Swine flu and environmental arsenic

Back in May there were some stories on the wires and flublogia regarding a new study about arsenic exposure and risk of flu. I didn't write about it at the time for purely arbitrary reasons (I was writing about other things), but I noticed it and in fact I know the senior author and his work fairly well. For reasons having nothing to do with flu I revisited the paper the other day, along with a bunch of others on arsenic toxicity from the same group up at Dartmouth (the senior author, Josh Hamilton, has now moved to the Marine Biological Laboratory at Woods Hole, Massachusetts, but I think most of the work was done while he and his students were still in New Hampshire). Before getting to the flu connection, let me put the series of papers, of which the May publication was a part, in context.

Arsenic is a metal and a notorious poison known since classical times. It was a favorite murder weapon in Renaissance Italy. It is only in the last couple of decades, however, that arsenic toxicology has been transformed by the techniques of modern molecular toxicology. What Hamilton's and other groups have found is that this ancient poison has widespread and fundamental biological effects on many different systems, often in very subtle ways we are only starting to unravel. Arsenic is not just a historical curiosity. Weathering and solution of natural mineral formations has put it in drinking water wells in many places and New Hampshire, where Dartmouth is located, is one of them. After a long tussle with a foot dragging Bush administration, public health advocates were finally able to get the standard for arsenic in community drinking water supplies reduced from 50 parts per billion (ppb) to 10 ppb, but private wells are not covered and many exceed it by wide margins. Almost half of New Hampshire residents get their water from private wells. Many other US states are also affected and arsenic contamination of water is a major problem in Bangladesh, Chile, Argentina and many other places. The levels in some places is high enough to cause frank poisoning, but in the US the levels are much lower. One significance of the work of Hamilton's lab and those of several other scientists is to show that arsenic can cause important biological effects in animal models at environmentally relevant levels, meaning, 10 ppb to 100 ppb in drinking water.

In a series of papers scientist have show arsenic to be an endocrine disrupting chemical (see our discussion here of endocrine disruption in connection with bisphenol A), interfering with steroid receptors, retinoic acid receptor and thyroid hormone receptor in rat cells at environmentally relevant levels (for an example and more references see here). Endocrine disruption is bad enough. But -- again at environmentally relevant doses -- arsenic also interferes with DNA repair. We're still trying to untangle the mechanisms, but it is clear that arsenic makes fixing some kinds of genetic mistakes more difficult, implicating this in arsenic's known carcinogenic effects (a paper on this subject here).

That's some of the context to the next couple of papers. In March Hamilton's group published a paper in Environmental Health Perspectives on arsenic exposure in drinking water and alteration of gene products associated with the immune response in mouse lung. Arsenic may be one of the few, if not the only, poison toxic to the lung by ingestion rather than inhalation. The cells in the lungs of mice who drank water with 10 or 100 ppb of arsenic for 5 weeks expressed messenger RNA and proteins associated with the innate immune response differently than mice not exposed. But what does this kind of molecular finding have to do with the ability of the mice to protect themselves against infection? It was with this background, not the swine flu outbreak, that the May paper on arsenic and influenza appeared.

Despite news reports that reported that experiments had shown that arsenic exposure increased the susceptibility to swine flu, it wasn't swine flu virus that was tested in the experiments. Close, perhaps, but not swine flu. The virus was a standard laboratory strain isolated in 1934 known as PR8. True, it is an H1N1 serotype descended from the same 1918 H1N1 as the recent swine flu (but via many twists, turns, divergences and stop overs in various species). Plausibly what we find out with PR8 has relevance to swine flu. But it wasn't an experiment directly testing an arsenic - swine flu connection. That's no surprise. These carefully done experiments must have been designed and conducted before the swine flu outbreak to have gone through review and publication by May 20 (just a month after the first swine flu H1N1 isolations).

So what did these experiments show (paper here)? Standard lab mice (C57BL/6J if you're interested) were inoculated intranasally with a sublethal dose of PR8. Some mice had partaken of 5 weeks of 100 ppb arsenic laced water before inoculation, others water without arsenic. Then the mice were studied for the way they responded to the virus. The arsenic treated mice were sicker (lost more weight) and had higher titers of influenza virus in their lungs as well fewer cells (dendritic cells) associated with preparing antigen for presentation to the immune system. The researchers interpreted this as showing that arsenic significantly compromised the lung's immune response to influenza infection.

That's basically the evidence at this point. The processes implicated are part of the early innate immune response and could determine whether viral replication is sufficient to exceed some critical threshold and lead to a catastrophic response. At this point, however, this is an example of an animal model that suggests how modulating host responses through the environment might affect outcome. Virulence is not just a property of a virus. It is something that depends on the virus, its interaction with the host and the environment both are part of.

There are lots of reasons not to be exposed to arsenic in your food and water. It's an endocrine disruptor, is implicated in cancer and may affect the immune system in subtle ways. And, yes, chronic exposure might make a difference in how you respond to a flu infection.

Lots of reasons.