Cyclooxygenase-1-Coupled Prostaglandin Biosynthesis Constitutes an Essential Prerequisite for Skin Repair
2003; Elsevier BV; Volume: 120; Issue: 5 Linguagem: Inglês
10.1046/j.1523-1747.2003.12140.x
ISSN1523-1747
AutoresHeiko Kämpfer, Lutz Bräutigam, Gerd Geißlinger, Josef Pfeilschifter, Stefan L. Frank,
Tópico(s)Dermatologic Treatments and Research
ResumoThis investigation demonstrated a functional coupling between cyclooxygenase-1 (cox) and prostaglandin E2/D2 biosynthesis in murine skin repair. Cyclooxygenase-1 expression decreased transiently after excisional wounding, and this was followed by a marked fall in the rate of prostaglandin E2/D2 biosynthesis at the wound site. Expression of cyclooxygenase-1, prostaglandin synthases, and prostaglandin E2/D2 production were colocalized in new tissue at the margin of the wound. Although cyclooxygenase-2 expression was strongly induced in granulation tissue on injury, this isoform did not contribute to high prostaglandin E2/D2 concentrations in wounds. Accordingly, wound tissue from SC-560-treated mice (selective cyclooxygenase-1 inhibitor) and diclofenac-treated mice (nonselective cyclooxygenase inhibitor), but not celecoxib-treated mice (selective cyclooxygenase-2 inhibitor), and wound tissue from cyclooxygenase-1-deficient animals exhibited a severe loss of prostaglandin E2/D2 at the wound site, and this change was associated with an impairment in the normal wound morphology. Topically administered prostaglandin E2 (dinoprostone) was able to restore normal wound repair to diclofenac-treated mice. In contrast to the presence of an injury-induced cyclooxygenase-2, these data constitute strong evidence that cyclooxygenase-1-coupled prostaglandin E2/D2 biosynthesis has a central role in skin repair. This investigation demonstrated a functional coupling between cyclooxygenase-1 (cox) and prostaglandin E2/D2 biosynthesis in murine skin repair. Cyclooxygenase-1 expression decreased transiently after excisional wounding, and this was followed by a marked fall in the rate of prostaglandin E2/D2 biosynthesis at the wound site. Expression of cyclooxygenase-1, prostaglandin synthases, and prostaglandin E2/D2 production were colocalized in new tissue at the margin of the wound. Although cyclooxygenase-2 expression was strongly induced in granulation tissue on injury, this isoform did not contribute to high prostaglandin E2/D2 concentrations in wounds. Accordingly, wound tissue from SC-560-treated mice (selective cyclooxygenase-1 inhibitor) and diclofenac-treated mice (nonselective cyclooxygenase inhibitor), but not celecoxib-treated mice (selective cyclooxygenase-2 inhibitor), and wound tissue from cyclooxygenase-1-deficient animals exhibited a severe loss of prostaglandin E2/D2 at the wound site, and this change was associated with an impairment in the normal wound morphology. Topically administered prostaglandin E2 (dinoprostone) was able to restore normal wound repair to diclofenac-treated mice. In contrast to the presence of an injury-induced cyclooxygenase-2, these data constitute strong evidence that cyclooxygenase-1-coupled prostaglandin E2/D2 biosynthesis has a central role in skin repair. arachidonic acid cyclooxygenase liquid chromatography tandem mass spectrometry prostaglandin Nonsteroidal anti-inflammatory drugs (NSAID) are among the most widely used therapeutic agents. It has long been known that these drugs exert both their wanted and unwanted effects by inhibition of prostaglandin (PG) synthesis (Vane et al., 1994Vane J.R. Mitchell J.A. Appleton I. Tomlinson A. Bishop-Bailey D. Croxtall J. Willoughby D.A. Inducible isoforms of cyclooxygenase and nitric-oxide synthase in inflammation.Proc Natl Acad Sci USA. 1994; 91: 2046-2050Crossref PubMed Scopus (1017) Google Scholar). Accordingly, currently available NSAID are now well established to inhibit the PG synthesizing enzymes cyclooxygenase (COX)-1 and COX-2 (Meade et al., 1993Meade E.A. Smith W.L. DeWitt D.L. Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammatory drugs.J Biol Chem. 1993; 268: 6610-6614Abstract Full Text PDF PubMed Google Scholar). In general, COX-1 is assumed to represent the constitutively expressed isoform that is present in most cells and nearly all tissues (Crofford, 1997Crofford L.J. COX-1 and COX-2 tissue expression: Implications and predictions.J Rheum. 1997; 24: 15-19Google Scholar). In contrast, the COX-2 isoenzyme is not abundantly expressed but can be induced, especially in immune cells (Masferrer et al., 1990Masferrer J.L. Zweifel B.S. Seibert K. Needleman P. Selective regulation of cellular cyclooxygenase by dexamethasone and endotoxin in mice.J Clin Invest. 1990; 86: 1375-1379Crossref PubMed Scopus (378) Google Scholar;Lee et al., 1992Lee S.H. Soyoola E. Chanmugam P. et al.Selective expression of mitogen-inducible cyclooxygenase in macrophages stimulated with lipopolysaccharide.J Biol Chem. 1992; 267: 25934-25938Abstract Full Text PDF PubMed Google Scholar;Smith and DeWitt, 1996Smith W.L. DeWitt D.L. Prostaglandin endoperoxide H synthases-1 and -2.Adv Immunol. 1996; 62: 167-215Crossref PubMed Google Scholar). COX activity contributes to the maintenance of body homeostasis, where the inducible COX-2 in particular appears to play a key part in a series of pathologic conditions such as inflammation, pain, Alzheimer's disease, and cancer (Dubois et al., 1998Dubois R.N. Abramson S.B. Crofford L. Gupta R.A. Simon L.S. Van De Putte L.B. Lipsky P.E. Cyclooxygenase in biology and disease.FASEB J. 1998; 12: 1063-1073Crossref PubMed Scopus (2222) Google Scholar). The occurrence of gastrointestinal side-effects are major limitations in NSAID therapy, and because of this COX expression and activity have been extensively studied in models of gastric ulceration and healing (Stenson, 1997Stenson W.F. Cyclooxygenase 2 and wound healing in the stomach.Gastroenterology. 1997; 112: 645-648Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar). Remarkably, gastric damage was restricted to enzymatic inhibition of both COX isoenzymes (Wallace et al., 2000Wallace J.L. McKnight W. Reuter B.K. Vergnolle N. NSAID-induced gastric damage in rats: requirement for inhibition of both cyclooxygenase 1 and 2.Gastroenterology. 2000; 119: 706-714Abstract Full Text Full Text PDF PubMed Scopus (642) Google Scholar). As inflammation is central to the tissue repair processes, a relationship between wound healing and COX-mediated PG synthesis might be expected. Accordingly, COX-2 has been shown to promote ulcer healing (Mizuno et al., 1997Mizuno H. Sakamoto C. Matsuda K. et al.Induction of cyclooxygenase 2 in gastric mucosal lesions and its inhibition by the specific antagonist delays healing in mice.Gastroenterology. 1997; 112: 387-397Abstract Full Text Full Text PDF PubMed Scopus (607) Google Scholar;Wallace et al., 1998Wallace J.L. Bak A. McKnight W. Asfaha S. Sharkey K.A. MacNaughton W.K. Cyclooxygenase 1 contributes to inflammatory responses in rats and mice: Implications for gastrointestinal toxicity.Gastroenterology. 1998; 115: 101-109Abstract Full Text Full Text PDF PubMed Scopus (294) Google Scholar). The role of COX isoenzymes in inflammation, however, could differ markedly. The constitutive COX-1 is able to trigger inflammatory responses whereas the inducible COX-2 contributes to both the onset and decline of the inflammatory processes (Smith and Langenbach, 2001Smith W.L. Langenbach R. Why there are two cyclooxygenase isoforms.J Clin Invest. 2001; 107: 1491-1495Crossref PubMed Scopus (533) Google Scholar); however, functional coupling of arachidonic acid (AA) availability, the activity of COX isoforms, and subsequent PG biosynthesis appears to be under subtle regulatory control. Immediate PG synthesis initiated by COX-1 is favored by high amounts of AA. On the other hand, it has been suggested that involvement of COX-2 rather than COX-1 in delayed PG responses is because COX-2 is able to metabolize low concentrations of AA (Murakami et al., 1999Murakami M. Kambe T. Shimbara S. Kudo I. Functional coupling between various phospholipase A2s and cyclooxygenases in immediate and delayed prostanoid biosynthetic pathways.J Biol Chem. 1999; 274: 3103-3115Crossref PubMed Scopus (337) Google Scholar). Thus, there is evidence that the COX isoenzymes are functionally associated with distinct components of the PG synthase machinery within the cell. In recent studies, COX-1 was shown to be associated with cytosolic PGE2 synthase (cPGES) in immediate PGE2 synthesis (Tanioka et al., 2000Tanioka T. Nakatani Y. Semmyo N. Murakami M. Kudo I. Molecular identification of cytosolic prostaglandin E2 synthase that is functionally coupled with cyclooxygenase-1 in immediate prostaglandin E2 biosynthesis.J Biol Chem. 2000; 275: 32775-32782Crossref PubMed Scopus (628) Google Scholar), whereas the membrane-associated PGE2 synthase (mPGES) is functionally coupled to COX-2 in delayed responses (Murakami et al., 2000Murakami M. Naraba H. Tanioka T. et al.Regulation of prostaglandin E2 biosynthesis by inducible membrane-associated prostaglandin E2 synthase that acts in concert with cyclooxygenase-2.J Biol Chem. 2000; 275: 32783-32792Crossref PubMed Scopus (854) Google Scholar). Although a recent study implicated a role for PG in skin morphology and homeostasis (Neufang et al., 2001Neufang G. Fürstenberger G. Heidt M. Marks F. Müller-Decker K. Abnormal differentiation of epidermis in transgenic mice constitutively expressing cyclooxygenase-2 in skin.Proc Natl Acad Sci USA. 2001; 98: 7629-7634Crossref PubMed Scopus (184) Google Scholar), the regulation of COX isoenzyme expression and the availability and functional role of PG after skin injury remains unknown. We therefore examined the regulation and potential function of COX isoenzymes and the PGE2/D2 biosynthetic machinery in skin repair. Comparable with gastric ulcer healing, cutaneous wound healing involves a series of well co-ordinated tissue changes such as re-epithelialization, granulation tissue formation, and angiogenesis, which together ensure an efficient closure of the wound (Martin, 1997Martin P. Wound healing: Aiming for perfect skin regeneration.Science. 1997; 276: 75-81Crossref PubMed Scopus (3730) Google Scholar;Singer and Clark, 1999Singer A.J. Clark R.A.F. Cutaneous wound healing.N Engl J Med. 1999; 341: 738-746Crossref PubMed Scopus (4645) Google Scholar). It is most important that these tissue changes be closely regulated by the inflammation process and the inflammation process itself be tightly controlled with regard to initiation, maintenance, and resolution phases (Martin, 1997Martin P. Wound healing: Aiming for perfect skin regeneration.Science. 1997; 276: 75-81Crossref PubMed Scopus (3730) Google Scholar;Singer and Clark, 1999Singer A.J. Clark R.A.F. Cutaneous wound healing.N Engl J Med. 1999; 341: 738-746Crossref PubMed Scopus (4645) Google Scholar;Wetzler et al., 2000Wetzler C. Kämpfer H. Stallmeyer B. Pfeilschifter J. Frank S. Large and sustained induction of chemokines during impaired wound healing in the genetically diabetic mouse: Prolonged persistence of neutrophils and macrophages during the late phase of repair.J Invest Dermatol. 2000; 115: 245-253Crossref PubMed Scopus (430) Google Scholar). Accordingly, the participation of COX enzymes with the subsequent release of PG might also be involved in the control of inflammation in wounds and skin repair. Here we report evidence for tight coupling of COX-1 expression to PGE2/D2 levels in the region of wounds. Moreover, COX-1 is shown to be colocalized with mPGE and hematopoietic PGD synthase (hPGDS) expression at the wound margins, where highest PGE2/D2 concentrations were determined. Impairment of skin repair was demonstrated in COX-1-dependent loss of PGE2/D2 production in inhibitor-treated as well as in COX-1-deficient mice. Thus, this study provides evidence that COX-1 coupled PGE2/D2 biosynthesis plays a crucial part in the regulation of the cutaneous wound healing process. Female BALB/C, or C57BLKS mice were obtained from Charles River (Sulzfeld, Germany). Female B6; 129P2-Ptgs-1tm1 mice (Langenbach et al., 1995Langenbach R. Morham S.G. Tiano H.F. et al.Prostaglandin synthase 1 gene disruption in mice reduces arachidonic acid-induced inflammation and indomethacin-induced gastric ulceration.Cell. 1995; 83: 483-492Abstract Full Text PDF PubMed Scopus (1041) Google Scholar) were purchased from Taconic (Germantown, NY). Diclofenac (Sigma, Deisenhofen, Germany), SC-560 (Witega, Berlin, Germany) or celecoxib (Celebrex®, Pharmacia AG, Erlangen, Germany) were administered orally twice daily by gastrogavage (2.5 mg per kg per 12 h). For systemic treatment, BALB/C mice were injected intraperitoneally (i.p.) with 7.5 mg diclofenac per kg (Sigma) in 0.5 ml phosphate-buffered saline (PBS) twice daily during healing. For PGE2 treatment, BALB/C mice were treated topically twice daily with 8 μg per wound of dinoprostone (Minprostin®, Pharmacia) in triacetine/SiO2. Triacetine/SiO2 alone was used as a control. Full-thickness excisional wounding of mice was performed as described previously (Frank et al., 1999Frank S. Stallmeyer B. Kämpfer H. Kolb N. Pfeilschifter J. The function of nitric oxide in wound repair: Inhibition of inducible nitric oxide-synthase severely impairs wound reepithelialization.FASEB J. 1999; 13: 2002-2014Crossref PubMed Scopus (214) Google Scholar;Stallmeyer et al., 1999Stallmeyer B. Kämpfer H. Kolb N. Pfeilschifter J. Frank S. The function of nitric oxide in wound repair: Inhibition of inducible nitric oxide-synthase severely impairs wound reepithelialization.J Invest Dermatol. 1999; 113: 1090-1098Crossref PubMed Scopus (195) Google Scholar). For each experimental time point, tissue from four wounds each from four animals (n=16 wounds, RNA analysis) and from two wounds each from four animals (n=8 wounds, protein analysis) were combined and used for RNA and protein preparation. Nonwounded back skin from four animals served as a control. All animal experiments were carried out according to the guidelines and with the permission from the local government of Hessen (Germany). RNA isolation and RNase protection assays were carried out as described previously (Chomczynski and Sacchi, 1987Chomczynski P. Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction.Anal Biochem. 1987; 162: 156-159Crossref PubMed Scopus (63149) Google Scholar;Frank et al., 1999Frank S. Stallmeyer B. Kämpfer H. Kolb N. Pfeilschifter J. The function of nitric oxide in wound repair: Inhibition of inducible nitric oxide-synthase severely impairs wound reepithelialization.FASEB J. 1999; 13: 2002-2014Crossref PubMed Scopus (214) Google Scholar). The murine cDNA probes were cloned using reverse transcription–polymerase chain reaction. The probes corresponded to nucleotides (nt) 1682–1946 (for COX-1, accession no. M34141), nt 796–1063 (for COX-2, accession no. M64291), nt 223–475 (for mPGES, accession no. AB041997), nt 792–947 (for cPGES, accession no. AK007198), nt 286–551 [for lipocalin-type PGD synthase (lPGDS), accession no. AB006361], nt 416–643 (for hPGDS, accession no. D82072), nt 139–585 [for vascular endothelial growth factor (VEGF), accession no. S38083], nt 816–1481 (for lipocalin, accession no. X81627), nt 425 (exon1)–170 (exon 2) (for lysozyme M, accession no. M21047), nt 481–739 (for interleukin-1β, accession no. NM008361), nt 541–814 [for tumor necrosis factor (TNF)-α, accession no. NM013693] or nt 163–317 (for GAPDH, accession no. NM002046) of the published sequences. Wound and cell culture lysates were prepared as described previously (Kämpfer et al., 1999Kämpfer H. Kalina U. Mühl H. Pfeilschifter J. Frank S. Counterregulation of interleukin-18 mRNA and protein expression during cutaneous wound repair in mice.J Invest Dermatol. 1999; 113: 369-374Crossref PubMed Scopus (70) Google Scholar;Stallmeyer et al., 1999Stallmeyer B. Kämpfer H. Kolb N. Pfeilschifter J. Frank S. The function of nitric oxide in wound repair: Inhibition of inducible nitric oxide-synthase severely impairs wound reepithelialization.J Invest Dermatol. 1999; 113: 1090-1098Crossref PubMed Scopus (195) Google Scholar). Fifty micrograms of total protein from skin or cellular lysates was separated using sodium dodecyl sulfate–gel electrophoresis. COX-1, COX-2, and CD31 protein was detected using polyclonal antibodies (anti-COX-1 sc-7950, anti-COX-2 sc-1746, and anti-CD31 sc-1506 were from Santa Cruz, Heidelberg, Germany). Complete wounds were isolated from the back, bisected, and frozen in tissue freezing medium. Six micrometer frozen sections were subsequently analyzed using immunohistochemistry as described (Stallmeyer et al., 1999Stallmeyer B. Kämpfer H. Kolb N. Pfeilschifter J. Frank S. The function of nitric oxide in wound repair: Inhibition of inducible nitric oxide-synthase severely impairs wound reepithelialization.J Invest Dermatol. 1999; 113: 1090-1098Crossref PubMed Scopus (195) Google Scholar). Polyclonal anti-sera against murine COX-1, COX-2, and CD31 (Santa Cruz), Ki67 (Gerdes et al., 1984Gerdes J. Lemke H. Baisch H. Wacker H.H. Schwab U. Stein H. Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67.J Immunol. 1984; 133: 1710-1715PubMed Google Scholar) (Dianova, Hamburg, Germany), monoclonal anti-sera against murine F4/80 antigen (Serotec, Eching, Germany), and murine Gr-1 (Ly-6G) (Pharmingen, Hamburg, Germany) were used for immunodetection. Total protein [50 μg diluted in lysis buffer (Kämpfer et al., 1999Kämpfer H. Kalina U. Mühl H. Pfeilschifter J. Frank S. Counterregulation of interleukin-18 mRNA and protein expression during cutaneous wound repair in mice.J Invest Dermatol. 1999; 113: 369-374Crossref PubMed Scopus (70) Google Scholar) to a final volume of 50 μl] from skin lysates was subsequently analyzed for the presence of immunoreactive VEGF by enzyme-linked immunosorbent assay using the Quantikine murine VEGF kit (R&D Systems, Wiesbaden, Germany). Quiescent murine 3T3 fibroblasts were stimulated with 20% serum, or a combination of cytokines (2 nm interleukin-1β, 2 nm TNF-α, 100 U per ml interferon-γ). Murine RAW264.7 macrophages were stimulated with 10 μg per ml lipopolysaccharide/100 U per ml interferon-γ. Lipopolysaccharide was purchased from Sigma, serum was from Life Technologies (Karlsruhe, Germany), and cytokines were from Roche (Mannheim, Germany). The LC unit consisted of a Jasco DG 1580–53 degasser, a Jasco LG-1580-02 ternary gradient unit, a Jasco PU-1585 pump, and a Jasco AS 1550 autosampler (Gross-Umstadt, Germany). Detection was performed using a PE Sciex API 3000 triple quadrupol mass spectrometer (Applied Biosystems, Langen, Germany) equipped with a turbo ion spray interface. Nitrogen (high purity) was supplied by a Whatman nitrogen generator (Parker Hannifin GmBH, Kaarst, Germany). Blood plasma samples were obtained from treated animals 3 h following the last drug administration for determination of diclofenac, celecoxib, and SC-560 plasma levels. Diclofenac Two hundred milliliters of plasma and 400 μl 0.83 m phosphoric acid in 10% sodium chloride solution were mixed and extracted with 4 ml methylene chloride. After centrifugation the organic layer was evaporated and the residue reconstituted with 200 μl mobile phase. Nifluminic acid was used as internal standard. The column used was a Nucleosil C8 HD (Macherey-Nagel, Dueren, Germany). The ultraviolet detector was set at a wavelength of 282 nm. The electronic data were processed using Jasco Borwin software (Godbillon et al., 1985Godbillon J. Gauron S. Metayer J.P. High-performance liquid chromatography determination of diclofenac and its monohydroxylated metabolites in biological fluids.J Chromatogr. 1985; 338: 151-159Crossref PubMed Scopus (71) Google Scholar). Celecoxib Plasma samples were extracted using acetonitrile/water/ammonium hydroxide (85: 14.9: 0.1 v/v/v). The column used was a Nucleosil C18 (Macherey-Nagel). Celecoxib was detected by LC/MS/MS. The mass transitions used were m/z 380→316 and m/z 366→302 for celecoxib and the internal standard (4-[5-phenyl-3-(trifluoromethyl)-1H-pyrazol-1-yl]benzenesulfonamide), respectively. The analytical data were processed using Analyst software (version 1.1) (Bräutigam et al., 2001Bräutigam L. Vetter G. Tegeder I. Heinkele G. Geisslinger G. Determination of celecoxib in human plasma and rat microdialysis samples by liquid chromatography tandem mass spectrometry.J Chromatogr B Biomed Sci Appl. 2001; 761: 203-212Crossref PubMed Scopus (47) Google Scholar). SC-560 Plasma samples were extracted by solid-phase extraction, and SC-560 was subsequently analyzed by high performance liquid chromatography (Jasco) as described previously (Rose et al., 2000Rose M.J. Woolf E.J. Matuszewski B.K. Determination of celecoxib in human plasma by normal-phase high-performance liquid chromatography with column switching and ultraviolet absorbance detection.J Chromatogr B Biomed Sci Appl. 2000; 738: 377-385Crossref PubMed Scopus (69) Google Scholar;Grösch et al., 2001Grösch S. Tegeder I. Niederberger E. Bräutigam L. Geisslinger G. COX-2 independent induction of cell cycle arrest and apoptosis in colon cancer cells by the selective COX-2 inhibitor celecoxib.FASEB J. 2001; 15: 2742-2744Crossref PubMed Scopus (457) Google Scholar). Mice were wounded as described above. After killing the mice, the wounds were isolated from the backs of the animals and snap frozen in liquid nitrogen. Frozen wound tissue was homogenized in PBS for 30 s, lysates cleared by centrifugation (2000×g for 5 min) and 100 μl used to determine protein concentrations (Bradford). Two milliliters of the supernatants were used for the extraction of PG with 5 ml ethyl acetate and 20 μl formic acid for 15 min. The organic phase was separated by centrifugation (4000×g for 15 min) and the solvent evaporated. The pellet remaining was dissolved in 200 μl acetonitrile/water (80: 20 v/v). Chromatographic separation of extracted samples was performed in isocratic mode with a Nucleosil C18 column. The mass transitions used were m/z 351.2→315.1, 351.2→271.1, and 351.2→189.1. The analytical data were processed using Analyst software (version 1.1) (Bräutigam et al., 2001Bräutigam L. Vetter G. Tegeder I. Heinkele G. Geisslinger G. Determination of celecoxib in human plasma and rat microdialysis samples by liquid chromatography tandem mass spectrometry.J Chromatogr B Biomed Sci Appl. 2001; 761: 203-212Crossref PubMed Scopus (47) Google Scholar). As PGE2 and PGD2 are stereoisomers and thus resulted in the same fragments, it was not possible to resolve both PG by MS/MS. Levels of PGE2 and PGD2 in the wound are based on total wound protein. Data were shown as mean±SD. Data analysis was carried out using the unpaired Student's t test with raw data. Statistical comparison between more than two groups was carried out by analysis of variance (anova, Bonferroni t test). As the tightly controlled inflammation process in wounds is associated with skin regeneration (Martin, 1997Martin P. Wound healing: Aiming for perfect skin regeneration.Science. 1997; 276: 75-81Crossref PubMed Scopus (3730) Google Scholar;Singer and Clark, 1999Singer A.J. Clark R.A.F. Cutaneous wound healing.N Engl J Med. 1999; 341: 738-746Crossref PubMed Scopus (4645) Google Scholar;Wetzler et al., 2000Wetzler C. Kämpfer H. Stallmeyer B. Pfeilschifter J. Frank S. Large and sustained induction of chemokines during impaired wound healing in the genetically diabetic mouse: Prolonged persistence of neutrophils and macrophages during the late phase of repair.J Invest Dermatol. 2000; 115: 245-253Crossref PubMed Scopus (430) Google Scholar), we first investigated expression patterns of COX isoenzymes and various PG synthases in normal skin and in skin after injury. Nonwounded skin was characterized by constitutively expressed COX-1, whereas COX-2 expression was absent (Figure 1a–c). After wounding, COX-1 and COX-2 expression appeared to be regulated biphasically and in an opposite manner. COX-1 protein expression rapidly declined shortly after wounding (Figure 1a), but COX-2 protein was markedly induced at this stage (Figure 1b,c). Next, we directed our attention to immunolocalize both COX isoforms in normal skin and at the wound site (Figure 1d). COX-1 was localized at epithelial sites in nonwounded skin (Figure 1d, upper left panel) and at the margins of the wound (5 d after injury) (Figure 1d, upper right panel). Moreover, we detected COX-1 protein in fibroblast-shaped cells spotted within the granulation tissue (Figure 1d, small panel). COX-2 protein was observed in neutrophils early after wounding (data not shown), and was strongly expressed in macrophages and blood vessels located in the granulation tissue (5 d wound). Moreover, COX-2 was expressed in keratinocytes of the developing wound margin epithelia. Note the perinuclear staining of COX-2 protein in a number of keratinocytes (Figure 1d, middle and lower panel). Recent in vitro studies have provided strong evidence for a functional coupling between COX isoenzymes and various PG synthases (Murakami et al., 2000Murakami M. Naraba H. Tanioka T. et al.Regulation of prostaglandin E2 biosynthesis by inducible membrane-associated prostaglandin E2 synthase that acts in concert with cyclooxygenase-2.J Biol Chem. 2000; 275: 32783-32792Crossref PubMed Scopus (854) Google Scholar;Tanioka et al., 2000Tanioka T. Nakatani Y. Semmyo N. Murakami M. Kudo I. Molecular identification of cytosolic prostaglandin E2 synthase that is functionally coupled with cyclooxygenase-1 in immediate prostaglandin E2 biosynthesis.J Biol Chem. 2000; 275: 32775-32782Crossref PubMed Scopus (628) Google Scholar). We addressed this point, and observed an induction of the mPGES and the hPGDS after skin injury (Figure 1e), but the expression of cPGES and the lPGDS was clearly attenuated during the first 7 d of repair. Next, we determined wound PGE2/D2 levels as an index of the overall wound COX/PG synthase enzyme activity. In order to do this, we established a LC/MS/MS method for the specific determination of PGE2/D2 levels in the wound. In line with our observation that COX-1 is expressed constitutively (Figure 1a), nonwounded skin was characterized by constitutively present PGE2/D2 (Figure 2). Although COX-2 and mPGES/hPGDS were induced in injured skin (Figure 1b,c,e), PGE2/D2 levels were reduced dramatically by 75% when compared with those in normal skin (Figure 2). As skin repair proceeds, PGE2/D2 concentrations returned to initial levels. It is of interest that skin and also wound PGE2/D2 biosynthesis was closely correlated with the expression pattern of the COX-1 isoenzyme. In order to confirm a correlation between COX-1 and wound PGE2/D2 biosynthesis, we analyzed COX and PGES/PGDS as well as PGE2/D2 concentrations in distinct compartments of wound tissue. Tissue was taken from 5 d old wounds and divided into the “wound margin” (which contained part of the nonwounded epidermis and dermis, and more importantly the complete developing wound margin epithelia) and the “inner wound” (which contained the complete developing granulation tissue consisting predominantly of macrophages, fibroblasts, and endothelial cells) compartments. Determination of COX-1 and COX-2 isoenzyme expression revealed a marked difference in the distribution of the enzymes in the wound tissue, where COX-1 comprised 80% of the COX expression at the wound margins and COX-2 80% of the expression within the inner wound (granulation tissue) (Figure 3a). Interestingly, although mPGES and hPGDS have been shown to be temporally coinduced with COX-2 after injury (Figure 1b,e), we observed that mPGES and hPGDS expression was significantly colocalized with COX-1 at the wound margins, suggesting a coupling of both synthases to COX-1 rather than COX-2 (Figure 3b). Moreover, a functional coupling at the enzyme activity level could be established using LC/MS/MS analysis of PGE2/D2 production (Figure 3c). The wound margins thus constitute sites with the highest PGE2/D2 biosynthetic activity in the wound. It is notable that the high levels of inducible COX-2 in granulation tissue of the inner compartment of the wound were not accompanied by any substantial change in PGE2/D2 concentrations (Figure 3a,c). Thus, a decreased substrate availability in the presence of elevated COX-2 protein expression might also contribute to the observed low levels of PGE2/D2 within the granulation tissue; however, in vitro experiments with cell lines, i.e., the murine macrophage (RAW264.7) and fibroblast (NIH 3T3) cell line, representing two predominant cell types of wound granulation tissue, confirmed the absence of coupling between induced COX-2 and PGE2/D2 synthesis in skin repair in vivo. With the exception of mPGES, which is coinduced with COX-2 in cytokine-stimulated 3T3 fibroblasts, cPGES, hPGDS, and lPGDS were not coexpressed in vitro with COX-2 (Figure 3d). At this point of this study we had to address the possibility that the observed COX-1 coupled PGE2/D2 production has no functional role in the repair process and that it represents an epiphenomenon with no pivotal involvement. Thus, we analyzed the ability of PGE2/D2 to drive the repair process, as it could still be reasonable to recognize highly induced COX-2 as the key enzyme to produce PGH2 substratum for additional prostanoid biosynthetic pathways and production of important functional messengers in tissue repair. Thus, we treated mice either with the selective COX-1 inhibitor SC-560, the selective COX-2 inhibitor celecoxib or the nonselective COX inhibitor diclofenac in order to differentiate the COX-1- and COX-2-specific function during repair. Animals received oral doses of the agents approximately equivalent to those used therapeutically in humans (2.5 mg per kg per 12 h), and LC/MS/MS analysis showed good systemic availability of the drugs (Figure 4a). The COX-2 selective inhibitor celecoxib, however, produced no significant reduction in PGE2/D2 levels in the wounds compared with mock-treated control animals (Figure 4b, left panel). This observation was not unexpected,
Referência(s)