Nitric oxide (NO) is a low molecular weight and highly reactive gas. In the atmosphere, it is a component of air pollution, but is also a free radical released by a variety of tissues by nitric oxide synthases (NOS). Since endothelium-derived relaxing factor (EDRF) and NO have been proved to have identical biological activity, stability and susceptibility to specifi c inhibitors, NO has been viewed with increasing interest [8–11].
Synthesis and release of nitric oxide
NO is enzymatically produced by NOS, then oxidised to nitrite and nitrate by several mechanisms including macrophage activation [12,13]. Mitchell et al. fi rst suggested that mammals produced oxides of nitrogen in 1916 and Furchgott et al. identifi ed EDRF in vessels with an intact endothelium in 1982 [14–18]. In 1988, Palmer and others suggested that L-arginine is the precursor for NO synthesis in vascular endothelial cells and confi rmed that NO is the intermediary of the L-arginine to nitrite and nitrate pathway [13,19].
The NOS enzyme includes three distinct isoforms representing three distinct gene products, including neuronal NOS (nNOS, NOS1), inducible NOS (iNOS, NOS2) and endothelial NOS (ecNOS, NOS3). The three isoforms vary according to their different chromosomal locations, amino acid sequences and functions. nNOS and ecNOS are termed constitutive NOS (cNOS) which generate small amounts of NO depending on calcium/calmodulin binding . All three isoforms are found in normal lung tissue, while iNOS is up-regulated in diseases of the airway and at the alveolar level. mRNA of the ecNOS and iNOS genes are expressed in human airway epithelial cells, and have been proved to be correlated with biochemical activity [21–23]. While it is clear that NO plays an important role in the normal and the diseased lung, it is becoming apparent that NOS genotypic differences are also associated with the development of lung diseases.
Polymorphisms of nitric oxide synthases in relation to the lung
Neuronal NOS in the lung
The first NOS was purifi ed from rat cerebellum in 1990 by Bredt and Snyder and was identifi ed as a calmodulin-requiring enzyme . This was later designated as nNOS and found to be constitutively expressed in human alveolar and bronchial epithelial cells in 1994 .
The gene for nNOS in humans is located on chromosome 12q24 and polymorphisms have been linked to the diagnosis of asthma in family studies. Allelic frequencies of a polymorphism in exon 29 of the nNOS gene were reported by Grasemann et al. to be signifi cantly different between asthmatics and controls. The C®T transition which is located 276 base pairs downstream of the translation termination site has been proved to be lower in asthmatic patients than that in normal controls, which suggests that variants of the nNOS gene may contribute to asthma . Grasemann et al. also reported high AAT repeat numbers in an intronic repeat polymorphism in the nNOS gene in cystic fi brosis (CF) patients in 2000 . This was associated with a decrease of nNOS expression in airway epithelial cells in those cystic fi brosis patients who were more likely than others to have low airway NO concentrations . In addition, dinucleotide GT repeats in the proximal region of the nNOS gene, associated with progression of lung disease in patients with CF were found by Texereau et al. in 2004. These polygenic associations may be important for understanding the phenotypic disparities of patients with the same cystic fi brosis transmembrane conductance regulator (CFTR) mutations, and implicate NOS in this lung disease .
In addition to cystic fibrosis, nNOS is thought to play a role in other diseases. Airway smooth muscle hyperplasia and hypertrophy are considered to contribute to airway infl ammatory diseases. Thus, enhanced airway smooth muscle cell proliferation is thought to be associated with the chronic stages of asthma and chronic obstructive pulmonary disease (COPD). Patel et al. demonstrated increased nNOS expression in airway smooth muscle cells, which is thought to inhibit these proliferative responses .
Inducible NOS in the lung
iNOS is generally not expressed unless the cells have been induced by certain cytokines, microbes, microbial products or infl ammatory cytokines in contrast to cNOS, which suggests that NO produced by expression of iNOS acts as a major effector molecule in host defence mechanisms [15,27]. Nearly every tissue, however, in the body has the ability to express iNOS when stimulated, and it produces sustained high concentrations (μM) of NO [30–32]. Koide et al.  reported that a high level of intracellular cAMP, especially if combined with infl ammatory cytokines, increases nitric oxide production via iNOS mRNA expression in vascular smooth muscle cells, suggesting that iNOS is involved in infl ammatory reactions. Likewise, Robbins et al. have shown that TNF and IL-1 up-regulate iNOS expression in airway epithelium [34,35].
Guo et al. demonstrated that NO synthesis is due in part to the continuous expression of the iNOS isoform in airway epithelial cells of normal subjects . Abundant iNOS mRNA expression was reported in normal airway epithelial cells, while it is not detected in other resting pulmonary cells, indicating that airway epithelial cells are unique in their continuous pattern of iNOS expression. In situ analysis reveals all airway epithelial cell types express iNOS. Expression is, however, strikingly decreased by inhaled glucocorticosteroids and b-adrenergic agonists, which are commonly used medications in the treatment of infl ammatory airway diseases .
The gene encoding iNOS is located at chromosome 17q11.2–q12 and Konno et al. reported that the (CCTTT)n repeat polymorphism in the promoter of the iNOS gene is inversely associated with atopy. This polymorphism affects promoter activity and is a risk factor for the development of atopy, which seems to be an independent risk factor for asthma . The (CCTTT)n polymorphism has been proved to be associated with increased NO production in healthy children, while no data were reported on asthmatic patients . Kharitonov et al. and Hansel et al. have demonstrated that a selective inhibitor of iNOS has a rapid onset and longterm suppression of exhaled NO levels asthmatic patients. This suggests that inhibition of iNOS may have some therapeutic potential for asthma, because selective iNOS inhibitors suppress eosinophil infi ltration to the lung, decrease lung chemokine expression and inhibit allergic airway infl ammation in murine models of asthma[7,39–41]. While decreasing iNOS activity may be advantageous in some situations, a defi ciency of nitric oxide production has been found in the bronchial epithelium of some CF patients, due to a reduction in iNOS expression whether they had received glucocorticosteroids or not. Such a reduction may play a role in susceptibility to pulmonary fungal, viral and bacterial infections [27,42,43].
Endothelial NOS in the lung
ecNOS and nNOS are calcium/calmodulin-dependent enzymes, which produce low concentrations of NO (nM), but which are critical for the maintenance of endothelial function, to inhibit the adhesion of platelets and to suppress the replication of smooth-muscle cells [44,45]. The human ec- NOS gene is located at 7q35–36, and in 1994, Nadaud et al. cloned the human endothelial NO synthase gene and determined its structure, being composed of 26 exons [46,48]. ecNOS polymorphisms are associated with many different diseases, principally relating to the cardiovascular system, which has been extensively studied in this regard. ecNOS polymorphisms comprise three recognised functional allelic variants including the ecNOS4a in intron 4, T-786C, and G5557T in exon 7.
This distribution of the ecNOS intron 4 polymorphism varies greatly in the normal population among different ethic groups [49,50]. The polymorphism in intron 4, known as ecNOS4a, has four tandem 27-bp repeats, while fi ve tandem 27-bp repeats is regarded as the common, wild-type allele ecNOS 4b . Genotypic frequencies of ecNOS 4b/b, ecNOS 4b/a and ecNOS 4a/a in an Australian population were 66.7%, 32.7% and 0.7%, respectively, as reported by Wang et al. in 1996, although the frequency of ecNOS 4b/b is about 80% in Japanese and 75% in normal South Koreans [49,52–54]. The 27-bp repeat polymorphism in intron 4 was identifi ed as being associated with coronary artery disease (CAD), and myocardial infarction (MI), smoking-associated coronary heart disease and hypertension [49,55–58]. In a small study that we conducted in normal Australian subjects, 3 of total 16 were a/b (18.75%), while 13 people were type b/b (81.75%), and no a/a were detected, probably related to the small sample size (unpublished data).
Homozygous ecNOS 4a/a subjects who have the smaller allele of four repeats, had a signifi cantly lower exhaled NO level than either heterozygous ecNOS 4a/b subjects or wildtype ecNOS 4b/b subjects. Also, ecNOS 4a/a patients were more likely to have more diseased coronary vessels, which suggests an association between a reduced ability of the ecNOS4a allele to generate NO and an increased risk of CAD [49,55]. No evidence of such as association, however, was found with CAD in a hospital-based Taiwanese population . The distribution of genotype ecNOS 4b/b was signifi cantly higher in subjects with asthma than normal controls in a South Korean population, but it did not segregate with asthmatic severity . The ecNOS 4b/b genotype was reported also to be more frequent in patients with lung cancer compared to a control group . Other polymorphisms of ecNOS have been identifi ed, one being the T-786C substitution in the promoter region and another the G5557T in exon 7, also known as exon 7 Glu298Asp.
In a disease involving vascular dysfunction, both the Glu298Asp variant and the ecNOS 4a allele were reported to be signifi cantly associated with high-altitude pulmonary edema, which may underlie impaired NO synthesis in the pulmonary circulation .
Exhaled nitric oxide (eNO)
Immunostaining has shown nNOS expression in the nerves and epithelial cells of normal human airways and iNOS and ecNOS expression in airway epithelial cells [23,61]. These sources of NO are thought to be responsible for the NO detected in the breath, although the principal contributors are thought to be airway epithethial cells and perhaps pulmonary macrophages. The measurement of exhaled nitric oxide is highly reproducible in both healthy and asthmatic adults and children . Exhaled NO is increased in asthmatic patients and decreases after glucocorticosteroid therapy [63,64]. Patients with severe stable COPD have reduced levels of exhaled NO compared with normal subjects. Exhaled NO levels are only to a minor extent related to the severity of airfl ow obstruction in this disease . A number of factors may affect exhaled NO levels, including the technique used for sample collection as well as some factors within the subjects. There are two methods of collection. On-line sampling measurement requires the subject to exhale directly to the NO analyser, while off-line sampling uses a gas impermeable bag to collect the exhaled breath for later analysis. This latter method allows exhaled NO measurements in large populations, especially in remote area [66,67].
It has been demonstrated that fl ow rate is a major factor affecting the level of exhaled NO. Deykin et al. have reported that the values of exhaled NO decreased with increasing expiratory fl ow rate within the range of 50 to 500 ml/s by the off-line method and 42 to 250 ml/s by the on-line measurement [66,68]. Different inspiratory fl ow rates did not appear to infl uence exhaled NO levels . Nasal NO contamination is also a common factor confounding exhaled NO levels. NO levels in the nasal cavity have been proved to be higher than other parts of respiratory tract [70,71]. Ensuring that the subject keeps a positive intra- oral pressure during exhalation has been reported as a way of avoiding nasal NO contamination due to vellum closure .
Caffeine decreases exhaled NO levels in normal subjects, although a study of asthmatic patients showed no effect [72,73]. Alcohol also decreases exhaled NO but only in asthmatic subjects and not in normal individuals which suggests that the effect is via iNOS . Cigarette smoke is another important factor affecting exhaled nitric oxide levels. Both active and passive smoking is able to reduce exhaled NO signifi cantly in normal subjects [75,76]. Diet probably also has an effect on NO production as either inhaled or ingested L-arginine, the substrate for NOS, increased the exhaled NO level in both normal and asthmatic subjects [77,78].