Tubercle and Lung Disease (2000) 80(4/5), 237I242 ^ 2000 Harcourt Publishers Ltd doi: 10.1054/tuld.2000.0251, available online at http://www.idealibrary.com.on Deletion of the putative antioxidant noxR1 does not alter the virulence of Mycobacterium tuberculosis H37Rv G.R. Stewart,* S. Ehrt, f L.W. Riley, f J.W. Dale,* J. McFadden* *School of Biological Sciences, University of Surrey, Guildford, Surrey GU2 5XH, UK f Division of Public Health Biology and Epidemiology, School of Public Health, 140 Warren Hall, UC Berkeley, Berkeley, CA 94720, USA Summary Setting : The cloned M. tuberculosis noxR1 gene has been shown to confer resistance to reactive nitrogen intermediates (RNI) and reactive oxygen intermediates (ROI) upon Escherichia coli and Mycobacterium smegmatis. Objective: To investigate the role of noxR1 in resistance to RNI and virulence of M. tuberculosis. Design: The noxR1 gene was deleted from M. bovis BCG and M. tuberculosis H37Rv by allelic exchange. The mutants were compared to wild type strains with respect to resistance to chemically generated RNI. The virulence of the M. tuberculosis mutant was investigated in a murine model of infection. Results: The NoxR1 mutants grew normally in Sautons and 7H9 broths. The BCG mutant demonstrated decreased resistance to in vitro generated RNI compared to the wild type. Resistance to RNI could be restored to the mutant by reintroduction of the noxR1 locus on a replicating plasmid. However, deletion of noxR1 from M. tuberculosis H37Rv did not result in decreased resistance to RNI nor a difference in growth and survival of the bacterium during murine infection. Conclusion: The noxR1 gene locus in M. bovis BCG bestows ability to resist RNI generated in vitro. In M. tuberculosis H37Rv, however, noxR1 is either not involved in RNI resistance and virulence or is better compensated for by other mechanisms. ^ 2000 Harcourt Publishers Ltd INTRODUCTION Central to the success of Mycobacterium tuberculosis is its ability to replicate within phagocytic cells. To achieve this the mycobacterium must avoid killing by the numerous antimicrobial mechanisms of the macrophage such as low pH, degradative enzymes and the production of reactive oxygen intermediates (ROI) and reactive nitrogen inter- mediates (RNI). Tubercle bacilli evade exposure to low pH and degradative enzymes by arresting processing along the normal endosomal/lysosomal pathway, instead re- maining in a non-acidi9ed endosomal compartment which does not fuse with lysosomes. The tubercle ba- cillus also has a number of defence mechanisms against ROI including: complement mediated uptake, which may avoid stimulation of ROI production; various sul- folipids, glycolipids and lipoarabinomannan in the cell Correspondence to: Graham Stewart, Medical Microbiology, Imperial College School of Medicine, St Mary’s, Norfolk Place, London W2 1PG, UK. Tel.: #44 171 594 3955; Fax: # 44 171 262 6299; E-mail: g.stewart@ic.ac.uk Accepted: 17 August 2000 wall which inhibit ROI production; and antioxidant enzymes such as catalase, alkylhydroperoxide reductase (AhpC) and superoxide dismutase. In testament to the effectiveness of these mechanisms, the antimicrobial ac- tion of host cell derived ROI appears limited. However, the importance of RNI in the control of M. tuberculosis in murine systems is well demonstrated and notably RNI-dependent killing and growth inhibition of M. bovis BCG and M. tuberculosis has been observed in human macrophages. Despite this, there is a paucity of knowledge concerning bacterial resistance mechanisms to RNI. Such mechanisms represent attractive targets for drug intervention strategies designed to weaken the bac- terial defence against macrophages. In a previous study, it was shown that heterologous expression of an open reading frame from M. tuberculosis, noxR1, conferred resistance to RNI and ROI generated in both macrophages and in host cell-free systems. This was the 9rst demonstration of a putative M. tuberculosis resistance mechanism against RNI and described an ap- parently novel antioxidant molecule. In the present study, we have deleted noxR1 from M. bovis BCG and M. tubercu- losis H37Rv, in an attempt to demonstrate whether this 237