Stereoselective Biosynthesis of Hepoxilin B3 in Human Epidermis
2000; Elsevier BV; Volume: 114; Issue: 3 Linguagem: Inglês
10.1046/j.1523-1747.2000.00903.x
ISSN1523-1747
Autores Tópico(s)Plant Toxicity and Pharmacological Properties
ResumoWe previously reported that normal human epidermis forms 12-oxo-eicosatetraenoic acid and hepoxilin B3 as major eicosanoids and that hepoxilins and trioxilins are dramatically elevated in psoriatic lesions. We also observed that normal epidermis only synthesized one of the two possible 10-hydroxy- epimers of hepoxilin B3, suggesting its enzymatic origin. This study investigated the enzymatic pathways involved in the formation of hepoxilin B3 in human epidermis. Human epidermal fragments or cell fractions were incubated with [14C]-arachidonic acid or authentic 12(S)-hydroperoxyeicosatetraenoic acid. Products were analyzed by high-performance liquid chromatography, gas chromatography–mass spectrometry or a combination of both techniques. Esculetin and nordihydroguaiaretic acid inhibited formation of hepoxilin B3, 12-oxo-eicosatetraenoic acid, trioxilins, and 12-hydroxyeicosatetraenoic acid in a concentration-dependent manner. 12-Lipoxygenase activity was mainly located in the microsomal fraction (100,000 × g pellet) and 12-hydroxyeicosatetraenoic acid, hepoxilin B3, and 12-oxo-eicosatetraenoic acid were formed. The hepoxilin B3-synthesizing activity was not observed in subcellular fractions incubated with authentic 12(S)-hydroperoxyeicosatetraenoic acid, although it was located at least in the microsomal fraction when incubated with arachidonic acid. Similar results were obtained using preparations of recombinant platelet-type 12-lipoxygenase that yielded 12-oxo-eicosatetraenoic acid and hepoxilin B3 in addition to 12-hydroxyeicosatetraenoic acid, when incubated with arachidonic acid but not when incubated with 12-hydroperoxyeicosatetraenoic acid. Nevertheless, recombinant 12-lipoxygenase produced a lower ratio of 12-oxo-eicosatetraenoic acid and hepoxilin B3-12-hydroxyeicosatetraenoic acid than epidermis. Our results support the concept that 12-lipoxygenase catalyzes the formation of hepoxilin B3 and 12-oxo-eicosatetraenoic acid. We previously reported that normal human epidermis forms 12-oxo-eicosatetraenoic acid and hepoxilin B3 as major eicosanoids and that hepoxilins and trioxilins are dramatically elevated in psoriatic lesions. We also observed that normal epidermis only synthesized one of the two possible 10-hydroxy- epimers of hepoxilin B3, suggesting its enzymatic origin. This study investigated the enzymatic pathways involved in the formation of hepoxilin B3 in human epidermis. Human epidermal fragments or cell fractions were incubated with [14C]-arachidonic acid or authentic 12(S)-hydroperoxyeicosatetraenoic acid. Products were analyzed by high-performance liquid chromatography, gas chromatography–mass spectrometry or a combination of both techniques. Esculetin and nordihydroguaiaretic acid inhibited formation of hepoxilin B3, 12-oxo-eicosatetraenoic acid, trioxilins, and 12-hydroxyeicosatetraenoic acid in a concentration-dependent manner. 12-Lipoxygenase activity was mainly located in the microsomal fraction (100,000 × g pellet) and 12-hydroxyeicosatetraenoic acid, hepoxilin B3, and 12-oxo-eicosatetraenoic acid were formed. The hepoxilin B3-synthesizing activity was not observed in subcellular fractions incubated with authentic 12(S)-hydroperoxyeicosatetraenoic acid, although it was located at least in the microsomal fraction when incubated with arachidonic acid. Similar results were obtained using preparations of recombinant platelet-type 12-lipoxygenase that yielded 12-oxo-eicosatetraenoic acid and hepoxilin B3 in addition to 12-hydroxyeicosatetraenoic acid, when incubated with arachidonic acid but not when incubated with 12-hydroperoxyeicosatetraenoic acid. Nevertheless, recombinant 12-lipoxygenase produced a lower ratio of 12-oxo-eicosatetraenoic acid and hepoxilin B3-12-hydroxyeicosatetraenoic acid than epidermis. Our results support the concept that 12-lipoxygenase catalyzes the formation of hepoxilin B3 and 12-oxo-eicosatetraenoic acid. gas chromatography–mass spectrometry hydroxyoctadecadienoic acid hydroperoxyeicosatetraenoic acid hepoxilin A3 (8(R,S)-hydroxy-11(S),12(S)-epoxy-5,9,14,-eicosatrienoic acid) hepoxilin B3 (10(R,S)-hydroxy-11(S),12(S)-epoxy-5,8,14-eicosatrienoic acid) lipoxygenase 12-oxo-eicosatetraenoic acid reverse phase-high performance liquid chromatography straight phase-high performance liquid chromatography trihydroxy-eicosatrienoic acid. TrXA3, trioxilin A3 (8,11,12-trihydroxy-5,9,14-eicosatrienoic acid) trioxilin B3 (10,11,12-trihydroxy-5,8,14-eicosatrienoic acid) 12-Lipoxygenase (12-LO) is the major arachidonic acid (AA) oxygenation pathway in epidermal cells, with total product formation generally exceeding cyclooxygenase activity (Holtzman et al., 1989Holtzman M.J. Turk J. Pentland A. A regiospecific monooxygenase with novel stereopreference is the major pathway for arachidonic acid oxygenation in isolated epidermal cells.J Clin Invest. 1989; 84: 1446-1453Crossref PubMed Scopus (51) Google Scholar;Solá et al., 1992Solá J. Godessart N. Vila L. Puig L. de Moragas J.M. Epidermal cell-polymorphonuclear leucocyte cooperation in the formation of leukotriene B4 by transcellular biosynthesis.J Invest Dermatol. 1992; 98: 333-339Crossref PubMed Scopus (51) Google Scholar). Platelet-type 12-LO has been found to be the predominant isoenzyme expressed in human and murine skin epidermis (Takahashi et al., 1993Takahashi Y. Ramesh Reddy G. Ueda N. Yamamoto S. Arase S. Arachidonate 12-lipoxygenase of platelet-type in human epidermal cells.J Biol Chem. 1993; 268: 16443-16448Abstract Full Text PDF PubMed Google Scholar;Hussain et al., 1994Hussain H. Shornick L.P. Shannon V.R. Wilson J.D. Funk C.D. Pentland A.P. Holtzman M.J. Epidermis contains platelet-type 12-lipoxygenase that is overexpressed in germinal layer keratinocytes in psoriasis.Am J Physiol. 1994; 266: C243-C253PubMed Google Scholar;Krieg et al., 1995Krieg P. Kinzig A. Ress-Löschke M. et al.12-Lipoxygenase isoenzymes in mouse skin tumor development.Mol Carcinog. 1995; 14: 118-129Crossref PubMed Scopus (65) Google Scholar) and an ‘‘epidermal’’-type 12-LO, which functionally resembles the platelet-type 12-LO is also present in murine epidermis (Van Dijk et al., 1995Van Dijk K.W. Steketee K. Havekes L. Frants R. Hofker M. Genomic and cDNA cloning of a novel mouse lipoxygenase gene.Biochim Biophys Acta. 1995; 1259: 4-8Crossref PubMed Scopus (23) Google Scholar;Funk et al., 1996Funk C.D. Keeney D.S. Oliw E.H. Boeglin W.E. Brash A.R. Functional expression and cellular localization of a mouse epidermal lipoxygenase.J Biol Chem. 1996; 271: 23338-23344Crossref PubMed Scopus (81) Google Scholar;Kinzig et al., 1997Kinzig A. Fürstenberger G. Bürger F. et al.Murine epidermal lipoxygenase (Aloxe) encodes a 12-lipoxygenase isoform.FEBS Lett. 1997; 402: 162-166Abstract Full Text Full Text PDF PubMed Scopus (38) Google Scholar). The initial product of any LO is a hydroperoxide with a predominant S configuration (Hamberg and Samuelsson, 1974Hamberg M. Samuelsson B. Prostaglandin endoperoxides: novel transformations of arachidonic acid in human platelets.Proc Natl Acad Sci USA. 1974; 71: 3400-3404Crossref PubMed Scopus (1304) Google Scholar;Nugteren, 1975Nugteren D.H. Arachidonate lipoxygenase in blood platelets.Biochem Biophys Acta. 1975; 380: 299-307Crossref PubMed Scopus (437) Google Scholar). 12-Hydroperoxide-eicosatetraenoic acid (12-HPETE) formed from AA by the action of 12-LO is reduced by peroxidases to the corresponding hydroxide, 12-hydroxy-5,8,10,14-eicosatetraenoic acid (12-HETE). Consistently, 12-HETE is the most abundant eicosanoid present in skin inflammatory lesions such as in psoriasis (Hammarström et al., 1975Hammarström S. Hamberg M. Samuelsson B. Duell A.E. Stawiski M. Voorhees J.J. Increased concentrations of nonesterified arachidonic acid, 12L-hydroxy-5,8,10,14-eicosatetraenoic acid, prostaglandin E2, and prostaglandin F2α in epidermis of psoriasis.Proc Natl Acad Sci USA. 1975; 72: 5130-5134Crossref PubMed Scopus (482) Google Scholar;Camp et al., 1983Camp R.D.R. Mallet A.I. Woollard P.M. Brain S.D. Kobza Black A. Greaves M.W. The identification of hydroxy fatty acids in psoriatic skin.Prostaglandins. 1983; 26: 431-447Crossref PubMed Scopus (132) Google Scholar;Fogh et al., 1987Fogh K. Kiil J. Herlin T. Ternowitz Th. Kragballe K. Heterogeneous distribution of lipoxygenase products in psoriatic skin lesions.Arch Dermatol Res. 1987; 279: 504-511Crossref PubMed Scopus (46) Google Scholar). We previously reported that in addition to 12-HETE, normal human epidermis incubated with exogenous AA produces 12-oxo-eicosatetraenoic acid (12-oxo-ETE), and 8-hydroxy-11,12-epoxy-5,9,14-eicosatrienoic acid and 10-hydroxy-11,12-epoxy-5,8,14-eicosatrienoic acid, termed hepoxilin A3 (HxA3) and hepoxilin B3 (HxB3), respectively, which can be further converted into 8,11,12-trihydroxy-5,9,14-eicosatrienoic acid [trioxilin A3 (TrXA3)] and 8,9,12-trihydroxy-eicosatrienoic acid (8,9,12-THETrE), and 10,11,12-trihydroxy-5,8,14-eicosatrienoic acid [trioxilin B3 (TrXB3)], respectively (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar). Hepoxilins exert action on plasma permeability on skin (Laneuville et al., 1991Laneuville O. Corey E.J. Couture R. Pace-Asciak C.R. Hepoxilin A3 increases vascular permeability in the rat skin.Eicosanoids. 1991; 4: 95-97PubMed Google Scholar;Wang et al., 1996Wang M.M. Demin P.M. Pace-Asciak C.R. Epimer-specific actions of hepoxilins A3 and B3 on PAF- and bradykinin-evoked vascular permeability in the rat skin in vivo.Adv Exp Med Biol. 1996; 416: 239-241Crossref PubMed Google Scholar), induce a specific receptor-dependent Ca2+ mobilization from endogenous sources (Dho et al., 1990Dho S. Grinstein S. Corey E.J. Su W.G. Pace-Asciak C.R. Hepoxilin A3 induces changes in cytosolic calcium, intracellular pH and membrane potential in human neutrophils.Biochem J. 1990; 266: 63-68Crossref PubMed Scopus (70) Google Scholar;Laneuville et al., 1993Laneuville O. Reynaud D. Grinstein S. Nigam S. Pace-Asciak C.R. Hepoxilin A3 inhibits the rise in free intracellular calcium evoked by formyl-methionyl-leucyl-phenylalanine, platelet-activating factor and leukotriene B4.Biochem J. 1993; 295: 393-397Crossref PubMed Scopus (36) Google Scholar) and release AA and diacylglycerol (Nigam et al., 1993Nigam S. Müller S. Pace-Asciak C.R. Hepoxilins activate phospholipase D in the human neutrophil.Dev Oncol. 1993; 71: 249-252Google Scholar). Recently, we also observed increased levels of hepoxilins and trioxilins in the psoriatic scales (Antón et al., 1998Antón R. Puig L. Esgleyes T. de Moragas J.M. Vila L. Occurrence of hepoxilins and trioxilins in psoriatic lesions.J Invest Dermatol. 1998; 110: 303-310Crossref PubMed Scopus (28) Google Scholar). Hepoxilins are formed by a intramolecular rearrangement of 12-HPETE (Pace-Asciak, 1984Pace-Asciak C.R. Arachidonic acid epoxides. Demonstration through [18O]oxygen studies of an intramolecular transfer of the terminal hydroxyl group of (12S) -hydroperxyeicosa-5,8,10,14,-tetraenoic acid to form hydroxyepoxides.J Biol Chem. 1984; 259: 8332-8337Abstract Full Text PDF PubMed Google Scholar). In a previous study (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar) we observed that normal human epidermis only synthesized one of the two possible 10-hydroxy-epimers of HxB3, which suggested an enzymatic origin. Interestingly, whereas (±)HxA3 and (±)HxB3 are both active in enhancing the bradykinin-evoked permeability in skin, only 10(R)-HxB3, which is probably the epimer synthesized by normal epidermis (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar), stereospecifically enhances the vascular permeability evoked by intradermal injection of platelet-activating factor (Wang et al., 1996Wang M.M. Demin P.M. Pace-Asciak C.R. Epimer-specific actions of hepoxilins A3 and B3 on PAF- and bradykinin-evoked vascular permeability in the rat skin in vivo.Adv Exp Med Biol. 1996; 416: 239-241Crossref PubMed Google Scholar). Enzymatic conversion of 12(S)-HPETE into 8(R,S)-HxA3 has been found in the rat pineal gland (Reynaud et al., 1994Reynaud D. Demin P. Pace-Asciak C.R. Hepoxilin A3 formation in the rat pineal gland selectively utilizes (12S) -hydroperoxyeicosatetraenoic acid (HPETE), but not (12R) -HPETE.J Biol Chem. 1994; 269: 23976-23980Abstract Full Text PDF PubMed Google Scholar). Enzymatic biosynthesis of HxB3, however, has not been fully demonstrated sincePace-Asciak et al., 1993Pace-Asciak C.R. Reynaud D. Demin P. Enzymatic formation of hepoxilins A3 and B3.Biochem Biophys Res Commun. 1993; 197: 869-873Crossref PubMed Scopus (17) Google Scholar observed that formation of a small quantity of a racemic mixture of HxB3 from 12-HPETE by different rat tissues was abolished by tissue boiling. Nevertheless, further works performed by these authors led them to the conclusion that, in contrast with that which occurs with HxA3, HxB3 is a nonenzymatic product (reviewed inPace-Asciak et al., 1995aPace-Asciak C.R. Reynaud D. Demin P.M. Hepoxilins: a review on their enzymatic formation, metabolism and chemical synthesis.Lipids. 1995; 30: 107-114Crossref PubMed Scopus (44) Google Scholar,Pace-Asciak et al., 1995bPace-Asciak C.R. Reynaud D. Demin P. Mechanistic aspects of hepoxilin biosynthesis.J Lipid Med Cell Signal. 1995; 12: 307-311Crossref PubMed Scopus (9) Google Scholar). All this prompted us to extend our previous investigation (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar,Antón et al., 1998Antón R. Puig L. Esgleyes T. de Moragas J.M. Vila L. Occurrence of hepoxilins and trioxilins in psoriatic lesions.J Invest Dermatol. 1998; 110: 303-310Crossref PubMed Scopus (28) Google Scholar) on the enzymatic pathways involved in the formation and further transformations of HxB3. 1-[14C]-arachidonic acid ([14C]-AA) 55–58 mCi per mmol was supplied by Amersham Ibérica (Madrid, Spain). Esculetin, nordihydroguaiaretic acid, metyrapone, bifonazole, and clotrimazole were from Sigma-Aldrich Química S.A. (Alcobendas, Spain). Unlabeled AA and authentic 12-HPETE were from Cayman (Ann Arbor, MI). Authentic (±)HxB3 was from Cascade Biochem (Berkshire, U.K.). 9-anthryldiazomethane (ADAM) was from Serva (Heidelberg, Germany). Recombinant human platelet-type 12-LO (1300 U per mg protein) was from Oxford Biomedical Research (Oxford, MI). All high-performance liquid chromatography (HPLC) solvents were supplied by Scharlau S.A. (Barcelona, Spain). Epidermis was isolated from normal skin, obtained from plastic surgery, using the Liu and Karasek technique (Liu and Karasek, 1978Liu S.C. Karasek M. Isolation and growth of adult human epidermal keratinocytes in cell culture.J Invest Dermatol. 1978; 71: 157-162Crossref PubMed Scopus (214) Google Scholar). Fragments of fresh human epidermis, obtained as described previously (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar) were used immediately. Epidermal fragments were placed in an Eppendorf containing RPMI-1640 plus 1 mM CaCl2 in a ratio medium/tissue of 450 μl per 100 mg. Afterwards, 20 μl of an ethanolic solution of [14C]-AA or unlabeled AA, as required, were added to yield the indicated substrate concentration. Incubation was performed at 37°C for the indicated period of time, after which 1 M HCl to yield pH 2–3 followed by half a volume of cold methanol were added. Incubation mixtures were centrifuged immediately at 15,000 × g for 5 min and the supernatant was recovered. The pellet was washed with another half a volume of cold methanol and the methanolic extract was added to the previous supernatant. Samples were kept at -80°C until analysis. Chromatography for quantitative analysis of 12-LO derived compounds was performed by injecting the samples directly into the column without further manipulation. Chromatography was performed isocratically with a mixture of methanol/water/trifluoroacetic acid/triethylamine 75:25:0.1:0.05 pumped at 1 ml per min. The column (Ultrasphere-ODS, 5 μm diameter particle, 4.6 × 250 mm, Beckman, San Ramón, CA) was coupled on line with a radioactivity detector (Beckman-171) equipped with a liquid scintillation cell. Eluents were mixed with a scintillation cocktail pumped at 3 ml per min. When required, ultraviolet absorption was monitored by means of a diode array detector Beckman-168 coupled between the column and the radioactivity detector. Data from detectors were processed with a System Gold Software Beckman in a PC-computer. When collection of eluted material was required the radioactive detector was equipped with a solid scintillation cell. For the analysis of the HxB3 epimers ADAM derivatives were obtained. Acidic water was added to the HPLC fractions containing HxB3 (14–20 min) to achieve a ratio methanol/water 1:1 (pH 2–3) and were then extracted three times with half a volume of ether/hexane 1:1. Extracts were evaporated under a N2 stream until dryness. Samples, dissolved in 30 μl of MeOH, were mixed with 30 μl of 0.2% (wt/vol) ADAM in ethylacetate, and left in the dark stirring for 40 h at room temperature under N2 atmosphere. The reaction mixture was evaporated and dissolved in 60 μl of CH3CN/H2O 72/28. A fraction was then injected into the chromatograph through a 20 μl loop and analyzed as previously described (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar). To analyze all samples, the electron impact mode was used as previously described (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar). The GC column was a TRB-1 fused silica capillary column (15 m length, 0.25 mm i.d., 0.25 μm film thickness, Tracer analítica S.A, Barcelona, Spain). Epidermis fragments (1–2 g) were placed in 2 ml of 50 mM Tris–HCl, pH 7.4, containing protease inhibitors (10 μM pepstatin A, 10 μM leupeptin, 1 mM sodium metabisulfite, 1 mM benzamidine, 1 mM phenylmethylsulfonylfluoride, 1 mM ethylenediamine tetraacetic acid, and 1 mM ethyleneglycol-bis(β-aminoethyl ether)-N,N,N ′,N ′,-tetraacetic acid). Tissue was then disrupted mechanically with a Potter-Elvehjem placed in an ice-water bath. Non-lysed cells were eliminated by centrifugation at 600 × g for 15 min. The supernatant was immediately centrifuged at 8000 × g for 30 min, and the resulting supernatant at 100,000 × g for 90 min at 4°C. The 100,000 × g pellet, corresponding to the microsomal fraction, was suspended in 1.5 ml of the above-mentioned buffer. The protein content was measured by the method ofBradford, 1976Bradford M. A rapid and sensative method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding.Anal Biochem. 1976; 72: 248-254Crossref PubMed Scopus (204527) Google Scholar using the Bio-Rad Protein Assay (Bio-Rad Laboratories S.A., Madrid, Spain). Aliquots of 700 μl of cell fractions containing 5 mM CaCl2 and 1 mM MgCl2 were incubated with 100 μM [14C]-AA or 50 μM 12-HPETE at 37°C for 30 min. Incubations with AA were stopped as described for epidermal fragments, and those with 12(S)-HPETE were stopped with half a volume of MeOH containing 5 mg SnCl2 per ml (final pH 2.5–3.0). 12-LO derived eicosanoids were analyzed as described. To observe the formation of 12-oxo-ETE and HxB3 catalyzed by 12-LO, 50 U of human recombinant platelet-type 12-LO in 250 μl of Tris–HCl 50 mM pH 8 were incubated with 200 μM [14C]-AA or 50 μM 12(S)-HPETE at 37°C for 5 min. Incubations with AA were stopped with one volume of cold methanol, and those with 12(S)-HPETE were stopped with one volume of MeOH containing 5 mg SnCl2 per ml (final pH 2.5–3.0). 12-LO derived eicosanoids were analyzed by RP-HPLC and the corresponding HxB3 fraction was collected, ADAM-derivatized and analyzed for epimer characterization as described above. Four peaks corresponding to 12-LO derived eicosanoids were observed after RP-HPLC analysis when human epidermis was incubated with 100 μM[14C]-AA, the identity of which was confirmed by GC-MS analysis: trioxilins, HxB3, 12-oxo-ETE, and 12-HETE were detected. The 12-LO origin of these eicosanoids was demonstrated by the concentration dependent inhibition of their formation by nordihydroguaiaretic acid, an unspecific LO inhibitor, and esculetin, a 12-LO inhibitor (Figure 1). No inhibition of any of these compounds by metyrapone, bifonazole, or clotrimazole, P450 inhibitors, was observed (not shown). The progression curves of formation of these compounds, obtained by incubating fragments of human epidermis with 100 μM [14C]-AA for several periods of time, showed that whereas the levels of HxB3 and 12-oxo-ETE achieve a plateau after 10 min of substrate addition, a continuous increase in the formation of 12-HETE and trioxilins was observed (Figure 2).Figure 2Trioxilins, HxB3, 12-oxo-ETE, and 12-HETE were formed in a time-dependent manner. Fragments of human epidermis were incubated with 100 μM of [14C]-AA at 37°C for distinct periods of time and the radioactive 12-LO derived compounds were analyzed by RP-HPLC. The total activity of 12-LO expressed as the sum of all 12-LO derived peaks is also depicted. Points are the mean ± SD, n = 4.View Large Image Figure ViewerDownload (PPT) To observe the location of the ‘‘HxB3-synthase’’ activity, supernatant and pellet obtained after centrifugation of homogenate epidermis at 100,000 × g were incubated with 100 μM [14C]-AA, and labeled 12-LO-derived eicosanoids were analyzed. The pellet of the 100,000 × g centrifugation produced 12-HETE, HxB3, and 12-oxo-ETE, and trioxilins were almost undetectable (Figure 3). No eicosanoids were detected in the supernatant of the 100,000 × g, indicating that 12-LO activity was associated to the microsomal fraction and that ‘‘HxB3-synthase’’ activity was at least in this subfraction. Fragments of fresh human epidermis produced only one of the two possible 10-hydroxy-epimers of HxB3 when incubated with AA (Figure 4a), whereas boiled epidermis did not produce an appreciable amount of HxB3 (Figure 4b). When 100,000 × g pellet and supernatant were incubated with authentic 12(S)-HPETE (n = 4) we failed to find appreciable amounts of the ‘‘epidermis epimer’’ of HxB3. A small amount of racemic HxB3 was occasionally observed, but was also observed in the incubation with buffer alone, indicating its nonenzymatic origin (not shown).Figure 4Human epidermis enzymatically produced one of the two possible 10-hydroxy-epimers of HxB3 when incubated with AA. Representative chromatograms of the ADAM-derivative of HxB3 from: (A) a sample of human epidermis incubated with exogenous 100 μM AA; (B) boiled epidermis incubated with 100 μM AA; (C) authentic (±)HxB3. These experiments were repeated three times with essentially identical results. Chromatography was performed isocratically with acetonitrile/water 72:28 pumped at 1 ml per min.View Large Image Figure ViewerDownload (PPT) To explore the possibility that 12-LO catalyzes the transformation of AA into HxB3 and 12-oxo-ETE, recombinant human platelet-type 12-LO was incubated with [14C]-AA. Figure 5 shows that both 12-oxo-ETE and HxB3 were formed by recombinant 12-LO, but to a lesser proportion than epidermis when compared with 12-HETE. Nevertheless, we failed to obtain any product other than 12-HETE when 12-LO was incubated with 12(S)-HPETE. Metabolism of AA in human epidermis through the 12-LO pathway results in the formation of hepoxilins, 12-oxo-ETE, and several triols in addition to 12-HETE (Antón et al., 1995Antón R. Abián J. Vila L. Characterization of arachidonic acid metabolites through the 12-lipoxygenase pathway in human epidermis by high-performance liquid chromatography and gas chromatography/mass spectrometry.J Mass Spectrom Rapid Commun Mass Spectrom. 1995: S169-S182Google Scholar). A high amount of HxB3 is produced by whole human epidermis. Unlike the hemin-catalyzed formation of HxB3, human epidermis only produced one of the two possible 10-hydroxy epimers. Whereas recent reports indicate that HxA3 can be synthesized enzymatically from 12(S)-HPETE and not from the enantiomer R (Reynaud et al., 1994Reynaud D. Demin P. Pace-Asciak C.R. Hepoxilin A3 formation in the rat pineal gland selectively utilizes (12S) -hydroperoxyeicosatetraenoic acid (HPETE), but not (12R) -HPETE.J Biol Chem. 1994; 269: 23976-23980Abstract Full Text PDF PubMed Google Scholar), it has been stated that HxB3 synthesis from 12-HPETE involves a nonenzymatic process (Pace-Asciak et al., 1995aPace-Asciak C.R. Reynaud D. Demin P.M. Hepoxilins: a review on their enzymatic formation, metabolism and chemical synthesis.Lipids. 1995; 30: 107-114Crossref PubMed Scopus (44) Google Scholar,Pace-Asciak et al., 1995bPace-Asciak C.R. Reynaud D. Demin P. Mechanistic aspects of hepoxilin biosynthesis.J Lipid Med Cell Signal. 1995; 12: 307-311Crossref PubMed Scopus (9) Google Scholar). Our results provide new evidence that the synthesis of HxB3 is enzymatically catalyzed in human epidermis. This concept is supported by the following experimental evidence: (i) progression curves showed that HxB3 levels reached a plateau; (ii) HxB3 synthesis was product stereoselective; and (iii) despite the fact that HxB3 was not formed by subcellular fractions from 12(S)-HPETE and only 12-HETE was detected, it was formed from AA by the microsomal faction. A radical mechanism for LO catalysis involving iron is generally accepted (reviewed inYamamoto, 1991Yamamoto S. ‘‘Enzymatic’’ lipid peroxidation: Reactions of mammalian lipoxygenases.Free Radic Biol Med. 1991; 10: 149-159Crossref PubMed Scopus (107) Google Scholar). Most work investigating the nature of the iron in LO has been done on soybean LO. Mossbauer spectral analysis of soybean 15-LO enriched in 57Fe revealed that the iron appears to cycle between Fe2+ and Fe3+ (Funk and Carrol, 1990Funk M.O. Carrol R.T. Role of iron in lypoxygenase catalysis.J Am Chem Soc. 1990; 112: 5375-5376Crossref Scopus (57) Google Scholar). Hydrogen abstraction in the substrate is probably mediated by Fe3+-enzyme that undergoes reduction to Fe2+-enzyme, yielding the substrate radical (AA·). The abstraction of a hydrogen atom from a double allylic methylene group takes place in the molecule of substrate yielding a carbon centered radical, which tends to be stabilized by a molecular rearrangement to form conjugate dienes. A further reaction of the lipid radical with molecular oxygen gives a peroxyl radical (AAOO·). This is an enzyme-bound intermediate in the formation of the corresponding hydroperoxide. The peroxyl radical can either be transformed into a hydroperoxide (AAOOH) by picking up a hydrogen atom, or it can be added to a double bond to form cyclic peroxides, as occurs in the prostaglandin-endoperoxides. The 12-LO-catalyzed reaction with AA as a substrate could be represented as: E(Fe3+) + AA + O2→[E(Fe3+) - AA] + O2→[E(Fe2+) - AA·] + O2→[E(Fe2+) - AAOO·]→[E(Fe3+) - AAOOH]⇄E(Fe3+) + AAOOH The equilibrium [E(Fe2+) - AAOO·]→[E(Fe3+) - AAOOH] would usually tend to the right side due to the peroxidase-mediated or spontaneous conversion of 12-HPETE into 12-HETE. Nevertheless, if peroxidase activity is limited or 12-HPETE is in some way protected, 12-HPETE could accumulate and [E(Fe2+)-AAOO·] could be present in a high enough concentration to allow AAOO· to undergo alternative rearrangements. The exact configuration of C10 in the HxB3 produced by human epidermis was not directly determined, but chromatographic data of these compounds reported by other authors (Vasiljeva et al., 1993Vasiljeva L.L. Manukina T.A. Demin P.M. Lapitskaja M.A. Pivnitsky K.K. Synthesis, properties, and identification of epimeric hepoxilins (-) -(10R) -B3 and (+) -(10S) -B3.Tetrahedron. 1993; 49: 4099-4106Crossref Scopus (36) Google Scholar;Demin et al., 1994Demin P.M. Pivnitsky K.K. Vasiljeva L.L. Pace-Asciak C.R. Synthesis of Methyl [5,6,8,9,14,15–3H6]-hepoxilin B3 and its conversion into methyl [5,6,8,9,14,15–3H6]-hepoxilin A3.J Label Compounds Radiopharms. 1994; XXXIV: 221-230Crossref Scopus (16) Google Scholar;Reynaud et al., 1994Reynaud D. Demin P. Pace-Asciak C.R. Hepoxilin A3 formation in the rat pineal gland selectively utilizes (12S) -hydroperoxyeicosatetraenoic acid (HPETE), but not (12R) -HPETE.J Biol Chem. 1994; 269: 23976-23980Abstract Full Text PDF PubMed Google Scholar) show that 10(R)– is more polar than the 10(S)-epimer. GC-MS and HPLC data strongly suggested that human epidermis produces exclusively 10(R)-hydroxy-11(S),12(S)-epoxy-isomer of HxB3. Figure 6 depicts the proposed putative mechanism of the 10(R) epimer of HxB3 based on the 12(S) configuration of the peroxyl radical. We postulate the formation of cyclic 10,12-peroxide through the addition of the 12-peroxyl radical, oriented in the pro-R side, to the double bond at C10 leading to the formation of another free radical centered at C11. A further rearrangement that involves the homolytic fission of O–O bond of this intermediate and the pick up of an atom of hydrogen would exclusively yield the 10(R) epimer of HxB3. The ‘‘hydroperoxidase’’ activity of 12-LO has also been proposed as a mechanism for the formation of 12-oxododeca-5,8,10-trienoic acid from 12-HPETE by porcine leukocytes (Yamamoto, 1991Yamamoto S. ‘‘Enzymatic’’ lipid peroxidation: Reactions of mammalian lipoxygenases.Free Radic Biol Med. 1991; 10: 149-159Crossref PubMed Scopus (107) Google Scholar;Glasgow et al., 1986Glasgow W.C. Harris T.M. Brash A.R. A. short-chain aldehyde is a major lipoxygenase product in arachidonic acid-stimulated porcine leukocytes.J Biol Chem. 1986; 261: 200-204Abstract Full Text PDF PubMed Google Scholar). In addition, the formation of the homolog of 10(R)-epimer of HxB3 from 13-hydroperoxide of linoleic acid by soybean LO has been reported (Garssen et al., 1976Garssen G.J. Veldink G.A. Vliegenthart J.F.G. Boldingh J. The formation of threo-11-hydroxy-trans-12: 13-epoxy-9-cis-octadecenoic acid by enzymic isomerisation of 13-L-hydroperoxy-9-cis,11-trans-octadecadienoic acid by soybean lipoxygenase-1.Eur J Biochem. 1976; 62: 33-36Crossref PubMed Scopus (87) Google Scholar). Nevertheless, we did not observe products other than 12-HETE when epidermis fractions or recombinant platelet-type 12-LO were incubated with 12-HPETE. In contrast, hematin and hemoglobin catalyze the formation of both 10(S) and 10(R) epimers of HxB3 from 12-HPETE (Pace-Asciak, 1984Pace-Asciak C.R. Arachidonic acid epoxides. Demonstration through [18O]oxygen studies of an intramolecular transfer of the terminal hydroxyl group of (12S) -hydroperxyeicosa-5,8,10,14,-tetraenoic acid to form hydroxyepoxides.J Biol Chem. 1984; 259: 8332-8337Abstract Full Text PDF PubMed Google Scholar) and the corresponding homologs from 13-hydroperoxy-linoleate (Hamberg, 1975Hamberg M. Decomposition of unsaturated fatty acid hydroperoxides by hemoglobin: structures of major products of 13L-hydroperoxy-9,11-octadecadienoic acid.Lipids. 1975; 10: 87-92Crossref PubMed Scopus (141) Google Scholar;Dix and Marnett, 1983Dix T.A. Marnett L.J. Hematin-catalyzed rearrangement of hydroperoxylinoleic acid to epoxy alcohols via an oxygen rebound.J Am Chem Soc. 1983; 105: 7001-7002Crossref Scopus (37) Google Scholar). This supports the hypothesis that epidermal HxB3 could be formed stereoselectively by rearrangement of the intermediate peroxyl radical at C12 in the bulk of 12-LO. 12-peroxyl radical could also pick up the hydrogen atom at C12 to form hydroperoxide and a C12 centered radical followed by the homolytic fission of O–O bond catalyzed by 12-LO by a ‘‘Fenton-like’’ reaction to yield 12-oxo-ETE. The putative mechanism that would yield 12-oxo-ETE by 12-LO is also depicted in Figure 6. Whether or not 12-peroxyl radical could be formed from 12-HPETE is not clear from our results. Apparently, peroxyl radicals were not generated in a sufficient amount to form 12-oxo-ETE and HxB3 from 12-HPETE when added exogenously to either cell fractions or recombinant platelet-type 12-LO. In contrast, when incubations of 100,000 × g or recombinant 12-LO were performed with exogenous AA, 12-HPETE was formed in situ in the bulk of 12-LO yielding HxB3 and 12-oxo-ETE in addition to 12-HETE. This concept is also supported by the fact that recombinant platelet-type 12-LO produced both 12-oxo-ETE and the epidermis epimer of HxB3 from AA. Differentiated and cornified epidermis are rich in proteins (keratins) and lipids (ceramides, phosphoglycerides, etc.). The enzyme environment in epidermis fragments differed from that in recombinant human platelet-type 12-LO preparations, as in the latter case the reaction occurred in an almost homogeneous aqueous phase, whereas in the former case the reaction occurred in a lipid-rich phase. Moreover, when we performed incubations of AA with epidermal cell suspensions formation of HxB3 and 12-oxo-ETE was very low and the ratio of HxB3 and 12-oxo-ETE to 12-HETE resembled that obtained from the recombinant human platelet-type 12-LO experiments rather than that obtained from the experiments using fragments of epidermis (not shown). The rate of transformation of 12-peroxyl radical into HxB3 and 12-oxo-ETE instead of 12-H(p)ETE was higher when keratinocytes (hence the enzyme) were ‘‘included’’ in a hydrated lipid–protein phase than when they were surrounded by water. This reasoning could be particularly relevant to explain results when 12-HPETE was added exogenously to preparations of recombinant human platelet-type 12-LO or cell fractions. In these incubations 12(S)-HPETE could be rapidly transformed into 12-HETE without the opportunity to reach the active site to generate 12-peroxyl radicals that further yield HxB3 and 12-oxo-ETE (see Figure 6). Support for the potential role of hepoxilins in the pathogenesis of inflammatory skin diseases, includes their potent action on plasma permeability when injected subcutaneously (Laneuville et al., 1991Laneuville O. Corey E.J. Couture R. Pace-Asciak C.R. Hepoxilin A3 increases vascular permeability in the rat skin.Eicosanoids. 1991; 4: 95-97PubMed Google Scholar;Wang et al., 1996Wang M.M. Demin P.M. Pace-Asciak C.R. Epimer-specific actions of hepoxilins A3 and B3 on PAF- and bradykinin-evoked vascular permeability in the rat skin in vivo.Adv Exp Med Biol. 1996; 416: 239-241Crossref PubMed Google Scholar) and the detection of a considerable amount in psoriatic lesions (Antón et al., 1998Antón R. Puig L. Esgleyes T. de Moragas J.M. Vila L. Occurrence of hepoxilins and trioxilins in psoriatic lesions.J Invest Dermatol. 1998; 110: 303-310Crossref PubMed Scopus (28) Google Scholar). The effect of HxA3 on Ca2+ mobilization has been demonstrated in neutrophils and whether or not this effect also occurs in epidermal cells should be the subject of further investigation. Hepoxilins could play an autocrine part as intracellular messengers and a paracrine role modulating leukocyte activation (reviewed inPace-Asciak, 1994Pace-Asciak C.R. Hepoxilins: a review on their cellular actions.Biochim Biophys Acta. 1994; 1215: 1-8Crossref PubMed Scopus (53) Google Scholar). The literature available on the physiologic role of HxB3 is limited. HxB3, however, is elevated in psoriatic lesions (Antón et al., 1998Antón R. Puig L. Esgleyes T. de Moragas J.M. Vila L. Occurrence of hepoxilins and trioxilins in psoriatic lesions.J Invest Dermatol. 1998; 110: 303-310Crossref PubMed Scopus (28) Google Scholar) andWang et al., 1996Wang M.M. Demin P.M. Pace-Asciak C.R. Epimer-specific actions of hepoxilins A3 and B3 on PAF- and bradykinin-evoked vascular permeability in the rat skin in vivo.Adv Exp Med Biol. 1996; 416: 239-241Crossref PubMed Google Scholar have observed that whereas HxA3 and HxB3 are both active in enhancing the bradykinin-evoked permeability in skin, only 10(R)-HxB3 stereospecifically enhances the vascular permeability evoked by intradermal injection of platelet-activating factor. The biologic role of these compounds on dermatoses is presently under research in our laboratory. This work was supported by grants from the Institut de Recerca of the HSCSP, DGICYT PM92–0183. The authors wish to thank the staff of Clínica Planas, Barcelona, for their contribution of human skin specimens, and also Cristina Gerbolés and Montserrat García for their technical assistance.
Referência(s)