Tryptophan catabolism during intracellular infection.

TO THE EDITOR—In a recent issue of the Journal, Divanovic and colleagues investigated the roles of the tryptophan-degrading enzyme indoleamine 2,3-dioxygenase (IDO) during murine toxoplasmosis, cutaneous leishmaniasis, and herpes simplex virus type 1 infection [1]. They concluded from their results that IDO plays important roles in the host response and that its antimicrobial or immunoregulatory activities are pathogen-specific. Although these findings are quite interesting and important, we still believe that some of the conclusions drawn can be questioned simply because IDO1 and IDO2 are not the only antimicrobial enzymes involved in the restriction of intercellular pathogen proliferation. Moreover, some of the critical events behind the expression and activity of IDO differ significantly between rodent and human animal models. Therefore, the experimental results of Divanovic and colleagues, which were obtained in mice, cannot directly be extrapolated to human physiology. In the discussion of their data, Divanovic et al touch briefly on the interaction between IDO and nitric oxide (NO) and mention that their studies do not provide much insight regarding any cross-regulatory influence between the enzymes. However, this is an important consideration when discussing the findings in relation to human biochemistry, where IDO activity seems to be of much greater relevance. It is well established that there exists a regulatory relationship between the activation of inducible NO synthase (iNOS) and IDO. The expression and activity of the latter are arrested by NO [2], and this fact may at least partly explain why anti–Toxoplasma gondii activity exerted by IDO and/or iNOS varies among tissues [3]. In C57BL/6 mice, iNOS was found to be crucial for the control of Leishmania infection [4]. Further, it is well established that iNOS functionality has been observed in human monocytederived macrophages and dendritic cells on infection; however, a protocol for ex vivo stimulation of iNOS in these cells by cytokines has not yet been developed. This may, in part, reflect the fact that stimulation of human macrophages with interferon γ does not lead to production of 5,6,7,8-tetrahydrobiopterin (BH4), the necessary cofactor for iNOS [5]; instead, such cells produce and release neopterin at the expense of BH4 [6,7]. Human and primate cells outside the monocytemacrophage lineage are still capable of producing BH4, and thus also forming NO, with iNOS stimulation [8]. Nevertheless, in patients suffering from, for example, Streptococcus pyogenes infection [9] or human immunodeficiency virus type 1 infection [10], parallel increases in blood neopterin concentrations and tryptophan breakdown have been observed and have been found to have predictive significance for the course of the infection. In conclusion, distinct from rats and mice, interferon γ–stimulated human macrophages and dendritic cells do not produce BH4, and, as a consequence, they do not produce excessive amounts of NO, even when iNOS is expressed. Thus, IDO activity in human macrophages is high, leading to increased levels of tryptophan degradation products with barely detectable NO and peroxynitrite. In contrast, rodent cells accumulate high amounts of NO that inhibit IDO activity and thus repress tryptophan breakdown [1]. Divanovic et al examined the murine expression of IDO, but it will still be necessary to demonstrate functional activity of the enzyme by measuring tryptophan breakdown and production of kynurenines before a definitive conclusion about IDO function will be possible. In contrast, in humans suffering from infectious diseases, IDO activity, as reflected by an increased kynurenine-tryptophan ratio, is very high and is associated with immune paralysis [9, 10]. This is especially true in viral infections.