One factor contributing to the development of antibiotic resistance is poor patient compliance - patients start feeling better after a few days, and don't finish their course of antibiotics.

And if taking medicine for 10 days seems hard, tuberculosis therapies would seem impossible - treatment regimens for that disease require daily treatment or monitoring for six to nine months.

Before making a comeback in recent decades as one of the opportunistic infections that follow in the wake of HIV infection, TB was considered in industrialized nations a problem of the past, conjuring up images of emaciated consumptives in Victorian-era infirmaries.

Today, the world's 1.9 billion TB-infected people are mostly poor, which has two consequences: They cannot afford expensive drugs, and they often cannot afford the time that the current treatment regimen necessitates, either. Taken together, those things have led to a glaring absence of tuberculosis drug development - the last new compound was approved in 1963.

But as HIV has paved the way for tuberculosis' sad comeback, interest in drug development for the disease has increased, as well. The Global Alliance for TB Drug Development lists a total of 16 compounds that now are being pursued, from the discovery stage to clinical trials, and the Dec. 9, 2004, issue of Sciencexpress - the early online edition of Science - contains a paper describing one of those compounds, which appears to have a novel mechanism of action.

An Anti-Inflammatory Failure Finds Its Calling

The new compound, known as R207910, originally was synthesized by scientists at Johnson & Johnson Pharmaceutical Research and Development labs in the U.S. and Belgium in search of an anti-inflammatory compound that would target cytokines, but was a no-starter in that indication. In fact, when asked whether the compound's origin might mean that immunological side effects could reveal themselves once the compound was administered to humans on a larger scale, lead author Koen Andries was succinct: "No. As an anti-cytokine compound, it was a complete failure."

The same can't be said for R207910's antimicrobial properties, and that's good news. J&J researchers tested it against a plethora of mycobacteria, including the international reference strain, drug-susceptible and drug-resistant strains for M. tuberculosis, and related mycobacteria, including M. avium, a virus that once was specialized on birds but has branched out into those HIV patients. R207910 turned out to be highly effective against all of them in cell culture.

The reason for R207910's range of effectiveness is the drug's novel mechanism of action: The drug appears to work by targeting the bacterial ATP synthase, cutting off the bacterium's energy supply. Luckily, human ATP synthase is sufficiently different from its bacterial counterpart that significant interactions were not observed.

Another advantage of R207910 is that eventually it could help make HIV and tuberculosis therapies more compatible. Rifampin, a current tuberculosis treatment that usually is given as part of a drug ring with isoniazid and pyrazinamide, is incompatible with HIV treatments due to drug-drug interactions - a most unfortunate incompatibility, since the Global Alliance for TB Drug Development pegs HIV as the most potent risk factor for converting latent TB into active transmissible TB, and TB is the leading cause of death in persons who are HIV positive.

The researchers next took their compound into animal studies, where it continued to perform. Several results suggested that R207910 might be able to simplify tuberculosis treatments, making compliance easier for patients. The compound has a long half-life in tissue, and administering the drug once a week at a higher dose showed almost the same effect as five times weekly at lower doses. Both results point to the possibility of less-frequent pill popping for patients, though Andries told BioWorld Today that "the need for companion drugs in combination therapy that have to be given more frequently will prevent us from using this asset for the time being."

Also, replacing one member of the current tri-drug therapy with R207910 led to complete disappearance of the tuberculosis bacterium from the animals' lungs in just one month, rather than two. If that result holds up in human testing, it could shave months off of treatment.

Andries said that in mice monotherapy with R207910 was as effective as the current standard-treatment combination of rifampin, isoniazid and pyrazinamide, though he stressed that "we are not envisioning a monotherapy, because that would inevitably lead to resistance."

The researchers have taken their compound into the clinic; Phase I studies have established safe doses, and the drug has shown good tolerability to date. The company intends to start Phase II trials soon.