Collaborating Author

With antibiotic-resistant strains of Mycobacterium tuberculosis on the rise, the development of new drugs that combat this bacterium and treat tuberculosis has become an urgent public-health need. Biophysicist Huilin Li of Brookhaven National Laboratory in Upton, New York, has been collaborating for several years on this problem with Carl Nathan, a researcher at Weill Cornell Medical College in New York City. Together they have identified a new class of compounds — oxathiazol-2-ones — that selectively target mycobacteria but not human cells (see page 621). He tells Nature about the implications.

How do the compounds work?

They target the mycobacterial ubiquitin-proteasome pathway, which degrades damaged proteins. We don't know exactly how or why, but agents that inhibit the function of proteasomes in mycobacteria kill these pathogens in host cells. It's clear that mycobacteria must get rid of their damaged proteins to survive the hostile environment of the human macrophage cells they infect.

What's the main hurdle to developing proteasome inhibitors of tuberculosis?

The difficulty is finding species-specific inhibitors that target the mycobacterial proteasome without affecting the human one. Proteasomes are protein complexes that are highly conserved from archaebacteria to humans. Dozens of proteasome inhibitors exist, but most inhibit all proteasomes from any cell and any species.

How did you beat the specificity problem?

Our earlier collaboration solved the crystal structure of the mycobacterial proteasome bound to another class of inhibitor. This revealed biochemical and structural differences between the bacterial and mammalian proteasomes, which we exploited to screen for chemical compounds that would bind to only the mycobacterial and not the human proteasome.

Did the compounds have other advantages?

Mycobacteria have a sticky lipid coating full of mycolic acid that prevents a lot of compounds from getting into the bacteria. Our inhibitors go through this cell wall very quickly. Another crucial issue is that many existing anti-tuberculosis drugs target ribosomes, which are not active if the bacteria are not replicating, or 'dormant', and thus not making any new proteins. But dormant bacteria still require active proteasomes. This means that our discovery could lead to compounds that would clear a person of mycobacteria in a way that current treatments do not.