Unlocking TB latency in spores and nitric oxide

Why does it take so long to cure TB?

(Editor's note: This is the second of a two-part series in which TB Monitor asks noted researchers to talk about their work in the field of TB latency.)

When William Bishai, PhD, considers latency, he thinks mostly of spores. Bishai, assistant professor at the Johns Hopkins School of Public Health in Baltimore, doesn't think latent TB bugs actually form spores, at least not in the technical sense, but he does believe something like that is going on.

That's why Bishai spends his days studying streptomyces, a group of organisms closely related to the TB mycobacterium yet far more researcher-friendly.

"We hypothesized that by looking at well-studied bugs, like streptomyces, we might find the genes they use to adapt to a hostile environment," he says. "That might lead us to the pathway TB uses in peoples' bodies during the process of latency."

So far, Bishai's idea seems to be correct. Working with clues from the gene he found in streptomyces that triggers the spore-forming process, he's discovered a counterpart in TB. "Now we're asking when TB turns this gene on, and what would happen if you deleted the gene," he says. "Though I don't think deep down that TB really forms a spore, I think this gene does help it respond to stress."

Maybe it does so by enabling the bug to alter the composition of its cell wall - which would account for the difficulty in staining TB cells in people latently infected with TB. More important, if the bug has somehow altered its cell wall - maybe by ceasing to produce mycolic acid - then a chemotherapeutic agent such as isoniazid, which targets mycolic acid, won't be as effective.

Once Bishai understands the gene well enough to mimic the on-off switch, exciting implications will result. "If someone had rip-roaring active TB, we could give a drug that mimicked the signal to become latent, and the bugs would stop growing," he says. By reversing the process, maybe the bugs in patients with latent infection could be switched on enough to make them susceptible to antibiotics.

That's approaching latency from the perspective of the bug. The host's point of view is what interests John Chan, MD, associate professor of medicine in the department of medicine, microbiology and immunology at Montefiore Medical Center in the Bronx, NY.

In mouse studies, Chan has established that a temperamental molecule called nitric oxide acts as a powerful weapon against the bacteria. T cells, alerted to the presence of an invader, crank out gamma interferon, which in turn teams with certain other cytochines to produce Nitric oxide.

Nitric oxide, the enforcer

Experimenting with various ways to disrupt nitric oxide production gives equal results, Chan has found. So far, the nitric oxide molecule seems to be the sole contender for antimycobacterial duty in the host. "Though it doesn't take a rocket scientist to see the body probably has other ways to kill the organism," he adds.

Now he's studying how nitric oxide does its dirty work. Clearly, it's highly reactive; remember the flap over the nitrites used to cure sausage and other meats but which caused illnesses? That was basically the same situation, Chan says. "Nitric oxide's a double-edged sword. It can kill TB, but too much is harmful to host cells."

As Bishai studies mycobacteria, and Chan explores the workings of cellular immunity, another part of the latency puzzle occupies John McKinney, PhD, a post-doctoral fellow who works with William Jacobs, PhD, associate investigator at the Howard Hughes Institute of the Albert Einstein College of Medicine in the Bronx.

McKinney's starting point is the Cornell model - the mouse model that most researchers use as a starting point for ways to explore latency in living systems. The model works like this: Mice infected with TB are given anti-TB drugs until detectable mycobacteria have vanished; once the mice receive a jolt of immunosuppresants, their so-called "latent" TB infections bounce back.

What strikes McKinney about the Cornell model isn't about latency, but rather a different question: How does TB survive the onslaught of drugs in the first place? Put more simply, why is it so hard to cure TB?

So what is it with TB, anyway?

"Most bacterial infections can be cured in a couple of weeks," McKinney says. "It's not like that with TB." People with latent infection undergo treatment for six months; to McKinney, it's worth noting that a full year of preventive treatment produces even better results.

It can take up to two years to rid the body of an active disease. And the surviving bugs aren't mutants, McKinney says. They haven't developed a new genetic trait; in a sense, they're just great at dodging bullets.

McKinney and Jacobs have dubbed that quality "persistence." It isn't exactly the same as latency, but the two may be linked.

Perhaps, he reasoned, the key to understanding the puzzle lies somewhere after the point where the body subdues the initial infection, and bug levels drop off and flatten out. "We thought, aha! What if there was a link between persistence and the host response?" Maybe by putting the bugs into a stationary, nondividing phase, the host's immune system was creating a persisting state.

Using a transgenic mouse missing the gamma interferon component of its immune system, McKinney tested his hypothesis. Absent the immune response, he hypothesized, persistence might vanish, and chemotherapeutic agents would swiftly eradicate all traces of infection.

Results were disappointing. The transgenic mice responded no faster to chemotherapy than regular mice. But with their infections treated to the so-called "vanishing" point, an interesting effect occurred: Mice lacking gamma interferon couldn't keep the remaining population of bugs at bay without continuous doses of chemotherapy. Instead, they swiftly relapsed.

That didn't surprise Chan, who had found the same thing happened when the nitric oxide synthase gene was knocked out. The molecule and the cytochine exist at two different points along the same pathway, it turns out.