Immunoregulatory and Antimicrobial Effects of Nitrogen Oxides

Abstract

The therapeutic effects of inhaled nitric oxide (NO) therapy are thought to be restricted to the pulmonary vasculature because of rapid inactivation of NO by hemoglobin in the bloodstream. However, recent data suggest that inhaled NO may not only be scavenged by the heme iron of hemoglobin but also may react with protein thiols in the bloodstream, including cysteine-93 of the hemoglobin B subunit. Reaction of NO with protein or peptide thiols is termed S-nitrosylation and results in the formation of relatively stable protein S-nitrosothiols that carry NO bioactivity to distal organs. Thus, inhaled NO-induced protein S-nitrosylation may allow inhaled NO to have multiple as yet undiscovered physiologic and pathophysiologic effects outside of the lung. Here we review the immunoregulatory and antimicrobial functions of NO and the potential effects of inhaled NO therapy on host defense.

Inhaled nitric oxide (NO) is used to treat pulmonary hypertension in both adults and infants. Unlike most vasodilators, inhaled NO selectively vasodilates the pulmonary vasculature without inducing systemic hypotension. The selectivity of inhaled NO has been attributed to the rapid scavenging and inactivation of NO by hemoglobin when NO diffuses from the lung into the bloodstream. However, data from the Stamler laboratory have challenged the idea that hemoglobin merely scavenges NO. Instead, the Stamler group has demonstrated that NO reacts not only with the heme iron but also with cysteine (Cys)-93 on the B subunit of hemoglobin (1, 2). Whereas reactions with the heme iron can inactivate NO, S-nitrosylation of Cys-93 converts hemoglobin into a carrier of NO bioactivity (3, 4). Not only hemoglobin, but also other intravascular proteins such as albumin, may be S-nitrosylated by inhaled NO and transport NO bioactivity to distal organs (Figure 1).

Figure 1.

Transport of NO bioactivity to distal organs after inhaled NO treatment. Inhaled NO diffuses from alveoli into smooth muscle, where it nitrosylates and activates soluble guanylate cyclase (sGC), leading to smooth muscle relaxation and vasodilation. In addition, NO diffuses into pulmonary capillaries, where it undergoes a variety of reactions including reaction with the heme iron of hemoglobin (Hb), leading to formation of nitrate (); reaction with superoxide to form peroxynitrite (ONOO⁻); reactions with oxygen to form higher oxides of nitrogen, such as N₂O₃; and reactions with thiols on proteins, such as hemoglobin and albumin to form relatively stable S-nitrosothiols such as S-nitrosohemoglobin (SNO-Hb) and S-nitrosoalbumin (SNO-Alb). Protein S-nitrosothiols transport NO to distal organs.

TRANSPORT OF INHALED NO TO THE INTESTINE DURING ISCHEMIA–REPERFUSION

Animal models of ischemia and reperfusion support the concept that NO bioactivity is transported to distant organs after inhaled NO therapy (5, 6). Exposure of feline intestine to 1 h of ischemia and 1 h of reperfusion results in a decrease in intestinal blood flow, an increase in leukocyte adhesion, and an increase in mucosal barrier leakiness. If animals are ventilated with 80 ppm NO either during or after reperfusion, the abnormalities in intestinal blood flow and leukocyte adhesion are corrected. These data indicate that inhaled NO bioactivity reaches distal organs. However, unlike systemic NO therapy (7), inhaled NO has no effect on mucosal dysfunction induced by ischemia–reperfusion. Thus, inhaled NO bioactivity may not reach the extravascular compartment of distal organs. In support of this hypothesis, S-nitrosothiol levels in the lymph draining the intestinal extravascular space are not increased by inhaled NO therapy. Inhaled NO may be unable to reach extravascular tissue because it is carried to distal organs on proteins such as S-nitrosohemoglobin or S-nitrosoalbumin, which are too large to traverse even an injured endothelial barrier (6). However, the documented effects of inhaled NO on the microvasculature of the gut raise the possibility that inhaled NO may have a variety of as yet undiscovered effects on intravascular physiology and pathophysiology outside of the lung.

EFFECTS OF NO ON THE FUNCTION AND SIGNALING OF IMMUNE CELLS

Immune cells in the vasculature of distal organs may be one of the targets of inhaled NO therapy. NO has multiple effects on immune cells. For instance, NO alters the T helper (Th)1–Th2 balance (8–10). The Th1 subset of helper T cells synthesizes the inflammatory cytokines IFN-γ and interleukin (IL)-2 whereas the Th2 subset synthesizes cytokines such as IL-4, IL-5, and IL-10. NO decreases Th1 proliferation and IL-2 synthesis while increasing IL-4 synthesis by Th2 cells (8–10). One mechanism by which NO inhibits IL-2 production may be disruption of zinc finger transcription factor Sp1, leading to release of zinc and decreased IL-2 transcription (11). NO-induced enhancement of the Th2 response and inhibition of the Th1 response may promote inflammation in allergic diseases such as asthma but inhibit the inflammatory response to viral and bacterial pathogens (12).

NO also modulates the immune response by regulating leukocyte adhesion and recruitment to sites of infection. As discussed above, inhaled NO has been shown to decrease leukocyte adhesion during ischemia and reperfusion of the gut (5, 6). Decreased leukocyte adhesion and recruitment may have deleterious effects on host defense during infections. However, in models of acute lung injury and Pseudomonas pneumonia, inhaled NO increased leukocyte recruitment when administered with high Fi_O2 (13, 14). Thus the effects of inhaled NO on leukocyte recruitment may depend on the coadministered Fi_O2 concentration as well as on the local redox environment of tissues.

One of the best-characterized mechanisms by which NO stimulates the immune response is via S-nitrosylation of the monomeric G protein Ras. Ras is a guanine nucleotide–binding protein that cycles between inactive guanosine diphosphate–bound and active guanosine diphosphate–bound states to stimulate a wide array of cellular processes including lymphocyte proliferation and cytokine production. S-nitrosylation of the redox-active Cys-118 on Ras leads to increased levels of active guanosine triphosphate–bound Ras. Of interest, guanine nucleotide exchange is stimulated not by S-nitrosylation of Cys-118 but by a radical intermediate formed during the process of S-nitrosylation (15–18).

In summary, NO regulates the function of immune cells by a variety of mechanisms, including alteration of the Th1–Th2 balance, regulation of leukocyte recruitment and adhesion, and activation of Ras. As is discussed in more detail below, the net effect of NO on the function of immune cells is likely to be dependent on the specific NO-related species generated within a cell, which in turn is dependent both on the intracellular concentration of NO and the intracellular redox environment.

EFFECTS OF NO ON APOPTOSIS

NO modulates the immune response not only by regulating the function of immune cells but also by regulating apoptosis. Both endogenous NO and inhaled NO have well-documented, generally antiapoptotic, effects in the lung. For instance, endogenous NO production inhibits apoptosis of pulmonary epithelial cells in culture (19) and inhibits pulmonary apoptosis induced by either bleomycin or LPS treatment in vivo (20, 21). In addition, inhaled NO has been shown to inhibit pulmonary apoptosis after hyperoxia (22) or ischemia–reperfusion injury (23).

Inhaled NO may also regulate apoptosis of immune cells in the intravascular compartment of distal organs. NO has both pro- and antiapoptotic effects on immune cells. The mechanisms underlying the proapoptotic effects of NO include inhibition of nuclear factor (NF)-κB through a variety of mechanisms, including S-nitrosylation of NF-κB and inhibitory κB kinase leading to decreased NF-κB–mediated transcription and decreased Bcl-2 expression (24–27), increased p53 expression secondary to NO-mediated inhibition of the proteasome or to direct DNA damage (28) and subsequent increased Bax expression (29, 30), S-nitrosylation–stimulated nuclear translocation of glyderaldehyde-3-phosphate dehydrogenase (GAPDH) (31), increased heme nitrosylation and release of cytochrome c from mitochondria (32, 33), and inhibition of inhibitor of apoptosis protein expression (Figure 2) (34). The mechanisms underlying the antiapoptotic effects of NO include S-nitrosylation and inhibition of caspases (35–37), stimulation of the antiapoptotic activity of thioredoxin via S-nitrosylation (38), and increased expression of heat shock proteins (39) and Bcl-2 (Figure 3) (40).

Figure 2.

Proapoptotic effects of NO. NO-related species stimulate apoptosis through a variety of mechanisms, including inhibition of nuclear factor-κB (NF-κB) and stimulation of p53 activity, leading to increased expression of the proapoptotic protein Bax and decreased expression of the antiapoptotic protein Bcl-2, S-nitrosylation–stimulated nuclear translocation of GAPDH, stimulation of cytochrome c (Cyc) release from mitochondria, enhancement of the proapoptotic activity of cytochrome c via heme nitrosylation, and inhibition of inhibitor of apoptosis (IAP) expression. C3 = caspase-3.

Figure 3.

Antiapoptotic effects of NO. NO-related species inhibit apoptosis through a variety of mechanisms, including a cGMP-induced increase in Bcl-2 expression; increased expression of heat shock protein-70 (HSP70), leading to decreased cytochrome c release and decreased apoptosome formation, inhibition of caspase (Casp-3) activity via S-nitrosylation of the caspase catalytic site cysteine, and stimulation of the antiapoptotic activity of thioredoxin via S-nitrosylation of Cys-69.

The of NO concentration plays a key role in determining whether NO stimulates or inhibits apoptosis. In general, higher concentrations of NO induce whereas lower concentrations of NO inhibit apoptosis. For instance, increased generation of NO after apoptotic stimulation elicits S-nitrosylation of GAPDH, which triggers its binding to Siah1 (an E3-ubiquitin ligase), nuclear translocation, and cell death (31). In contrast, low constitutive levels of NO production in nonapoptotic cells lead to S-nitrosylation of the catalytic site cysteine of caspase members (37). Caspases are a family of cysteine proteases that play an essential role in the initiation and execution of apoptotic death pathways. S-nitrosylation of the catalytic site cysteine of caspase members inhibits caspase activity and thereby prevents apoptosis in resting cells.

Thus, NO has multiple targets that can either stimulate or inhibit both the function and apoptotic death of immune cells. The precise effect of inhaled NO on the immune response is likely to depend not only on the concentration of inhaled NO that reaches immune cells but also on the specific redox environment of the cells. The redox environment determines the specific NO-related species that are generated from inhaled NO. Different NO-related species have different biochemical targets and therefore elicit different biological responses. For instance, the presence of electron acceptors, such as transition metals, allows the generation of NO⁺ equivalents from inhaled NO that S-nitrosylate reduced thiols on caspases and thereby inhibit apoptosis (41). In contrast, superoxide generated intracellularly will react with NO to form the potent oxidant peroxynitrite, which damages DNA, leading to increased p53 levels and stimulation of apoptosis (28). To complicate matters further, inhaled NO can alter NO synthase (NOS) activity in tissues (42–44). Consequently, the intracellular concentration of NO-related species generated after inhaled NO therapy will depend on whether inhaled NO has stimulated or inhibited endogenous NOS activity. Further in vivo studies are needed to help predict the net effect of inhaled NO on the immune response in specific disease states.

ANTIMICROBIAL EFFECTS OF NO

NO not only has immunoregulatory effects but also direct effects on microbes. In general, NO either kills pathogens or inhibits their replication. NO has been reported to inhibit the growth of a wide variety of organisms including viruses, bacteria, parasites, and fungi (45). NO seems to play a particularly important role in the innate immune response to intracellular organisms such as Listeria monocytogenes, Salmonella, and Mycobacterium tuberculosis (46). There are multiple mechanisms involved in the antibacterial properties of nitrogen oxides including inhibition of proteins in the bacterial respiratory chain (47, 48) and disruption of iron–sulfur clusters in bacterial proteins leading to the release of free iron that catalyzes toxic oxidative reactions (49, 50). Moreover, NO and related species inhibit bacterial DNA replication by disrupting zinc metalloproteins involved in DNA replication and by inhibiting ribonucleotide reductase via reactions with a tyrosyl radical in the enzyme (51, 52).

NO also inhibits viral replication via a diverse array of mechanisms (Figure 4). As discussed above, reactions of NO with superoxide lead to the production of the potent oxidant peroxynitrite (ONOO⁻). Peroxynitrite reacts with capsid proteins on coxsackievirus, leading to the inhibition of viral entry into cells (53). NO also S-nitrosylates and inhibits a variety of viral proteinases that are required for viral replication (54, 55). Finally, NO inhibits transcription factors involved in viral replication (56, 57).

Figure 4.

Antiviral mechanisms of NO. NO reacts with superoxide to form the potent oxidant peroxynitrite. Peroxynitrite nitrates core proteins expressed on the surface of coxsackievirus, leading to the inhibition of viral RNA entry into cells. NO also S-nitrosylates and inhibits viral proteinases that are required for viral replication. In addition, NO inhibits transcription factors involved in viral replication.

In addition to static and cidal effects on pathogens, NO may play a critical role in maintaining pathogen latency. Many organisms, including M. tuberculosis, Leishmania major, Toxoplasma gondii, and Epstein-Barr virus, persist in host cells in a nonreplicative dormant state. Maintenance of latency requires an active host response because latent infections reactivate in immunosuppressed individuals. NO production by host cells may be critical for maintaining latency because inhibition of NO synthesis either by NOS inhibitors or by targeted disruption of the inducible NOS (iNOS) gene leads to reactivation of M. tuberculosis, L. major, T. gondii, and Epstein-Barr virus (56, 58, 59).

Although in general NO exerts antimicrobial effects, not all pathogens are inhibited by NO. Moreover, in some infectious processes NO contributes to disease pathogenesis. For instance, NO may contribute to the pathology of influenza pneumonia because mice with a targeted deletion of iNOS or treated with NOS inhibitors have less severe disease than do control mice (60). Similarly, NOS inhibition improves the course of Mycobacterium avium pneumonia and herpes simplex virus encephalitis in mouse models (61, 62). In these diseases, cytotoxic effects of NO on host cells may contribute to disease pathogenesis.

The effect of inhaled NO on infections in humans is just beginning to be evaluated. In one study, patients with acute lung injury were randomized to receive either placebo or 5 ppm inhaled NO (63). Patients receiving inhaled NO had higher rates of infection than did control subjects. Thus, in some clinical scenarios, inhaled NO may have deleterious effects on host defense. On the other hand, inhaled NO has been reported to decrease bacterial loads in animal models of Pseudomonas pneumonia (64).

Pulmonary tuberculosis is an infectious disease in which inhaled NO is likely to be of therapeutic benefit for several reasons. First, inhaled NO reaches M. tuberculosis in the lung. Second, 90 ppm NO gas kills M. tuberculosis in culture (65). Finally, NOS inhibition or targeted deletion of inducible NOS exacerbates M. tuberculosis infection in animal models, indicating that NO plays an important role in controlling the disease (59, 66). In another study, patients with tuberculosis were treated with 80 ppm inhaled NO for 3 days. Although the inhaled NO therapy was well tolerated, no decrease in bacterial load was noted (67). However, patients with tuberculosis often do not have a therapeutic response to conventional treatment at this early time point. Therefore additional studies using longer courses of inhaled NO therapy are needed to assess its therapeutic efficacy in tuberculosis.

CONCLUSIONS

In summary, NO has multiple immunoregulatory and antimicrobial functions that are likely to be of relevance to inhaled NO therapy. The precise effects of inhaled NO on host defense are likely to be concentration, redox environment, and pathogen dependent. The infections in which inhaled NO is most likely to have therapeutic benefit are diseases such as pulmonary tuberculosis, in which (1) inhaled NO is known to reach the target organ (i.e., the lung or the intravascular compartment of distal organs), (2) NO kills or inhibits the growth of the pathogen, and (3) NO does not contribute to disease pathogenesis. Future studies analyzing inhaled NO treatment in a variety of other infectious diseases are warranted to determine whether inhaled NO is an efficacious antimicrobial agent as well as a pulmonary vasodilator.