The next molecule of the year?: Carbon monoxide as a regulator of host inflammatory responses.

Abstract

Like nitric oxide, carbon monoxide is not only a gas but it serves as an endogenous molecule in mammals which mediates a variety of physiologic functions. Via activation of intracellular guanylate cyclase, carbon monoxide has a variety of effects on organ function including, for example, smooth muscle relaxation, inhibition of platelet aggregation, and modulation of neuronal activity (Proc Natl Acad Sci USA 2000;97:1851–5). Heme oxygenase is an enzyme that converts heme into biliverdin and iron, with the release of carbon monoxide. Like nitric oxide synthase producing nitric oxide, the heme oxygenase exists in two isoforms: a constitutive enzyme (referred to as HO-2) and an inducible isoform (HO-1) that is upregulated by a variety of stress-inducing factors including cytokines, chemokines, bacteria-derived endotoxin and hypoxia (FASEB J 1998;2:2554–8). Increasing evidence indicates that carbon monoxide, produced via HO-1, also plays a role in host defense against proinflammatory stimuli (J Clin Invest 1999;103:1047–54). Indeed, HO-1 activity is present in immune cells and increased HO activity modulates neutrophil migration. In a recent issue of Nature Medicine (2000;6:422–8), Otterbein et al. show that carbon monoxide results in less tumor necrosis factor [alpha] production and increased levels of interleukin 10 in mice challenged with lipopolysaccharide. The observed effects were independent of either activation of guanylate cyclase or nitric oxide liberation. Knockout mice were used to show the importance of the MAP kinase signal transduction cascade in mediating the observed anti-inflammatory effects of carbon monoxide. These exciting findings demonstrate that carbon monoxide can selectively modulate the inflammatory cascade of cytokines and that these effects are not mediated by the same pathways producing effects in either neuronal or vascular tissues. The direct anti-inflammatory effects of carbon monoxide have been shown in vivo by demonstrating that low concentrations can provide protection against lung injury in experimental animals (Am J Physiol 1999;276:L688–L694). If confirmed by other investigators when extended to models of injury in the gastrointestinal tract, the potential for using carbon monoxide as a therapeutic agent in humans will have to be tested. Safety issues are unlikely to be a major concern because concentrations of carbon monoxide used are 10-to 50-fold lower that the doses administered to humans to assess the diffusion capacity of the lung in a standard pulmonary function. Carbon monoxide could be considered for use in a number of clinical settings including, for example, modulating host responses to immune challenges, promoting long-term graft survival, and in a variety of inflammatory disease states.