Blueprint for the white plague

With annual deaths from Mycobacterium tuberculosis estimated at around three million, this single pathogen claims more human lives than any other. What we now learn from the sequence of its genome should help in devising new strategies to fight it.

“Now — under the microscope the structures of the animal tissues⃛ are brown, while the tubercle bacteria are a beautiful blue.”

More than a century after Robert Koch's historic lecture1, we are presented with a new perspective on Mycobacterium tuberculosis — the complete sequence of its genome, reported by Stewart Cole and his colleagues2 on page 537 of this issue. Although no longer blue, it retains a sense of beauty, and the sequence marks a new phase in the battle against one of mankind's most successful predators.

In the final frantic years of the twentieth century, with newly emerging pathogens adapting to a lifestyle dictated by the time it takes to cook a hamburger, it is remarkable that it is the tubercle bacillus — a ponderous microbe more often associated with the nineteenth century — that still claims the title of “captain of all the men of death”. Mycobacterium tuberculosis opted for residence in the inner recesses of the human lung probably around the time that cattle were first domesticated, some 10,000 years ago3. It took a long view of life, eschewing the rapid, go-for-growth strategies of more acute pathogens in favour of a leisurely cycle restricted to a single daily round of cell division even in optimal culture conditions. For the most part, it adopted a pattern of peaceful co-existence with its human host in the form of a quiescent or dormant infection, establishing a massive reservoir of infected individuals — possibly including as much as one-third of the world's population. But it can also trigger progressive and deadly destruction of the lungs in the unfortunate minority who suffer from clinical disease (Fig. 1) and act as the source for further dissemination.

Figure 1


Attempt at treating tuberculosis by Francisque Crotte, 1901.

Over subsequent millennia M. tuberculosis has taken a close professional interest in the evolution of human society. It has exploited opportunities to spread during periods of urbanization and social upheaval, retreating to the margins of society in times of affluence and improved hygiene. The tubercle bacillus has also weathered the antibiotic revolution, in part through mankind's natural reluctance to comply with the requisite six-month intensive course of treatment, and, more recently, by the development of drug-resistant forms. Moreover, it has forged a deadly partnership with the human immunodeficiency virus (HIV) — increased susceptibility to tuberculosis is associated with the early stages of HIV infection, and tuberculosis in turn accelerates the progression to AIDS.

This history, together with the clues to conquer the tubercle bacillus, is written in its genome. Thanks to Cole et al.2, we now have the sequence of every potential drug target and of every antigen we may wish to include in a vaccine. But can we convert this mass of information into a useful understanding? It will not be a simple task. It is a good-sized genome — with 4.4 million base pairs, it is surpassed in size only by Escherichia coli in the list of bacterial genomes completed to date. It has the metabolic potential to exist in a variety of environments, including pathways associated with anaerobic metabolism. This is important because, although the bacteria require oxygen for growth in the laboratory, prolonged anaerobic survival might be necessary for long-term persistence in tissues4. The authors have also identified a series of genes that encode proteins involved in transcriptional regulation, indicating a potentially flexible response to the changing fortunes encountered during infection.

Koch was the first to comment on the unusual cell wall of M. tuberculosis, and its importance in mycobacterial physiology is reflected in a vast repertoire of genes involved in lipid and polysaccharide metabolism. The cell wall has been analysed biochemically by generations of outstanding researchers5, and there is real poetry in seeing how the complex structures of the components that make up the cell wall are mirrored in the myriad of genes that encode the corresponding biosynthetic machinery. Information about the genes involved in synthesis of the cell wall provides a cornucopia of potential drug targets.

The ability of M. tuberculosis to cause disease has not been ascribed to any well-defined virulence factors. However, Cole et al. have identified a repeated DNA sequence that encompasses a putative cell-entry protein, hinting at a parallel with pathogenicity islands found in other bacterial pathogens6. The genetic information required to cause disease is often exchanged between bacteria by a process termed ‘horizontal transfer’, and there is evidence that such transfer can occur in M. tuberculosis. The genome contains copies of mobile genetic elements known as insertion sequences, as well as genes derived from viruses that can attack bacteria, although there is no indication of a role in virulence. By comparing different strains of M. tuberculosis, the authors suggest that it probably acquired these genes before becoming established in its relatively isolated niche in the lung.

There is almost no diversity between one set of genes from a range of clinical isolates3, indicating a remarkably high level of sequence conservation in the M. tuberculosis genome. But DNA-fingerprinting techniques reveal that other parts of the genome are highly variable7,8. An unexpected and exciting finding by Cole and colleagues is that one set of variable elements — the so-called polymorphic G+C-rich sequences — corresponds to a family of sequences that encode proteins with short peptide motifs made up of common repetitive domains. These proteins, which account for an astonishing 10% of the genome, are reminiscent of those implicated in antigenic variation in other microbes. By altering the pattern in which these proteins are expressed, pathogens present the immune system with a moving target, thereby increasing their chance of survival. Could the polymorphic G+C-rich sequences play an analogous role for M. tuberculosis? Because the pathology of tuberculosis infection is probably mediated mainly by the misguided actions of the host immune response, rather than by bacterial toxins, the concept of some form of antigenic ‘smokescreen’ is intriguing.

Figure 2: Electron micrograph of Mycobacterium tuberculosis, the genome of which has now been sequenced by Cole et al.2


Also known as the white plague, tuberculosis was given the title “captain of all the men of death” by John Bunyan, towards the end of the seventeenth century.

The M. tuberculosis sequence is the first product of the Wellcome Trust Pathogen Genome Unit at the Sanger Centre, Cambridge, UK. The sequence of its cousin, Mycobacterium leprae, is close to completion, and scientists at the Institute for Genome Research (TIGR) are working on other mycobacterial genomes, including a second isolate of M. tuberculosis. By comparing these genomes we should be able to identify genes associated with particular biological properties. Moreover, the recent development of tools for genetically manipulating M. tuberculosis9,10 will be invaluable in translating these bioinformatic ideas into the ‘wet lab’ products of new drug targets and attenuated strains for vaccine testing. Promising results with DNA vaccination against tuberculosis open up additional strategies in the search for protective antigens by whole-genome screens11,12. The sequence data are being matched with maps of protein expression by two-dimensional gel electrophoresis and mass spectrometry, and with transcriptional analysis using microchip technologies. After several decades in the slow lane of classical microbiology, M. tuberculosis is once again at the cutting edge of science.


  1. 1

    Koch, R. Berliner Klin. Wochenschr. 19, 221–230 (1882).

    Google Scholar 

  2. 2

    Cole, S. T. et al. Nature 393, 537–544 (1998).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Sreevatsan, S. et al. Proc. Natl Acad. Sci. USA 94, 9869–9874 (1997).

    Google Scholar 

  4. 4

    Wayne, L. G. & Hayes, L. G. Infect. Immun. 64, 2062–2069 (1996).

    Google Scholar 

  5. 5

    Brennan, P. J. & Nikaido, H. Annu. Rev. Biochem. 64, 29–63 (1995).

    Google Scholar 

  6. 6

    Covacci, A. et al. Drug Dev. Res. 41, 180–192 (1997).

    Google Scholar 

  7. 7

    van Embden, J. D. et al. J. Clin. Microbiol. 31, 406–409 (1993).

    Google Scholar 

  8. 8

    Poulet, S. & Cole, S. T. Arch. Microbiol. 163, 87–95 (1995).

    Google Scholar 

  9. 9

    Bardarov, S. et al. Proc. Natl Acad. Sci. USA 94, 10961–10966 (1997).

    Google Scholar 

  10. 10

    Pelicic, V. et al. Proc. Natl Acad. Sci. USA 94, 10955–10960 (1997).

    Google Scholar 

  11. 11

    Huygen, K. et al. Nature Med. 2, 893–898 (1996).

    Google Scholar 

  12. 12

    Tascon, R. E. et al. Nature Med. 2, 888–892 (1996).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Young, D. Blueprint for the white plague. Nature 393, 515–516 (1998).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.