The susceptibility to hydrogen peroxide of Indian and British isoniazid-sensitive and isoniazid-resistant tubercle bacilli.
Subbaiah, T.V.; Mitchison, D.A.; Selkon, J.B.
Tubercle; 1960; 41; 323-333.
Cohn and others (1954), Bonicke (1954) and Meadow and Worssam (1956) have shown that isoniazid-resistant, catalase-negative strains of tubercle bacilli, whether obtained by in vitro selection or from treated patients were inhibited by hydrogen peroxide at lower concentrations than their isoniazid-sensitive, catalase-positive parent strains. These authors also demonstrated that hydrogen peroxide was more rapidly bactericidal to the catalase-negative than to the catalase-positive strains.
Kreis and Le Joubioux (1957a) described a quantitative bactericidal test, in which a large inoculum (0.8 mg./ml.) of a dispersed culture was exposed to 0.03 per cent w/v hydrogen peroxide for 3 hrs. at 37°C and a viablecount of the surviving bacterial population was then made. Strains composed of catalase-negative organisms were completely killed, whereas there was little fall in the counts of those with high catalase activity. Such a test should provide an estimate of the proportion of catalase-positive organisms in a strain composed of a mixture of catalase-positive and catalase-negative bacilli. However, the authors showed that a small inoculum of catalase-positive organisms was killed by the prolonged contact with the peroxide employed in the test; the reason for the survival of the bacterial populations in the large inoculs of these organisms normally used was that they were capable of destroying much of the peroxide itself during the period of exposure. Thus the method could not be relied upon to detect the presence of small numbers of catalase-positive bacilli in a predominantly catalase-negative strain, nor was it certain to what extent any of the estimated proportions of these populations would correspond accurately with the true proportions. No experiments were described in which the size of the inoculum, the concentrations of peroxide or the exposure period were varied to obtain optional conditions.
The present work describes an attempt to modify the method of Kreis and Le Joubioux (1957a) so that it would accurately estimate the relative proportions of catalase-positive and catalase-negative organisms in strains containing mixtures of the two types. A bactericidal test was chosen in preference to a bacteriostatic test, since it is difficult to obtain quantitative measurement with the latter technique. In performing a bactericidal test residual peroxide must be inactivated or removed by dilution so that it does not inhibit the growth of surviving organisms. Knox, Meadow and Worssam (1956) removed peroxide by centrifugation and washing, but this method was considered impracticable if this test were to be used on a large scale, and likely to produce inaccurate counts on the surviving organisms. In the present work the method of removal of peroxide was studied as well as the determination of the optimal peroxide concentration and period of exposure which would kill all catalase-negative organisms, but would leave catalase-positive organisms unaffected. In addition, the method of Kreis & Le Joubioux (1957a) was modified by reducing the inoculum of organisms exposed to peroxide so that catalase-positive bacilli would not be able to destroy peroxide during the test itself. The standardised bactericidal test was then employed in comparing the susceptibility to peroxide of isoniazid-sensitive strains from British and Indian patients, and in investigating the relationship between the peroxide susceptibility and the catalase activity of their isoniazid-resistant mutant strains.
Back to List of publications / Home |