Inflammation & Allergy - Drug Targets, 2006, 5, 115-119
115
Nitric Oxide and Inflammation Giuseppe Cirino*,1, Eleonora Distrutti 2 and John L. Wallace3 1
Department of Experimental Pharmacology, Università di Napoli –Federico II, Napoli, Italy, 2Dipartimento di Medicina Sperimentale, Medical School, Università di Perugia, Perugia, Italy 3
Mucosal Inflammation Research Group, University of Calgary, Calgary, Alberta, Canada Abstract: There are several pre-clinical studies on the involvement of NO in inflammation. From this large amount of information it is clear that virtually every cell and many immunological parameters are modulated by NO. Thus, the final outcome is that NO cannot be rigidly classified as an anti-inflammatory or pro-inflammatory molecule. This peculiar aspect of the pathophysiology of NO has hampered the development of new drugs based on the concepts developed. Recent therapeutic approach are targeted to increase endogenous NO by activating the gene and some promising early data are available. At the present stage one of the most promising approach in the inflammation field is represented by a new class of NO-releasing compounds namely NO-NSAIDs that have recently enrolled in phase 2 clinical studies.
Keywords: NO, inflammation, NO-NSAIDs, pre-clinical data, clinical data. NO CHEMISTRY Unlike other biological mediators, the chemistry of NO determines its biological properties and there are several chemical reactions related to NO. There are direct chemical reactions that are defined as those in which NO interacts with biological targets. The most common reaction is between NO and heme-containing proteins. Such reactions are generally rapid, requiring low concentrations of NO and most likely account for the majority of the physiological effects of NO. Conversely, indirect effects involve reactive nitrogen oxide species (RNOS) derived from the reaction between NO and either O2 or O2- rather than NO itself. As direct effects have to be remembered the direct reaction of NO with the metal center, redox reactions with dioxygen complexes and redox reactions with high-valence oxo-complexes. The indirect effects of NO are often associated with pathophysiological conditions, and higher nitrogen oxides are thought to be the chemical species responsible. Nitric oxide is produced by a specific enzyme called nitric oxide synthase (NOS) starting from the precursor L-arginine. There are three NOS and they are all constituted by an oxygenase and a reductase domain. The reductase domain shares an high homology with cytochrome P-450 reductase. Two of the three NOS are constitutively expressed in cells and they synthesise NO in response to an increase in calcium or in some cases to calcium independent stimuli such as shear stress [1-2]. The constitutive enzyme are known as neuronal or brain NOS (bNOS) and endothelial NOS (eNOS) or NOS I and NOS III respectively and they are so called since they were first isolated in rat neurons and bovine endothelial cells. The third isoform is the inducible NOS (iNOS or NOS II) that is *Address correspondence to this author at the Department of Experimental Pharmacology, Università di Napoli – Federico II, 80131, Napoli, Italy; Tel: 39-081-678442; Fax: 39-081-678403; E-mail:
[email protected]
1871-5281/06 $50. 00+. 00
constitutively expressed only in some tissues such as lung epithelium and is typically synthesised in response to inflammatory or pro-inflammatory mediators. NO AND INFLAMMATION There is a large body of evidence that NO is involved in several inflammatory disorders. Indeed, virtually every cell and many immunological parameters are modulated by NO. It has been shown that NO can be pro-inflammatory (immunostimulatory, anti-apoptotic) or anti-inflammatory (immunosuppressive, pro-apoptotic), host-protective or hostdamaging during infections. For these reasons NO has been described as “double edge sword mediator” and this phenomenon is often referred to as the NO paradox. The papers produced in this field can be broadly divided into two categories i) studies performed with exogenous NO ii) studies performed with NO synthesis inhibitors such LNAME or L-NMMA in order to modulate endogenous NO release. The first category of experiments has a major flaw in the fact that the amount of exogenous NO delivered in most cases is far from what a cell can would be exposed to endogenously in a pathological condition. The second category of experiments has some other major problems that should be considered: i) these inhibitors are false substrates and not real enzymatic inhibitors; ii) the local injection of these inhibitors will cause vasoconstriction and thus an antiinflammatory effect by itself iii) administration by oral route can cause an hypertension that can be transient or stable depending on the dose and the administration time-frame used. Thus, the final outcome is that NO cannot be rigidly classified as a anti-inflammatory or pro-inflammatory molecule. This peculiar aspect of the pathophysiology of NO has hampered the development of new drugs. Inhibitors of neuronal NOS have been proposed to be useful in stroke and Parkinson while inhibitor of iNOS in inflammatory diseases, shock and cancer where a role for endothelial NOS has been © 2006 Bentham Science Publishers Ltd.
116
Inflammation & Allergy - Drug Targets, 2006, Vol. 5, No. 2
also postulated. Undoubtedly, several organic nitrates and other NO donors have been synthesised but they are mainly used in basic research as pharmacological tools. In addition, so far none of the compounds that have been enrolled in clinical studies have gone beyond a phase 1. Another therapeutic approach is targeted to cause an increase in endogenous NO and this enclose i) promotion of NO synthesis by activating the gene or increasing the enzymatic activity; ii) prevent breakdown by increasing superoxide breakdown; iii) Gene therapy where preclinical data on iNOS and eNOS have shown promising results in postangioplasty restenosis. However, in recent years a new class of NSAID that enclose in their structure an NO-releasing moiety (so-called NO-NSAIDs) has been developed and it is now in advanced clinical development representing at the present stage the only direct therapeutic application of NO other than the organic nitrates used in cardiovascular therapy. NO-NSAIDs The development of cyclooxygenase (COX)-inhibiting nitric oxide donors (NO-NSAIDs) was based on an intention to reduce the gastrointestinal and renal toxicity associated with COX inhibition [3,5]. Of course, it was also very important that the derivatization of the COX inhibitor to include an NO-releasing moiety did not interfere with its ability to suppress inflammation and pain. A variety of preclinical models of pain and inflammation have been employed to characterize the effects of NO-NSAIDs, and in particular to compare their effects to the parent drugs. These studies suggest that NO-NSAIDs exert anti-inflammatory and analgesic effects beyond those seen with native NSAIDs.
Cirino et al.
(~85%), indicating that the drugs were equally effective in suppressing COX-2 activity. Celecoxib was found not to suppress whole blood thromboxane synthesis, confirming its ability to selectively inhibit COX-2 at the dose tested. Thus, in the sponge model, PGE2 synthesis was suppressed by naproxen and NO-naproxen but not by celecoxib, indicating that it must have been derived primarily from COX-1. Carrageenan Paw Oedema and Hyperalgesia Given that NO-NSAIDs inhibit PG synthesis at sites of inflammation as effectively as the parent NSAIDs, it would be predictable that they would exert comparable antiinflammatory effects. Fig. 1 illustrates the dose-dependent reduction of carrageenan-induced paw oedema that can be produced by administration of diclofenac or a NO-NSAIDs (MET-157). It is clear that the two drugs exhibit very similar potency and efficacy in this model. However, it is important to note that in this study [2], the same doses of the two drugs (i.e. on a mg/kg basis) were compared. As the molecular weight of MET-157 is approximately 50% more than that of diclofenac, the amount of diclofenac delivered by MET-157 is ~ 33% less than by the same mg/kg dose of diclofenac itself. This suggests that the NO-NSAIDs may actually be more potent than the parent NSAID in reducing oedema in this model.
PRE-CLINICAL DATA Carrageenan-Sponge and Airpouch Models Some of the earliest experiments with NO-NSAIDs involved an assessment of whether or not these drugs inhibited prostaglandin synthesis at sites of inflammation as effectively as the parent NSAIDs. Rats in which carrageenan-soaked sponges had been implanted subcutaneously were treated orally with an NSAID (diclofenac; 20 mg/kg) or a NO-NSAIDS (MET-157; 20 mg/kg) [4]. Both compounds markedly suppressed (by ~70%) the levels of prostaglandin E2 in the sponge, which was harvested 1 hour after drug administration. The level of suppression of PGE2 synthesis did not differ significantly between the NSAID and the NO-NSAIDS. Muscara et al. [5] examined the effects of an NSAID (naproxen), an NO-NSAID (NO-naproxen) and a COX-2 selective NSAID (celecoxib) in a sponge implant model in rats that did not involve prior soaking of the sponge in carrageenan. The rats were treated with the test drugs once daily for five days. At the end of that period, the sponges were harvested and PGE2 levels in the exudates were determined. Interestingly, naproxen and NO-naproxen suppressed PGE2 synthesis to the same extent (~70%), but celecoxib had no effect. When the same drugs were tested in the carrageenan-airpouch model, PGE2 synthesis was suppressed by all three of the test drugs to the same extent
Fig. (1). Reduction of carrageenan-induced paw edema by oral pretreatment with diclofenac (top panel) or with a nitric oxidereleasing derivative of diclofenac, MET-157 (bottom panel). Both drugs dose-dependently reduced the magnitude of the inflammatory reaction. This graph was constructed using data reported previously [2].
Nitric Oxide and Inflammation
A nitric oxide-releasing derivative of aspirin, NCX-4016, was found to be more effective at reducing carrageenaninduced oedema than was aspirin [6]. The ED50s for reducing oedema for NCX-4016 and aspirin were, respectively, 64 and > 555 umol/kg. Both drugs also caused dose-dependent reduction of mechanical hyperalgesia, and once again the NO-releasing derivative (NCX-4016) exhibited significantly greater potency (the ED50 was about 50% less than that for aspirin).
Inflammation & Allergy - Drug Targets, 2006, Vol. 5, No. 2
117
Acetic Acid-Induced Abdominal Constrictions NO-naproxen has also been compared to naproxen for analgesic effects in the acetic acid-induced abdominal constriction model in mice. Davies et al. [9] tested the effectiveness of acute administration of these drugs (i.e., 1 hour prior to intraperitoneal injection of acetic acid. As shown in Fig. 3, NO-naproxen was more potent than naproxen at reducing the number of abdominal constrictions induced by acetic acid.
It is interesting to note that NO-NSAIDs exert analgesic and anti-inflammatory effects more potently than the parent NSAIDs, while only producing comparable suppression of PG synthesis. The most likely explanation for this finding is that the nitric oxide released by NO-NSAIDSs contributes to their analgesic and anti-inflammatory effects. Evidence to support this notion comes from studies of an NO-releasing derivative of a drug that is not a COX inhibitor, acetaminophen. Acetaminophen exerts little if any antiinflammatory effects. However, an NO-releasing derivative of acetaminophen (NCX-701) was found to dosedependently reduce carrageenan-induced paw oedema [7]. This compound also exerted more potent analgesic effects than the parent drug, as discussed in more detail below. Adjuvant-Induced Arthritis Cicala et al. [8] evaluated mechanical hyperalgesia in the adjuvant-induced arthritis model in rats. The rats were treated orally each day with vehicle, naproxen (1, 3 or 10 mg/kg) or an equimolar dose of NO-naproxen. On day 15 after adjuvant administration, mechanical hyperalgesia was determined in the contralateral hindpaw (i.e. the paw that was not injected with adjuvant). Fig. 2 shows the threshold for pain sensation in this assay, and clearly shows that NOnaproxen was more potent than naproxen in terms of producing analgesia. Naproxen was only effective at the highest dose tested (10 mg/kg), while NO-naproxen produced significant analgesia at one-third that dose, on a molar basis.
Fig. (3). Reduction of the number of abdominal constrictions induced in mice by intraperitoneally administered acetic acid by naproxen and NO-naproxen. The shift to the left of the doseresponse curve with NO-naproxen indicates a marked increase in potency relative to naproxen. This graph was constructed using previously reported data [16].
A similar increase in potency was reported by al-Swayeh et al. [6], who compared NCX-4016 to aspirin in this model. With aspirin, the ED50 for reduction of abdominal constrictions was 243 umol/kg, while for NCX-4016, the ED50 was 155 umol/kg. The possibility that the NO released from NO-NSAIDs accounts for the increase in analgesic potency is supported by the work of al-Swayeh et al. on NCX-701 [7], the NOreleasing derivative of acetaminophen. As shown in Fig. 4, acteminophen dose-dependently reduced the number of abdominal constrictions induced in mice by acetic acid. However, NCX-701 was approximately 20-fold more potent in producing analgesia, despite having no effect on COX activity (Wallace, unpublished). Intravital Microscopy
Fig. (2). Reduction of pain in rats with adjuvant-induced arthritis by treatment over a 15-day period with naproxen or NO-naproxen. The pain threshold (in grams applied to the paw) of the contralateral hindpaw (not injected with adjuvant) was determined. *p
