Cytokine storm in avian flu

New evidence of cytokine storm in avian flu cases
Jun 14, 2006 (CIDRAP News) – Scientists in Hong Kong have reported new experimental evidence that avian influenza infections in human cells are more likely to trigger a destructive immune-system overreaction, or "cytokine storm," than human flu viruses are.

Writing in the July 1 Journal of Infectious Diseases, the researchers report that two avian flu viruses, a 1997 strain of H5N1 and a 1997 H9N2 strain, caused immune system cells in lab cultures to produce much greater levels of certain chemokines (a class of cytokine, or messenger protein) than such cells did when infected with an ordinary human flu virus.

"In general, the chemokines and chemokine-receptor responses of MDMs [monocyte-derived macrophages, a type of immune cell] to avian influenza viruses were much stronger than those to human virus, which may account for the high pathogenicity of avian viruses," the report states.

In addition, the H5N1 strain caused immune cells from adults to produce higher levels of certain cytokines than similar cells from newborn babies did. The authors say that finding may help explain why Hong Kong's human H5N1 outbreak in 1997 killed 5 of 9 infected adults (older than 12) but only 1 of 9 infected children. That sharp difference in adult and child mortality rates has not been seen in the current wave of H5N1 cases dating to late 2003.

Scientists have suggested that the cytokine storm played a role in 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 H5N1 avian flu victims in Vietnam and elsewhere have revealed lungs choked with debris from excessive inflammation triggered by the virus. Similar severe lung damage was frequently reported in victims of the 1918 pandemic, which disproportionately killed people with the strongest immune systems—young, healthy adults.

The new study was conducted by a University of Hong Kong team that has previously reported experimental evidence of a cytokine storm in H5N1 avian flu. The team includes J. S. Malik Peiris and, as first author, Jianfang Zhou.

In view of the severe lung damage caused by H5N1 in humans, the team decided to measure the expression of four chemokines and two chemokine receptors induced in MDM cells by avian and human flu viruses. They also sought to compare the chemokine production induced by these viruses in adult-derived MDMs and in neonatal MDMs derived from umbilical cord blood.

Three viruses were used: a strain from the 1997 human outbreak in Hong Kong, a 1997 strain of H9N2 from quail, and a 1998 strain of H1N1 human flu from Hong Kong. The H9N2 virus is a precursor of the H5N1 virus, with which it shares six internal proteins, the report says. Because of the safety risks involved in working with H5N1 viruses, the team first did the experiment with the H9N2 virus and then repeated it with the H5N1 virus in a biosafety level 3 facility.

The investigators found that all three viruses replicated at similar rates in both adult and neonatal MDMs, as indicated by similar numbers of viral matrix gene copies in the cells. That suggested that differences in chemokine production are not due to greater growth of the avian viruses.

The MDMs generally showed much greater chemokine responses to the avian flu viruses than to the human flu virus, and the differences were often greater for the adult MDMs than the neonatal MDMs, the report says.

For example, the adult MDM responses to the H5N1 strain were roughly 20-fold greater than their responses to the H1N1 virus. For one particular chemokine, called CCL3, the increase for adult MDMs was about 25-fold, but for neonatal MDMs, it was significantly lower—less than 10-fold, as shown on a graph in the report. For the other three measured chemokines, the responses of adult MDMs to H5N1 also exceeded those of the neonatal MDMs, but the differences were smaller.

Also, compared with the H1N1 virus, the H5N1 virus caused adult MDMs to express 6- to7-fold greater levels of the two chemokine receptors (CCR1 and CCR5). But the H5N1 strain induced no significant increase in expression of chemokine receptors by the neonatal MDMs.

"We have demonstrated that human MDMs have differential responses to human influenza virus H1N1/98 and avian viruses H9N2/G1 and H5N1/97, in spite of their similar infectivity and viral replication," the authors write. "Moreover, stronger chemokine and chemokine-receptor responses to avian influenza viruses were detected in adult MDMs than in neonatal MDMs."

They add that the higher CCL3 response to H5N1 by adult MDMs, as compared with neonatal MDMs, may be "one of the important factors" in the higher adult mortality rate in Hong Kong's 1997 outbreak. They note that higher levels of CCL3, along with several other chemokines, have been found in plasma from people who died of H5N1 disease than in people who survived it.

Overall, the authors conclude, "These data suggest that host factors may influence the disease process or outcome."

The latest findings parallel evidence that Peiris and colleagues reported last November concerning the cytokine storm hypothesis. In that study, lung cells growing in a lab culture reacted much more intensely to two strains of H5N1 virus than to an ordinary human flu virus (see link below).

Zhou J, Law HKW, Cheung CY, et al. Differential expression of chemokines and their receptors in adult and neonatal macrophages infected with human or avian influenza viruses. J Infect Dis 2006 Jul 1;194:61-70 [Abstract]

See also:

Nov 16, 2005, CIDRAP News story "Lab study supports idea of 'cytokine storm' in H5N1 flu"