omplete genomes are getting a lot of citations, but that isn't, as the leader of the latest team to enter the lists modestly tells Science Watch, "because they contain a lot of genes for people to cite." It is because they make sense of genuine biological mysteries. Mycobacterium tuberculosis, whose sequence is at #6, makes the case perfectly.

   John Bunyan characterized tuberculosis, the disease caused by the M. tuberculosis bacterium, as the captain of all the men of death," and more than 300 years later TB is still responsible for more deaths than any other single infectious agent. With the sequence comes a real chance to demote TB. The bare facts are these: M. tuberculosis, the H37Rv strain, was sequenced by a joint team from France's Institut Pasteur and the U.K.'s Sanger Centre. It has a genome 4,411,529 base pairs long, about 4,000 genes. That makes it a mite smaller than E. coli, but a lot more interesting.

   For example, one reason TB has proved such a scourge is that it grows very slowly and is, for the most part, benign for much of the course of the infection. The disease becomes dormant, during which time the bacterium is holed up in tubercles within the lungs, perhaps in response to an immune response that can contain, but not eradicate, the infection. As the host's immune system loses its vigor as a result of age or other infections, tuberculosis flares up again, killing the host but in the meantime infecting many new cases.

   Historically, the big question was what the bacterium was doing inside the tubercles. Researchers opined that M. tuberculosis was an obligate aerobe; at least, they were unable to grow it anaerobically in the laboratory. Because the inside of the tubercle contains little or no oxygen, M. tuberculosis must therefore be inert when inside the tubercles. That view, the sequence reveals, is probably wrong.

   "We found a set of genes that would enable it to grow anaerobically," says Julian Parkhill, a member of the team at the Sanger Centre, outside Cambridge. These genes include a previously unknown reductase that seems to represent a rearrangement of conventional nitrate reductase genes. M. tuberculosis clearly is capable of growth inside the tubercle; whether that capability is medically important remains to be seen.

   Another reason TB is difficult to treat is that it is naturally resistant to many antibiotics. That too is reflected in its sequence. Many resistance genes are encoded in the genome, including enzymes to destroy penicillins, and others that pump antibiotics out of the cell as fast as they enter. The cell envelope is possibly the main protection, being extremely hydrophilic, and M. tuberculosis boasts one of the most diverse sets of lipid metabolism genes yet uncovered.

   Stewart Cole, director of the French part of the team, believes that evidence that the PPE proteins are antigenic may in time lead to better vaccines. And the PE proteins are potentially immuno-suppressive. But for him, perhaps the biggest surprise was that M. tuberculosis has almost no virulence factors like the toxins seen in other pathogens. It is a surprise, however, that makes sense. M. tuberculosis is not very virulent. It sits there, slowly growing away, but does little apart from antagonize the host's immune system. "The pathology results from an excessive inflammatory reaction as the host tries to control the infection," Cole tells Science Watch. "It can be so extensive, you get tissue damage. The host kills itself, in a way."