Research Paper Mediators of Inflammation, 7, 93–98 (1998)

Nitric oxide (NO), produced by alveolar macrophages (AM) is used as a marker of respiratory tract inflammation. Lipocortin 1 (Lc-1) is an anti-inflammatory, glucocorticoid-inducible protein. The current aims were to determine whether (a) an Lc-1-derived peptide, Ac2-26, inhibited lipopolysaccharide (LPS)-induced NO release by primary AM in vitro and (b) the inhibitory action of dexamethasone was Lc-1-dependent. LPS treatment stimulated NO release from rat AM. Ac2-26 had little effect on unstimulated release, but suppressed LPS-stimulated release at concentrations > or =320 nM (320 nM, 10 +/- 3%; 3.2 microM, 15 +/- 3%; 32 microM, 27 +/- 4% NO inhibited, mean +/- SEM, n = 6). Inhibition by dexamethasone of NO release was unaffected by neutralizing anti-Lc-1 indicating that this action is Lc-1-independent in primary AM. Nevertheless inhibition of NO release by Ac2-26 (80 microM) was similar to that of 1 microM dexamethasone (Ac2-26, 40 +/- 6%; dexamethasone, 48 +/- 6% NO inhibited, mean +/- SEM, n = 6).


Introduction
Lipocortin-1 (Lc-1) is a 37 kDa anti-inflammatory, glucocorticoid-inducible member of the annexin superfamily of proteins. Its synthesis is upregulated by glucocorticoids both in vitro and in vivo in a number of species, including man. 1 There is considerable evidence to suggest that intact Lc-1 and synthetic peptides derived from its primary sequence possess intrinsic anti-inflammatory properties in a range of experimental models. [2][3][4][5][6][7] Lc-1 is a potential mediator of the anti-inflammatory action of glucocorticoids. Increased levels of cellular Lc-1 protein have been observed in vivo in human alveolar macrophages (AM) exposed to oral glucocorticoid 8 and in human mononuclear cells following i.v. administration of hydrocortisone. 9,10 Studies on adrenalectomized rats have shown a decrease in Lc-1 mRNA and protein in the lung, spleen, liver and kidney, suggesting not only a pharmacological modulation, but endogenous physiological steroidal control on Lc-1 expression and turnover. 11 Immunoneutralization studies have confirmed that a number of the anti-inflammatory actions of exogenous glucocorticoids are, in part, dependent on Lc-1, both in vivo 5 -7,12 and in vitro. 13 -17 Previous studies have demonstrated that the lung is a rich source of Lc-1 18 -20 w hich is detectable in a number of different cell types, including human AM 21 and epithelial cells. 22 Lc-1 is present in bronchoalveolar lavage fluid, 23 in w hich its levels are increased by oral glucocorticoids in healthy volunteers, 24,25 patients with lung disease, 24 and control and carrageenin-treated rats. 26 Increased Lc-1 concentrations in lung lavage also occur in response to non-specific stimuli, such as cigarette smoke 27 and LPS; 28 this may reflect increased expression of Lc-1, but may also be a result of increased cell number, cell death or injury.
Inflammation in the respiratory tract is associated with the release of numerous inflammatory mediators, including cytokines and eicosanoids. Recently, nitric oxide (NO) has received much attention as a potential modulator of lung function in health and disease. Both constitutive nitric oxide synthase (cNOS) and inducible nitric oxide synthase (iNOS) are present in the respiratory tract. The former is found in endothelial cells and peripheral nerves, w here it acts as a smooth muscle relaxant and neurotransmitter 29 and the latter, which is probably involved in lung defence, is localized to AM and epithelial cells 30,31 and is increased in acute and chronic inflammation. 32 Systemic administration of LPS or generalized sepsis results in high levels of exhaled NO in the lungs of rats. 33 Wood smoke-exposed rats have elevated concentrations of NO metabolites in the plasma w hich correlates with increased lung epithelial permeability, a process reversible with NOS inhibitors. 34 This suggests that smoke-induced damage may be due to reactive nitrogen radicals in the lung. Intra-tracheal instillation of LPS to rats also causes an increase in epithelial leak, elevated levels of NO in respiratory tract secretions and increased NO release by AM 35 and is associated with induction of iNOS mRNA. 36 Raised iNOS activity and NO release also occur in rat macrophages and epithelial cells in response to exposure to cytokines, such as IFN-g , TNFa and IL-1. 31 Unlike cNOS, which produces NO at a constant low level, activation of iNOS results in high output NO production. 37 Thus, many pulmonary inflammatory conditions are accompanied by increased NO release. Although it is unclear w hether the NO is damaging or protective, NO release is used as a marker of lung inflammation.
Glucocorticoids inhibit NOS activity in the J774.2 macrophage-like cell line, 38 an effect which is Lc-1 dependent in these cells. 15 Similarly, inhibition by glucocorticoid of LPS-induced iNOS activity in rat lung homogenates is Lc-1-dependent, the AM being suggested as a potential target for the glucocorticoid action. 15 Therefore, we hypothesized that NO release from AM in vitro could be inhibited by glucocorticoids in a Lc-1-dependent manner. Primary AM represent a useful in vitro model system for examining the relationship between glucocorticoids, Lc-1 and pulmonary NO, since they possess functional glucocorticoid receptors, 39 release NO via the activity of iNOS and express Lc-1. 21 In addition they are the predominant cell type in the airspaces with potential anti-and pro-inflammatory functions. The AM therefore encapsulates all the components under examination and, furthermore, is easily accessible and readily isolated.
Thus the aims of the current study were to determine the effect of a synthetic peptide derived from the N-terminal of Lc-1 (Ac2-26), 4,6,40,41 on NO release by primary rat AM in vitro and to examine whether inhibition of NO release by glucocorticoids in this system is Lc-1 dependent.

Animals and removal of lungs
Pathogen-free male Wistar rats (Charles River), weighing 200-250 g were killed by a lethal intra-peritoneal injection of pentobarbitone (1 g/kg body weight) and heparin (1000 U/kg body weight). Prior to excision, the trachea was cannulated and the lungs perfused free of blood via the right ventricle and pulmonary artery using a gravity feed of sterile 0.15 M NaCl.

Isolation and culture of alveolar macrophages (AM)
Excised lungs were fully inflated via the tracheal cannula with 10 ml aliquots of sterile 0.15 M NaCl, emptied, and lavage repeated until approximately 50 ml of fluid had been recovered. The lavage fluid was pooled and centrifuged at 300 3 g for 10 min at 4°C. Pelleted cells were resuspended in Hanks's balanced salt solution, without Ca 2+ or Mg 2+ (Sigma) and centrifuged as before. BAL cells (approximately 95% AM) were resuspended in low protein hybridoma medium, containing 2% foetal calf serum, glutamine (2 mM), penicillin (100 U/ml) and streptomycin (100 m g/ml). Cells were counted and seeded in to 96-well plates (Nunc) at a density of 1 3 10 5 cells per well. AM were allowed to adhere for 3 h in a humidified chamber at 37°C in 5% CO 2 in air. All cell culture reagents were obtained from Gibco.

Effect o f dexa m etha s o ne a nd peptide Ac2-26 on NO relea s e
Following the 3-h adherence period, non-adherent cells and conditioned media were aspirated and replaced by fresh medium containing graded concentrations of Lc-1 peptide, Ac2-26 (3.2 nM-80 m M), or dexamethasone (10 nM-1 m M). Adherent cells were pre-incubated for 1 h, following which lipopolysaccharide (0.1-10 m g/ml, serotype 055:B5 from Escherichia coli, Sigma) was added to all wells except controls and the cells incubated for a further 24 h. Conditioned media were harvested and analysed for nitrite. The competitive NOS inhibitor N G -monomethyl L-arginine (L-NMMA 4 mM, Sigma) was added to selected wells to determine the proportion of measured nitrite which originated from the activity of NOS.

Effect o f Lc-1 im muno-neutra liza tion o n dexa m ethas o ne a ctivity
Non-adherent cells and conditioned media were removed as above and replaced with fresh medium containing a sheep polyclonal neutralising antibody to Lc-1 (LC01, the generous gift of Professor R. J. Flower, William Harvey Research Institute, London) used at 1:50 dilution. The cells were pre-treated w ith this antiserum for 3 h, then incubated w ith graded doses of dexamethasone for 1 h and stimulated with lipopolysaccharide for 24 h as described above.

Measurement of nitrite
Spontaneous ox idation of the NO radical in aqueous solutions leads to the formation of nitrite (NO 2 -) as the predominant stable breakdown product. 29,42 Nitrite accumulation in conditioned media was therefore used as an index of NO release by AM, 24 h post-LPS. This was measured using the colorimetric Greiss reaction. 43 Nitrite concentrations were determined by comparison w ith standard solutions of sodium nitrite (0.78-100 m M) prepared in culture medium.

Data analysis
Data were analysed by the Wilcoxon's signed rank test for paired data; a probability of P<0.05 was regarded as statistically significant. Percentage inhibition of

Effect of lipopolysaccharide on NO release
Median basal release of NO was 56, range 47-61 nmol NO/10 6 cells/24 h in culture (Fig. 1, n = 4); 86% was inhibited by 4 mM L-NMMA. Stimulation of AM with LPS produced a concentration-dependent increase in nitrite accumulation in the culture medium comparable with that observed in previous studies 31,38,44 (Fig.  1). At a concentration of 10 m g LPS/ml, 85% of NO release was inhibited by L-NMMA. Based on these observations, an LPS dose range of 0.1-10 m g/ml was used in all subsequent experiments.

Effect of peptide Ac2-26 on NO release
The peptide had no significant effect on basal NO release in unstimulated cells (Fig. 2) except at the highest concentrations of Ac2-26 tested (32 m M and 80 m M) where basal release was significantly inhibited (31% and 28% respectively). Ac2-26 significantly inhibited LPS-induced NO release by AM at concentrations of 320 nM or more (Fig. 3). The maximum inhibition observed was 40% by a concentration of 80 m M Ac2-26. The action of Ac2-26 was similar irrespective of the LPS concentration used (Table 1) and therefore only data for LPS at 10 m g/ml are illustrated.

Effect of dexamethasone on NO release
Dexamethasone significantly inhibited basal NO release at all concentrations of drug and LPS-stimulated release at concentrations above 10 nM dexamethasone (Table 1 and Figs 2 and 3). Maximal inhibition by dexamethasone was 48%.

Effect of anti-Lc-1 neutralizing antibody on NO release
Pre-treatment of rat AM with a neutralizing antibody to Lc-1, had no effect on dexamethasone-mediated inhibition of NO release, irrespective of the concentration of dexamethasone used (Fig. 4).

Discussion
This study demonstrates that the N-terminal peptide of Lc-1, Ac2-26, inhibits LPS-induced NO release from rat AM in a concentration-dependent manner. To our knowledge, this is the first such in vitro demonstration in a primary cell w ith a known pro-inflammatory function. By increasing the dose of Ac2-26 to 80 m M it was possible to achieve the same degree of inhibition as with 1 m M dexamethasone. Interestingly use of a neutralising anti-Lc-1 antiserum indicated that the inhibitory action of dexamethasone itself on NO release was independent of endogenous Lc-1. Dow nregulation of LPS-induced NO release by a longer Lc-1 peptide, Lc-1 1-188 , at a concentration of 20 m g/ml has been reported in the J774.2 cell line. 15 The observation that it was effective at a substantially lower concentration than Ac2-26 may reflect differences between the responsiveness to Lc-1 of the cell line and the primary AM used in the current study. Another possibility is that the Ac2-26 was partially degraded by AM-derived products during the long incubation period. Alternatively, there may be differences between the peptides themselves, as shown previously with murine neutrophils; 4 it is possible that the Ac2-26 does not activate the signal transduction machinery needed to trigger the cell response as effectively as does the larger molecule. The optimal size for inhibition of NO release by Lc-1-derived A. M. Ka mal et al.

96
Mediators of Inflammation · Vol 7 · 1998 peptides remains to be established. Provided that the peptides are as effective as dexamethasone and low in toxicity, the absolute concentration required to achieve inhibition is probably not a major consideration for their use in vivo . Dexamethasone was an effective inhibitor of NO release by AM in this study, although interestingly, use of neutralizing antiserum indicated that its action was Lc-l-independent. Pulmonary AM treated w ith dexamethasone are capable of increasing their Lc-1 expression and release. 21 However, studies in our laboratory indicate that at the highest concentration of dexamethasone used in the current investigation (1 m M) the increase in cellular Lc-1 above baseline is likely to be modest in rat AM. 45 Since the antiinflammatory actions of glucocorticoids are mediated via multiple pathways, the effect of dexamethasone on NO release is likely to be via another mechanism, perhaps downregulation of iNOS. 36 Similarly, the mode of action of dexamethasone in the lung in vivo is not fully known. Previous publications by us and others have demonstrated increases in cellular and extracellular pulmonary Lc-1 following oral glucocorticoid 24,25 and also after non-specific stimuli including LPS 15 and carrageenin. 26 A previous study in vivo by Wu et al. 15 showed that neutralizing antisera to Lc-1 blunted the action of dexamethasone against LPS-induced NO release from lung homogenates, suggesting that increases in endogenous Lc-1 contribute to the steroid action. However, in vivo, there are many other potential sources of Lc-1 in the lung w hich may be targets for the dexamethasone, such as epithelial and gland cells. 19 Our previous work indicates that the type II epithelial cell (TII) releases enhanced levels of Lc-1 in response to lower doses of dexamethasone than AM. Allied w ith their high number in the lung, 46 -48 this makes TII likely cellular mediators of the Lc-1-dependent action of dexamethasone observed by Wu et a l. 15 Similarly, in vivo , extracellular Lc-1 may inhibit NO release by a variety of cell types not present in our simple in vitro model.
In summary, we have shown that Ac2-26 causes the same degree of inhibition of LPS-induced NO release by AM as dexamethasone, but at a much higher concentration and by a different mechanism. This and other peptides derived from Lc-1 may have therapeutic potential for suppression of pulmonary inflammation, since they could be delivered topically to the target organ.