Eur Respir J 1993, 6, 258-266 REVIEW

L-arginine-dependent nitric oxide synthase: a new metabolic pathway in the lung and airways

P.G Jorens, P.A Vermeire, A.G. Herman

£-arginine-dependent nitric oxide synthase: a new metabolic pathway in the lung Depts of Respiratory Medicine and and airways. P.G. Jorens, P.A. Vermeire, A.G. Herman. Pharmacology, UIA, University of Ant­ ABSTRACT: An L-arginine-dependent pathway, metabolising L-arginine to cit­ werp, Belgium. rulline and nitrogen oxides, has been described in many cell types in different Correspondence: P.G. Jorens species, including man. Two subtypes of this nitric oxide synthase have been Dept of Respiratory Medicine reported: a constitutive enzyme type, releasing nitric oxide after stimulation, UIA, University of Antwerp is typically found in endothelial and neural cells; another subtype can be in­ Universiteitsplein I duced in macrophages after cytokine treatment. B-2610 Wilrijk (Antwerp) This review summarizes the literature on the known and proposed roles of Belgium this L-arginine-dependent nitric oxide production in different pulmonary pro­ cesses. Keywords: (L)-arginine Nitric oxide has been reported to act as a neurotransmitter in the inhibitory asthma nonadrenergic, noncholinergic nerves in the airways of guinea-pig and man. It nitric oxide synthase is released in cytostatic processes by immune-stimulated alveolar macrophages. nitrite macrophage Recent data on the role of L-arginine-dependent processes in immune-complex­ guanylate cyclase mediated lung injury, histamine-induced activation of guanylate cyclase or cytokine networks in the lung are also discussed. Finally, similarities and dif­ Received: August 5 1992 ferences between tracheal epithelium-derived relaxing factor and nitric oxide Accepted: October 15 1992 are analysed. The details of the role and distribution of nitric oxide synthase in the (hu­ man) lung and airways are not yet completely understood. Nitric oxide is be­ lieved to play a role in various pulmonary physiological processes, such as bronchodilation and the cytotoxic action of certain cells. The modulation of nitric oxide release will therefore, most probably lead to application of novel therapies in diseases such as asthma and innammatory pulmonary diseases. Eur Respir J., 1993, 6, 258-266.

In 1987, a new metabolic pathway was described tissue. The other subtype is an inducible, both in the vascular endothelium [1], and in activated calcium-independent enzyme, which is found follow­ murine macrophages [2, 3]. The products of this ing induction with cytokines, in macrophages, vascu­ enzymatic pathway are L-citrulline and the highly re­ lar smooth muscle cells, cardiac myocytes, endothelial active and unstable nitric oxide (NO), which decom­ cells, and fibroblasts, as well as certain epithelial and -), poses into other nitrogen oxides, such as nitrite (N02 adenocarcinoma cells [4, 5]. -) nitrate (N01 and the potent oxidising agent pero­ Nitric oxide is a highly reactive and unstable sub­ xynitrite (ONOO) (fig. 1). The enzyme responsible stance with a very short half-life. Except for on-line for the production of these nitrogen oxides has been chemiluminescence, it is mainly measured by indirect named nitric oxide synthase [4]. From the first mo­ techniques (fig. I). Since NO accounts for the ments of its discovery, it was seen to be L-arginine­ biological activity of endothelium-derived relaxing fac­ dependent. Indeed, NO is synthetised from the tor (EDRF), and also inhibits platelet aggregation, both terminal guanidino nitrogen atom of the L-arginine vascular tissue and platelets can be used as indicators (and not the D-arginine) molecule [4]. for the NO production by other cells. Haemoglobin Soon it became clear that this metabolic pathway and other haeme proteins, which absorb nitric oxide exists in various cells from different embryological to form paramagnetic detectable nitroso-haeme prod­ origin, and that two forms of NO synthase can be dis­ ucts, and superoxide dismutase, which prolongs tinguished, based on their mode of NO release [5]. NO action by preventing its breakdown to other A first subtype is a constitutive, cytosolic calcium­ nitrogen oxides, are then used to further modulate dependent enzyme, releasing nitric oxide after receptor pharmacological effects. Another indirect method is or physical stimulation: this has been found in the spectrophotometric determination of nitrite and ni­ the vascular endothelium, platelets, neutrophils neural trate, two nitrogen oxides rapidly formed during NO PULMONARY NITRIC OXIDE SYNTHASE 259

oxidation. Considerable evidence has emerged sug­ The role of nitric oxide in the pulmonary vascular bed, gesting that the effects of NO in some physiological i.e. the impairment of its release in chronic obstruc­ processes are mediated through the activation of tive lung disease [8], and its possible therapeutic use guanylate cyclase, resulting in an increased level of in pulmonary hypertension [9], has recently been re­ cyclic guanylate monophosphate (cGMP) in the target viewed in The European Respiratory Journal [10]; it cell [4, 5]. Therefore, cGMP accumulation in cultured will, not therefore, be considered in this review. rat lung fibroblasts is frequently used as a model to study the ability of cells to produce NO [6, 7]. Fi­ nally, a class of L-arginine analogues is known to A role for nitric oxide in nonadrenergic, inhibit nitric oxide synthase in either a competitive or noncholinergic neural control in the airways irreversible way, rendering them useful for exploring the role of L-arginine dependent nitric oxide produc­ In addition to the classical cholinergic bronchocon­ tion. strictor and adrenergic bronchodilator neural mecha­ Recent observations suggest that this metabolic path­ nisms, evidence supports the existence, in the airways way is present in different pulmonary cell-types, and of different animal species and man, of neural path­ that it may play a role in various regulatory mechanis­ ways which are neither adrenergic nor cholinergic, the ms in airway and lung tissue. This review will focus so-called nonadrenergic, noncholinergic (NANC) on the known and proposed mechanisms of the regula­ nerves [11, 12]. NANC stimulation may either induce tion and role of nitric oxide production in the lung and contraction (excitatory NANC), or relaxation (inhibi­ airways, in humans, as well as in other animal species. tory NANC) [1 2]. While excitatory NANC responses direct stimulation induction by cytokines Generator cell Inhibitors glucocorticoids DODO NO Synthase Type I NO Synthase·-----­ Type 11 L-arginine L-NMMA /•------..... : =: =: : =: : : L-NA L-NAME L-citrulline NO

NO • _ ~ _ ___ Haemoglobin Superoxide

/

Target cell Methylene blue guanylate cyclase ....,.._

cGMP• cGMP phosphodiesterase•

Fig. 1. - Schematic diagram of the proposed L-arginine-dependent nitric oxide synthase pathway. NO is formed in the generator cell from L-arginine, either after direct stimulation (NO synthase type 1), or after induction with cytokines (NO synthase type Jl). After dif­ fusion out of the cell, NO can react with oxygen radicals and is then vulnerable to degradation to nitrogen oxides. NO can stimulate soluble guanylate cyclase activity in certain target cells or inhibit deoxyribonucleic acid (DNA) synthesis and aconitase activity (citric acid cycle) in tumour cells. The induction of NO synthase type ll is suppressed by glucocorticoids. Both subtypes of the enzyme NO synthase are suppressed by certain analogues of L-arginine (such as L-NMMA, L-NA, L-NAME). The pathway is also inhibited by hae­ moglobin and superoxide anion, which inactivate NO, and by methylene blue, a guanylate cyclase inhibitor. L-NMMA: L-NG monomethyl arginine; L-NA: L-NG nitro-arginine; L-NAME: L-NG nitro-arginine methylester; ONOO·: peroxynitrite. 260 P.G. JORENS, P.A. VERMEIRE, A.G. HERMAN

are believed to be mediated by the release of differ­ responses [21]. BAI et al. [22] found that nitric oxide ent neuropeptides, including substance P, neurokinin A mediates a component of the inhibitory response after and calcitonin gene-related peptide, the inhibitory stimulation of inhibitory NANC neurons in bronchial NANC responses are believed to be mediated, at least smooth muscle, although this inhibition (by L-NAME) in part, by the release of vasoactive intestinal peptide was less than 50% of the NANC relaxation in (VIP) and related peptides, such as peptide histidine precontracted human bronchial tissues, implying that isoleucine/methionine. However, ELus and FARMER other mediators are involved in the relaxant response. [13] have shown that a-chymotrypsin, an enzyme These data indirectly suggest that NO is a (or the?) capable of degrading peptides, only reduces inhibitory neurotransmitter of bronchodilator nerves in human responses to NANC stimulation in the guinea-pig airways. airways by about 35%; raising the possibility that at Nitric oxide-containing vasodilators such as glyceryl­ least part of the response is mediated by a molecule trinitrate and isosorbide dinitrate act as nonspecific other than a protein. Therefore, it seemed reasonable smooth muscle relaxants, activating guanylate cyclase to assume that NO might play a role in inhibitory and raising cGMP levels [5]. Nitrates have been NANC neurotransmission in the airways, comparable reported to be effective as relaxants of bovine airway to its proposed role as a neurotransmitter of the NANC smooth muscle in vitro [23], and intravenously admin­ nerves in the anococcygeus muscle [14], and the gut istered glyceryl trinitrate was reported to relax the [ 15]. tracheal smooth muscle in nonasthmatic, anaesthetized In the airways of guinea-pigs, the peptidase­ humans undergoing cardiac surgery [24]. However, resistant component of the NANC inhibitory response reports about the clinical efficacy of sublingual is attenuated in a concentration-dependent manner by glyceryl trinitrate and isosorbide dinitrate in the treat­ the L-arginine analogue, L-N° nitro-arginine (L-NA), ment of asthma are conflicting [25-30]. but not by D-N° nitro-arginine, and can be reversed S-nitrosothiols are bioactive NO-adducts, which fmm by L-but not D-arginine. It is suggested that this ef­ readily, under physiological conditions, from the reac­ fect is presynaptic, because at a concentration produc­ tion of endogenously-derived nitrogen oxides with ing a 74% reduction in the electrically-induced NANC reduced thiols [31]. These compounds have half-lives relaxation, L-N° nitro-arginine is without effect on that are significantly greater than the unstable NO­ relaxations produced by submaximal concentrations of radical itself. They have been shown to relax both VIP [16]. precontracted guinea-pig tracheal rings [32], and Lr and RAND [17] demonstrated that L-N° mono­ human peripheral bronchi [33], an effect which is methyl arginine (L-NMMA) and L N° nitro-arginine mediated in part by activation of guanylate cyclase. methylester (L-NAME), two more competitive inhibi­ A study published by DuPUY et al. [34] demonstrated tors of NO synthase, also reduce relaxations of guinea­ that inhalation of low concentrations of NO rapidly pig tracheal smooth muscle elicited by electrical reversed methacholine-induced bronchoconstriction in stimulation of NANC nerves; with L-NAME being 10- anaesthetized and mechanically-ventilated guinea-pigs. 30 times more potent than L-NMMA. Recent data Inhaling S-nitroso-N-acetyl-penicillamine, an NO do­ also suggest that NO could mediate NANC relaxation nor molecule, also induced prompt and long-lasting of tracheal smooth muscle in other species, such as the bronchodilation. Nitric oxide and NO-donors have, pig [18]. Moreover, in guinea-pig airways, L-NAME therefore, potential for pharmacological application as significantly enhances cholinergic nemo-transmission in bronchodilators, and their utility in the treatment of the trachea. Therefore, NO could also be an endog­ airflow obstruction certainly deserves further consid­ enous modulator of the cholinergic neurotransmission eration. in guinea-pig airways which co-exists with NANC in­ All of these investigations indicate that NO, or a nervation in this species [19]. Thus, NO may be eo­ NO-like product, might play a role in NANC-induced released with acetylcholine, and act as a braking relaxations of tracheal smooth muscle in different mechanism for cholinergic reflex bronchoconstriction species. They give no answer yet to the quest­ [19]. ions whether NO: l) may mediate the smooth muscle Recently, these studies have been extended to relaxant activity of other putative NANC nemo­ human airways, where the prominent inhibitory NANC transmitters, such as VIP [35]; 2) may be eo-released is the only neural bronchodilator mechanism known. with transmitters such as VIP from the same neurons BELVISI et al. [20] initially reported that L-NMMA [36]; or 3) may even be released from other than neu­ only partially blocked these inhibitory NANC ronal pulmonary cells. responses in human trachea, but later reported that L-NAME inhibits the tetrodotoxin-sensitive (i.e. neu­ ral) dilator response to neural stimulation at all fre­ Immune-stimulated alveolar macrophages release quencies tested in tracheal rings from normal donors L-arginine derived nitric oxide for heart-lung transplantation [21]. L-arginine, but not D-arginine, reversed this inhibitory effect of L-NAME. KNOWLES et al. [37] found that a calcium-independ­ Moreover, a-chymotrypsin, which degrades VIP and ent NO synthase is induced in the lungs of rats treated related peptides, had no inhibitory effect in humans with lipopolysaccharide (LPS), but it was not clear in [21], contrasting with guinea-pig tracheal neural which cell type this NO synthase was induced. PULMONARY NITRIC OXIDE SYNTHASE 261

Recent work in our laboratory shows that pulmonary instillation of a single fibrogenic dose of bleomycin, rat macrophages, alveolar as well as pleural, can are able to express cytostatic activity for eo-cultured produce L-arginine derived nitrite in a dose- and neoplastic cells. This is linked to the release of sig­ time-dependent manner, after activation with endo­ nificant amounts of nitrites. Erythrocytes inhibit this toxin, rat recombinant interferon-y (IFN-y) and opson­ alveolar macrophage cytostatic activity, by preventing ized zymosan in vitro [38]. The L-arginine reactive nitrogen intermediates from reaching target analogue L-NMMA is effective in inhibiting this cells, because haemoglobin serves as a sink for reac­ production. tive nitrogen oxides [48] . Therefore, it seems that re­ Consistent with the fact that glucocorticoids are able active nitrogen oxides are important in the cytostatic to inhibit the induction of nitric oxide synthase in the mechanism of macrophages in certain in vivo models lung in vivo after LPS treatment [39], we were able of lung injury. to demonstrate that these anti-inflammatory agents Although fungistasis in mice peritoneal macrophages block the induction of nitrite in alveolar macrophages and tumour cell killing in mice and rat macrophages in vitro by all of the stimuli mentioned [38]. The in vitro requires L-arginine, human alveolar macro­ mechanism underlying this inhibition is suggested to phage-mediated fungistasis in vitro does not require L­ be receptor-mediated, since the addition of equimolar arginine, and human alveolar macrophages do not concentrations of cortexolone (the partial agonist of the produce amounts of nitrite and nitrate detectable by glucocorticoid receptor) [38] or RU 38 486 (a com­ colorimetry in eo-cultures with Cryptococcus neo-for­ plete antagonist of the receptor) [40] to the incubations mans [49]. This absence of L-arginine-dependent with dexamethasone partially reverse this inhibitory nitrogen oxides in human alveolar macrophages, (com­ effect. These data confirm reports on the ability of pared to mice peritoneal macro-phages), during con­ glucocorticoids to inhibit nitric oxide synthase in ditions under which fungistasis occurs suggests that murine macrophages [41], and suggest that part of the this phenomenon is species-specific, rather than spe­ anti-inflammatory effect and immunosuppressive cific to the tissue origin of the macrophage. Various effects of glucocorticoids are due to their inhibition of explanations are possible for this observed difference. the induction of NO-synthase. In the meantime, we The lack of demonstrable NO in human macrophages, have been able to extend these observations by show­ suggests that NO synthase may be absent in these ing that two constituents of food components, soybean cells, may respond to alternative stimuli, or may not trypsin inhibitor and ~-amylase, are capable of induc­ produce extracellular nitric oxide [50]. Reduced ing this L-arginine dependent nitrite production in biopterin (5,6,7 ,8 tetrahydrobiopterin) is required for rat alveolar macrophages in vitro in a dose- and full-activity of cytokine-induced NO synthase in the time-dependent manner [42] . Whether this in vitro ob­ murine peritoneal [51] and the rat alveolar [52] mac­ servation may explain, in part, the previously observed rophage, but human macrophages lack the enzymatic anti-carcinogenic effects of certain food products, system necessary to produce reduced biopterin [53]. by inducing the production of the cytotoxic mole­ This does not exclude the existence of this system in cule NO in phagocytic cells, remains to be estab­ alveolar macrophages in vivo, since reduced biopterin lished. or other eo-factors could be obtained in vivo from an Many authors [2, 43] have shown that mouse peri­ exogenous biological source. This may not completely toneal macrophages are cytotoxic to tumour cells and account for the difference between human and murine microorganisms in vitro, via this L-arginine-dependent macrophages, since human macrophages treated with pathway, by inactivating iron/sulphur centres in key interferon-y plus endotoxin, and reconstituted with enzymes associated with cellular respiration and deox-­ intracellularly-delivered biopterin, produce no nitrite or yribonucleic acid (DNA) synthesis. Few studies have nitrate [54]. been conducted to investigate whether this pathway Recently, SHERMAN et al. [55] reported that human plays a role in alveolar macrophages. This is of prac­ recombinant interferon-y increases the citrulline con­ tical importance, since alveolar macrophages are tent, but not the nitrite content, in the in vitro culture known to play an important role in pulmonary host media of human alveolar macrophages; they propose defence mechanism through the release of numerous that citrulline production is a main indicator of L­ secretory products, such as oxygen radicals, enzymes, arginine oxidation. In contrast to this cytokine, anti-proteases, cytokines and bioactive lipids [44]. Pneumocystis carinii when placed in eo-culture with A recent study by BARRATT et al. [45] reported that human alveolar macrophages does induce a small but simultaneous application of the immune-activating detectable nitrite production. [55]. This raises the pos­ agent, muramyl dipeptide, and endotoxin, to rat alveo­ sibility that NO may indeed play a role as a micro­ lar macrophages in vitro, induced cytostatic activity bicide, produced by human alveolar macrophages, against a rat fibrohistiocytoma line, mediated by the against certain opportunistic pathogens; a mechanism L-arginine NO-generating pathway. comparable to the one proposed for human blood Ozone inhalation increases nitric oxide release from monocytes in vitro inactivation of Mycobacterium rat alveolar macrophages harvested 48 h following in­ avium [56]. halation. HuOT and HACKER [47] were able to prove Chloromethyl ketone derivatives, a class of mol­ that alveolar macrophages, harvested by broncho­ ecules co-valently binding to the active site of serine alveolar lavage from rats exposed to intratracheal proteases, are able to block nitrite production of 262 P.G. JORENS, P.A. VERMEIRE, A.G. HERMAN

immune-stimulated rat alveolar macrophages in vitro high concentrations in th e bronchoalveolar lavage [57]. These inhibitors may, therefore, be useful in (BAL) fluid of animals undergoing immune complex further defining the role of nitric oxide synthase deposition, supports this hypothesis [63]. production in (alveolar) macrophages. Histamine seems to be responsible for the produc­ tion of cGMP both in human [64], and guinea-pig lung tissue [65]. LEURS et al. [65] showed that this H - 1 Cytokine networks and nitric oxide receptor-mediated cGMP response is potently inhibited in the guinea-pig lung by haemoglobin. Their experi­

The development of pulmonary inflammation ments suggest that after an initial H 1 receptor-mediated involves a series of cellular events controlled by a phospholipase C-dependent, calcium mobilization, the variety of mediators, including cytokines. During pul­ enzymatic conversion of L-arginine to NO is stimu­ monary inflammation, a number of these cytokines are lated. This NO production may then cause the in­ expressed and released into the lung and airways. creased cGMP production, extending data reported by Therefore, the interaction between these cytokines in MAYER and BoHME [66] that the (bovine) lung is ca­ cytokine networks may play an important role in the pable of converting L-arginine to an activator of modulation of inflammation in the local environment. guanylate cyclase. Since cGMP has been reported to Granulocyte macrophage-colony-stimulating factor modulate the immunologically induced release of (GM-CSF) and muramyl dipeptide, a constituent of the mediators from mast cells, through an effect on the bacterial cell wall, are able to enhance IFN-y-induced microtubular assembly [67], this might suggest a role nitrite production in rat alveolar macrophages in vitro, for NO-mediated guanylate cyclase activation in aller­ with GM-CSF serving as a priming factor [58]. The gic diseases. interleukins 1, 3 and 4 and tumour necrosis factor are Endogenous nitric oxide has been found in expired devoid of such activities in rat alveolar macrophages air samples from rabbit, guinea pig and human, uti­ [58]. lising different techniques including chemiluminescence Results from our laboratory indicate that not only rat [68]. Although the exact origin remains to be deter­ alveolar macrophages, but also rat lung fibroblasts, are mined, it is likely that NO is produced in the lung capable of producing nitrite upon stimulation with during different processes, such as vascular regulation IFN-y. This effect is markedly enhanced in fibroblasts and in host defence. Indeed, many cells apart from by the addition of endotoxin and interleukin-1~ [59]. endothelium and macrophages, known to play a role Interleukin-1 ~ can also serve as an efficient priming in lung defence and injury, have been described as us­ signal for this IFN-y induced nitrite production. The ing L-arginine for their cytotoxic action (lymphocytes nitric oxide synthase pathway, therefore, provides an [69]), or to release NO in vitro (neutrophils [70]; excellent tool for exploring the interaction between platelets [71]; monocytes [70], as well as mast cells cytokines in the lung. Several recent observations sug­ [72, 73]). gest that nitric oxide may play a role in pulmonary cell-to-cell interactions and critical defence mecha­ Epithelium-derived t·elaxing factor is nisms of the host: pyocyanine, a pigment produced by probably not nitric oxide Pseudomonas aeruginosa, can inhibit the release of reactive nitrogen intermediates from murine alveolar In the past, the airway epithelium was regarded as macrophages [60]; peroxynitrite inhibits rat alveolar mainly passive, thereby functioning as a barrier. It is type II cell respiration [61]; the immunosuppressive now known that the epithelium actively participates in effect of cultured rat alveolar macrophages on mi­ many airway functions, and that various agents can in­ togen-induced proliferative responses of splenic duce the release of an inhibitory substance from lymphocytes involves the L-arginine dependent nitric airway epithelium, the so-called epithelium-derived oxide pathway [62] . relaxing factor (EpDRF). This substance is capable of relaxing certain tissues, and regulating the tone of the Other L-arginine dependent processes in the lung underlying airway smooth muscle, therefore attenuat­ ing airway smooth muscle responsiveness [74]. Acute or chronic lung injury has been linked to Such a system may, therefore, function in a manner the presence of proteases and oxygen radicals released analogous to that described for the influence of the from activated phagocytic cells. Recent data suggest endothelium on vascular smooth muscle by releasing that the free radical NO may be added to the list EDRF or nitric oxide. Recent studies on the effect of cytotoxic molecules involved in lung injury. of removal of the airway epithelium on underlying Indeed, L-NAME protects rats against immune­ smooth muscle contraction strongly suggest the exist­ complex-induced lung and skin vascular injury [63], a ence of such an EpDRF, but its identity remains process which is believed to be polymorphonuclear­ unclear. In some studies, the production of guinea­ and oxygen-radical-dependent. This protective effect pig and human tracheal bronchial epithelium-derived is reversed by the presence of L- but not of D-ar­ relaxing factor has been demonstrated in the presence ginine. Additionally, in the absence of L-NAME, this of inhibition of cyclo- and lipoxygenase, ruling out a observed injury is enhanced by the presence of L- but prostaglandin as the relaxant responsible [74, 75]. A not D-arginine. That nitrite and nitrate are found in saturated solution of nitric oxide is able to cause PULMONARY NITRIC OXIDE SYNTHASE 263

concentration-dependent and complete relaxation of Secondly, although NO looks promising for measur­ nonvascular smooth muscle relaxation including bovine ing diffusion capacity in humans [81, 82], the main epithelium denuded airway strips [76]. Since it is limitation is the fact that the toxicity of both the speculated that the airway epithelium has a similar role nitric oxide radical and the L-arginine analogues have as endothelium in vascular tissue, nitric oxide is a not been investigated in humans or animals. When good candidate for being the epithelium-derived relax­ considering the investigation of the role of NO in the ing factor. human lung, one should, indeed, be aware that NO is However, epithelium-dependent relaxation of vascu­ one of the main constituents of the gas phase of ciga­ lar or gastrointestinal smooth muscle, used as a rette smoke at a concentration even >50 ppm [83-85], bioassay for EpDRF, is not affected by methylene blue and is recognized as a toxic air pollutant [86]. [77, 78], which inhibits the effects of EDRF on solu­ Nitric oxide is extremely cytotoxic, which explains its ble guanylate cyclase. EpDRF must, therefore, be an toxicity to malignant cells. In addition, nitric oxide agent that produces endothelium-independent relaxation is known to cause nitrosative deamination of in these systems. Moreover, L-NMMA does not DNA, resulting in genomic alterations in bacteria inhibit osmotic-induced epithelium-dependent relaxation which are comparable to those alterations that have in guinea-pig airways [79]. These results suggest that been linked, in humans, to a variety of disorders re­ NO is probably not responsible for causing epithelium­ sulting from base deamination [87]. Nitric oxide and derived relaxation of airway smooth muscle, although the superoxide anion can react to form peroxynitrite, a EpDRF and vascular EDRF have some pharmacologi­ a powerful oxidant [88] and inacti vator of 1-protein­ cal similarities. More work is required to identify the ase inhibitor [89], the most abundant and effec­ precise characteristics of EpDRF. tive extracellular antiprotease in the lower respiratory tract. Discussion It is likely that multiple functions of the L­ arginine-dependent nitric oxide production exist in It is clear that full exploration of the role of the lung, since multiple cell types of the lung, in­ nitric oxide suffers various limitations. Firstly, the volved in different pathological processes, are capable measurement and detection of nitric oxide, due to of producing L-arginine-dependent nitric oxide (table its instability, is based mainly on indirect detection 1). Although details about the role and distribution techniques. However, the recent cloning of the of nitric oxide synthase in the (human) lung are not nitric oxide synthase enzyme will make it possible yet known, the wide array of stimuli that are in­ to look for its distribution and expression [80]. volved in both the immediate release or the induction

Table 1. - Evidence for stimulus-dependent generation of NO by pulmonary cells Cell type Animal species Stimulus First Author Year [Ref.] NO synthase type I (direct stimulation) NANC neurons Guinea-pigs Electrical stimulation L1 1991 [17] NANC neurons Guinea-pigs Electrical stimulation TUCKER 1990 [16] NANC neurons Pig Electrical stimulation KANNAN 1992 [18] NANC neurons Human Electrical stimulation BELVISI 1992 [21] NANC neurons Human Electrical stimulation BAI 1992 [22] NO synthase type 11 (induction by cytokines) Pulmonary cells, cultured in l'itro Alveolar macrophages Rat Endotoxin, interferon-y, opsonized zymosan }OR ENS 1991 [38] Alveolar macrophages Rat Endotoxin BARRATT 1991 [45] Alveolar macrophages Rat Soybean trypsin inhibitor ~-amylase JORENS 1992 [42] Alveolar macrophages Human Co-culture with Pneumocystis carinii SHERMAN 1991 [55] Alveolar macrophages Rat Co-culture with splenic lymphocytes KAWABE 1992 [62] Alveolar macrophages Mice Interferon-y SHELLJTO 1992 [60] Pleural macrophages Rat Endotoxin, interferon-y JORENS 1991 [38] Lung fibroblasts Rat Interferon-y }OR ENS 1992 [59] In vivo models of lung injury Whole lung Rat After intravenous endotoxin KNO\VLES 1990 [37] Alveolar macrophages Rat After intratracheal bleomycin HUOT 1990 [47} Alveolar space, cell type? Rat Immune complex lung injury MULLIGAN 1991 [63] Alveolar macrophages Rat Following ozone inhalation PENDING 1992 [46]

NANC: nonadrenergic, noncholinergic. 264 P.G. JORENS, P.A. VERMEIRE, A.G. HERMAN

of nitric oxide production indicate that this pathway 17. Li CG, Rand MJ. - Evidence that part of the NANC may play an important role in the pathogenesis of dif­ relaxant response of guinea-pig trachea to electrical field ferent pulmonary disorders. stimulation is mediated by nitric oxide. Br J Pharmacal 1991; 102: 91-94. 18. Kannan MS, Johnson DE. - Nitric oxide mediates the neural nonadrenergic, noncholinergic relaxation of pig tra­ References cheal smooth muscle. Am J Physiol 1992; 262: L511- L514. 1. Palmer RMJ, Ferrige AG, Moncada S. - Nitric ox­ 19. Belvisi MG, Stretton D, Barnes PJ. - Nitric oxide ide accounts for the biological activity of endothelium­ as an endogenous modulator of cholinergic neuro­ derived relaxing factor. Nature 1987; 327: 524-526. transmission in guinea-pig airways. Eur J Pharmacal 1991; 2. Hibbs JB, Taintor RR, Vavrin Z. - Macrophage cy­ 198: 219-221. totoxicity: role for L-arginine deaiminase and imino nitro­ 20. Belvisi MG, Stretton D, Verleden GM, Yacoub MH, gen oxidation to nitrite. Science 1987; 235: 473-476. Barnes PJ. - Inhibitory NANC nerves in human tracheal 3. Stuehr DJ, Marletta MA. - Induction of nitrite/nitrate smooth muscle: involvement of VIP and NO. Am Rev synthesis in murine macrophages by BCG infection, Respir Dis 1991; 143: A355. lymphokines or interferon-y. J lmmunol 1987; 139: 518- 21. Belvisi MG, Stretton D, Yacoub M, Barnes PJ. 525. Nitric oxide is the endogenous neurotransmitter of bron­ 4. Moncada S, Palmer RMJ, Higgs EA. - Biosynthesis chodilator nerves in humans. Eur J Pharmacal 1992· 210· 221-222. , . of nitric oxide from L-arginine. A pathway for the regula­ tion of cell function and communication. Biochem 22. Bai TR, Bramley AM, Schellenberg RR, Pare PD. - Pharmacal 1989; 11: 1709-1715. Nitric oxide mediates nonadrenergic, noncholinergic inhibi­ 5. Moncada S, Palmer RMJ, Higgs EA. - Nitric oxide: tory neural responses in human bronchial smooth muscle. physiology, pathophysiology and pharmacology. Pharmacal Am Rev Respir Dis 1992; 145: A383. Rev 1991; 443: 109-142. 23. Gruetter CA, Childers CE, Bosserman MK, et al. 6. Leitman DC, Agnost VL, Tuan JJ, Andresen JW, Comparison of relaxation induced by glyceryl trinitrate, Murad F. - Atrial natriuretic factor and sodium nitroprus­ isosorbide dinitrate and sodium nitroprusside in bovine air­ side increase cyclic GMP in cultured rat lung fibroblasts by ways. Am Rev Respir Dis 1989; 139: 1192-1197. activating different forms of guanylate cyclase. Biochem J 24. Byrick RJ, Hobbs EG, Martineau R, Noble WH. 1987; 244: 69-74. Nitroglycerin relaxes large airways. Anesth Analg 1983; 62: 7. Schroder H, Leitman DC, Bennett BM, Waldman SA, 421-425. Murad F. - Glyceryl trinitrate-induced desensitization of 25. Hirschleifer L, Arora Y. - Nitrates in the treatment guanylate cyclase in cultured rat lung fibroblasts. J of bronchial asthma. Dis Chest 1961; 39: 275-283. Pharmacal Exp Ther 1988; 245: 413-418. 26. Miller WC, Schultz TF. - Failure of nitroglycerin as 8. Dinh-Xuan AT, Higenbottam TW, Clelland CA, et al. a bronchodilator. Am Rev Respir Dis 1979; 120: 4 71. - Impairment of endothelium-dependent pulmonary artery 27. Kennedy T, Summer WR, Sylvester J, Robertson D. - relaxation in chronic obstructive lung disease. N Engl J Airway response to sublingual nitroglycerin in acute asthma. Med 1991; 324: 1539-1547. JAmMed Assoc 1981; 246: 145-147. 9. Pepke-Zaba J, Higenbottam TW, Dinh-Xuan AT, Stone 28. Tullett WM, Pate! KR. Isosorbide dinitrate and D, Wallwork J. Inhaled nitric oxide as a cause of isoxsuprine in exercise-induced asthma. Br Med J 1983; selective pulmonary vasodilatation in pulmonary hyperten­ 286: 1934-1935. sion. Lancet 1991; 338: 1173-1174. 29. Goldstein JA. Nitroglycerin therapy of asthma. 10. Dinh-Xuan AT. - Endothelial modulation of pulmo­ Chest 1984; 85: 449. nary vascular tone. Eur Respir J 1992; 5: 757-762. 30. Okayama M, Sasaki H, Takishima T. - Bronchodila­ 11. Barnes PJ. - New concepts in the pathogenesis of tor effect of sublingual isosorbide dinitrate in asthma. Eur bronchial hyperresponsiveness and asthma. J Allergy Clin J Clin Pharmacal 1989; 26: 151-155. Immunol 1989; 83: 1013-1026. 31. Oae S, Shinhama K. Organic thionitrites and 12. Stretton D. - Nonadrenergic, noncholinergic neural related substances. A review. Org Prep Proc Int 1983; control of the airways. Clin Exp Pharmacal Pathol 1991; 15: 165-198. 18: 675-684. 32. Jansen A, Drazen J, Osborne JA, et al. - The relax­ 13. Ellis JL, Farmer SG. Effects of peptidases on ant properties in guinea-pig airways of S-nitrosothiols. J rionadrenergic, noncholinergic inhibitory responses of Pharmacal Exp Ther 1992; 261: 154-160. tracheal smooth muscle, a comparison with effects on VIP­ 33. Gaston B, Drazen JM, Jansen A, et al. - Relaxation and PHI-induced relaxation. Br J Pharmacal 1989; 96: of human airways in vitro by S-nitrosothiols. Am Rev 521-526. Respir Dis 1992; 145: A383. 14. Gillespie JS, Liu X, Martin W. The effects of 34. Dupuy PM, Shore SA, Drazen JM, et al. Bron­ L-arginine and N° monomethyl L-arginine on the response chodilator action of inhaled nitric oxide in guinea-pigs. J of the rat anococcygeus muscle to NANC nerve stimulation. Clin Invest 1992; 90: 421-428. Br J Pharmacal 1989; 98: 1080-1082. 35. Lilly CM, Stamler JS, Gaston BM, Loscalzo J, 15. Bult H, Boeckxstaens GE, Pelckmans PA, et al. - Ni­ Drazen JM. - NO-synthase modulates vasoactive intesti­ tric oxide as an inhibitory nonadrenergic noncholinergic neu­ na~ pep~ide (VIP) pulmonary relaxation in superfused rotransmitter. Nature 1990; 345: 346-347. gumea-p1g lungs. Am Rev Respir Dis 1992; 145: A382. 36. Sharaf HH, Said SI. VIP relaxation of airway, 16. Tucker JF, Brave SR, Charamlambous L, Hobbs AJ, pulmonary artery and aortic vascular smooth muscle is Gibson A. L-N°-nitro-arginine inhibits non­ unaffected by blockade of nitric oxide synthase. Am Rev adrenergic, noncholinergic relaxations of guinea-pig isolated Respir Dis 1992; 145: A382. tracheal smooth muscle. Br J Pharmacal 1990; 100: 663- 37. Knowles RG, Merrett M, Salter M, Moncada S. 664. Differential induction of brain, lung and liver nitric PULMONARY NITRIC OXIDE SYNTHASE 265

oxide synthase by endotoxin in the rat. Biochem J 54. Werner-Felmayer G, Werner ER, Fuchs D, et al. 1990; 270: 833-836. On multiple forms of NO synthase and their occurrence in 38. Jorens PG, Van Overveld FJ, Bult H, Vermeire PA, human cells. Res lmmunol 1991; 142: 555-561. Herman AG. L-arginine dependent production of 55. Sherman MP, Loro ML, Wong VZ, Tashkin DP. nitrogen ox ides by rat pulmonary macrophages. Eur J Cytokine- and Pneumocystis carinii induced L-arginine Pharmacal 1991; 200: 205-209. oxidation by murine and human pulmonary alveolar 39. Knowles RG, Salter M, Brooks SL, Moncada S. macrophages. J Protozoal 1991; 38: 2349-2369. Anti-inflammatory glucucorticoids inhibit the induction of 56. Denis M. - Tumor necrosis factor and granulocyte nitric oxide synthase in the lung, liver and aorta of the rat. macrophage-colony stimulating factor stimulate human Biochem Biophys Res Commtm 1990; 172: 1042-1048. macrophages to restrict growth of virulent M. avium. Kill­ 40. Jorens PG , Van Overveld FJ, Bult H, Vermeire PA, ing effector mechanism depends on the generation of reac­ Hennan AG. - Corticosteroids prevent the induction of tive nitrogen intermediates. J Leukoc Bioi 1991; 49: nitric "oxide synthase activity in different pulmonary cells. 380-387. In: De Deyn PP, Marescau B, Stalon V, Qureshi I, eds. 57. Jorens PG, Van Overveld FJ, Bult H, Vermeire PA, Guanidino Compounds in Biology and Medicine London, Herman AG. - Serine protease inhibitors modulate nitric Great Britain, John Libbey Publishers, 1992; 97-101. oxide synthase activity in rat alveolar macrophages. Agents 41. Di Rosa M, Radomski M, Carnuccio R, Moncada S. Actions 1992; 36: 243-247. - Glucocorticoids inhibit the induction of nitric oxide syn­ 58. Jorens PG, Van Overveld FJ, Bull H, Vermeire PA, thase in macrophages. Biochem Biophys Res Commun 1990; Herman AG. - Muramyl dipeptide and GM-CSF enhance 172: 1246-1252. interferon-y-induced nitric oxide production by rat alveolar 42. Jorens PG, Van Overveld FJ, Bult H, Vermeire PA, macrophages. Agents Actions (in press). Herman AG. Soybean trypsin inhibitor and beta­ 59. Jorens PG , Van Overveld FJ, Vermeire PA, Bult H, amylase induce alveolar macrophages to release nitrogen ox­ Herman AG. - Synergism between interleukin- 1~ and the ides. Biochem Plwrmacol 1992; 44: 387-390. nitric oxide synthase inducer interferon-y in rat lung 43. Stuehr DJ, Nathan CF. - Nitric oxide. A macrophage fibroblasts. Eur J Pharmaco/ 1992; 224: 7-12. product responsible for cytostasis and respiratory inhibition 60. Shellito J, Nelson, Sorensen RU. Pyocyanine, a in tumor target cells. J Exp Med 1989; 169: 1543-1555. product of Pseudomonas aeruginosa, inhibits the re­ 44. Sibille Y, Reynolds HJ. - Macrophages and polymor­ lease of reactive nitrogen intermediates from murine phonuclear neutrophils in lung defense and injury. Am Rev alveolar macrophages. Am Rev Respir Dis 1992; 145: Respir Dis 1990; 141 : 471-501. A551. 45. Barratt GM, Raddassi K, Petit JF, Tenu JP. - MDP 61. Hu P, Zhu L, lschiropoulos H, Matalon S . and LPS act synergistically to induce arginine-dependent Peroxynitrite inhibition of oxygen consumption and ion cytostatic activity in rat alveolar macrophages. lnt J transport in alveolar type II (AT JI) pneumocytes. Am Rev lmmunopharmaco/ 1991; 13: 159-165. Respir Dis 1992; 145: A568. 46. Pendino K, Punjabi C, Lavnikova N, et al. - Inhala­ 62. Kawabe T , Isobe KI, Hasegawa Y, Nakashima I, tion of ozone stimulates nitric oxide production by pulmo­ Shikomota K. - Immunosuppressive activity-induced by nary alveolar and interstitial macrophages. Am Rev Respir nitric oxide in culture supernatants of activated rat alveolar Dis 1992; 145: A650. macrophages. Immunology 1992; 76: 72-78. 47. Huot EA, Hacker MP. - Role of reactive nitrogen 63 . Mulligan MS, Hevel JM, Marletta MA, Ward PA. - intermediate production in alveolar macrophage-mediated Tissue injury caused by deposition of immune complexes cytostatic activity induced by bleomycin lung damage in is L-arginine-dependent. Proc Natl A cad Sci USA 1991; 88: rats. Cancer Res 1990; 50: 7863-7866. 6338- 6342. 48. Huot EA, Kruszyna H, Kruszyna R, Smith RP, Hacker 64. Platshon LF, Kaliner M. - The effect of immuno­ MP. - Formation of nitric oxide hemoglobin in erythro­ logic release of histamine upon human lung cyclic nucle­ cytes eo-cultured with alveolar macrophages taken from otide levels and prostaglandin generation. J Clin Invest bleomycin-treated rats. Biochem Biopltys Res Commun 1978; 62: 1113- 1121. 1992; 182: 151-157. 65. Leurs R, Brozius MM, Jansen W, Bast A, Timmennan 49. Cameron ML, Granger DL, Weinberg JB , Kozumbo H . Histamine H 1-receptor-mediated cyclic GMP WJ, Koren HS . Human alveolar and peritoneal production in guinea-pig lung tissue is an L-arginine­ macrophages mediate fungistasis independently of L-arginine dependent process. Biochem Plwnnaco/ 1991; 42: 271- oxidation to nitrite and nitrate. Am Rev Respir Dis 1990; 277. 142: 1313-1319. 66. Mayer B, Bohme E. - Calcium-dependent formation 50. Hite RD, Smith RM. Inducibility of human and of an L-arginine-derived activator of soluble guanylate murine mononuclear phagocyte nitric oxide synthase by cyclase in bovine lung. FEES Lell 1989; 256: 211-214. lipopolysaccharide and y-interferon. Am Ret• Respir Dis 67. Kaliner M. Human lung tissue and anaphylaxis. 1992; 145: A284. Evidence that cyclic nucleotides modulate the immunologic 51. Tayeh MA, Marletta MA. Macrophage oxidation release of mediators through effects on microtubular assem­ of L-arginine to nitric oxide, nitrite and nitrate. Tetra­ bly. J Allergy Cl in /mm uno/ 1977; 60: 951 - 959. hydrobiopterin is required as a co-factor. J Bioi Cltem 1989; 68. Gustafsson LE, Leone AM, Persson MG, Wiklund NP, 264: 19654-19658. Moncada S. Endogenous nitric oxide is present in 52. Jorens PG, Van Overveld FJ, Vermeire PA, Bult H, the exhaled air of rabbits, guinea-pigs and hu­ Herman AG. - Pterins inhibit nitric oxide synthase activ­ mans. Biochem Biophys Res Commu11 1991; 181: 852-857. ity in rat alveolar macrophages. Br J Pltarmacol 1992; 107: 69. Park KGM. Hayes PD, Garlick PJ , Sewell H, Eremin 1088-1091. 0. Stimulation of lymphocyte natural cytotoxicity by 53. Nichol CA, Smith GK, Ouch DS. - Biosynthesis and L-arginine. Lancet 1991; 337: 645-646. metabolism of tetrahydrobiopterin and molybdopterin. Ann 70. Salvemini D, De Nucci G, Gryglewski RJ, Vane JR. Ret• Biochem 1985; 54: 729- 764. Human neutrophils and mononuclear cells inhibit 266 P.G. JORENS, P.A. VERMEIRE, A.G. HERMAN

platelet aggregation by releasing a nitric oxide-like factor. factor from nitric oxide. J Appl Physial 1990; 69: 665-670. Prac Natl Acad Sci USA 1989; 86: 6328-6332. 80. Bredt OS, Hwang PM, Glatt CE, et al. - Cloned and 71. Radomski MW, Palmer RMJ, Moncada S. An expressed nitric oxide synthase structurally resembles L-arginine, nitric oxide pathway present in human platelets cytochrome P-450 reductase. Nature 1991; 351: 714- regulates aggregation. Proc Natl Acad Sci USA 1990; 87: 718. 5193- 5197. 81. Borland COR, Higenbottam TW. - A simultaneous 72. Bissonnette EY, Hogaboam CM, Wallace JL, Befus single breath measurement of pulmonary diffusing capacity AD . Potentiation of tumor necrosis factor-alpha­ with nitric oxide and carbon monoxide. Eur Respir J 1989; mediated cytotoxicity of mast cells by their production of 2: 56- 63. nitric oxide. J lmmunal 1991 ;147: 3060-3065. 82. Meyer M, Piper J. - Nitric oxide (NO), a new test 73. Salvemini D, Masini E, Anggard E, Mannaioni PF, gas for study of alveolar-capillary diffusion. Eur Respir J Vane J. - Synthesis of a nitric oxide-like factor from 1989; 2: 494-496. L-arginine by rat serosal mast cells: stimulation of guanylate 83. Norman V, Keith CM . - Nitrogen oxides in tobacco cyclase and inhibition of pl atelet aggregation. Biachem smoke. Nature 1965; 205: 915-916. Biaphys Res Cammun 1990; 169: 596- 601. 84. Borland C, Higenbottam T. - Nitric oxide yields of 74. Goldie RG, Fernandes LB, Farmer SG, Hay DWP. - contemporary UK, US and French cigarettes. In! J Airway epithelium-derived inhibitory factor. TIPS 1990; 11: Epidemial 1987; 16: 31-34. 67-70. 85. Alving K, Lundberg JM. - Nitric oxide, a possible 75. Barnes PJ, Cuss FM, Palmer JB. - The effect of mediator of cigarette smoke-induced bronchial and pulmo­ airway epithelium on smooth muscle contractility in bovine nary vasodilatation in the pig. Am Rev Respir Dis 1992; trachea. Br J Pharmacal 1985; 86: 685-691. 145: A264. 76. Buga GM, Gold ME, Wood KS , Chaudhuri G, 86. Lippmann M. Airborne acidity: estimates of lgnarro LJ. - Endothelium-derived nitric oxide relaxes exposure and human health effects. Enviran Health nonvascular smooth muscle. Eur J Pharmacal 1989; 161 : Perspect 1985; 63: 63-70. 61-72. 87 . Wink DA, Kasprzak KS, Maragos CM, et al. 77. Guc MO, Ilham M, Kayaalp SO. The rat DNA deaminating ability and genotoxicity of nitric anococcygeus muscle is a convenient bioassay organ for oxide and its progenitors. Science 1991; 254: 1001-1003 the airway epithelium-derived relaxant factor. Eur J 88. Beckman JS, Beckman TW, Chen J, Marshall PA, Pharmacal 1988; 148: 405-409. Freeman PA. Apparent hydroxyl radical production 78. Fernandes LB, Paterson JW, Goldie RG. - Co-axial by peroxynitrite: implications for endothelial injury bioassay of a smooth muscle relaxant factor released from from nitric oxide and superoxide anion. Prac Not/ Acad guinea-pig tracheal epithelium. Br J Pharmacal 1989; 96: Sci USA 1990; 87: 1620- 1624. 117- 124. 89. Moreno JJ, Pryor WA. Inactivation of a) ­ 79. Munakata M, Masaki Y, Sakuma I, et al. - Pharma­ proteinase inhibitor by peroxynitrite. Chem Res Toxicol cological differentiation of epithelium-derived relaxing 1992; 5: 425-431.