A killing hunger

Malnutrition may be to blame not only for the spread of infectious disease but for the rise of ever more dangerous strains. Laura Spinney reports on a controversial new theory

In 1989, Russia withdrew aid from Cuba. The Cuban economy went into crisis, and the health of the population quickly followed. With malnutrition rife, Cubans were buffeted by one epidemic of neurological disease after another.

In 1989, Russia withdrew aid from Cuba. The Cuban economy went into crisis, and the health of the population quickly followed. With malnutrition rife, Cubans were buffeted by one epidemic of neurological disease after another. First, in 1990 and 1991, came an outbreak of meningitis, followed two years later by a disease that affected the vision of 50,000 people, mostly young men, in some cases leaving them blind. Finally, in 2000, the exhausted Cubans took a second battering from meningitis.

All three epidemics were caused by viruses, and at first glance this story is easily explained in terms of conventional immunological theory: malnutrition weakened the hosts' immune defences, leaving them exposed and vulnerable to the invading pathogen. But one group of American researchers has studied the chronological sequence of events a little more closely, and come up with a controversial, alternative interpretation.

Melinda Beck of the Department of Pediatrics at the University of North Carolina at Chapel Hill believes malnutrition not only damages the host, it also drives genetic changes in the virus, causing it to become more virulent. If so, malnutrition could play a far greater part in the spread of infectious disease than anyone ever suspected, and the presence of significant malnutrition in a population could even predict the advance of new and more dangerous strains.

Beck thinks that 1990s' Cuba provides the perfect human laboratory in which to test her theory. She and her colleagues have studied blood samples taken from Cubans before, during and after the epidemics, analysed the viruses isolated from those samples and constructed a map or tree to show just how similar they are, both in terms of their genetic material and the amino acid building blocks from which they are made.

The strains responsible for the 2000 and 1993 epidemics, although years apart, turned out to be far more closely related than those responsible for the 1990 and 1991 outbreaks. Those associated with the 1993 epidemic of optic and peripheral neuropathy - the disease that impaired the young men's vision - appeared to have gone through a period of very rapid mutation in the two years since the first, related meningitis outbreak.

Next, the researchers married up their findings to historical events in Cuba. The two years that saw the most rapid viral change also saw the worst malnutrition and in 1994 the Cuban government took steps to tackle the problem by issuing vitamin supplements. Beck says this helped reverse the dietary deficiencies and slow the rate of viral change - perhaps explaining the small genomic difference between the strains that contributed to the 1993 and 2000 epidemics.

She also made an intriguing observation. Among the eight strains that contributed to the first meningitis epidemic, there was one anomaly: a strain that in terms of its amino acid composition lay halfway between its seven "siblings" and those that caused the epidemic two years later. "That indicates a transition in viral virulence that began with the onset of malnutrition," she says.

"It's very exciting and the implications are profound," says Kevin Fritsche of the University of Missouri in Columbia, who works on the link between nutrition and bacterial virulence. To date, he says, the human evidence is largely circumstantial. But it is tantalising, because Beck has already published a series of studies demonstrating the effect in mice. She also has some preliminary data on Keshan disease, which affects heart muscles in children, and is found in parts of China where diets are low in selenium.

She has shown that if mice deficient in selenium are exposed to mild strains of influenza A or members of the coxsackievirus family, they get much sicker than healthy mice exposed to the same strains. Moreover, the viruses replicate faster in the malnourished mice, and become more virulent, so that healthy mice later exposed to those viruses also show worse symptoms than when exposed to the initial, mild strains. In other words, that increased virulence is not reversed even when the malnutrition is corrected, although the rate of change may slow down.

Selenium is essential for the body to produce anti-oxidant enzymes for breaking down harmful free radicals - charged molecules produced by oxidative processes. "When you have an infection, the infection process itself creates some oxidative stress in the host," explains Fritsche. Macrophages and other immune system cells generate free radicals as weapons to fight the invading pathogens. The body's own anti-oxidant enzymes then mop these up once the job is done. But in selenium-deficient mice, these enzymes are depleted and the oxidative stress runs rampant.

Beck believes it is this stress that brings about changes in viral genes. Rather than allowing the virus to continue to outwit the host's immune system, the genetic changes may in fact be rendering it more contagious. She thinks something like this is also happening in Keshan disease: a nutritional deficiency and a virus interact to produce the symptoms of an infectious disease.

But natural selection may also play a part. Both coxsackieviruses and influenza A are RNA viruses - they lack an in-built mechanism for correcting errors that crop up through replication so those errors or mutations accumulate. In turn, any virus isolated from a patient is not one virus but a "cloud" of related mutants. This diversity renders the virus adaptable to changing environmental conditions, with selection promoting the fittest in any given context. If the host is malnourished, says Dieter Ebert of the University of Fribourg in Switzerland - a leading expert on the evolution of pathogen virulence - selection might favour the viral strains with the highest growth rates, which also turn out to be the most virulent.

Could Beck's theory explain why the 1918 pandemic of Spanish flu killed 40 million people in the wake of the First World War, with so many victims among the malnourished survivors? Or why HIV/AIDS has ravaged the African continent? Understandably, the experts balk at projecting such an apparently straightforward process onto the minefield of human epidemiology.

In the case of HIV, Ebert says there is very little consensus on the origins of differences in the virus's virulence, and one must bear in mind the changes in behaviour that go with malnutrition. "Very often, malnutrition is a consequence of war," he says. "Hygienic conditions go down, hospital care is bad, people are travelling. Each of those factors could play a role." It is also essential to treat each disease separately, he says. HIV is, after all, transmitted predominantly through sexual contact. There may be a link between malnutrition and sexual behaviour in relation to its spread, but if so, nobody has explored it.

Nevertheless, John Oxford, a medical microbiologist at Queen Mary's School of Medicine and Dentistry in London and an expert on the 1918 flu pandemic, says it's interesting to speculate. "The state of malnutrition varied quite a lot in 1918 according to the different countries," he says. "But where there was huge malnutrition, that is in peripheral groups, they suffered very badly indeed - in Aboriginal groups such as the Inuit, for example."

Even if the association between malnutrition and infectious disease does turn out to be impenetrably complex, Ebert says Beck's work could be valuable from a public health perspective. "If you can say that malnutrition is an indicator that diseases will spread, then this is an extremely important message," he says.