HIV's Integrase Protein Has a Second Job Inside the Capsid

Every time an HIV particle successfully infects a human cell, it follows a tightly choreographed sequence. The outer envelope fuses with the cell membrane. A protective protein shell called the capsid enters the cytoplasm, carrying the virus's genetic payload and the enzymes needed to replicate. Eventually, the capsid disassembles, the viral RNA is reverse-transcribed into DNA, and a protein called integrase splices that DNA into the host cell's chromosomes - cementing the infection permanently.

Integrase has been a drug target for decades. Three classes of FDA-approved antiretroviral medications interfere with its DNA-splicing activity. But a study published in Nature by researchers at the University of Delaware, the Francis Crick Institute, and six other institutions reveals that integrase does something else entirely during an earlier stage of the viral life cycle - something that, until now, nobody knew it did.

Filaments That Anchor the Genome

Using high-resolution cryo-electron microscopy (cryo-EM), the research team discovered that integrase proteins assemble into elongated filaments along the inner surface of the HIV capsid. These filaments do not float freely inside the shell. Each segment slots precisely into the hexagonal tiles that make up the capsid's geometry, and the filaments simultaneously grip the viral RNA genome, holding it in an organized configuration within the capsid interior.

The result is a zipper-like arrangement that packages the virus's genetic material and prepares the entire assembly to function as an infectious particle. Without the filaments, the genome is not properly anchored, and the virus loses the ability to infect cells.

"Integrase plays a structural role inside the HIV capsid - nobody expected that," said Professor Juan R. Perilla of the University of Delaware's Department of Chemistry and Biochemistry, who co-led the computational side of the work. "This protein forms filaments that anchor the RNA to the capsid. Without these filaments, the virus is non-infective."

How Cryo-EM Made It Visible

Seeing inside an HIV particle is a formidable technical challenge. The capsid measures roughly 120 nanometers across - about 1/800th the diameter of a human hair - and is densely packed, fragile, and structurally variable. Cryo-EM works by freezing samples in milliseconds to temperatures below minus 150 degrees Celsius, then imaging them with electron beams rather than light. The facility used for imaging at the Francis Crick Institute sits 20 meters below ground to isolate it from vibrations and magnetic fields that would degrade image quality at atomic resolution.

The process generates millions of two-dimensional images of frozen particles in random orientations. Sorting, averaging, and aligning those images into a coherent three-dimensional model requires high-performance computing on a scale that would have been impractical even a decade ago. Once the structural shapes were established, Perilla's team built atom-by-atom molecular models that fit the cryo-EM data, providing chemical detail about how the integrase filaments interlock with the capsid tiles and grip the RNA strand.

A New Vulnerability No Existing Drug Exploits

Earlier laboratory experiments using compounds called ALLINIs - allosteric integrase inhibitors - offered a clue that this structural role existed. ALLINIs disrupt the formation of large integrase assemblies during viral maturation, producing non-infectious particles. The new cryo-EM data explains why: by preventing integrase oligomerization, ALLINIs also break the interactions between integrase filaments and the capsid, leaving the RNA unanchored.

Critically, no currently approved antiretroviral therapy targets this structural role. Existing integrase inhibitors block the DNA-splicing step that occurs after the capsid has already disassembled. The filament-capsid-RNA interaction that the new study characterizes happens earlier, during viral maturation - the process by which newly assembled HIV particles become infectious. This earlier stage represents a distinct vulnerability in the HIV life cycle that therapeutic development has not yet addressed.

Perilla's group has spent more than a decade mapping HIV capsid structure and function with the explicit goal of finding new drug targets. University of Delaware PhD candidate Juan S. Rey contributed to the molecular modeling work. Co-authors Peter Cherepanov at the Francis Crick Institute and Alan Engelman at Harvard have collaborated with Perilla for nearly a decade; additional partners from the Dana-Farber Cancer Institute, University of Oxford, Birkbeck College, Harwell Science and Innovation Campus, and Imperial College London contributed to the cryo-EM and structural analysis.

Early-Stage Findings and the Path to Therapeutics

This is a structural biology discovery, not a clinical result. Demonstrating that a protein interaction exists and is essential for viral infectivity is a necessary step toward drug development - but a step that precedes by many years the testing of compounds that might interfere with that interaction in patients. The cryo-EM work establishes the target's existence and defines its structural geometry. Identifying molecules that can disrupt the filament-capsid-RNA interaction specifically, without off-target toxicity, is a separate and substantial undertaking.

Pre-clinical ALLINI compounds that disrupt integrase oligomerization already affect the interaction indirectly, which demonstrates that pharmacological disruption is at least conceptually feasible. Whether compounds can be developed that target this interaction more specifically and with a safety profile suitable for human use remains to be determined.

The research was funded by the National Science Foundation, the National Institutes of Health, and the U.S. Department of Energy.