© 2002 American Thoracic Society
Inducible Nitric Oxide Synthase in the Tuberculous Human LungWeill Medical College of Cornell University New York, New York "In vivo veritas": pioneers of tuberculosis research followed this adage when they pursued pathologic analysis of the disease. Such studies fell from fashion as molecular and genetic approaches began to generate a wave of new hypotheses about pathogenesis. Thus it has come to pass that much of contemporary tuberculosis research is based on experiments in vitro, in mice, and to a lesser extent in human bronchoalveolar lavage cells, but not in human tissues. For this and other reasons discussed subsequently, it is noteworthy that, in this issue of AJRCCM (pp. 178–186), Choi and colleagues have now analyzed tuberculous human lungs for the expression of nitric oxide synthases (NOSs) (1). To appreciate their contribution, it is useful to reflect on the immunologic puzzle posed by tuberculosis, the questions that Choi and coworkers (1) have answered, and the questions that they've left hanging. One hundred twenty years after Koch's discovery of the tubercle bacillus, we still do not understand how it is that infection by Mycobacterium tuberculosis is usually controlled, and yet viable organisms usually persist; nor why disease develops in 5–10% of infected individuals who are immunocompetent and immunoreactive. During subclinical infection and at the outset of disease, most tubercle bacilli reside in macrophages. Thus, the question can be distilled: How do macrophages kill tubercle bacilli, and why don't they do a better job of it? The NOS2 isoform, often called iNOS, is a leading candidate to account for the incompletely successful bactericidal activity of macrophages toward M. tuberculosis. The designation "i" denotes that production of nitric oxide by iNOS is independent of elevated intracellular Ca2+ (2), a biochemically distinctive and biologically important feature that explains how the enzyme can produce nitric oxide for days after its transcription is induced by immunologic and inflammatory stimuli. Nitric oxide is the only molecule known to be produced by mammalian cells that can kill tubercle bacilli in vitro with a molar potency comparable to that of chemotherapy (3). That the primary product of iNOS is mycobactericidal provides one type of evidence consistent with a role for iNOS in controlling tuberculosis. There are four more lines of evidence: (1) immunologically activated, iNOS-expressing mouse macrophages can kill M. tuberculosis in vitro, but not if the macrophages are treated with iNOS inhibitors (4) or bear disrupted NOS2 alleles (5); (2) iNOS is expressed in infected mouse tissues in which the growth of M. tuberculosis is restrained, but iNOS is scant or absent when immunosuppressive drugs or genetic interventions impair host resistance (reviewed in [6]); (3) healthy mice that harbor tubercle bacilli succumb abruptly to tuberculosis after ingestion of specific iNOS inhibitors (6, 7); and (4) mice with disrupted NOS2 alleles die with fulminant tuberculosis in a few weeks, whereas wild-type mice survive the infection for about 9 months (6, 8). It is difficult to determine how much of this is relevant to human tuberculosis. No one has knowingly administered iNOS inhibitors to people infected with tubercle bacilli, and no primary genetic deficiency of iNOS has been identified in humans. Thus, only two experimental avenues have been open: the impact of iNOS inhibitors on the mycobactericidal activity of human macrophages in vitro, and a search for iNOS at sites of tuberculosis. Studies of human macrophages in vitro have been more frustrating than informative, because the macrophages tested to date have rarely exerted a bactericidal effect against M. tuberculosis, precluding a determination as to the contribution of iNOS to killing. Both mouse (5) and human macrophages (9) that lack iNOS can exert a bacteriostatic effect against M. tuberculosis by an unknown mechanism, but with respect to iNOS these cells do not model human macrophages at sites of infection or inflammation, which are often intensely iNOS-positive. Thus, one of the major unfulfilled goals for research in the immunology of tuberculosis is to learn how to obtain or culture human macrophages that are as iNOS-positive in vitro as they are in vivo and to test whether they kill M. tuberculosis and use iNOS to do so. With so many experimental avenues blocked, intense interest devolves on the one remaining: whether iNOS is expressed in human tuberculosis. The answer has been affirmative for cells obtained by bronchoalveolar lavage. Macrophages lavaged from each of 11 patients with tuberculosis, but not those from normal subjects, expressed active iNOS as assessed by immuno- and cytochemistry (10). Tuberculosis patients exhaled more nitric oxide than healthy control subjects, and their bronchoalveolar macrophages contained iNOS and released nitric oxide in vitro (11, 12). Choi and coworkers (1) have now carried this inquiry into lung specimens resected from eight patients with tuberculosis. Immunohistochemistry demonstrated iNOS in the inflammatory zone of granulomas and surrounding pneumonitic regions. The enzyme was abundant in epithelioid macrophages, multinucleated giant cells, alveolar macrophages, and epithelial cells (1). Nitrotyrosine was detected in the same cells (1). Tyrosine residues become nitrosated when peroxynitrite arises in their immediate vicinity from the interaction of nitric oxide and superoxide, or when nitrite (arising from the spontaneous oxidation of nitric oxide) is oxidized by hydrogen peroxide through the agency of myeloperoxidase. By whichever of these routes tyrosine residues became nitrated, the cellular colocalization of nitrotyrosine and iNOS implies that at some point in the development of the granulomas, there must have been enough oxygen to sustain the catalytic activity of iNOS. Finally, Choi and coworkers (1) detected macrophage expression of NOS3, an enzyme first cloned from endothelial cells. NOS3 has rarely been detected in macrophages and never implicated in their function. In endothelial cells the enzyme is intermittently active in accord with its regulation by transient elevations in intracellular Ca2+ or stimulus-induced serine phosphorylation. The amounts of nitric oxide made by NOS3 in endothelial cells are probably too low to exert mycobactericidal activity. What might activate NOS3 in infected macrophages, and to what effect? Perhaps the Chan laboratory will complete the circle, testing the significance of the observation they made in humans by studying M. tuberculosis infection in NOS3-/- mice (13). The work of Choi and colleagues (1) does not establish, but strongly supports, the possibility that iNOS may be functionally important in human tuberculosis. Their report is timely, because we urgently need to understand the biochemistry of the dialog between the macrophage and this peristent pathogen. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
and Mycobacterium tuberculosis: signaling roles of nitric oxide synthase-2 and phagocyte oxidase. J Exp Med 2001;194:1123–1140.
