Antiviral response promotes bacterial infection


by Melissa Lee Phillips, The Scientist, 10 Oct 2006

An innate immune response to viral infection can kill white blood cells needed to fight off bacteria, according to a study published online this week in PNAS. This effect could explain why bacterial "superinfections" can take hold in the body more easily when a viral pathogen is already present, the study authors say.

"Viral-bacterial synergism is something that is a significant clinical issue in both human and veterinary medicine and we don't have a detailed understanding of what's going on," said Charles Czuprynski of the University of Wisconsin-Madison School of Veterinary Medicine, who was not involved in the study.

Researchers led by Alexander A. Navarini and Mike Recher of the University Hospital Zurich in Switzerland examined superinfection in mice by first infecting them with an RNA virus called lymphocytic choriomeningitis virus (LCMV) and then with the bacterium Listeria monocytogenes. Three days later, the mice infected with both pathogens showed 1000 times higher bacterial concentration in the liver and spleen than did mice infected with bacteria only.

Since the mice showed susceptibility to bacterial superinfection in the first three days, the authors examined key members of the early innate immune response: white blood cells called granulocytes. In humans, granulocyte number is the most important predictor of susceptibility to bacterial infection, Navarini told The Scientist. He and his colleagues found that granulocytes began undergoing apoptosis about two days after viral infection and that granulocyte numbers in bone marrow of mice infected with LCMV were considerably below normal.

"There's probably some window of time here in which viral infection leads to increased susceptibility to bacterial infection, and they're providing a mechanistic explanation for why this might be occurring," Czuprynski said.

The researchers next looked for antiviral factors that might be responsible for granulocyte death. They found that levels of type I interferon -- a cytokine known to become up-regulated in response to most viruses, including LCMV -- inversely correlated with levels of granulocytes. To see if there was a causal connection, the researchers infected mice lacking type I interferon receptors with LCMV. These mice did not show granulocyte cell death or sensitivity to L. monocytogenes infection – demonstrating that type I interferon activity is required for virus-induced granulocyte death.

"This link to the interferon is really interesting," said Jonathan McCullers of St. Jude Children's Research Hospital in Memphis, who was not involved in the study. "That, to me, is the important part of the paper."

It's not yet clear if this mechanism will underlie other cases of bacterial superinfection after viral disease, Czuprynski told The Scientist, but, because type I interferon activation is a generalized body response to viral infection, "this might be something that happens with other viral agents as well," he said.

"All agents causing high interferon type I levels are in our opinion likely to cause such effects and facilitate superinfection," Navarini told The Scientist in an Email.

According to McCullers, "it's going to take a little work to generalize it to human viruses. Influenza, for example, causes an increase in human granulocyte number and so probably does not encourage bacterial infection through this mechanism, but "it might be a very reasonable model for HIV," McCullers told The Scientist. Researchers will "have to apply it to each virus in turn and see if it fits."