The potentially deadly association between AIDS and tuberculosis represents an enormous public health problem which has been compounded by the emergence of strains of M. tuberculosis resistant to two, or more, frontline drugs. It is now known that, as in other eubacteria, resistance to rifampicin, streptomycin and fluoroquinolones results from missense mutations to essential chromosomal genes encoding the drug targets. In contrast, at least two novel mechanisms are responsible for resistance to the potent tuberculocidal drug, isoniazid. High level resistance is associated with mutations which inactivate the katG gene, or result in greately reduced activity of its product, the hem-containing enzyme catalase-peroxidase which is believed to activate the drug inside the bacterium. Lower resistance levels, and cross-resistance to ethionamide, result from over-expression of the inhA gene, encoding a novel fatty acid synthase that may be involved in mycolic acid production. Unlike for many bacteria, multidrug resistance is not due to the acquisition of resistance plasmids or transposons but to the accumulation of mutations in the genes encoding the respective drug targets.