BioWorld International Correspondent

LONDON - The treatment of tuberculosis could be revolutionized by the discovery of a new antibiotic that belongs to a completely novel class of drugs.

The compound, R207910, worked in laboratory tests even against multidrug-resistant strains of Mycobacterium tuberculosis. In an animal model, its action is fast, suggesting that its use could cut the length of treatment for TB in humans to just three or four months.

R207910 already has been tested in healthy human volunteers. It caused only mild side effects. Phase II trials to measure the drug's efficacy in people with active pulmonary TB are under way. If those go well, researchers plan to do Phase III trials to compare the drug with standard treatments.

Worldwide, TB causes 2 million deaths a year. M. tuberculosis infects one-third of the world's population, and each year there are almost 9 million new cases of active TB. No new drugs specific for TB have become available for 40 years, and, increasingly, strains of the bacterium that are resistant to multiple drugs are emerging.

Koen Andries, distinguished research fellow in antimicrobial research at Johnson & Johnson Pharmaceutical Research and Development, where the compound was discovered, said: "This is the first antibiotic, to my knowledge, shown to be active against any bacteria by inhibiting their energy supply. If this drug is ultimately approved for humans, it could lead to a change in treatment paradigm for tuberculosis."

Further development of R207910 will take place at Tibotec, a Belgian subsidiary of Johnson & Johnson.

Andries and colleagues and a multinational group of collaborators, based in France, the U.S., the UK and Sweden, report the discovery of R207910 online in the Dec. 9, 2004, issue of Sciencexpress. The title of the paper is "A Diarylquinoline Drug Active on the ATP Synthase of Mycobacterium tuberculosis."

Researchers at Johnson & Johnson had been screening compounds from a chemical library for their effect on a member of the Mycobacterium family. The research team identified a family of compounds called diarylquinolones, or DARQs, the most promising of which is R207910.

Studies in the laboratory showed that R207910 could kill several different mycobacteria, including drug-resistant strains of M. tuberculosis. The drug did not target other bacteria.

Until now, known antibiotics have targeted four actions: the bacteria's cell-wall synthesis, protein synthesis, folate biosynthesis or nucleic acid replication. The team found that R207910 operates in a completely different way. It attacks the proton pump of the enzyme ATP synthase, which manufactures adenosine triphosphate (ATP), the cell's energy source.

When the researchers tested the compound in a mouse model of TB infection, they found that treatment with a drug cocktail that included R207910 greatly reduced the amount of bacteria in the animals' lungs after one month (compared to two months for the same cocktail without the new drug), and after two months, the lungs of the mice appeared to be completely clear of bacteria - a finding that the researchers called "unprecedented."

Those studies showed that R207910 quickly entered the animals' bloodstream and even became concentrated in the cells of the lung, where TB lives. The compound also remained in the body for days, even when given to the animals only once a week.

Andries and colleagues believe that the new drug might be able to reduce treatment time of human TB by about 50 percent.

If that could be achieved, it could help to make treatment more effective: The cocktail of drugs recommended by the World Health Organization has to be taken for six to nine months to eradicate infection. Patients feel better within a couple of months, and it can be difficult for doctors to persuade them that they must continue with the therapy.

During the study in humans, the drug levels in the volunteers' plasma reached levels that were about seven times higher than in the mice that had been treated.

Andries concluded: "Barring any unsuspected side effects, we really do think we have a very interesting compound."