Key influenza-severity risk factor found hiding in plain sight on our antibodies

Some new strains are rougher than others. The 1918 flu pandemic killed 50 million people and infected one-fifth of the world's population. Influenza pandemics also occurred in 1957, 1968 and 2009.

"Influenza remains an incredibly dangerous risk to global health," said Taia Wang, MD, PhD, associate professor of infectious diseases and of microbiology and immunology.

A team led by Wang has found that the relative abundance of a certain kind of sugar molecule on our antibodies -- the specialized proteins our immune systems cook up to block viruses and other microbes from entering our cells and replicating profusely once inside -- plays an outsized role in whether we become mildly ill rather than severely ill from a flu infection.

They've also figured out why. And they've demonstrated, in mice, how to head off severe flu symptoms regardless of which strain is on the march. That would come in handy in the event of the next large-scale flu outbreak -- and it may apply to other infectious diseases.

The findings, described in a study to be published online Nov. 13 in Immunity, may even help explain why older people are more susceptible to severe flu and many other diseases, infectious or otherwise.

Wang, a faculty member in Stanford Medicine's Institute for Immunity, Transplantation and Infection, is the study's senior author. Lead authors of the study are basic life research scientist Saborni Chakraborty, PhD; postdoctoral scholar Bowie Cheng, PhD; graduate student Desmond Edwards; and former graduate student Joseph Gonzalez, PhD.

Inflammation over replication

Sitting on the surface of some of our immune cells is a receptor, called CD209, that the study showed can dial down inflammation in response to a flu infection. Wang and her associates were able to kick that anti-inflammatory receptor into gear by fiddling with the composition of antibodies.

This didn't stop the virus from getting into lung cells and making copies of itself while inside the cells it invades. But it didn't have to.

Viruses furiously self-replicating inside our lung cells is not a good thing, obviously. But fatal cases of influenza infection are usually caused by an overwhelming inflammatory response to the infection, which exacerbates lung damage and prevents gas exchange, rather than by the virus alone, Wang said.

"We've discovered a new way to protect against severe influenza by shutting down this follow-on inflammation, regardless of ongoing viral replication," she said.

The experimental inflammation-lowering technique isn't limited to a single flu strain.

The antibody molecules circulating in our blood and known to immunologists as IgG (the acronym stands for "immunoglobin G") are roughly Y-shaped. The Y's horns are customized to cling to specific surface features of particular pathogens and, if the fit is snug enough and ties up the right part of the invading pathogen, prevent it from getting into cells.

The stalk of an antibody's Y-shaped structure is oblivious to whatever the horns are binding to. That stalk's job is to tell the rest of the immune system what to do. It can have differing affinities for various immune cells. And it can exert different effects on whichever immune cells it makes contact with, depending on the chemistry of a couple of long, bifurcated chains attached to its surface.

These chains' links are made of sugar -- granted, not the kind you'd find in a candy store. To carbohydrate scientists, the term "sugar" refers to nearly a dozen distinct but chemically similar substances our own bodies produce. Most of these sugars have names few of us have ever heard of. Sewed into or stapled onto larger molecules, they provide structural support, stability or signaling capability.

As many as four molecules of a particular sugar called sialic acid may get snapped into place as final links on an IgG molecule's branching sugar chains. How many actually do can make a big difference.

Enter the alveolar macrophage

Wang's study began by characterizing antibodies from people who did or did not become very sick after infection by H1N1, a common seasonal influenza subtype. The only significant difference the scientists observed between those who became mildly ill and those who were hospitalized was in the amount of sialic acid on infected individuals' antibodies. High levels correlated with mild symptoms; sicker patients' antibodies sported fewer sialic acid links.

Wang and her colleagues explored further, using bioengineered mice whose cells expressed human receptors for antibodies.

"We gave the mice human antibodies differing only in their sialic acid content," Wang said. Then the mice received what would ordinarily be a lethal dose of either of two very different seasonal influenza virus subtypes.

The sialic acid-rich antibodies, but not the others, protected the animals from both types of flu, evidently due to markedly reduced lung inflammation.

"The reduced inflammation led to better oxygen and carbon dioxide exchange," Wang said. "The lungs could keep doing their job."

The difference in sialic acid abundance had no effect on the virus's ability to replicate inside infected lung cells.

The scientists found that the high- and low-sialic acid antibodies were binding to entirely different receptors on the surfaces of immune cells called alveolar macrophages. These fierce sentinel cells, among other things, patrol the alveoli -- the tiny, delicate air sacs studding the surfaces of our lungs, through which oxygen from the air we breathe and carbon dioxide, a respiration byproduct, are exchanged.

When alveolar macrophages spy a pathogen, they gobble it up. They also signal the immune system to send in more troops. Usually that influx of additional immune cells is enough to quell the microbial invasion. But sometimes, ironically, the surplus of fired-up immune cells and the noxious substances they squirt out -- the essence of inflammation -- do more harm than good: They not only expunge virally infected cells but tear up healthy ones, too. That may cause even more inflammation.

Antibodies usually bind to pro-inflammatory receptors on alveolar macrophages, spurring downstream inflammatory activity. But higher levels of sialic acid on an antibody's stalk, the scientists proved, induces the antibody to bind to CD209 instead, shifting alveolar macrophages' mood to anti-inflammatory.

"CD209 has been shown to be anti-inflammatory in autoimmunity," Wang said. "But it's never before been implicated in calming our immune response to an infectious disease."

Analyses of alveolar macrophages' gene-activation levels showed that the same set of genes whose activity levels differed in flu-infected mice receiving high- versus low-sialic acid antibodies could be used to divide influenza patients into "mild" and "severe" categories.

Many of these genes are associated with mounting an inflammatory response. In particular, sialic acid-rich antibodies' binding to CD209 shut down activity of a famous inflammatory sparkplug called NF-kappa-B.

Who needs horns?

An antibody's horns can bind to only one or, at most, an extremely narrow range of pathogens. Yet, sialic acid-rich antibodies' beneficial dialing down of disease severity wasn't limited to a single flu strain. Nor did their benefits come from any clearing of the virus. It was purely the anti-inflammatory response that was reducing disease severity.

Wang said, "We wondered: Does this protection we see against highly different influenza-virus subtypes require the whole antibody? Are its horns even necessary? Or might the stalk alone be enough to protect against flu severity?

Fortunately, just such high-sialic acid antibody stalks are available, as they're under active clinical investigation for treating autoimmune disorders, which are also inflammatory in character. These stalks proved effective in preventing severe symptoms in flu-infected mice.

Wang's team is doing longitudinal studies in humans to see if sialic acid-enriched antibody stalks can predict risk of disease progression in influenza patients.

The findings' applications may extend beyond influenza, or lung infections in general, to numerous infectious diseases and even to a broad range of inflammatory conditions, she said.

"Age is the major factor differentiating people whose antibodies characteristically have high versus low sialic acid content," Wang said. The age-associated decline in sialic acid's abundance on people's antibodies may account in part for the observed high incidence of chronic low-level inflammation on older people, predisposing them to conditions ranging from heart trouble and strokes to Alzheimer's and Parkinson's disease to cancer and many other aging-associated diseases.

Wang is a consultant for Nuvig Therapeutics Inc., which is testing antibody stalks for treating autoimmunity and provided these reagents for use in this study.

Researchers from the University of California, San Diego; Nuvig Therapeutics Inc.; the University of Colorado; National Jewish Health; Washington University School of Medicine; the National Institutes of Health; and the Howard Hughes Medical Institute contributed to the work.

The study was funded by the National Institutes of Health (grants R01AI150214, R01AI173203, 75N93029C00051, 5U01AI144616, R01AI178298 and UL1TR001866) and the Howard Hughes Medical Institute.