Characterization of an 8-Lipoxygenase Activity Induced by the Phorbol Ester Tumor Promoter 12-0-Tetradecanoylphorbol- 13-acetate in Mouse Skin in Viuo*

An enzymatic activity has been found in cytosolic preparations from mouse epidermis which catalyzes the formation of 8-hydroperoxyeicosatetraenoic acid/ 8-hydroxyeicosatetraenoic acid (8-HPETE/8-HETE) from arachidonate. In contrast to 12-lipoxygenase this enzyme activity was not detectable in normal (un- treated) mouse skin but only after in vivo treatment with the phorbol ester tumor promoter TPA (12-0- tetradecanoylphorbol- 13-acetate). The induction showed a maximum at 24 h after TPA treatment strictly depended on the age of the mice and the TPA dose and was prevented by cycloheximide. The pri- mary product formed from arachidonic acid was 8- HPETE, and the enzyme seems not to possess a signif-icant peroxidase activity. This result as well as studies with specific inhibitors and its cytosolic localization indicates this enzyme to be a member of the lipoxygen- ase family. Most of the 8-lipoxygenase activity is located in cells of the suprabasal compartment of the epidermis. In to be was activated by enzyme and showed a pH optimum at 7.5-8.0. of the as in in Epidermal cells trypsinization. as top of consisted of squamous and large granular cells. With increasing density of the fraction consisted of early granular and spinous cells and fractions 3 and 4 contained basal keratinocytes.

An enzymatic activity has been found in cytosolic preparations from mouse epidermis which catalyzes the formation of 8-hydroperoxyeicosatetraenoic acid/ 8-hydroxyeicosatetraenoic acid (8-HPETE/8-HETE) from arachidonate. In contrast to 12-lipoxygenase this enzyme activity was not detectable in normal (untreated) mouse skin but only after in vivo treatment with the phorbol ester tumor promoter TPA (12-0tetradecanoylphorbol-13-acetate). The induction showed a maximum at 24 h after TPA treatment strictly depended on the age of the mice and the TPA dose and was prevented by cycloheximide. The primary product formed from arachidonic acid was 8-HPETE, and the enzyme seems not to possess a significant peroxidase activity. This result as well as studies with specific inhibitors and its cytosolic localization indicates this enzyme to be a member of the lipoxygenase family. Most of the 8-lipoxygenase activity is located in cells of the suprabasal compartment of the epidermis.
In spite of being a cytosolic enzyme 8-lipoxygenase appeared to be lipophilic to some extent and was activated by lecithin. The enzyme did not require calcium ions or ATP and showed a pH optimum at 7.5-8.0. 8-HPETE/8-HETE levels in mouse epidermis in viuo were determined by gas chromatography-mass spectrometry and found to be strongly increased after phor-bo1 ester treatment, in agreement with the induction of 8-lipoxygenase observed.
Lipoxygenases represent a family of structurally and functionally related enzymes which catalyze the dioxygenation of polyenoic fatty acids carrying one or more 1,4-cis,cis-pentadiene systems (1-3). Using eicosa-5,8,11,14-tetraenoic acid (arachidonic acid) as substrate, stereospecific incorporation of molecular oxygen occurs at carbons 5, 8, 9, 11, 12, and 15 leading to the corresponding regioisomeric hydroperoxyeicosatetraenoic acids (HPETEs).' These products are short-lived and are transformed into various families of biologically active metabolites, including their reduced analogues the hydroxyeicosatetraenoic acids (HETEs,. Lipoxygenases with different regiospecificities have been detected in plant and animal tissues. Although being ubiquitous their distribution has been shown to exhibit some tissue specificity (2).
Lipoxygenase activities have also been detected in skin epidermis derived from various mammalian species. 12-Lipoxygenase has been partially characterized in epidermis of mice (4, 5), rats (6), guinea pigs (7), and humans (6,8) and in keratinocytes (9). Interestingly, large quantities of 12-HETE were detected in psoriatic epidermis as compared to normal tissue (10). Moreover, psoriatic scales were found to contain 12-(R)-HETE rather than 12-(S)-HETE (ll), the enantiomer formed by all 12-lipoxygenases characterized so far including those from homogenates of normal human (12) and mouse (13) epidermis. 12-(R)-HETE appears to be a product of an epidermal cytochrome P-450 monooxygenase rather than a 12-lipoxygenase (14). In contrast to the predominant 12lipoxygenase only very low 5-lipoxygenase activity was identified in epidermis from mice (9) and humans (15) generating 5-HETE and leukotriene B,. In addition, a 15-lipoxygenase activity has been shown to be expressed in human keratinocytes in uitro (6,16). In a previous paper we have demonstrated the generation of 8-HETE in cell-free preparations of mouse epidermis treated with phorbol ester tumor promoters such as 12-0-tetradecanoylphorbol-13-acetate (TPA) in uiuo but not in extracts of normal mouse epidermis (17,18). The present experiments were designed to characterize this pathway of epidermal arachidonate oxygenation, in order to describe the enzymatic properties of a novel 8-lipoxygenase and its inducibility by irritant and hyperplasiogenic stimuli of mouse epidermis as well as the ontogeny of this response.
Animals-Female NMRI mice (Deutsche Versuchstieranstalt, Hannover, F. R. G.) aged 1-10 days post natum or 7-weeks old ("adult") were used in all experiments. The animals were kept under an artificial day/night rhythm and were fed Altromin Standard food pellets (Altromin, Lage, F. R. G.) with sterile water available ad libitum. Phorbol esters or other chemicals were dissolved in 0.05 ml of acetone (for application to 1-to 10-day-old mice) or 0.1 ml of acetone (for application to adult animals) and topically applied onto either untreated (neonatal mice) or shaved (adult mice) back skin. Shaving of the back skin by electrical clippers was performed 3 days prior to treatment.
Determination of Protein-Determination of protein was carried out by the method of Lowry et al. (19).
Preparation of Subcellular Epidermal Fractions-Epidermis was prepared from frozen back skin of mice of different age (20) and homogenized on ice with a Polytron PT-10 homogenizer (setting 6, five strokes for 6 s each) in buffer containing 50 mM Tris, 2 mM EDTA, pH 8.0 (TE) or in TE buffer containing 100 pg/ml phosphatidylcholine. The homogenate was centrifuged at 3000 X g at 4 "C for 10 min. The supernatant was used for incubations or subjected to a second centrifugation at 105,000 X g at 4 "C for 60 min. The supernatant (cytosol) was decanted and used for incubations. The sediment (particulate fraction) was resuspended in the homogenization buffer, and aliquots were taken for incubations.
Lipoxygenuse Assay Conditions-Epidermis-derived homogenate, cytosol, and particulate fractions were used as a source of enzyme. Incubations were carried out at 37 "C after addition of 50 pl of homogenate (about 10 mg of protein/ml) or cytosol (about 5 mg of protein/ml) to 200 pl of TE buffer (50 mM Tris, 2 mM EDTA, pH 8.0) or TE/phosphatidylcholine buffer (50 mM Tris, 2 mM EDTA, 100 pg/ml phosphatidylcholine, pH 8.0) containing 20-40 p~ [l-"C] arachidonic acid for varying time periods. The incubations were terminated by acidification with formic acid to pH 3 and addition of 3 ml of ice-cold ethanol. To reduce the hydroperoxy acids to hydroxy acids, 5 pl of triphenylphosphite were added. A final ethanol concentration of 15% was obtained by adding 17 ml of H20. A HETEenriched fraction was obtained from this sample by solid-phase extraction according to Powell (21). The samples were applied to SEP PAK C-18 cartridges preconditioned with 20 ml of ethanol followed by 20 ml of water. After application of sample, columns were washed with 10 ml of 15% aqueous ethanol and 10 ml of HzO. Residual water was removed from the column by a gentle stream of argon for 3 0 s followed by elution with 10 ml of n-hexane. HETEs were then eluted in a 10-ml fraction of ethyl acetate, the recovery being 65-70% as measured by tracing tritiated HETE added to the ethanolic tissue preparation.
Lipid extract fractions obtained from SEP-PAK C-18 cartridges were concentrated in U~C U O at 25 "C, redissolved in 50 pl of n-hexane, and injected onto the column. The effluent was monitored either at 229 nm with a UV detector (Kontron model 430) or by determination of radioactivity using a radioactivity monitor (Berthold LB 505, Wildbad, F. R. G.), and 0.5-ml fractions were collected.
Determination of Tissue Leuels of HETEs by Gas Chromatography/ Mass Spectrometry-Epidermis was prepared from frozen back skin of neonatal and adult mice as published (20). Aliquots of 500 mg of 12-HETE (17). epidermis were disintegrated for 10 s at liquid nitrogen temperature using a dismembrator. The pulverized tissue was suspended in 2 ml of ice-cold ethanol, vortexed for 20 s, and centrifuged 10 min at 3000 X g and 4 "C. The pellet was discarded and aqueous 0.1 M ammonium formate buffer, pH 3.1, was added to the supernatant to obtain a final 15% (v/v) ethanol solution. A HETE-enriched fraction was obtained by solid-phase extraction as described above with a recovery of 65-70%. HETEs were converted to methyl esters with diazomethane in ether at 25 "C for 10 min. The solvents were removed under a gentle stream of nitrogen. Then 100 p1 of BSTFA + 1% TMCS were added, and the mixture was incubated for 60 min at 60 "C yielding the corresponding trimethsilyl ethers (22). Again, the sample was evaporated under a stream of nitrogen and redissolved in 100 pl of hexane. Aliquots of 1-2 ~1 of this solution were injected into the gas chromatograph. 12[carboxyl-'*02]HETE was used as internal standard. It was prepared by enzyme-catalyzed carboxyl oxygen exchange, according to published procedures (23), yielding a reaction product which contained >95 atom % "0 in the carboxyl group. The solution of the "0-labeled material was calibrated by stable isotope dilution using GC-electron-impact/MS and negative fast atom bombardment mass spectrometry against a solution of nonlabeled 12-(S)-HETE, the concentration of which had been determined by UV spectroscopy ( e = 23,000 at 237 nm).
A gas chromatograph type 5890 (Hewlett-Packard) equipped with a fused-silica capillary column (10 m, 0.25-mm inner diameter, 0.25pm film thickness, OV1, Macherey & Nagel, Duren, F. R. G.) interfaced with a double-focusing mass spectrometer type MAT 90 (Finnigan MAT, Bremen, F. R. G.) was used for GC-MS investigations. A temperature-programmable injection system (Gerstel KAS 2, Mulheim, F. R. G.) was used for sample injection in the splitless mode. A linear GC temperature program from 150 to 200 "C at a rate of 5 "C/ min was used. After each run the column was cleaned by heating to 300 "C for 5 min.
Preparation of Liposomes-800 p1 of lecithin (100 mg/ml phosphatidylcholine Sigma Type I11 E in hexane) were used for the preparation of liposomes. The solvent was removed in a gentle stream of argon. The lipid was suspended in 10 ml of 25 mM Tris-HC1, pH 7.7, containing 0.5% dimethyl sulfoxide using a sonifier (Branson B15). 8 ml of 25 mM Tris-HC1, pH 7.7, were added. The mixture was vortexed for 1 min, left at room temperature for 1 h, and centrifuged at 400,000 X g for 20 min. The pellet was resuspended in 10 ml of 25 mM Tris-HC1, 134 mM NaCl, pH 7.7.
15 microunits of 8-lipoxygenase were incubated for 3 min at 37 "C with 1 mg of liposomes in the presence or absence of Ca" in a total volume of 1 ml. The incubate was centrifuged for 20 min at 400,000 X g. The pellet was washed once with Tris buffer and resuspended in 0.3% sodium deoxycholate. The supernatants and resuspended pellets were assayed for lipoxygenase activity.
Separation and Fractionation of Epidermal Cells from Newborn Mouse Skin-For determination of 8-lipoxygenase activity in the individual layers of epidermis, 4-5-day-old mice were treated with 10 nmol of TPA for 24 h. After decapitation the back skin of the mice was dissected as described in detail in Ref. 24. Epidermal cells were obtained from the skin by trypsinization. The keratinocyte suspension was then centrifuged on a discontinuous Percoll density gradient, thus obtaining a separation into four fractions as described in Ref.
24. Fraction 1 on top of the gradient consisted mainly of squamous and large granular cells. With increasing density of the gradient, fraction 2 consisted of early granular and spinous cells and fractions 3 and 4 contained basal keratinocytes.

Qunntitation of 8-HETE in Cell-free Preparations and in
Mouse Epidermis in Viuo-In a preceding paper (17), we had shown that upon incubation with saturating concentrations of [l-'4C]arachidonic acid cell-free preparations of mouse epidermis (NMRI) generated 12-HETE and 8-HETE, the latter only when the skin had been pretreated in vivo with the phorbol ester TPA. Accordingly, the determination by quantitative GC/MS analysis of the HETE methyl esters trimethyl silyl ethers using 12[~arboxyl-'~O~]HETE as internal standard revealed a 50-150-fold elevation of 8-HETE formation in incubations with TPA-treated epidermal cytosol as compared with normal epidermal cytosol, whereas the content of 12-HETE was approximately equal in both incubations (Table I).
In order to prove the in vivo relevance of 8-HETE generation, the eicosanoid composition of murine epidermal tissue was investigated. GC/MS analysis (Table 11) demonstrated 8-HETE as the major HETE species in TPA-treated mouse epidermis in situ, whereas only small amounts could be detected in normal epidermis. However, this effect was strictly age-dependent; neonatal mice did not respond to TPA, whereas 3-and 6-day-old animals showed epidermal 8-HETE levels that were increased by factors of 5 and 7 upon phorbol ester treatment (Table 11). Moreover, topical treatment of normal and TPA-treated mice with arachidonic acid for 30 min prior to killing increased the tissue levels of 8-HETE in TPA-treated more than in normal epidermis corresponding to the low level of 8-HETE generating activity in normal mouse epidermis (data not shown). On the other hand, 12-HETE formation was not significantly modulated upon TPA treatment.
8-HETE Is the Product of 8-Lipoxygenase Activity-8-HETE could be either the product of a 8-lipoxygenase or that of a cytochrome P-450-dependent monooxygenase (25). Three lines of evidence suggest an 8-lipoxygenase to be the responsible enzyme in epidermis. First, 8-HETE generation was suppressed by typical lipoxygenase inhibitors such as eicosa-5.8.11.14-tetraynoid acid, nordihyroguaiaretic acid, and quercetin (26-28) but remained unaffected by inhibitors of monooxygenases such as naphtoflavone (29) (Table 111). Moreover, NADPH did not cause activation of 8-HETE formation   when applied in a dose range of IO-' to M in contrast to the findings with HETE-producing enzymes of the cytochrome P-450 family (29). Finally, HPLC analyses of unreduced extracts from incubation mixtures clearly demonstrated the presence of 8-HPETE, in addition to the more polar 8-HETE as identified by comigration with authentic standards, and the disappearance of 8-HPETE upon addition of reducing agents such as triphenylphosphine or trimethylphosphite to the extracts (Fig. 1). HPETEs are HETE precursors only in the dioxygenase(1ipoxygenase)-but not in monooxygenase P-450-catalyzed oxygenation of arachidonic acid.
Enzymatic Properties of Epidermal 8-Lipoxygenase-The enzyme activity was assayed at 37 "C, optimum pH, and an arachidonic acid concentration of 35 p~, corresponding to about five times K. The assay was validated by varying the relevant parameters. The product formation was proportional to the protein concentration up to 2 mg/ml when using cytosol in 5-min incubations. Beyond 4 mg of protein/ml, 8-HETE formation could not be further increased by the addition of more cytosol. The time course of 8-HETE formation (Fig. 2) showed that the product formation was linear for about 25 min under typical assay conditions (e.g. 0.34 mg of cytosolic protein/ml), leveling off at 45-60 min at a ratio of 8-HETE to arachidonic acid of 5 X By varying the concentration of arachidonic acid between 2 and 45 p~, the Michaelis constant for this substrate was determined to 8.0 p~ according to Lineweaver and Burk. The standard concentration in the lipoxygenase assay was chosen to be 35 pM; truly saturating concentrations would create experimental problems due to the lipophilicity of arachidonic acid. The pH dependence of 8-lipoxygenase from epidermal cytosol (Fig. 3) showed an optimum around pH 7.5-8.
The 8-lipoxygenase appears to be inhibited by divalent cations that are present in the cytosol since EDTA increased the activity. Dialyzed enzyme preparations were not susceptible to activation nor to inactivation by Ca2+ up to a concentration of 10 p~ which slightly inhibited enzyme activity. P-Mercaptoethanol was found to inhibit 8-lipoxygenase activity by two-thirds when present in the assay at a concentration of 2.0 mM. P-Mercaptoethanol in the homogenization buffer did not show this effect when the enzyme was subsequently assayed using mercaptane-free buffer. Glutathione also inhibited enzyme activity up to 50% after a 5-min incubation at a concentration of 200 pM.
Subcellular Distribution-Studies on the distribution of 8lipoxygenase in the cytosol, microsomal, and mitochondrial fractions showed 80% of the activity localized in the cytosol, i.e. the 105,000 X g supernatant. A smaller amount was equally distributed between the 12,000 X g (mitochondrial) pellet and the 105,000 x g (microsomal) pellet (Table IV). The subcellular distribution was not affected by the concentration of Ca2+ ions in the range of to M. Although 8-lipoxygenase appears to be a cytosolic enzyme, surprisingly high levels of activity were found in the skimmed lipid floating on the surface of the 105,000 x g supernatant. Addition of lecithin or CHAPS to the homogenization buffer was shown to increase the cytosolic 8-lipoxygenase activity in a dosedependent manner (Fig. 4). This effect is most probably attributable to an increased amount of enzyme extracted into the cytosol. In addition, phosphatidylcholine also elicited a  direct stimulatory effect on enzyme activity when added to the incubation mixture (Table V). An association of the enzyme with phosphatidylcholine micelles could not be found. Detergents including Triton X-100, Tween 20, or CHAPS decreased the enzyme activity at concentrations above 0.2% (data not shown).

TABLE IV
Intracellular localization and Ca2+ independence of 8-lipoxygenase actiuity 5 ml of epidermal homogenate was prepared from TPA-treated skin of 7-day-old mice as described under "Experimental Procedures" using 50 mM Tris-HC1, pH 7.5, with or without 10 mM CaC12 and centrifuged at 12,000 X g for 10 min at 4 "C. The pellet was resuspended in 50 mM Tris-HC1, 1 mM EDTA, and referred to as mitochondrial fraction. The supernatant was recentrifuged at 105,000 X g for 1 h a t 4 "C. The microsomal fraction was obtained by resuspending the pellet in 0.5 ml of Tris-EDTA, and the supernatant was defined cytosol. 50-200-jtl aliquots were taken for the lipoxygenase activity assay as described. The numbers are mean values from three experiments. The sum of the three numbers was defined as loo%, representing 90% (no Ca2+) and 78% (with Ca2+) of the activity in the homogenate. Effect of phosphatidylcholine on 8-HETE formation in epidermal cytosol. Epidermis from TPA-treated mice was homogenized in TE buffer containing 0-300 pg/ml phosphatidylcholine that had been dispersed using a Branson Sonifier. After centrifuging at 105,000 X g, the supernatant (black bars) and the lipid emulsion floating on top of the supernatant (white bars) were assayed separately. Incubations were performed as described in the legend of Fig.  2 for 5 min. 8-Lipoxygenase activity is expressed as multiples of the specific activity of cytosol obtained in the absence of phosphatidyl choline in the homogenization buffer (15 microunits/mg protein = 1) (mean values from three experiments with duplicate assays).
Localization of Enzyme Activity within the Epidermk-Mouse epidermis is a multilayered tissue consisting of four cell layers: the stratum basale, i.e. a monolayer of less differentiated replicating cells situated just above the dermis, the stratum spinosum, and the stratum granulosum both containing post-mitotic cells of different degrees of terminal differentiation and a stratified sheet of dead horny cells at the surface. By centrifugation using a discontinuous Percoll density gradient keratinocytes obtained from mouse epidermis can be subfractionated according to their state of differentiation (24). Using this approach, 8-lipoxygenase activity was found to be associated with suprabasal rather than basal keratinocytes (Fig. 5).  Induction of 8-Lipoxygenase-While 8-lipoxygenase activity is very low in untreated mouse skin, the enzyme can be induced by a single topical treatment with 10 nmol of the phorbol ester TPA dissolved in acetone. 3 h after treatment the activity became apparent, reached a maximum between 18 and 36 h (depending on the age of the animals), and disappeared after 3-5 days (Fig. 6). 3 nmol of TPA elicited a half-maximal response while a dose of 10 nmol was found sufficient to yield the maximum response (Fig. 7). This dose was therefore used in all experiments.
The inducibility of epidermal 8-lipoxygenase by TPA undergoes a dramatic change during ontogenesis. While newborn mice were found to be completely insensitive, 4-5-dayold mice became responsive to TPA (Fig. 8). Maximum inducibility was reached at the age of 6-7 days, followed by a subsequent drop to about one-third of the maximum, an effect which could at least partially be attributed to a shift in the induction kinetics with aging of the mice (Fig. 6). The induction of 8-lipoxygenase was found to depend on protein biosynthesis. Topical application of cycloheximide to T I 20 40 60 FIG. 6 . Induction of 8-lipoxygenase by TPA. Cytosol was prepared from the epidermis of 7-day-old mice (plain black circles) and 10-day-old mice (open circles) at defined intervals after topical treatment with 10 nmol of TPA. Aliquots were assayed for 8-lipoxygenase activity as described in the legend of Fig. 2. Each value represents the mean S.D. from three independent experiments performed in triplicate. mouse skin inhibited both epidermal protein synthesis and 8lipoxygenase activity to a similar degree (Table VI). The strongest inhibition was obtained when cycloheximide was applied prior to or concomitantly with TPA treatment, whereas the inhibitory effect of cycloheximide was significantly less pronounced when the drug was applied 6 h after TPA application.  Not determined DISCUSSION 8-HPETEIHETE Is the Product of an Epidermal 8-Lipoxygenase-The skin of various species displays potent activities of arachidonic acid-metabolizing enzymes including lipoxygenases (for review see Ref. 30). In addition to 5-, 12-, and 15-lipoxygenases reported as keratinocyte-derived enzyme activities, we have characterized an 8-HETE-generating activity which was originally described by us (17,18), and later on by others (31), as being due to 8-lipoxygenase activity. This enzyme activity is only barely detectable in normal skin but highly expressed in phorbol ester-treated hyperproliferative epidermis of various mouse strains (17,31). In this respect 8lipoxygenase activity is clearly different from 12-lipoxygenase which is constitutively expressed in mouse epidermis and not increased upon treatment with phorbol esters.
Evidence for the enzyme representing an 8-lipoxygenase rather than a P-450-dependent monooxygenase is provided by the identification of the intermediate hydroperoxide 8-8-Lipoxygenme Activity Induced by Phorbol Ester TPA in Mouse Skin in Vivo HPETE, the cytosolic localization of the enzyme activity, and inhibitor studies. In a recent elegant study, Hughes and Brash (13) demonstrated that the epidermal 8-HETE is almost exclusively the 8-(S)-hydroxy enantiomer and that its biosynthesis is associated with stereoselective removal of hydrogen from C-10 of the arachidonic acid molecule and antarafacial oxygen insertion at C-8. These data provide further evidence for the occurrence of a specific 8-(S)-lipoxygenase in phorbol ester-treated mouse skin. The cyclooxygenase inhibitor indomethacin (32) increased 8-HPETE/8-HETE formation in epidermal homogenates, indicating substrate competition between the cyclooxygenase and lipoxygenase pathways.
Only brief reports have been published on the enzymatic production of 8-HETE in other tissues and cell types such as human leukocytes (33), mouse and guinea pig intraperitoneal macrophages and monocytes (34), mouse and rat liver (25), rat kidney glomeruli (35), psoriatic skin (36), human tracheal cells (37), and human primary squameous carcinoma cells (38). Interestingly, the specific generation of 8-(R)-HETE catalyzed by lipoxygenases has been observed in corals (39,40) and in starfish oocytes (41). However, a more detailed characterization of the 8-lipoxygenase pathway of arachidonic acid metabolism has not been reported so far.
Enzymatic Properties-The distribution as well as the activity of the epidermal 8-lipoxygenase and its interaction with phospholipid (data not shown) were independent of the Ca2+ concentration in the homogenization buffer, thus characterizing 8-lipoxygenase as a cytosolic enzyme which is not translocated to membranes or activated by Ca2+ ions as observed for the 5-lipoxygenase of human leukocytes (42) and the 12lipoxygenase from rat platelets (43).
Lipoxygenases have been reported to require various stimulatory factors, some of which are still unidentified. Hogaboom et al. (44) observed an increase in recovery of 5-lipoxygenase activity after ammonium sulfate precipitation and gel filtration in the presence of 10-50 pg/ml of sonicated phosphatidyl choline. Our experiments revealed a stimulatory effect of phosphatidylcholine on 8-lipoxygenase due to both a more efficient extraction of the enzyme from cell homogenates into the homogenization buffer and to a direct effect on enzyme activity. The question arises whether lecithin acts by facilitating the uniform dispersion of hydrophobic substrates in an aqueous environment or interacts directly with the protein, for instance helping to maintain a hydrophobic enzyme in a more active conformation. On the other hand, micelles prepared from phosphatidylcholine did not exert a stimulatory effect on 8-lipoxygenase (data not shown) such as observed for 5-lipoxygenase (45).
8-HPETE/8-HETE formation leveled off at a ratio of 8-HETE to arachidonic acid of 5 x This may be due to a kind of suicide inactivation of the enzyme by peroxide species or free radicals resulting from arachidonate oxidation as reported for other lipoxygenases (46).
The substrate specificity of epidermal 8-lipoxygenase deserves further investigation, especially since the physiological substrates of lipoxygenases are still a matter of debate. In addition to arachidonic acid, other unsaturated fatty acids may serve as lipoxygenase substrates as well. Thus, in human and rat epidermis, linoleic acid, which is converted to 13hydroxyoctadecadienoic acid by a 15-lipoxygenase, appears to be the major substrate (6). Other possible substrates include phospholipids, i.e. unsaturated fatty acids esterified in the position sn-2 of diacylglycerol phosphodiesters, e.g. for the reticulocyte 15-lipoxygenase (47).
Induction of 8-Lipoxygenase by Phorbol Esters-The dosedependent induction of 8-lipoxygenase activity by TPA most probably proceeds via de novo synthesis of the enzyme as recently also has been detected for the 12-lipoxygenase in TPA-treated human erythroleukemia cells (48). The idea is easily at hand that the 8-lipoxygenase gene belongs to the group of phorbol ester-responsive genes which are thought to be regulated via a protein kinase C-dependent pathway. The remarkable ontogenetic variation of 8-lipoxygenase inducibility correlates with a similar pattern of postnatal development of other TPA effects such as the induction of ornithine decarboxylase activity and mitotic activity as well as of the conversion stage of skin carcinogenesis (49). The implications of this phenomenon, e.g. a possible incompetence in cellular signaling, as well as eventual postnatal changes in epidermal target cell populations for TPA, have not been elucidated. With respect to the latter it may be of interest that suprabasal epidermal cells were found to be most active in producing 8-HPETE/&HETE. This finding is supported by data on variation of arachidonic acid metabolism with terminal differentiation of mouse keratinocytes (50).
Tissue Levels and Putative Pathophysiological Function of 8-HPETEIHETE-The demonstration of an epidermal 8lipoxygenase activity in cell-free preparations from TPAtreated mouse epidermis showed an excellent correlation with corresponding in vivo data. Thus, only small amounts of 8-HETE (upon reduction of 8-HPETE/8-HETE) in untreated epidermis were found in situ. These values could not be increased upon application of arachidonic acid, ruling out the possibility of a lack of substrate being the cause for the low levels observed. Upon treatment with phorbol esters increased levels of 8-HETE could be demonstrated in situ in epidermis from mice treated on days 3 and 6 after birth but not in that from neonatal mice, again supporting the data obtained with cell-free preparations. These results suggest some (patho)physiological function of the 8-lipoxygenase pathway.
Transient or permanent up-regulation of arachidonic acid metabolism has been frequently found to be associated with inflammatory and regenerative processes and with benign or malignant hyperproliferative skin diseases such as psoriasis (51) or cancer (52). Pathogenic functions of eicosanoids in general and lipoxygenase-derived arachidonic acid metabolites in particular are furthermore indicated by their proinflammatory effects in skin (53) and other tissues, their elevated tissue levels associated with inflammatory skin diseases (10,30,51) and epidermal tumors: and the therapeutic effects of inhibitors of arachidonic acid metabolism on such diseases and on tumor development (30,54). Thus, 12-and 5-HETE as well as leukotriene B4 have been shown to exert leukotactic effects (53), to affect the vascular tone and permeability (55), and, in the case of 12-HETE and leukotriene B4, to induce epidermal hyperproliferation (51).
By using mouse strains differing in phorbol ester sensitivity (31) and by studying postnatal development of mice (see above), a clear correlation between the inducibility of 8lipoxygenase and the degree of the inflammatory and hyperplastic responses of mouse skin to phorbol ester treatment was observed. In particular, the conversion stage of skin carcinogenesis (56) has been found to depend on lipoxygenasecatalyzed arachidonic acid metabolism in epidermis. The same is true for the induction of chromosomal damage by converting tumor promoters which correlates with conversion (56). elucidation of the role of 8-lipoxygenase in skin (patho)-28. Nakadate, T., Aizu, F., Yamamoto, s., and Kato, R. (1985) sponding gene. This work is presently carried out in our