INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nitric oxide (NO), a free radical gas previously considered merely an atmospheric pollutant, is now recognized to be one of the key mediators in many physiological and pathological processes. NO is generated from arginine by a family of enzymes, the NO synthases, and acts as an autocrine and paracrine messenger (1). NO also plays a major role in nonspecific host defense due to its cytotoxicity toward tumor cells and microorganisms. Inducible NO synthase (type II NO synthase) can be expressed after induction by proinflammatory cytokines or by bacterial lipopolysaccharide in many cell types (4). Thus, NO appears as a second line of defense against hard to kill pathogens in most mammalian tissues (4). The first line of nonspecific host defense is the NADPH-oxidase of the phagocytic cells (monocyte-macrophage and polymorphonuclears), which, upon activation, generate large amounts of superoxide anion (O₂^·) and derived reactive oxygen species (ROS, as hydrogen peroxide and hydroxyl radical) (5, 6).

However, this schematic view probably does not apply to all sites of the organism. The epithelia of the respiratory tract, bronchial (7) as well as paranasal sinuses (8), constitutively express the inducible NO synthase. As a consequence of the low air renewal in paranasal sinuses, very high concentrations of NO gas (5-20 parts per million [ppm]) have been measured in these cavities (8). Thus, NO could play a critical role in the physiology and pathology of the upper respiratory tract, because in addition to its role in nonspecific host defense (9), NO also stimulates ciliary motility (10). In the paranasal sinuses, NO appears to represent the first line of host defense, contributing to the sterility of these cavities (11). Inflammation of the sinus (i.e., sinusitis) induces the recruitement of phagocytic cells, which, in this particular case, represent a second line of defense (11). Thus, the strategy for host defense in the paranasal sinuses appears to be the reverse of that encountered in other cavities (such as the pulmonary alveola and peritoneal cavity) where resident macrophages are constitutively present to kill pathogens, whereas basal levels of NO are very low (12).

Because NO and O₂^· both contain an unpaired electron, they rapidly react together, leading to their reciprocal inactivation and eventually to peroxynitrite (ONOO) (13, 14). Peroxynitrite can, in turn, elicit protein tyrosine nitration, although other mechanisms, such as oxidation of nitrites by myeloperoxidase, can also elicit similar protein alterations (15). The in vivo interaction between NO and O₂^·, as well as the consequences for host defense and cell toxicity, are hard to predict for the following reasons: (1) in vivo concentrations of NO cannot be easily determined with the technologies currently available, and the estimates in the literature vary widely according to the authors, (2) the generation of O₂^· is even more difficult to determine, due to the very short half-life of this ROS. Moreover, the validity of the most sensitive and commonly employed technique, lucigenin-enhanced chemiluminescence, has recently been questioned (16), and (3) the very rapid interaction between NO and O₂^· further complicates the assessment of each radical species (13, 14).

In the present study, we first applied electron spin resonance (ESR) spectroscopy using DMPO as spin trap, ferricytochrome c reduction, and lucigenin-enhanced chemiluminescence to assess the O₂^· production of a macrophage cell line RAW 264.7 and validate the latter technique. We also compared the effect of three NO donors on ESR and chemiluminescent signals. Fragments of nasal polyps from patients who underwent surgery for nasal polyposis were then studied by histological examination and lucigenin-enhanced chemiluminescence. The effect of various concentrations of an NO donor, DETA NONOate, on chemiluminescent signals from nasal polyps fragments was therefore investigated, and the NO concentrations generated by DETA NONOate were calculated and compared with those reported in maxillary paranasal sinuses. The final goal of this work was to determine to what extent the concentration of NO present in paranasal sinuses inactivates the O₂^· production by phagocytes.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell Culture and Materials

A mouse macrophage cell line (RAW 264.7; American Type Culture Collection, Rockville, MD) was cultured in Dulbecco's modified Eagle medium with 10% heat-inactivated fetal calf serum (GIBCO) supplemented with gentamycin (0.1 mg/ml) and amphotericin B (125 ng/ml) in a 5% CO₂-containing atmosphere. RAW 264.7 were used when cells approached confluency (100,000/cm²) in the flasks or in 96 microwells.

Except when specified, all reagents were purchased from Sigma (St. Louis, MO): superoxide dismutase (SOD, ref. S-2515; phorbol 12-myristate 13-acetate (PMA), identical to 12-O-tetradecanoylphorbol 13-acetate [TPA]), catalase (ref. C-40), 5,5-dimethyl-1 pyrroline-N-oxide (DMPO), diethylenetriaminepentaacetic acid (DTPA, ref. D-751), and activated charcoal. DETA/NO NOC-18 and S-nitroso-N-acetyl-D,L-penicillamine (SNAP) were purchased from Alexis Biochemicals (San Diego, CA). DMPO was purified before use by treatment with activated charcoal for 15 min as reported in the literature (19), passed through a membrane filter (0.2 µm; Sartorius, Göttingen, Germany), aliquoted, protected from light and stored frozen at 20° C.

Patients with Nasal Polyposis

The procedures employed in this study were reviewed and approved by the local Ethics Committee (Comité consultatif de protection des personnes). The study was performed in 24 patients with nasal polyposis. The diagnosis of patients with nasal polyposis was based on the endoscopic observation of nasal polyps. Patients characteristics were mean age 46 ± 12 yr and sex ratio 16 males-8 females. Clinical symptoms were assessed the day of the NO measurement. The presence of symptoms (nasal obstruction, rhinorrhoea, and sneezing) the day of the NO measurement was noted, and the intensity (0, 1, 2 or 3) of each symptom was evaluated. The symptom scores were nasal obstruction 2.91 ± 0.29, rhinorrhea 1.09 ± 0.73, and sneezing 0.7 ± 0.7.

All patients underwent surgery for nasal polyposis, and the nasal polyps were removed surgically and stored at 4° C in Dulbecco's modified Eagle medium supplemented with gentamycin (0.1 mg/ml) and amphotericin B (125 ng/ml) for less than 1 h before being processed in the laboratory. The nasal polyps were then cut with a razor blade into fragments weighing 10 mg on average.

A nasal polyp from each patient was cut into two large fragments, one for histological examination, the other being segmented, depending on its size, into 15 to 25 fragments of approximately 10 mg.

Assessment of O₂^· Production

ESR measurements. RAW 264.7 were cultured in a 25-cm² flask, washed with phosphate-buffered saline (PBS), and then incubated with a mix containing 150 mM DMPO, 1 g/L glucose, 0.2 g/L CaCl₂, 0.0059 g/L DTPA, 0.15 g/L NaCl, and 0.37 g/L KCl in sodium phosphate buffer (2.35 g/L NaH₂PO₄/7.61 g/L Na₂HPO₄, pH 7.4) containing phorbol myristate acetate (PMA) 1 µM for 10 min. The entire procedure was performed at 37° C. The supernatant was then transferred to a flat quartz cell that was inserted in a TM 110 Bruker cavity. ESR spectra were recorded at room temperature with an ER 200 D Bruker spectrometer by starting a 3 min scan 5 min after the end of the incubation with the cells. The ESR spectrometer operated at 9.66 GHz with high frequency at 100 kHz, modulation amplitude 1 G, time constant 0.5 s, microwave power 10 mW, field: midrange at 3,500 G, and scan range 20 G/min. The intensity of the ESR signal was calculated by adding the height of the four peaks, and expressed in arbitrary units as previously described (20). In some experiments, 5 min after the addition of a NO donor, RAW 264.7 were stimulated with 1 µM PMA.

SOD inhibitable reduction of cytochrome c. O₂^· production was measured as the SOD inhibitable reduction of cytochrome c (21). RAW 264.7 (400,000 cells/well) were preincubated with RPMI for 15 min at 37° C, washed once with RPMI, and incubated with 1 ml RPMI containing 1 mg/ml cytochrome c with or without SOD (200 IU/ml) in humidified air on a shaking table. Cytochrome c reduction was determined at zero time to obtain basal values and after the indicated time, the absorbance of the medium was read spectrophotometrically at 550 nm against a distilled water blank. Reduction of cytochrome c in the presence of SOD was subtracted from the values without SOD: the proportion of superoxide specific reduction of cytochrome c (i.e., SOD inhibitable) was between 30 and 50% depending on the experiments. In some experiments, RAW 264.7 were stimulated with 1 µM PMA, 5 min after the addition of an NO donor.

Lucigenin-enhanced chemiluminescence. RAW 264.7 were cultured in 96-microwells. The production of reactive oxygen intermediates was measured using the chemiluminogenic probe lucigenin. The medium was removed from the wells and the cells washed twice with Krebs-Henseleit bicarbonate (KHB) (composition [mM]: NaCl 99.0, KCl 4.69, CaCl₂ 1.87, MgSO₄ 1.2, NaHCO₃ 25, K₂HPO₄ 1.03, Na-HEPES 20, D-glucose 11.1) with 1 µM PMA for 10 min, and lucigenin (250 µM final) was then added. The 96 microwells were thermostatically (37° C) controlled, and chemiluminescence was triggered with 1 µm PMA through an automatic injector (200 µl of final volume in each well), continuously monitored for 30 min and expressed in relative light unit/min. An average O₂^· production was calculated from that of the different fragments, and this value was then used in correlation with histological scores. In some experiments, RAW 264.7 were stimulated with 1 µm PMA 5 min after the addition of an NO donor.

Calculation of the Steady State Concentration of NO Generated by DETA/NO NOC-18

Decomposition of NONOates was monitored spectrophotometrically, at 37° C, in 0.1 M phosphate buffer, pH 7.4. For simulation of NO production, it was considered that NONOates release two equivalents of NO and that NO auto-oxidizes in aqueous solutions according to the following equations (22):

with

[O₂] being constant and equal to 220 µM.

The second integration method of Runge-Kutta was used (23). A small integration step was used to simulate the reactions at the initial times; as a rule, the initial integration step size was of the order of one-hundredth of the reaction time of the fastest reaction. The calculations were performed using the spreadsheet Excel 98 (Microsoft Office).

Histological Examination

The abundance of inflammatory cells was estimated using a semiquantitative score: 1 for a low number of inflammatory cells, 2 for a moderate infiltration by inflammatory cells on the chorion surface, and 3 for an intense infiltration by inflammatory cells on the chorion surface. The percentage of eosinophils in the inflammatory cells was also estimated using a semiquantitative score: 1 for < 10% of eosinophils, 2 for 10-50% of eosinophils, and 3 for > 50% of eosinophils. Finally, fibrosis was semiquantified as follows: 1 for chorion's edema, 2 for normal chorion, and 3 for constituted sclerosis of the chorion.

Statistical Analysis

The data were expressed as mean of triplicate values. The relation between the variables was calculated using Pearson's rank-correlation coefficients. p values of < 0.05 were considered significant. Analysis was performed with the SPSS program.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Validation of O₂^· Measurements

ESR signals detected using DMPO as spin trap. After 15 min incubation of DMPO with RAW 264.7 stimulated by 1 µm PMA, a typical ESR signal was obtained resulting from the DMPO-OH adduct (Figure 1A). This adduct could result from hydroxyl radical trapping by DMPO in the extracellular medium. However, it is well known that the DMPO-OOH adduct, resulting from the trapping of O₂^·, rapidly decomposes into a more stable adduct: DMPO-OH (24). To identify the ROS released by the RAW 264.7 and initially trapped by DMPO, the effects of SOD and catalase were tested as previously reported (25, 26). When SOD 30 U/ml was coincubated, the ESR signal was completely suppressed (not shown). In contrast, neither denatured SOD (15 min boiling) nor catalase (2,000 U/ml) affected the ESR signal (not shown). Together, these results demonstrate that the ESR adduct DMPO-OH detected in the supernatant of RAW 264.7 originated from the trapping of extracellular O₂^·. The ESR signal given by unstimulated (basal) cells was about 10-fold the baseline (Figure 1). As shown in Figure 2, three NO donors (sodium nitroprusside, DETA/NO NOC-18, and SNAP) dose dependently inhibited the amplitude of the ESR adduct DMPO-OH from RAW 264.7. 

View larger version (14K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Figure 1.   Representative tracings of electron spin resonance (ESR) spectra (A) and lucigenin-enhanced chemiluminescence (B) obtained from RAW 264.7, and of lucigenin-enhanced chemiluminescence (C ) obtained from nasal polyp. (A) The cells (2 × 106) were incubated in the presence of the spin trap DMPO and stimulated with 1 µM 12-O-tetradecanoylphorbol 13-acetate (TPA) for 15 min. (B) The cells (5 × 104) were incubated in the presence of lucigenin. Two right tracings: basal. Two left tracings: stimulation with 1 µM TPA, being applied just before the begining of the 15 min recording. (C ) A fragment of nasal polyp was incubated in the presence of lucigenin. Two right tracings: basal. Two left tracings: stimulation with 1 µM TPA, being applied just before the begining of the 45 min recording. RLU = relative light units, mg = mg tissue, s = second.

View larger version (37K): [in this window] [in a new window]   Figure 2.   The dose-response effect of SNP, DETA/NO NOC-18, and SNAP was studied on the ESR spectra generated (hatched bars) and on the lucigenin-enhanced chemiluminescence (black bars) by TPA-stimulated RAW 264.7. Data are the averages of triplicate incubations and representative of two different experiments.

Cytochrome c reduction technique. RAW 264.7 stimulated by 1 µm PMA induced an immediate SOD-inhibitable cytochrome c reduction, which was suppressed if SOD was previously boiled for 15 min. Catalase (2,000 U/ml) had no effect (not shown). These results demonstrate that SOD-inhibitable cytochrome c reduction signal elicited by PMA-stimulated RAW 264.7 is mainly due to extracellular O₂^·.

Lucigenin-elicited chemiluminescent signals from macrophage cell line. RAW 264.7 stimulated by 1 µm PMA induced an immediate lucigenin-elicited chemiluminescent signal. Representative tracings are shown in Figure 1B. SOD 100 U/ml and diphenyleneiodonium (DPI) 100 µM inhibited  90% of the lucigenin-enhanced chemiluminescent signal. In contrast, coincubation with denatured SOD (15 min boiling) or with catalase (2,000 U/ml) did not affect the morphology or the amplitude of the chemiluminescent signal (not shown). Altogether, these results demonstrate that the lucigenin-enhanced chemiluminescent signal elicited by PMA-stimulated RAW 264.7 is mainly due to extracellular O₂^·. As shown in Figure 2, the three NO donors dose dependently inhibited the amplitude of the lucigenin-enhanced chemiluminescent signal from RAW 264.7.

Histological Examination of Nasal Polyps and Relation to Lucigenin-elicited Chemiluminescent Signals

Eight fragments of each polyp were used to study the reproducibility of lucigenin-elicited chemiluminescence. For a given polyp, some variations were found for both basal and TPA-stimulated lucigenin-elicited chemiluminescence (Figures 3A and 3B). A mean value was calculated for each polyp and used for subsequent correlations with histology. Large variations in both basal and TPA-stimulated lucigenin-elicited chemiluminescence were also found between patients, and these are ordered from lowest to highest O₂^· producer in Figures 3A and 3B. Histological examination revealed the presence of various degrees of inflammation as well as fibrosis (Figure 4).

View larger version (21K): [in this window] [in a new window]   Figure 3.   (A, B) Reproducibility of lucigenin-elicited chemiluminescence (RLU = relative light units, mg = mg tissue, s = second) from the differents fragments of the polyp patients.

View larger version (140K): [in this window] [in a new window]   View larger version (142K): [in this window] [in a new window]   View larger version (141K): [in this window] [in a new window]   Figure 4.   Representative picture of polyp eosinophil infiltration and fibrosis. (A, B) High infiltration (score 3, > 50% eosinophils). Eosinophils are characterized by a bilobar nucleus and the presence of red granules in the cytoplasm (hematoxilin-eosin staining). (A) Original magnification: ×100; (B) original magnification: ×40. (C) Low infiltration (score 1, < 10% eosinophils) and abundant fibrosis. The hemalun- eosine-safran staining colors the collagen fibers yellow and identifies fibrosis. Original magnification: ×100.

Basal lucigenin-elicited chemiluminescence (calculated mean from the eight fragments) was significantly correlated with eosinophils abundance (r = 0.37, p < 0.05), fibrosis abundance (r = 0.3, p < 0.05) but not with cell abundance (r = 0.01, p = 0.8) (not shown). Similarly, TPA-stimulated lucigenin-elicited chemiluminescence was significantly correlated with eosinophils abundance (r = 0.67, p < 0.01) (Figure 5A), fibrosis abundance (r = 0.43, p < 0.05) (Figure 5B), but not with cell abundance (r = 0.27, p = 0.18) (Figure 5C).

View larger version (15K): [in this window] [in a new window]   Figure 5.   Correlation between TPA-stimulated lucigenin-elicited chemiluminescence and (A) polyp cell abundance, (B) polyp eosinophils abundance, and (C) polyp fibrosis.

Characterization of the Lucigenin-elicited Chemiluminescent Signals from Nasal Polyps

Polyp fragments stimulated by 1 µM PMA induced an immediate lucigenin-elicited chemiluminescent signal. Representative tracings are shown in Figure 1C. SOD 100 U/ml and DPI 100 µM inhibited = 90% of the lucigenin-enhanced chemiluminescent signal (Figure 6). In contrast, coincubation with denatured SOD (15 min boiling, data not shown), with catalase (2,000 U/ml), or with N^w-nitro-L-arginine methyl ester (L-NAME) (100 µM) did not affect the morphology or the amplitude of the chemiluminescent signal. Altogether, these results demonstrate that the lucigenin-enhanced chemiluminescent signal elicited by PMA-stimulated polyp fragment is mainly due to extracellular O₂^·, a result analogous to that obtained with RAW 264.7. 

View larger version (18K): [in this window] [in a new window]   Figure 6.   Characterization of the lucigenin-elicited chemiluminescent signals from nasal polyps. SOD = superoxide dismutase (100 U/ml), DPI = diphenyleneiodonium (100 µM), cat = catalase (2,000 U/ml), L-NAME = Nw-nitro-L-arginine methyl ester (100 µM). Data are means of triplicate incubations and representative of two separate experiments.

Effect of a DETA NONOate on the O₂^·-elicited Signals from Fragments of Nasal Polyps

Various concentrations of DETA NONOate were added 5 min before the addition of 1 µm PMA. As shown in Figure 7, DETA/ NO NOC-18 dose dependently inhibited the lucigenin-enhanced chemiluminescent signal with an EC₅₀ of approximately 1.5 mM.

View larger version (18K): [in this window] [in a new window]   Figure 7.   The dose-response effect of DETA/NO NOC-18 was studied on the lucigenin-enhanced chemiluminescence generated by TPA-stimulated polyp fragments. Indicated concentrations of DETA NONOate were added 5 min before the addition of 1 µM TPA. Data are expressed as percentage of TPA-stimulated polyp fragments obtained in the absence of NO donor.

Calculation of the Steady-State Concentration of NO Generated by DETA/NO NOC-18

Calculations revealed that 0.75 mM and 1.5 mM of the DETA NONOate generate steady-state NO concentrations of 0.5 µM and 2.5 µM, respectively.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The major findings of the present study are as follows: (1) Electron spin resonance (ESR) spectroscopy using DMPO as spin trap, ferricytochrome c reduction, and lucigenin-enhanced chemiluminescence can apparently detect the O₂^· production from a macrophage cell line RAW 264.7. Three different NO donors dose dependently inhibited these ESR and lucigenin-enhanced chemiluminescent signals. (2) The basal and stimulated O₂^· production in nasal polyps largely varied among patients, and were correlated to eosinophilic abundance. (3) A slow releasing NO donor, DETA NONOate, dose dependently inhibited lucigenin-enhanced chemiluminescence from TPA-stimulated polyp fragments. (4) The NO concentrations generated by DETA/NO NOC-18 were calculated, and those inhibiting lucigenin-enhanced chemiluminescence were found to be within the range of those in the maxillary paranasal sinuses.

The first goal of the present study was to validate a technique allowing the evaluation of O₂^· generation in cells and ex vivo tissues. Indeed, several factors account for the difficulty in assessing the generation of O₂^· in biological systems. O₂^· has a very short half-life due to its spontaneous dismutation, which is accelerated by SOD, and also to its rapid reaction with other radical species such as NO (6, 14). In addition, two of the three major techniques used to assess the production of O₂^· have recently been questioned. The simplest and easiest method for the detection of O₂^· is the reduction of ferricytochome c. However, Thomson and coworkers recently reported the oxidation of cytochrome c by peroxynitrite (27). These authors pointed out that O₂^· formation as measured with the cytochrome c reduction technique can be underestimated when NO is simultaneously generated at a comparable rates. This is why we did not determine this technique to study the effect of NO donors on O₂^· production. Lucigenin luminescence has been widely used to assess the production of O₂^·. However, the reliability of this technique was also recently questioned (16). Far fewer studies have used ESR spectroscopy because this technique requires an expensive tool and is time consuming compared with the two previous ones.

In the absence of NO, we found that ESR spectroscopy using DMPO as spin trap, ferricytochrome c reduction, and lucigenin-enhanced chemiluminescence were apparently correlated and quite suitable for assessing the high O₂^· production from a macrophage cell line RAW 264.7. In addition, three NO donors (sodium nitroprusside, DETA/NO NOC-18, and SNAP) dose dependently inhibited the amplitude of the ESR and of the lucigenin-enhanced chemiluminescence signals from RAW 264.7.

Thus, lucigenin-enhanced chemiluminescence, which is more sensitive than ESR, was subsequently applied to assess O₂^· production in fragments of nasal polyps. Large variations of lucigenin-elicited chemiluminescence were found among patients, together with heterogeneous inflammatory cell and eosinophil abundance. Interestingly, both basal and TPA-stimulated lucigenin-elicited chemiluminescence were significantly correlated with eosinophils abundance, but not with cell abundance. This correlation is probably the consequence of the high O₂^· production of eosinophils, this type of phagocytic cells being the most rich in NADPH oxidase (28).

NO influences O₂^· availability by at least two mechanisms. First, because NO and O₂^· both contains an unpaired electron, both species inactivate each other. Second, NO prevents the activation of NADPH oxidase by inhibiting its assembly process (29). In the present study, the NO donor was added before stimulating the phagocytic cells with TPA to mimic the in vivo conditions (where both mechanisms are acting in concert). Concentrations of DETA/NO NOC-18 applied before TPA stimulation of phagocytic NADPH oxidase dose dependently inhibited the TPA-stimulated lucigenin-elicited chemiluminescence from nasal polyp fragments, with an EC₅₀ of approximately 1.5 mM. At equilibrium, NO concentrations of 10-20 ppm were measured in normal maxillary sinuses (8) and are equivalent to a concentration of 0.45-0.90 µM in the aqueous phase (30). Concentrations of 0.75 mM and 1.5 mM DETA/NO NOC-18 were calculated to generate steady-state NO concentrations of 0.5 µM (10 ppm) and 2.5 µM (50 ppm), respectively. Thus, NO concentrations present in maxillary sinuses partially inhibit (approximately 20-40%) the TPA-stimulated lucigenin-elicited chemiluminescence, that is, phagocytic O₂^· production. The high concentration reached in the paranasal sinus cavities is probably due to the fact that the cavity is almost closed, and thus, the NO concentration can attain values close to the µM range. In contrast, in our experiments, the endogenous production of NO by the fragments of nasal polyp rapidly diffused into the air. Under these in vitro conditions, the concentration of NO due to endogenous production did not reach a level high enough to inhibit endogenous superoxide production, as demonstrated by the lack of effect of the NO synthase inhibitor L-NAME (Figure 6).

It has been shown that the NO concentrations found in normal sinuses are sufficient to inhibit the growth of several bacteria (9, 31). This reinforces the idea that under normal conditions, NO represents the first line of defense of the paranasal sinuses. Indirect evidence is supported by the dramatic decrease of nasal NO concentration in patients with Kartagener's syndrome (referred to as an "immobile cilia syndrome" and characterized by situs inversus, sinusitis, and bronchiectasis) (32, 33). However, the sensitivity of the various pathogens to NO, ROS, peroxynitrite, and the synergism and/or antagonism between these two mechanisms of defense require further study (31), and understanding of their respective roles in the pathophysiology of paranasal sinuses probably requires refined modeling. Finally, myeloperoxidase of eosinophils represents an alternative and potentially prominent mechanism of protein nitration (15). This important question about the link between these mechanisms and the pathobiology of the paranasal sinuses also deserves specific treatment.

In conclusion, it appears that the paranasal sinuses are an unusual anatomic site where NO accounts for the first line of nonspecific host defense, which could interfere and counteract the second line of defense constituted by the phagocytic cells.The NO concentrations found in paranasal sinuses appear to be within the critical range to inactivate the production of O₂^· by phagocytes. In vivo, this scheme could be complicated by additional events such as (1) allergic rhinitis increases in nasal NO concentration (33, 34), which can be normalized upon glucocorticoid treatment (35); (2) the potential generation of peroxynitrite, as the respective amounts of NO and O₂^· reach stoichiometry near 1:1, a condition favorable for the generation of peroxynitrite (14); and (3) the inhibition of apoptosis of certain phagocytic cells such as eosinophils by NO (36). All three mechanisms could contribute to the chronicity of the inflammatory process and thus to the pathophysiology of polyposis. All these events should be analyzed in future studies.