activates and starts cleaving the Gag polyprotein at specific sites in a defined order. The cleavage products — MA, CA, NC, and several small peptides — then reorganize. MA stays at the membrane. NC condenses onto the RNA. And the freed CA proteins reassemble into the conical fullerene core.¹⁹ This is not a process of disassembly followed by free-floating reassembly. Studies suggest it happens through a non-diffusional phase transition: the immature CA lattice gradually peels away and rolls into what becomes the surface of the mature cone, without the individual subunits ever fully dissociating into solution.²⁰ This conversion from a spherical shell into a conical core is what transforms a harmless particle into an infectious virus. Drugs called maturation inhibitors aim to block this step.

Accessory Proteins in the Virion

Beyond the major structural proteins and enzymes, a small number of accessory proteins are packaged into HIV particles. The best-studied virion-associated accessory protein is Vpr, which is recruited into budding particles through direct interactions with the p6 domain at the tail end of Gag. Two specific motifs in p6 serve as binding sites for Vpr, and Vpr’s own tendency to form oligomers amplifies the interaction through multivalency: disrupting oligomerization together with removal of either binding motif reduces Vpr incorporation by 25- to 50-fold.²¹ Once inside a new target cell, though, both p6 and Vpr are rapidly shed from the viral core, suggesting their jobs are largely completed by that point.

Vpr’s exact structural role remains a matter of active research. It has been implicated in transporting the viral DNA complex into the nucleus of non-dividing cells, in cell-cycle arrest, and in modulating the host’s DNA damage response. Structural studies using NMR have mapped how p6 and Vpr physically interact, identifying specific residues whose mutation can dramatically change infectivity.²²

How the Capsid Meets the Host Immune System

The conical capsid is not just a container; it is also the viral surface that host defense proteins first encounter after the virus enters a cell and sheds its envelope. One of the best-characterized innate immune sensors of the capsid is TRIM5α, a protein found in primates that can recognize the repeating pattern of CA hexamers on the capsid surface. TRIM5α assembles into its own hexagonal lattice that wraps around the outside of the cone, forming an extensive net. Imaging studies have captured these nets decorating intact viral cores, with arm-like densities about 19 nanometers long arranged in a roughly hexagonal pattern and, in some cases, enveloping the entire capsid.²³

The assembly process is hierarchical. TRIM5α dimers first bind individually to the capsid surface, diffuse across it, and contact other bound dimers to nucleate a growing lattice. As the lattice spreads, it incorporates non-hexagonal defects at regions of high capsid curvature, including pentameric defects along the broad end of the cone and tetrameric defects near the narrow tip.²⁴ The end result is that the host protein builds a cage around the viral capsid, marking it for destruction. In some primate species, TRIM5α is potently active against HIV-1 and essentially prevents infection. In humans, the protein recognizes HIV-1 capsids only weakly, which is one reason the virus can replicate in human cells. Understanding why human TRIM5α fails where other primate versions succeed has been a long-running question in HIV biology, and the structural details of how TRIM5α’s lattice matches or mismatches different capsid geometries are central to answering it.²⁵

How HIV-1 and HIV-2 Differ Structurally

Most structural work focuses on HIV-1, the strain responsible for the global pandemic. HIV-2, which is found primarily in parts of West Africa and progresses more slowly to disease, shares the same general architecture but diverges in some telling details. One of the most studied differences involves the V3 loop of the envelope glycoprotein, a region involved in binding co-receptors on target cells. In HIV-2, the V3 loop is significantly retracted compared to HIV-1, and it potentially folds back to interact with neighboring regions of the protein in a way not seen in HIV-1.²⁶ This structural difference may help explain why HIV-2 is generally harder to transmit and why it responds differently to some drugs and antibodies. The retracted V3 loop means that HIV-2’s envelope exposes different surfaces to the immune system, which has implications for both vaccine design and diagnostic testing.

Imaging Techniques That Revealed the Architecture

Much of what we know about HIV’s structure comes from cryo-electron tomography, a technique that flash-freezes individual virus particles and images them from many angles to build three-dimensional reconstructions. Unlike conventional electron microscopy, which produces flat projections, tomography can reveal the interior of each particle. Early tomographic studies of HIV showed that while every particle’s core had a unique shape and position, the general features were consistent: a cone with a characteristic angle, a broad base, and a defined spacing from the envelope.¹⁰

More recently, combining cryo-electron tomography with subtomogram averaging has allowed researchers to resolve structural intermediates during maturation by imaging viruses carrying mutations that stall protease cleavage at specific steps.²⁷ These snapshots of partially matured particles have been crucial for understanding the order in which the immature lattice disassembles and the mature cone assembles. X-ray crystallography and NMR spectroscopy have filled in atomic-level detail for individual proteins and protein-protein interfaces. The convergence of these methods over the past fifteen years has moved HIV structural biology from coarse cartoons to near-complete atomic models, fundamentally changing how drugs and vaccines are designed against the virus.