Gene Profiles May Predict TB Prognosis

Information could be used to guide treatment.

+ More

By Tina Hesman Saey, Science News

Thanks to molecular profiling, scientists now have a better idea about how a mass killer selects its victims. And the new analysis suggests that the killer, TB, may use a different murder weapon than researchers previously believed.

Mycobacterium tuberculosis infects one-third of people worldwide. But only about 10 percent of people infected will actually get sick with a debilitating lung disease. Until now, scientists had no way to predict who would become ill.

Now, an international consortium of researchers has compiled profiles of genetic activity in the blood of people with dormant TB infections, people with active infections, and healthy people. Those profiles show how the immune system deals with tuberculosis and point to some surprising culprits responsible for awakening a slumbering infection. Such profiles may help predict who will succumb to TB, the researchers report in the Aug. 19 Nature.

“This is literally the way to tell who is going to get sick,” says Clifton Barry, chief of the tuberculosis research section at the U.S. National Institute of Allergy and Infectious Diseases in Bethesda, Md. This study could revolutionize TB diagnosis in the same way that breast cancer treatments were forever changed by the discovery that some tumor cells make molecules that respond to estrogen, and that those molecules can serve as targets for chemotherapy, Barry says.

In the new study, researchers drew blood from TB patients and from healthy people in London and analyzed gene activity in the blood cells. People who had active infections had 393 genes with activity different from that seen in healthy people. The team could classify people into groups—no infection, latent infection or active illness—just by looking at the gene activity profiles in their blood. The findings were replicated in a separate group of patients from Cape Town, South Africa. The TB signature disappeared as people were treated with antibiotics.

About 10 to 25 percent of people with latent infections had signatures similar to those of people with active infections, indicating that people with the active profile may go on to develop the disease even if their infection is currently dormant, says study coauthor Matthew Berry of the MRC National Institute for Medical Research in London. The researchers are planning to follow people with latent infections to see if those with the signature really are the same ones who develop active infections later. If the results hold up, the blood profiles could be the first means of predicting who is likely to get sick from TB.

That could spare people from developing a lung-damaging infection, but may also mean that people who aren’t likely to get sick won’t need to take anti-TB drugs that can damage the liver.

Blood cells called neutrophils also appeared in the new study to be important for spreading the disease. Previously, researchers didn’t think that short-lived neutrophils could play any role in such a long-term infection as tuberculosis. The dogma in the field was that the bacterium infected only immune cells called macrophages, says study leader Anne O’Garra, also of the MRC National Institute for Medical Research. The new study indicates that genes turned on by a protein known as type 1 interferon become active in the neutrophils of people with full-blown TB. Interferon helps to fight off viral infections but may actually make bacterial infections such as TB even worse, O’Garra says.

These findings fit well with recent data from mouse studies implicating both neutrophils and interferon in serious disease caused by tuberculosis, says Andrea Cooper, an infectious disease immunologist at the Trudeau Institute in Saranac Lake, N.Y.

“We’re at a watershed here in changing what we think the disease is about,” she says.

TB’s molecular signature was distinct from the profiles of blood taken from people with autoimmune diseases, such as lupus, and from those with other infectious diseases like Streptococcus or Staphylococcus infections, the researchers found. The discovery was unexpected, as most researchers thought that different types of bacteria might change the activity of specific genes at the site of the infection but that those differences would not show up in the blood, Cooper says. The variety of signatures indicates that the immune system has developed multiple ways of dealing with infectious organisms.