Skip Navigation
Leading research to understand, treat, and prevent infectious, immunologic, and allergic diseases
Skip Website Tools
​​​ ​​​​​​​​​
Skip Content Marketing
  • Share this:
  • submit to facebook
  • Tweet it
  • submit to reddit
  • submit to StumbleUpon
  • submit to Google +

Mapping Flu’s Trek Through Our Cells

With just 11 proteins of its own, an influenza virus particle needs a lot of help from you in its quest to make more of itself. The cells in your respiratory tract are unwilling hosts, forced to aid the virus at each stage of its progress: getting in; making copies of itself (replication); and getting all the new viruses out of the infected cell. Cells fight this hostile take-over and the flu virus, in turn, has ways to avoid the defensive maneuvers. Now, new details of the jousting match between virus and cell are revealed in research by three groups of NIAID-supported scientists.

How the Host Helps

graphical representation of flu virus protiens

A team led by Sumit Chanda, Ph.D., of Burnham Institute for Medical Research, and Megan Shaw, Ph.D., of Mount Sinai School of Medicine, used a technique called genome-wide RNA interference (RNAi) to detect the human genes involved in the flu infection process. RNAi allows scientists to see which of the many thousands of human proteins the flu virus uses at various points in the infection cycle.

The investigators screened more than 19,000 human genes and turned up nearly 300 that the flu virus couldn’t do without. They used a modified influenza virus that could not complete its full infection cycle, so their RNAi screen revealed the parts of the host cell’s machinery that get tapped in early stages of the cycle, such as when a virus attaches to an uninfected host cell.

In theory, drugs that could temporarily dampen the activity of the identified factors would also put the brakes on any flu virus trying to take advantage of them. Current flu drugs are aimed directly at the influenza virus. But the flu virus mutates readily and these frequent changes allow it to gain resistance to antiviral drugs. However, if a drug were to be targeted to factor in the human host instead of being aimed directly at the virus, the pathogen’s ability to escape through mutation would be thwarted.

Many of the host factors revealed in the screen had not previously been known to be exploited by the influenza virus in its replication. Of note, the scientists found that they could greatly inhibit the growth of influenza virus by blocking certain of the newly identified host factors with small molecules. This work was done in test tube experiments, not in animals or people, but the investigators now have fresh avenues to explore in developing host-directed antiviral drugs.

“Armed with this large panel of required host factors, our next challenge is to characterize their involvement in the flu virus life cycle in more depth,” says Dr. Shaw. “Those factors that possess enzyme activities are prime candidates to use as targets for new antiviral drugs for flu, so this is an area we will be exploring further.”

The paper was published in the journal Nature in February 2010.

Found: Flu-Fighting Proteins

diagram of IFITM proteins blocking influenza infection

In another NIAID-supported genetic screening effort, scientists discovered that a family of human proteins—called interferon-inducible transmembrane (IFITM) proteins—plays a dramatic role in protecting our cells from attack by flu virus. IFITMs were discovered 25 years ago, but until now no one knew that they form part of an inherent anti-viral defense system. The team was led by Abraham L. Brass, M.D., Ph.D, of the Ragon Institute of Massachusetts General Hospital, Massachusetts Institute of Technology (MIT), and Harvard University, and Stephen J. Elledge, Ph.D., of Brigham and Woman’s Hospital and Harvard Medical School (HMS) in collaboration with Michael Farzan, Ph.D., of the New England Primate Research Center, HMS.

In a study published in Cell in 2009, the researchers used a whole-genome RNAi screen to find more than 130 host proteins needed by flu, and also found that lowering the levels of the IFITM proteins enhanced viral infections. They then selectively raised or lowered IFITM protein levels in the presence of influenza and other viruses. When the proteins weren’t present, viruses could easily enter the cell and multiply rapidly. Conversely, “when we boosted protein levels, virus replication went down strikingly,” says Dr. Brass.

The scientists focused on one member of the protein family, IFITM3, which not only hindered flu virus entry or replication, but also protected cells from other viruses, including those that cause West Nile fever and dengue fever. As their name indicates, the interferon-inducible proteins react to interferon, a potent immune system protein. Interferon is produced in response to viral or other infections and orchestrates a wide array of additional immune responses. Interferon does help you fight off infection, but it also makes you feel achy. According to Dr. Elledge, if levels of IFITM3 could be artificially boosted in the absence of interferon, it might be possible to fend off certain viral illnesses without suffering interferon’s unpleasant effects.

One of the biggest surprises, says Dr. Brass, is that IFITM3 is just as effective against flu virus strains that originally circulated in the 1930s as it is against viruses that emerged in the past few years. Why hasn’t the virus evolved any defenses against the human protein? “That’s one question we will explore next,” says Dr. Brass.

Although many details of IFITM virus-blocking activity are still unknown, the proteins do appear to act very early in the infection cycle and may be a potential new way to stop viral invasion, notes Dr. Brass. “We may have found a key gatekeeper,” he says. In fact, variations in the amount of IFITM production between individuals may help explain why some people are more resistant to influenza illness than others, Dr. Brass adds. If the IFITMs are frontline defenders, as these findings suggest, individuals who naturally produce more of these proteins may be able to avoid full-blown illness because flu viruses are unable to gain a foothold.

In 2012, Dr. Brass collaborated with Paul Kellam, Ph.D., and Aaron Everitt, B.Sc., from the Wellcome Trust Sanger Institute, together with Peter Openshaw M.D., Imperial College London, and J. Kenneth Baillie M.D., University of Edinburgh, to further define the role of IFITM3 in modulating flu severity. In one set of experiments, the team took mice engineered to lack the Ifitm3 gene and exposed them to a strain of influenza virus that typically does not cause serious disease. The ordinarily mild virus produced severe disease symptoms in mice without the Ifitm3 gene demonstrating that this single protective protein is needed to stop the virus.

The team also sequenced the Ifitm3 genes of 53 people who had been hospitalized following infection with either seasonal or pandemic 2009 H1N1 influenza. Almost 6 percent of the hospitalized patients had a variant form of the ifitm3 gene that may produce a shorter than normal protein. In contrast to the high percentage of hospitalized people with the variant, only 0.3 percent of all Northern Europeans are believed to have this variant ifitm3 gene. Further experiments, conducted with cells grown in the lab, showed that this shorter IFITM3 protein did not perform as well as normal IFITM3 in preventing influenza virus from infecting cells. Collectively, these data point to IFITM3 as being a key first-line defender against severe influenza infection in both mice and humans.

The research was published in Nature.

A Global View of Flu

photo of Aviv Regev
Aviv Regev, Ph.D., of the Broad Institue of MIT and Harvard.
Credit: Broad Institute

A third team of scientists recently completed the most comprehensive map to date of the many thousands of interactions between flu virus and host cell. The group was led by NIAID grantee Nir Hacohen, Ph.D., of Massachusetts General Hospital and Harvard Medical School and computational biologist Aviv Regev, Ph.D., of the Broad Institute of MIT and Harvard.

Because flu virus has so few of its own proteins, each one must be multifunctional, like a Swiss army knife, says Dr. Regev. One layer of the new map highlights the physical interactions between these multitasking flu proteins and the human proteins they contact. Amazingly, the flu’s proteins connect, directly or indirectly, with more than 1,750 human proteins. There are twice as many flu-human protein interactions in the course of the infection cycle as there are human-human protein contacts in a typical cellular process.

Once invaded, the cell reacts. The researchers sketched another map layer to identify the regulatory pathways triggered by entry of a flu virus into a lung cell. They used siRNA to switch off, one by one, the host genes identified in the first screen as playing some role in the influenza infective process. By measuring how seriously the virus was hurt by the loss of access to each human gene, the researchers gradually developed a “responsiveness” map of the most important pathways in the flu-host interplay.


photo of Nir Hacohen
Nir Hacohen, Ph.D., of Massachusetts General Hospital and Harvard Medical School
Credit: Massachusetts General Hospital
“Our screens turned up some expected and some unexpected findings,” says Dr. Hacohen. “Some flu proteins were previously known to interact with human proteins, but this new map showed that flu proteins we didn’t expect would interact with human proteins actually did.”


Finally, the investigators integrated the map of physical interactions and the one of regulatory pathways and revealed a picture with some surprising features. For example, the combined map showed that flu virus taps into a set of human proteins involved in embryonic development. The map allowed researchers to see a connection that was otherwise unpredicted.

“We also discovered that certain flu proteins appear to ‘moonlight.’ Their main job is to assist in replication but, as the functional map revealed, they also have a second job of manipulating signaling pathways in a way that makes the cell more hospitable to the virus.”

The scientists stress that the new map is a rough first draft. There is much more exploration ahead, says Dr. Hacohen. With this guide in hand, he says, investigators will have more confidence in filling in the missing pieces.

The research was published in the journal Cell in December 2009.


R Konig et al. Human host factors required for influenza virus replication. Nature DOI: 10.1038/nature08699 (Feb. 11, 2010)

AL Brass et al. The IFITM proteins mediate cellular resistance to influenza A H1N1 virus, West Nile virus, and dengue virus. Cell 139: 1243-54. (Dec. 24, 2009)

AR Everitt et al. IFITM3 restricts the morbidity and mortality associated with influenza. Nature DOI: 10.1038/nature10921 (2012)

SD Shapira et al. A physical and regulatory map of host-influenza interactions reveals pathways in H1N1 infection. Cell 139:1255-67. (Dec. 24, 2009)

back to top​​

Last Updated April 03, 2012

Last Reviewed June 15, 2010