HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 17/9/1171 02-0777fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by FIORUCCI, S. Articles by WALLACE, J. L Search for Related Content PubMed PubMed Citation Articles by FIORUCCI, S. Articles by WALLACE, J. L

Relative contribution of acetylated cyclo-oxygenase (COX)-2 and 5-lipooxygenase (LOX) in regulating gastric mucosal integrity and adaptation to aspirin¹

STEFANO FIORUCCI², ELEONORA DISTRUTTI, OCTAVIO MENEZES DE LIMA^*, MARIO ROMANO, ANDREA MENCARELLI, MIRIAM BARBANTI, ERNESTO PALAZZINI, ANTONIO MORELLI and JOHN L WALLACE^*

Clinica di Gastroenterologia ed Epatologia, Dipartimento di Medicina Clinica e Sperimentale, Università degli Studi di Perugia, Perugia, Italy;
^* Department of Pharmacology, University of Calgary, Alberta, Canada;
Dipartimento di Scienze Biomediche, Università "G. D’Annunzio," Chieti; and
Alfa Wassermann, Bologna, Italy

²Correspondence: E-mail: fiorucci{at}unipg.it

SPECIFIC AIMS

The primary aim of this study was to investigate the relative contribution of acetylated cyclo-oxygenase (COX) -2- and 5-lipooxygenase (LOX) -derived prostanoids to the pathogenesis of gastric damage induced by aspirin.

PRINCIPAL FINDINGS

1. COX inhibitors, but not dual COX/5-LOX inhibitors, potentiate gastric damage caused by aspirin
Treating rats with aspirin (10–100 mg/kg) caused a dose-dependent injury of gastric mucosa and increased gastric myeloperoxidase (MPO) activity, a measure of polymorphonuclear leukocyte (PMN) recruitment in the gastric mucosa (n=5–6 rats per group, P<0.01 vs. control). In contrast to aspirin, the COX/5-LOX inhibitor licofelone (10, 50, and 100 mg/kg) spared the stomach and did not increase MPO activity (n=6 rats per group, nonsignificant [NS] vs. control). The gastric protection afforded by licofelone was not due to a lack of activity on COX isoenzymes, since licofelone dose-dependently inhibited gastric PGE₂ and serum TXB₂ generation (n= 6, P<0.01 vs. control), a measure of COX-1 activity, and reduced systemic PGE₂ and LTB₄ synthesis induced by in vivo injection of bacterial endotoxin, an overall measure of COX-2 and 5-LOX activity (n=5–6, P<0.01). As shown in Fig. 1 , administration of licofelone at doses of 10, 50, and 100 mg/kg did not exacerbate gastric damage caused by aspirin, whereas it dose-dependently reduced gastric MPO activity (n=5–6 group, P<0.01 vs. aspirin alone). In experiments using intravital microscopy, licofelone (100 mg/kg) completely reversed the leukocyte adherence to mesenteric postcapillary endothelial cells caused by aspirin, 50 mg/kg (n=5–6 per group, +P<0.05 vs. aspirin). In contrast to licofelone (Fig. 1) , selective and nonselective COX-2 inhibitors celecoxib, indomethacin, and ketorolac administered in combination with aspirin exacerbated mucosal damage caused by aspirin and increased PMN margination (n=4–6 rats per group, P<0.05 vs. aspirin alone). These effects were significantly reduced in rats pretreated with the 5-lipoxygenase inhibitor zileuton (n=4 group, P<0.05 vs. aspirin plus celecoxib, indomethacin, or ketorolac).

View larger version (20K): [in this window] [in a new window]   Figure 1. a, b) 5-LOX inhibition reduces the mucosal damage and PMN margination into the gastric mucosa caused by coadministration of selective or nonselective COX-2 inhibitors and aspirin.c) Aspirin induces COX-2 expression in the stomach. M, molecular masses; -, negative control; +, positive control (i.e., COX-1, COX-2, or 5-LOX-positive cDNA; 0, 1, 3, and 12 gastric samples obtained before and 1, 3, and 12 h after aspirin (30 mg/kg) administration. Data are representative of at least 3 other experiments. d) Aspirin-induced ATL formation is abolished by coadministration of selective or nonselective COX-2 inhibitors and by 5-LOX inhibitors. P<0.05 vs. vehicle. e) Selective and nonselective COX inhibitors trigger LTB4 production in aspirin-treated rats.. *P<0.01 vs. vehicle. **P<0.01 zileuton vs. aspirin plus another COX inhibitor.

2. Combined exposure to aspirin and COX inhibitors prevents ATL formation and stimulates LTB₄ production
We then examined whether COX inhibitors, as well as licofelone or zileuton, affected aspirin-triggered lipoxin (ATL) formation induced in the gastric mucosa by exposure to aspirin. ATL or 15-epi-lipoxin A₄ (15-epi-LXA₄) is formed at the endothelial–neutrophil interface in the gastric microcirculation through oxidation of arachidonic acid (AA) by acetylated COX-2. We first assessed whether exposure to aspirin induces COX-2 expression into the mucosa. As shown in Fig. 1c , whereas COX-2 was not detected in control rats, expression of COX-2 mRNA became evident at 1 h after administration of 50 mg/kg aspirin and remained up-regulated for at least 12 h after a single dose of aspirin. In contrast, aspirin had no effect on COX-1 and 5-LOX mRNA expression. Although aspirin did not induce LTB₄ synthesis, coadministration of celecoxib, indomethacin, and ketorolac, but not licofelone, to aspirin-treated rats triggered an four- to fivefold increase in mucosal LTB₄ concentrations (Fig. 1d ; n=6, P<0.01 vs. aspirin alone). At a dose of 100 mg/kg, zileuton, a selective 5-LOX inhibitor, completely inhibited these changes (n=4–6, P<0.05 vs. aspirin plus each one of these nonsteroidal anti-inflammatory drugs, or NSAIDs). As illustrated in Fig. 1e , aspirin administration stimulated ATL formation (n=4–6, P<0.01 vs. control). Cotreating rats with selective and nonselective COX-2 inhibitors, as well as with selective and nonselective 5-LOX inhibitors licofelone and zileuton, completely inhibited ATL synthesis (n=5–6, P<0.01 vs. aspirin).

3. LTB₄/ATL ratio modulates gastric adaptation to aspirin
Daily administration of 50 mg·kg^-1·day^-1 aspirin for 2 wk resulted in a time-dependent adaptation of the gastric mucosa to damage. Thus, as shown in Fig. 1 , after 2 wk of daily administration of the same dose of aspirin the extent of gastric damage was significantly reduced relative to that seen in the group of rats treated with a single dose (n=4–6, P<0.05 repeated vs. single dose of aspirin). Gastric adaptation to aspirin was abolished by cotreating rats with celecoxib, indomethacin, or ketorolac (n=5–6, P<0.01 vs. aspirin). In contrast to COX inhibitors, licofelone, 100 mg·kg^-1·day^-1, did not prevent gastric adaptation to aspirin. Adaptation to aspirin was associated with a significant increase in mucosal ATL production. Administration of COX inhibitors to aspirin-treated rats inhibited ATL formation (n=4–6, P<0.01 vs. aspirin alone) but increased LTB₄ concentrations, resulting in 30-fold increase in the LTB₄/ATL ratio vs. rats treated with aspirin alone. In contrast, the gastric mucosa from rats treated with aspirin and licofelone displayed a fivefold increase in the LTB₄/ATL ratio (n=4–6, P<0.01 vs. rats treated with aspirin in combination with another COX inhibitor). These findings indicate that an increased LTB₄/ATL ratio in the gastric mucosa is associated with gastric damage.

CONCLUSIONS

Gastric mucosal injury caused by administration of NSAIDs is a neutrophil-driven process regulated by small lipid and protein signaling molecules. In the present study, we demonstrated that inhibition of ATL formation in conjunction with stimulation of LTB₄ synthesis mediates some of the in vivo impact of a COX inhibitor on aspirin-treated rats, a finding that provides new understanding of the role of COX-2- and 5-LOX-derived lipid mediators in the pathogenesis of "NSAID-gastropathy." However, in contrast to the classical hypothesis of "substrate shunting," the role of 5-LOX in the pathophysiological chain of events that leads to gastric injury in rats exposed to aspirin is largely dependent on the "state" of COX-2 activity. Interaction of aspirin with the two isoforms of COX differs in an important way from the interaction of most other NSAIDs with these enzymes. Aspirin covalently modifies, through acetylation, a serine residue near the active site of COX. In the case of COX-1, acetylation of a serine residue (Ser530) results in conformational changes in the enzyme such that it can no longer oxidize AA. In the case of COX-2, aspirin also acetylates a serine residue (Ser516), but COX-2 is still active and converts AA to 15-HETE, which carries its C15 alcohol in the R configuration. This molecule can be further metabolized by PMN and/or other 5-LOX expressing cells to 15-epi-LXA₄. Crucially, the biosynthesis of ATL does not arise from a metabolic shunt but represents the effect of aspirin on the oxygenating function of COX-2. However, when a second NSAID is coadministered with aspirin, complete inhibition of acetylated and nonacetylated COX-2 is achieved, 15(R)-HETE is not generated, and AA may be preferentially metabolized by gastric PMN via a 5-LOX-dependent pathway to form LTB₄. Because LTB₄ is a potent chemoattractant, its presence in the gastric microenvironment could further stimulate PMN activation and adherence to gastric endothelial cells. Supporting this scenario, we demonstrated that coadministration of selective and nonselective COX-2 inhibitor with aspirin exacerbates the gastric mucosal damage caused by aspirin. However, since the doses of aspirin used in this study fully inhibit COX-1 activity, exacerbation of gastric damage most likely reflects the ability of COX inhibitors to block the activity of acetylated COX-2 (i.e., ATL formation).

Given the crucial role played by 5-LOX products in the development of gastric lesions when aspirin is administered together with NSAIDs, we evaluated the effect of a dual COX/5-LOX inhibitor, licofelone, as well as the selective 5-LO inhibitor zileuton, on aspirin-induced gastric mucosal damage and gastric PMN recruitment. We demonstrated that administration of licofelone to rats preexposed to aspirin did not exacerbate acute mucosal damage nor did it interfere with gastric adaptation to aspirin. Furthermore, results from the intravital microscopy experiments demonstrated that licofelone at a dose of 100 mg/kg potently inhibited the PMN adherence to the endothelium of mesenteric venules induced by aspirin. At this dose, licofelone is an effective COX-1 and COX-2 inhibitor; since in contrast to licofelone, selective and nonselective COX inhibitors exacerbate the damage caused by aspirin, the protection afforded by this drug most likely is due to inhibition of 5-LOX activity. Indeed, whereas licofelone (similar to ketoprofen, indomethacin, and celecoxib) inhibits ATL formation, in sharp contrast, it also inhibited LTB₄ generation. The relevance of 5-LOX in the protection afforded by licofelone was further highlighted by results obtained using zileuton: treating rats with this compound protected against gastric damage caused by administration of aspirin in combination with another NSAID.

Adaptation of the gastric mucosa to aspirin is characterized by a marked reduction in mucosal injury after repeated administration of a damaging agent over several days. In the present study we have documented that adaptation to aspirin was abrogated by coadministration of a COX inhibitor but not by licofelone. Because development of gastric adaptation to aspirin is associated with enhanced ATL formation, whereas its reversal correlates with LTB₄ generation and ATL suppression, it appears that the LTB₄/ATL balance modulates adaptation to aspirin. Together with the finding that licofelone did not interfere with adaptation even though it inhibits ATL formation, these data strongly suggest that the relative abundance of ATL and LTB₄ in the gastric microenvironment dictates whether or not the mucosa will adapt to repeated aspirin administration.

Clinical trials suggest that selective COX-2 inhibitors produce severe gastrointestinal ulcer complications at about half the rate of conventional NSAIDs. However, such a safety advantage may be significantly abrogated if the patient treated with a selective COX-2 inhibitor is also using low-dose aspirin, such as for cardioprotection. The results of the present study might therefore have clinical relevance since they indicate that dual COX/5-LOX inhibitors represent a safe alternative to selective and nonselective COX-2 inhibitors in patients taking low doses of aspirin.

In conclusion, the results presented herein support the notion that a balance in the production of LTB₄ vs. ATL is involved in controlling both acute and chronic responses to aspirin. ATL protects the stomach via multilevel regulation of proinflammatory signals at the PMN/endothelial cell interface. We also provide evidence that licofelone, a dual COX/5-LOX inhibitor, did not exacerbate acute gastric damage caused by aspirin or interfere with gastric adaptation to aspirin.

View larger version (48K): [in this window] [in a new window]   Figure 2. Schematic diagram depicting the findings of this study. Aspirin administration inhibits gastric COX-1 activity and induces COX-2 in the gastric mucosa. Aspirin-acetylated COX-2 is unable to generate PG but is still able to carry on oxidative reactions generating ATL or 15-R-epi-LXA4. Addition of selective or nonselective COX-2 inhibitors blocks acetylated and nonacetylated COX-2, thereby inhibiting ATL formation. This is associated with exacerbation of acute gastric injury and increased formation of LTB4 in the gastric mucosa. The dual COX/5-LOX inhibitor licofelone prevents ATL formation but also inhibits LTB4 accumulation and therefore does not interfere with adaptation. The inset shows transcellular pathways involved in ATL/LTB4 formation at the neutrophil/endothelial interface in the gastric mucosa. Hypothetical shunts are shown in red and inhibitory steps in blue.

FOOTNOTES

¹ To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0777fje; to cite this article, use FASEB J. (April 22, 2003) 10.1096/fj.02-0777fje

This article has been cited by other articles:

				E. Rahme, M. Bardou, K. Dasgupta, Y. Toubouti, J. Ghosn, and A. N. Barkun Hospitalization for gastrointestinal bleeding associated with non-steroidal anti-inflammatory drugs among elderly patients using low-dose aspirin: a retrospective cohort study Rheumatology, February 1, 2007; 46(2): 265 - 272. [Abstract] [Full Text] [PDF]

				Y. Birnbaum, Y. Ye, Y. Lin, S. Y. Freeberg, S. P. Nishi, J. D. Martinez, M.-H. Huang, B. F. Uretsky, and J. R. Perez-Polo Augmentation of Myocardial Production of 15-Epi-Lipoxin-A4 by Pioglitazone and Atorvastatin in the Rat Circulation, August 29, 2006; 114(9): 929 - 935. [Abstract] [Full Text] [PDF]

				N. Chiang, E. A. Bermudez, P. M. Ridker, S. Hurwitz, and C. N. Serhan Aspirin triggers antiinflammatory 15-epi-lipoxin A4 and inhibits thromboxane in a randomized human trial PNAS, October 19, 2004; 101(42): 15178 - 15183. [Abstract] [Full Text] [PDF]

				T. Ohira, G. Bannenberg, M. Arita, M. Takahashi, Q. Ge, T. E. Van Dyke, G. L. Stahl, C. N. Serhan, and J. A. Badwey A Stable Aspirin-Triggered Lipoxin A4 Analog Blocks Phosphorylation of Leukocyte-Specific Protein 1 in Human Neutrophils J. Immunol., August 1, 2004; 173(3): 2091 - 2098. [Abstract] [Full Text] [PDF]

				J. L. Wallace, S. R. Zamuner, W. McKnight, M. Dicay, A. Mencarelli, P. del Soldato, and S. Fiorucci Aspirin, but not NO-releasing aspirin (NCX-4016), interacts with selective COX-2 inhibitors to aggravate gastric damage and inflammation Am J Physiol Gastrointest Liver Physiol, January 1, 2004; 286(1): G76 - G81. [Abstract] [Full Text] [PDF]

				S. Fiorucci, L. Santucci, J. L. Wallace, M. Sardina, M. Romano, P. del Soldato, and A. Morelli Interaction of a selective cyclooxygenase-2 inhibitor with aspirin and NO-releasing aspirin in the human gastric mucosa PNAS, September 16, 2003; 100(19): 10937 - 10941. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 17/9/1171 02-0777fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by FIORUCCI, S. Articles by WALLACE, J. L Search for Related Content PubMed PubMed Citation Articles by FIORUCCI, S. Articles by WALLACE, J. L