Study: Inhibiting cytokine response might not reverse H5N1 infections

Jul 16, 2007 (CIDRAP News) – Scientists have hoped that disabling the body's destructive immune-system overreaction to the H5N1 avian influenza virus, known as "cytokine storm," could lead to new lifesaving treatments, but according to a new study, trials testing the strategy didn't protect mice infected with the disease.

The study, to be published ahead of print this week in the Proceedings of the National Academy of Sciences (PNAS), was conducted by Robert G. Webster, PhD, and colleagues at St. Jude Children's Research Hospital in Memphis. They report that groups of mice genetically programmed to lack one of three important inflammatory cytokines died after they were exposed to a Vietnamese H5N1 virus strain.

Also, they found that mortality rates were high in nonaltered mice that were given glucocorticoids to suppress cytokine production after exposure to the H5N1 virus.

Scientists have suggested that the cytokine storm contributed to the high death rate in the 1918 Spanish flu pandemic and is playing a similar role in human cases of H5N1 infection today. Autopsies of some H5N1 avian flu victims in the current outbreaks have revealed lungs were choked with debris from excessive inflammation triggered by the virus. Similar severe lung damage was often reported in victims of the 1918 pandemic, which disproportionately killed young, healthy adults—those with the strongest immune systems.

The authors used mice that were genetically altered to lack one of three cytokines: TNF-alpha, a proinflammatory molecule that appears to direct lymphocyte-mediated lung injury during influenza infection; IL-6, which is known to be elevated during H5N1 virus infection; and CC chemokine ligand 2 (CCL2), a chemocytokine that recruits leukocytes to infection sites.

They found that H5N1-infected mice deficient in TNF-alpha or its receptors had mortality rates and weight loss similar to infected wild-type mice. After H5N1 inoculation the IL-6–deficient mice lost nearly the same amount of weight as the wild-type mice, and all died. Similar mortality and weight loss results were seen in the mice with the CCL2 defect.

Infected mice that were treated with glucocorticoids didn't lose as much weight, but the researchers found that the mortality rates were similar to those in the other groups.

The researchers concluded that deficiency of any of the three key cytokines alone does not protect mammal hosts from H5N1 avian influenza death and that glucocorticoids do not reduce the lethality of the H5N1 infection.

Though the findings are negative, they help guide the search for therapies for H5N1 infections, the authors write.

"These results refute the popular paradigm that the cytokine storm is the cause of death during H5N1 infection," the authors state, adding that more research is needed to examine the contribution of each cytokine to factors other than weight loss and mortality.

A more promising therapeutic strategy may be early inhibition of H5N1 polymerase activity to decrease viral replication, they wrote.

However, an internationally known virologist who discussed the study's findings with CIDRAP News and asked not to be named disagreed with the authors' conclusions, but said their data were important.

Cytokine cascades are pleiotropic and multiply redundant, and once one starts, it is unlikely that blocking a single cytokine will have a major effect, the virologist said.

"Thus, these results do not disprove a role for cytokines in the pathogenesis of H5N1," said the expert, though the data may support the argument that therapy to block a single cytokine may not be clinically beneficial.

The virologist pointed out that mice with TNF receptor defects lost less weight than the wild-type mice (9.1% vs 17.1%), though the difference may not be statistically significant because there were only three mice in each study group. The scientist also said the glucocorticoid-treated mice lost less weight than the untreated group.

"If inflammatory responses are not contributing to pathology, how is glucocorticoid therapy having such a strong effect on weight loss?" the virologist asked.

Mortality patterns in the mice are not unexpected, the expert said, noting that in mice the H5N1 virus primarily kills by neuroinvasion rather than by respiratory pathology.

"In any event, these data cannot be ignored, nor can it be concluded that these results refute the role for cytokines in the pathogenesis of human H5N1 lung disease," the virologist commented. "The role of inflammatory responses in the pathogenesis of influenza is still far from understood."

Michael T. Osterholm, PhD, MPH, director of the University of Minnesota Center for infectious Disease Research and Policy (CIDRAP), publisher of CIDRAP News, also disagreed with the study findings and said the clinical picture in human and animal models suggests that age distribution is a critical finding that supports a role for the cytokine cascade in the pathology of the disease.

"Any lab finding that is going to be meaningful has to match up with the clinical disease severity and distribution by age," Osterholm said.

Anthony S. Fauci, MD, director of the National Institute of Allergy and Infectious Diseases (NIAID), told CIDRAP News that though the new study appears to create some doubt about the role of high cytokine levels in the pathogenesis of H5N1 infection, "It probably tells us a bit of truth about both hypotheses. When the researchers genetically manipulated the system, the mice still did badly," he said.

"You have to be careful when interpreting mouse studies, but this leaves an open question," Fauci commented.

Salomon R, Hoffmann E, Webster RG. Inhibition of the cytokine response does not protect against lethal H5N1 influenza infection. Proc Natl Acad Sci 2007 (published online Jul 17) [Abstract]

Newsletter Sign-up

Get CIDRAP news and other free newsletters.

Sign up now»


Unrestricted financial support provided by

Bentson Foundation 3MAccelerate DiagnosticsGilead 
Grant support for ASP provided by


  Become an underwriter»