By taking an in-depth look at the transcription factor network of Mycobacterium tuberculosis, researchers have gained new insights into how the bacterium adapts to environmental changes, including conditions of low oxygen.

For now, the team has looked at about a quarter of the more than 180 known transcription factors in the M. tuberculosis genome – 50 in all to date, though co-corresponding author Kyle Minch told BioWorld Today that "we are on our way to doing all of them."

But even that number has allowed them to gain new insights into how the bacterium adapts to drops in the level of oxygen – an ability that enables it to hide out from the human immune system.

The ability to deal with low levels of oxygen is the key to M. tuberculosis' success, which is staggering – one-third of the global population is infected with the bacterium. Those infections can remain latent for long periods of time because the host response to M. tuberculosis consists partly of walling off the bacterium, which leaves it at a stalemate with the host – the bacterium doesn't damage the lungs, but the host also can't get rid of the bacterium.

The work, which was published in the July 3, 2013, advance online issue of Nature, combined mapping of transcription factor binding with profiling of messenger RNA, proteins, lipids and metabolites to both look at changes in gene expression that resulted from changes in oxygen level, and to build a model that could predict how the bacterium would respond to environmental changes.

One specific finding that might yield new approaches to fighting tuberculosis is that low oxygen conditions appear to have a strong effect on lipid metabolism.

More generally, the work has shown just how intricate gene expression control for M. tuberculosis is.

"I went into expecting that we would find lots of connections," co-author David Sherman told BioWorld Today. But even with those expectations, "there are many more than I expected." Sherman is at the Seattle Biomedical Research Institute, and the work was performed in collaboration with a number of academic institutions.

Sherman said that from a scientific perspective, "this paper is the first time I've been convinced that we are able to predict, on a molecular level, the behavior of a cell."

"Not that we're exactly there yet," he added. "But when we finish [this work], we'll be able to predict what happens when the oxygen drops . . . or when the immune system suddenly realizes [tuberculosis] is there and starts producing lots of interferon-gamma."

Sherman hopes that such an ability to predict cell behavior can be used to find vulnerabilities, and from there, drug targets.

Finding such targets is "not the only bottleneck by any means," Sherman said. "But it's certainly a big one."

The approach used by Sherman and his colleagues is the logical extension of the genome sequencing studies of the last two decades.

As those studies became possible through advances in sequencing technology, Sherman said, there was "a widespread assumption that we would do that work – and then everything would become clear."

After the work was actually done, however, the main thing that became clear was that with a sequenced genome, "you get the raw pieces of some machine, but not how they are wired together. . . . What we are doing now is working on the wiring.

"If we could determine where every one of these bound," he added, "then in theory, we would have the wiring diagram."