Rejection of the Peroxide Accumulation Hypothesis of Isoniazid Action

Abstract

FROM three quite different lines of investigation it has been postulated that isoniazid kills tubercle bacilli through the intracellular accumulation of hydrogen peroxide^1–3. In Micrococcus lysodeikticus, which contains an exceptionally high concentration of catalase, Chance has been able to demonstrate a peroxidase–catalase complex and to show that the peroxide concentration is controlled by the presence of suitable oxygen acceptors within the cell or added externally⁴. Sensitive though it is, Chance's spectro-photometric method would be difficult to apply to Mycobacterium tuberculosis, which has a very low catalase activity⁵, and direct chemical determination of intracellular peroxide does not seem possible. It seemed reasonable, however, to test the theory of peroxide accumulation in the presence of isoniazid by growing the bacilli in an excess of a stable exogenous oxygen acceptor capable of penetrating the osmotic barrier of the cell but not inhibiting cell growth. Sodium nitrite appears to fulfil all the requirements in that it reduces free peroxide and reacts rapidly with the catalase–peroxide complex, both in vitro⁶ and intracellularly⁴.