A new technique may lead to faster, more precise ways to detect M. tuberculosis (scanning electron micrograph shown), the bacterium responsible for tuberculosis. Credit: NIAID CC BY 2.0

A molecule made in the lab may change the way tuberculosis is diagnosed in the field.

Chemically tweaking a sugar molecule known as trehalose lets it slip inside the bacterium that causes tuberculosis (TB) and glow. The method offers a quick, simple way to detect the pernicious bug, and may help counter TB, a deadly lung infection that’s particularly common in developing countries.

Howard Hughes Medical Institute (HHMI) Investigator Carolyn Bertozzi and colleagues report the work February 28, 2018, in the journal Science Translational Medicine.

Despite its devastating toll on health, the bacterium behind TB, Mycobacterium tuberculosis, can be hard to spot. Current tests rely on chemical stains that have been around for decades and can be finicky. Estimates put the sensitivity of these stains anywhere from 32 percent to 94 percent.

Better detection methods are sorely needed to combat TB, which killed more than 1.7 million people worldwide in 2016, says Bertozzi, of Stanford University. “If you can’t even get an accurate diagnosis, how do you treat people?”

The lab-made molecule DMN-Tre lets researchers detect M. tuberculosisbacteria (green) in patients’ sputum samples (eight shown). Credit: M. Kamariza et al./Science Translational Medicine 2018

As a chemist, Bertozzi, along with her colleagues, studies the molecules that make up bacterial cell walls. Early discoveries by her lab revealed that some bacteria use sugar molecules called trehalose as building blocks. Bertozzi found the cell walls of M. tuberculosisparticularly compelling. “There’s some really interesting biology there.” She began working with a team of scientists who held personal stakes in tuberculosis research. After a chance encounter at a meeting at HHMI’s Janelia Research Campus, Bertozzi decided to collaborate with study coauthor Professor Bavesh Kana of the University of Witwatersrand in Johannesburg, South Africa.

The researchers realized that trehalose molecules – those cell wall building blocks – might offer a way to flag living M. tuberculosiscells. But first, the team needed to find a chemical beacon that would make the flag visible. One chemical, called DMN, seemed to fit the bill. DMN can glow under certain wavelengths of light­ – but only when it is out of water. Because the M. tuberculosis cell wall contains a membrane that’s a “thick layer of grease,” Bertozzi says, it’s the perfect place for DMN to light up.

That insight – that DMN was “off” until a cell tucks it into its membrane – was key, Bertozzi says. “It’s such a simple thing, but simple things like that make all the difference between something that can be deployed or not.”

After linking trehalose to DMN in the lab, the researchers tested their hybrid molecule, called DMN-Tre, on an M. tuberculosis relative. Just as they had hoped, the bacteria grabbed the molecule and, within minutes, incorporated it into their cell membranes, where it began to glow under a fluorescent microscope.

A new labeling molecule called DMN-Tre detects live M. tuberculosis cells that haven’t been treated with drugs (green, top left), but not those that have been hit with a drug cocktail (bottom left). An existing stain called Auramine marks both untreated (red, top right) and treated cells (red, bottom right). Scale bar represents 5 micrometers. Credit: M. Kamariza et al./Science Translational Medicine 2018

In tests on sputum samples from 16 people with TB, DMN-Tre picked up M. tuberculosiscells in all of the samples. The new technique performed similarly to the standard – but more complex and time-consuming – labeling method based on the Auramine O stain, a dye that sticks to acids in bacterial cell walls.

Other tests showed that DMN-Tre is selective to Actinobacteria, the bacterial phylum that includes M. tuberculosis. Human cells and other types of bacteria, both of which are plentiful in sputum samples, don’t incorporate the molecule, the researchers found.

Unlike existing TB detection methods, DMN-Tre can also distinguish cells that are metabolically active from those that are not. Because the molecule relies on bacteria to actively incorporate it into the membrane, only healthy cells are labeled, whereas cells that are compromised by drug treatment do not label as well. That property may allow clinicians to monitor how well treatments are working in people, and perhaps even test whether certain mixtures of drugs would work against specific strains of M. tuberculosis.

More work remains before the molecule is ready for use in the field, Bertozzi says. But she’s optimistic that the new method could prove useful in the global fight against TB. Her team can get fluorescent images in about an hour, but the researchers would need field-ready portable fluorescence microscopes­ to detect the bug. Still, Bertozzi points out that may be possible one day with work from Stanford collaborator Manu Prakash, an HHMI-Gates faculty scholar who debuted an origami-based paper microscope in 2014.


M. Kamariza et al. “Rapid detection of Mycobacterium tuberculosis in sputum with a solvatochromic trehalose probe.” Science Translational Medicine. Published online February 28, 2018. doi: 10.1126/scitranslmed.aam6310