What Is Influenza A RNA and How Does It Replicate?

Influenza A virus stores its genetic information not in DNA but in RNA, specifically eight separate strands of negative-sense, single-stranded RNA that together encode everything the virus needs to hijack a host cell and produce copies of itself. Because the RNA is “negative-sense,” it cannot be directly read by the cell’s protein-making machinery; the virus must carry its own enzyme, an RNA-dependent RNA polymerase, to first copy each strand into a usable form.¹ That segmented design and the error-prone polymerase at the heart of replication together explain much of what makes influenza so persistent and so unpredictable.

Eight Segments, Eight Jobs

Unlike many viruses that carry a single continuous stretch of genetic material, influenza A splits its genome across eight distinct RNA segments. Each segment is wrapped around multiple copies of a protein called nucleoprotein and capped at one end by a copy of the viral polymerase, forming a structure known as a viral ribonucleoprotein complex, or vRNP.² The vRNP is the basic working unit for both reading and copying the genome.³

Recent cryo-electron microscopy work has clarified the shape of these complexes in fine detail. The RNA wraps around a ring of nucleoprotein subunits, with the two ends of each RNA strand held together inside the polymerase. One nucleoprotein sits right at the fork where double-stranded and single-stranded regions of the RNA meet, anchoring the whole assembly.⁴ At full scale, the vRNP forms a right-handed, antiparallel double helix, with the viral RNA tucked into the minor groove and individual nucleoprotein subunits linked by flexible loops that slot into a conserved pocket on their neighbor.⁵ That flexibility matters: during copying, the polymerase needs to pull RNA through itself while the overall helical architecture stays intact, accomplished by a process researchers describe as strand sliding.

How the Virus Gets Its RNA into a Cell

Replication begins with the virus latching onto a host cell. The viral surface protein hemagglutinin (HA) binds to sugars on the cell surface, triggering the cell to swallow the virus through endocytosis. This can happen through clathrin-coated pits or through a less selective engulfing process called macropinocytosis.⁶ Either way, the virus ends up inside an endosome, a small acidic compartment within the cell.

The low pH inside the endosome sets off a cascade of changes. It opens the virus’s M2 ion channel, letting protons flood into the viral particle. That acid bath loosens the connections between the vRNPs and the matrix protein M1 that holds them in place. Simultaneously, HA undergoes a dramatic shape change that exposes a hidden “fusion peptide,” which punches into the endosomal membrane and merges the viral and cellular membranes together.⁶ The freed vRNPs spill out individually into the cell’s cytoplasm. Real-time imaging has shown that roughly 30% of virus particles complete this uncoating step by fusing with late endosomes near the nucleus, typically 30 to 90 minutes after infection. Blocking either the M2 channel or endosome acidification prevents uncoating entirely.⁷

Into the Nucleus

Most RNA viruses replicate in the cytoplasm, but influenza is an exception. Once the vRNPs are released, they hijack the cell’s normal nuclear import machinery to shuttle themselves through the nuclear pore and into the nucleus, where both transcription and replication take place.⁸ This nuclear dependency is unusual among RNA viruses and has practical consequences: it means the virus is deeply entangled with host-cell processes that happen inside the nucleus, from mRNA processing to protein transport.

Cap-Snatching and Messenger RNA Production

Once inside the nucleus, the viral polymerase has to produce messenger RNA (mRNA) that the cell’s ribosomes can translate into viral proteins. But it faces a problem: the cell’s translation machinery requires mRNAs to carry a chemical “cap” at their front end, and the viral polymerase cannot make one from scratch. Instead, the virus steals caps from the cell’s own freshly made mRNAs in a process called cap-snatching.

The PB2 subunit of the polymerase grabs the capped end of a host pre-mRNA. Then the PA subunit, which contains an endonuclease active site, slices the host RNA about 10 to 13 nucleotides downstream of the cap.⁹ That short capped fragment becomes the primer for viral mRNA synthesis. The polymerase then extends the primer using the negative-sense vRNA as a template, producing a positive-sense mRNA.

The polymerase also needs to add a poly(A) tail to the end of each mRNA, which is essential for stability and translation. It does this through a stuttering mechanism: near the end of each vRNA template, there is a short stretch of uridine residues. When the polymerase reaches this stretch, physical obstacles within the polymerase complex block the template from moving forward. The polymerase repeatedly copies the same uridines, generating a long run of adenines on the mRNA.¹⁰ Structural studies have pinpointed specific amino acid residues in the PB1 and PB2 subunits whose side chains physically prevent the template from advancing, forcing the stuttering that produces the poly(A) tail.¹¹

Switching from Transcription to Replication

Making mRNA is only half the polymerase’s job. It also has to copy the genome so that new virus particles have their own set of eight RNA segments. Genome replication requires a fundamentally different mode of operation. Instead of using a stolen cap as a primer, the polymerase initiates without any primer at all, starting from positions 1 and 2 at the very end of the vRNA template.¹² The product of this step is a full-length positive-sense copy called cRNA, which then serves as the template for producing new negative-sense vRNA.

The second step, making new vRNA from cRNA, uses a distinctive priming and realignment trick. The polymerase first positions itself internally on the cRNA template and synthesizes a short dinucleotide. It then realigns this dinucleotide to the true starting position of the template before continuing to copy the full length of the segment.¹³

How the polymerase decides whether to transcribe mRNA or replicate the genome has been a long-standing puzzle. Small viral RNAs derived from the ends of each segment appear to help tip the balance toward replication as infection progresses.¹⁴ Structural work has also identified an “intermediate state” of the polymerase, characterized by a blocked cap-binding domain and a contracted core, that appears to serve as a gateway between the two modes. The polymerase passes through this intermediate before committing to either transcription or replication.¹⁵ The viral protein NS2 may further nudge the switch by interacting with the polymerase on the vRNP and shifting how the promoter binds, favoring genome copying over mRNA production.¹⁶

Nuclear Export and Genome Packaging

Newly made vRNPs need to leave the nucleus and reach the cell surface, where new virus particles bud off. The virus accomplishes this by coating each vRNP with a layer of matrix protein M1, then using a viral adaptor protein called NEP (also known as NS2) that contains nuclear export signals recognized by the cellular export receptor CRM1. This CRM1-dependent pathway ferries the M1-vRNP complexes out through nuclear pores.¹⁷ NEP carries more than one export signal, and both contribute to efficient vRNP export.¹⁸

Once in the cytoplasm, the eight vRNPs must be gathered into a single virus particle. This is not left to chance. Fluorescence studies have confirmed that a high percentage of virus particles contain exactly one copy of each of the eight segments, meaning the packaging process is selective rather than random.¹⁹ The segments recognize each other through a network of RNA-RNA interactions involving both their terminal regions and internal sequences, though the full map of these interactions is still being worked out.²⁰

Why Influenza A Mutates So Quickly

The viral RNA polymerase has no proofreading ability. Every time it copies a genome, it makes mistakes. Measurements using fluctuation tests put the overall mutation rate at roughly 1.8 × 10⁻⁴ substitutions per nucleotide per genome replication for H1N1 and about 2.5 × 10⁻⁴ for H3N2, meaning each replicated genome picks up an average of two to three mutations.²¹ Those mutations are not evenly distributed across all possible types. Transitions, in which a purine swaps for the other purine or a pyrimidine for the other pyrimidine, are about three to four times more common than transversions. This constant stream of small genetic changes is what drives antigenic drift, the gradual evolution of surface proteins that forces annual updates to the flu vaccine.

Reassortment and the Segmented Advantage

The segmented genome creates a second, much faster route to genetic change. When two different influenza A viruses infect the same cell at the same time, their segments can mix and match during packaging, producing offspring with a patchwork of segments from both parents. This process, called reassortment, generates enormous diversity in a single step.²² While point mutations accumulate gradually, reassortment can swap out an entire surface-protein gene at once, creating a virus with a surface the human immune system has never encountered. This is the mechanism behind antigenic shift, which has been responsible for pandemic influenza strains throughout history.²³

Reassortment is not entirely unconstrained, however. Not every combination of segments produces a functional virus. Mismatched segments can reduce the fitness of reassortant viruses, meaning natural selection filters out many of the possible combinations before they spread.²⁴

The Species Barrier and Host Adaptation

Influenza A circulates naturally in wild birds, and the jump from birds to mammals is not straightforward. One of the most critical barriers involves a host protein called ANP32A. The viral polymerase needs ANP32A to help assemble the replication machinery, specifically to form a dimer of two polymerases that is required for genome copying.²⁵ Avian and mammalian versions of ANP32A differ enough that a bird-adapted polymerase works poorly with the human version of the protein.²⁶

The virus can overcome this restriction through adaptive mutations in its PB2 polymerase subunit. The most well-known of these is the E627K mutation, a single amino acid change that enables avian polymerases to efficiently replicate in the presence of human ANP32 proteins.²⁷ This mutation has appeared repeatedly across different phylogenetic branches of avian viruses that have crossed into humans, suggesting it is a common evolutionary route to mammalian adaptation.²⁸ An alternative adaptation at position 701 (D701N) can serve a similar function. Work with both H3N2 and H5N1 viruses has shown that either 627K or the combination of 627E with 701N can boost transmission in mammals, pointing to these polymerase mutations as a common gateway to pandemic potential.²⁹

How the Immune System Detects Viral RNA

The very features that define influenza RNA also make it visible to the immune system. Inside infected cells, a sensor called RIG-I recognizes the 5′-triphosphate group on replicating viral RNA, a chemical signature absent from normal cellular mRNA. In certain immune cells, a different sensor called TLR7 detects viral RNA carried by incoming virus particles, while TLR3 in airway cells and immune cells picks up RNA associated with infected cells.³⁰ These detection pathways trigger interferon production and launch the innate immune response that acts as the first line of defense before antibodies and T cells mobilize.

Antiviral Drugs That Target Replication

Understanding the replication machinery has led directly to antiviral drugs. The cap-snatching step is a particularly attractive target because it has no equivalent in normal human biology. Baloxavir marboxil, approved in many countries, works by blocking the PA endonuclease that cleaves host mRNAs during cap-snatching. Inhibiting that endonuclease activity prevents the virus from producing any functional mRNA.³¹ Newer compounds continue to refine this approach, with some experimental inhibitors reaching potency in the low-nanomolar range against the polymerase complex.³² The older class of drugs, the neuraminidase inhibitors like oseltamivir (Tamiflu), target a different step entirely: they block the release of new virus particles from the cell surface rather than interfering with RNA replication itself. The M2 ion channel blockers (amantadine and rimantadine) target the uncoating step, but widespread resistance has made them largely obsolete for current influenza A strains.

Chemical Marks on Viral RNA

Even viral RNA picks up chemical modifications inside the host cell. One of the most significant is m⁶A, a methyl group added to adenosine residues by the cell’s own enzymes. Influenza A RNA carries m⁶A marks on both its positive-sense mRNAs and its negative-sense genomic vRNAs, and these modifications are not just bystander events. When researchers engineered mutant viruses that lacked the m⁶A sites on the HA gene segment, expression of HA protein dropped by roughly two- to three-fold compared to wild-type virus, while other viral proteins were unaffected.³³ This means the virus has co-opted the cell’s own RNA-decorating machinery to boost production of specific proteins, adding yet another layer of complexity to how influenza RNA functions inside an infected cell.

Defective Interfering Particles

Not every round of replication produces a fully functional virus. The error-prone polymerase sometimes generates truncated versions of genome segments, producing what are known as defective interfering particles (DIPs). These particles carry incomplete RNA that cannot sustain infection on its own but can compete with full-length “standard” virus for the replication and packaging machinery, reducing the output of functional virus.³⁴ DIPs can also stimulate the innate immune response by triggering interferon production, which is partly why researchers have explored using engineered DIPs as a potential antiviral strategy. That said, the picture is not entirely straightforward: experiments combining two different defective RNA segments into a single particle did not enhance the interferon-stimulating effect beyond what a single defective segment achieved.³⁵ DIPs are a natural byproduct of the replication cycle’s inherent sloppiness, and they may serve as a built-in brake on runaway viral replication during infection.