Crave chicken soup when you have a cold? There may be a good reason for that. Research in mice has found that changing eating habits could be crucial for surviving the body’s own immune responses to different types of infection.
Ruslan Medzhitov at Yale University and his team have found that giving mice with flu glucose saved their lives, but it killed those that were infected with bacteria. Amazingly, this effect worked in the absence of the actual pathogens – glucose had the same effect on mice injected only with inflammation-triggering molecules either from bacteria or viruses.
Protecting the brain
Inflammation is a general activation of the immune system that occurs when an invader is detected. It causes most disease symptoms, and can damage and even kill the host it is trying to save. Researchers increasingly believe that surviving an infection is as much about tolerating your own immune response as it is about killing the invaders.
Mice seem to survive their own immune responses thanks to feeding strategies. Like sick humans, all the infected mice initially lost their appetites, but the mice with flu quickly resumed eating. This could be because bacteria and viruses trigger different inflammatory responses, and feeding is helpful for surviving the viral response, but harmful when fighting off bacteria.
To test this, the team administered or blocked the sugar glucose, or interfered with various metabolic processes within the mice, while giving them molecules that triggered either a viral-like or bacterial-like inflammation response. They found that, when responding to a virus, the mice needed glucose to protect their brain cells from being damaged by inflammation. Without glucose, one specific anti-viral response killed cells in their brains.
But when mice were in bacterial defence mode, they benefited from a lack of sugar. As many dieters know, not eating sugar pushes the body to metabolise fat instead, generating chemicals called ketones.
This “ketogenic” switch seems to benefit mice with bacterial inflammation. If these mice were given glucose, or their ketogenic metabolism was blocked in some other way, they died from epileptic-like seizures caused by neuron damage. Medzhitov believes this was because too many highly reactive free radicals were generated both by digestion of glucose and by inflammation due to bacteria, and that the radicals damaged the neurons. Inflammation due to viruses, however, does not produce radicals.
Intriguingly, evidence from brain disorders in people suggests that abandoning glucose also helps our neurons when they are stressed. A ketogenic diet seems to protect brain cells in those who have epilepsy, and some are trying it as a way to fight brain cancer.
The findings may help explain the ancient adage that it’s best to feed a cold, but starve a fever. Colds are usually caused by viruses, while fevers would traditionally have been more likely to be down to a bacterial infection. Most diets were historically heavy on carbohydrates, which release glucose in our bodies.
The discovery may also save lives. Sepsis – a severe systemic inflammation of the body that often occurs in response to an infection – kills around a third of those who develop it. Efforts to fight sepsis with fasting or feeding have yielded no clear results. That could be because so far, none of these studies distinguished or recorded whether patients had bacterial or viral sepsis.
New diagnosis methods might help. “We are planning to conduct our own clinical trial where we will separate patients based on causes of sepsis,” says Medzhitov. It may be that feeding those infected with bacteria or viruses differently boosts their survival rates.
As the northern hemisphere’s cold and flu season approaches, what does this mean? “We think during illness one changes food preferences to support the appropriate metabolic program,” says Medzhitov.
He believes chicken soup does help in some way, but says that, when he has flu, he prefers tea with honey more than anything else. His mice would agree.
Journal reference: Cell, DOI: 10.1016/j.cell.2016.07.026