The Flu Virus Does Not Just Invade Your Cells. It Rearranges Their Inner World

A flu virus enters a human cell carrying only a small collection of proteins and genetic instructions. It cannot reproduce on its own, so it must turn the cell's machinery into a temporary virus factory. Scientists have long known this general strategy. What they have struggled to see is exactly which viral and human proteins touch one another inside an intact infected cell, and what those contacts do to the cell's internal organization.

A study published on 20 July 2026 in Nature Microbiology provides one of the most detailed maps yet. Researchers combined in-cell chemical cross-linking, mass spectrometry, structural modelling and functional experiments to capture protein contacts during influenza A infection. Their results show that the virus does not merely borrow isolated components. It reorganizes whole cellular compartments, including structures inside the nucleus that normally manage RNA and proteins.

Freezing millions of molecular handshakes

Traditional protein studies often remove molecules from cells before analysing them. That can destroy weak interactions or strip proteins from the membranes and liquid-like compartments that give those contacts meaning.

The team instead used structural host-virus interactome profiling, known as SHVIP. Newly made viral proteins were chemically labelled inside infected human cells. A cross-linking molecule then acted like molecular glue, joining proteins that were physically close. After the cells were broken open, mass spectrometry identified the linked protein fragments and revealed which amino-acid regions had met.

The researchers combined these measurements with AlphaFold-based structural models and targeted biological tests. This allowed them to move beyond a list of possible partners and place many contacts inside specific stages and locations of infection.

Following the virus through the cell's protein factory

One part of the map followed haemagglutinin, the spike-like protein that influenza uses to attach to and enter cells. Before new virus particles can leave an infected cell, haemagglutinin must be folded and chemically modified as it moves through the endoplasmic reticulum and Golgi apparatus.

The new data identified host proteins contacting different forms of haemagglutinin along this route. Some of those host factors had not previously been assigned a clear role in influenza infection. The work also identified an interaction between the viral M2 protein and LAT1, a human amino-acid transporter embedded in cell membranes.

These discoveries may help researchers distinguish human proteins that the virus truly depends on from those that merely appear nearby in simplified experiments.

The most surprising takeover happens in the nucleus

The study's most striking result concerns paraspeckles, small membraneless compartments inside the nucleus. They are assembled around a long RNA molecule called NEAT1_2 and hold collections of RNA-binding proteins. Think of them less as sealed rooms and more as temporary droplets that organize molecules through their chemical properties.

As influenza infection progressed, the researchers saw these compartments disassemble across several cell types. Viral proteins NP and NS1 interacted with paraspeckle components. The viral enzyme PA-X degraded the organizing RNA, while infection also inhibited the cellular machinery that transcribes it.

Once the compartments dissolved, RNA-binding proteins were released. Functional tests indicated that some of these liberated human proteins then supported the production of viral RNA. The virus had effectively dismantled part of the nucleus and reused its contents.

Why this could matter for future antivirals

Most influenza medicines target viral proteins. That is logical, but viruses mutate quickly and can develop resistance. Mapping critical interactions with human proteins offers another strategy. A drug might disrupt a specific contact or protect a cellular compartment without trying to kill the virus directly.

The detailed contact sites are particularly useful because blocking an entire human protein can be toxic, while interrupting only the surface used by the virus may be more selective. The map can now guide experiments that test which interactions are indispensable and which can be blocked safely.

What the study does not prove

This is not a new flu treatment, and it does not show that targeting paraspeckles will be safe or effective in people. Much of the work used a laboratory-adapted H1N1 strain and cultured human cells. The researchers validated key observations in primary lung epithelial cells, but they did not test a drug, an animal therapy or a patient intervention.

Influenza strains differ, and the exact network may change in seasonal viruses or strains with pandemic potential such as H5N1. Human proteins also perform essential jobs, so any attempt to interfere with them must avoid damaging healthy cells.

The achievement is therefore a map, not a cure. But maps often determine where the most productive search begins. By showing influenza's takeover inside the architecture of a living cell, this work gives antiviral researchers a more precise set of places to look.