Identification and subcellular localization of leukotriene A4-hydrolase activity in human epidermis

JOU R KA L OF

ELSEVIER

Journal

of Dermatological

Dermatological Science

Science 7 t 1994) 191-201

Identification and subcellular localization of leukotriene A,-hydrolase activity in human epidermis Lars Iversen* a Vincent A. Zibohb, Takao Shimizud, Nobuya Ohishi”, Olof Ridmark’, Anders Wetterholmc, Knud Kragballe” qf

UDepartment

Dermatology.

Marselisborg

hDepartmenr

‘Departmentof ‘Department

(Received

Hospital,

of Dermafology, Physiological

of Biochemistry,

8 September

Chemistry.

FaraIr?

1993; revision

Universiry of Aarhus,

University

qf

California.

Karolinska

of Medicine,

DK-8000

Instituret,

The linirersiry

received 15 December

Aarhus

C. Denmark

Davis, C.4, US.4 Stockholm, of Tokyo,

1993: accepted

Sweden Tokyo, Japan

17 January

19941

Abstract

The purpose of this study was to determine whether normal human epidermis could produce leukotriene B, (LTB,) from leukotriene A., (LTA,) ex vivo. and to localize this LTA,-hydrolase activity. Epidermis obtained by suction blister technique incubated with human polymorphonuclear cells. resulted in a 54% increase in LTB, formation when compared to polymorphonuclear cells incubated alone. Furthermore. human epidermis transformed exogenous LTA4 into LTB,, and this reaction obeyed Michaelis-Menten kinetics with an apparent K,,, of 6 PM. Subcellular fractionation of homogenized epidermis localized the LTA,-hydrolase activity mainly in the 105 000 x g supernatant fraction (cytoplasmic fraction). This activity was inhibited by two inhibitors of LTA,-hydrolase (bestatin and captopril). Western blot analysis of the 105 000 x g fraction of homogenized epidermis and cultured keratinocytes supported the presence of a LTA,-hydrolase. Thus, normal human epidermis possesses LTA,-hydrolase activity which can transform exogenous LTA, and polymorphonuclear cell-derived LTA, into LTB,. The identification of LTA,-hydrolase in the cytoplasmic fraction of human epidermis indicates that epidermal cells may play a more active role in the enzymatic process leading to formation of the proinflammatory compound LTB, than previously expected. Kej’ rr,ords: Leukotriene

A,-hydrolase;

Human epidermis;

* Corresponding author. Abbreviations: AA. arachidonic dihydroxy-7,9-frans-I

acid; 5.6-DiHETE, 5iS),6(R)I ,14-cis-eicosatetraenoic acid; ECL, enhanc-

ed chemiluminescence: HETE, hydroxyeicosatetraenoic acid; HSA, human serum albumin: HTAB. hexadecyltrimethylammonium bromide: KCs. keratinocytes; KGM. keratinocyte growth medmm: S-LO. Wipoxygenase; LT. leukotriene; MeOH. methanol; PBS. phosphate-buffered saline (Ca” and Mg’+ free phosphate buffer. pH 7.4); PMN. polymorphonuclear leukocyte(s); RIA. radioimmunoassay: RP-HPLC. reversed-phase high performance liquid chromatography.

Transcellular

metabolism

1. Introduction B4 (LTBJ is formed via the 5(5LO) pathway [ 11from arachidonic acid (AA). which is released from the cell-membrane phospholipids by phospholipase Al [2]. The initial step in the 5-LO pathway is the formation of S-HPETE, which is further transformed into LTA4 [3,4]. LTA4 can be enzymatically converted Leukotriene

lipoxygenase

0923-181 l/941$07.00 0 1994 Elsevier Science Ireland Ltd. All rights reserved. SSDI 0923- I8 I I (94)00280-R

192

to LTB, (by LTA4-hydrolase), either in the cell of origin or after transfer to another cell [5,6], i.e. by transcellular metabolism. In addition to these enzymatic transformations, LTA4 is also nonenzymatically hydrolyzed into 6-trans-LTB4 and 12-epi-6-trans-LTB4 [1,7] and two isomers of 5,6DiHETE [8,9]. Furthermore, LTA4 can be transformed into one isomer of 5,6-DiHETE by a cytosolic epoxide hydrolase [8,9]. The concept of transcellular metabolism of AA metabolites has recently generated much interest. Interaction of monocytes-lymphocytes [lo] and polymorphonuclear cells (PMNs) with other cells such as endothelial cells and erythrocytes [ 11,121 have all been reported to result in the transformation of LTA4 into LTB4, by cells not possessing 5LO activity. Recently, we and others reported that co-incubation of cultured human keratinocytes and human PMNs resulted in an increased LTB4 formation compared to each cell type incubated separately [13,14]. Cultured keratinocytes alone did not generate detectable amounts of LTB4. Also, LTA4 incubated with cultured human keratinocytes was transformed into LTB4 indicating the existence of a keratinocyte LTA4-hydrolase [ 13,141. LTArhydrolase has been purified from several sources [15]. The amino acid sequence of the human enzyme has been deduced from cDNA clones isolated from spleen and human lung cDNA libraries [ 16,171. Furthermore, immunological quantitation and immunochemical localization of the LTAd-hydrolase has been carried out for different guinea pig tissues [18,19], as well as for some human tissues [20]. Despite the detection of LTB4 in lesions of psoriasis [21-231, the presence of LTA4-hydrolase in human epidermis has not been investigated. As an extension of our recent study, in which LTA4-hydrolase activity was demonstrated in human cultured keratinocytes [13], we present evidence that LTA,hydrolase activity is present in human epidermis, and that it is localized in the cytosolic fraction. 2. Materials and methods 2. I. Materials LTA,-methyl ester was obtained from Cascade Biochem Limited, Reading, UK. All the HPLC

L. Iversen et al. /J.

Dermarol. Sci. 7 (1994)

191-201

graded organic solvents were from Merck, Darmstadt, Germany. Trizma base, hexadecyltrimethylammonium bromide (HTAB), o-dianisidine dihydrochloride, bestatin, captopril, 3,3’-diaminobenzidine chloride and calcium ionophore (A 23 187) were obtained from Sigma Chemicals. Authentic standards of LTB4, 6-trans-LTB4, 12-epi-6- transLTB.+ 12-hydroxyeicosatetraenoic acid (1ZHETE) and 15-hydroxyeicosatetraenoic acid (15-HETE) were obtained from Cayman Chemical, European Division, Paris, France. 5,6-DiHETE was from Biomol Research Laboratories, PA, USA. Fatty acid-free albumin was from Novo Nordisk, Bagsvaerd, Denmark. The protein assay kit was obtained from Bio-Rad, Munich, Germany. The low calcium keratinocyte growth medium (KGM) (Gibco, Life Technologies A/S, Roskilde, Denmark) was fatty acid- and serum-free. The radioimmunoassay (RIA) kit for LTB4, the hyperfilm enhanced chemiluminescence (ECL), the hybond ECL nitrocellulose membrane and the ECL Western blotting detection reagents were purchased from Amersham (Buckinghamshire, UK). Peroxidase conjugated affinity isolated anti rabbit, goat immunoglobulins were obtained from Dakopatts A/S Denmark. 2.2 Polymorphonuclear cell preparation Blood was obtained from the same healthy human volunteers who provided suction blister skin. Polymorphonuclear cells (PMNs) were prepared from the collected blood as previously described [24]. The isolated cells were suspended in Tris-HCl buffer, pH 7.5 (containing 0.87 mM CaClJ or in 50 mM potassium phosphate buffer pH 6.0 containing 0.5% HTAB. 2.3. Cultures of human keratinocytes Cultures of normal human keratinocytes were prepared as previously described [13]. Briefly, the cells were grown in low calcium, serum free keratinocyte growth medium. The medium was changed 3 times weekly. When the primary cell cultures were sub-confluent the cells were released by scraping in Tris-HCl buffer (pH 7.5) and then treated as indicated below. 2.4. Preparation of epidermis by suction blister

To obtain viable epidermis, a suction cup of acryl-

193

L. Iversen CI al. /J. Dernzatol. Sri. 7 (1994) 191-201

dermal sample was centrifuged at 3000 x g for 30 min at 4°C and the resulting supernatant investigated for myeloperoxidase activity. Specifically, 300 ~1 of the supematant was incubated in 2.7 ml 50 mM potassium phosphate buffer pH 6.0 containing 0. I67 mg/ml o-dianisidine dihydrochloride and 0.0005% hydrogen peroxide for 10 min at 37°C. Change in absorbance (at 460 nm) was measured with a Shimadzu UV-160A spectrophotometer. For control, myeloperoxidase activity (MPO) was measured after homogenization of varying numbers of PMNs (1.5 x lo3 to 2 x lo5 cells). A linear relationship was always observed in the range of PMNs indicated, when the absorbance was plotted as a function of the number of cells (compare Fig.

ic plastic with an exchangeable, pierced acrylic plate was placed on the abdominal skin. Then a vacuum of 250-300 mmHg was applied resulting in 30 separate blisters, each with a diameter of 5 mm, within 90-120 min. The epidermis from each blister was removed using sterile forceps and scissors and immediately placed on ice. The donors were normal human volunteers who had not taken steroids or non-steroidal antiinflammatory drugs during the preceding 2 weeks. This procedure was approved by the local ethics committee. 2.5. Assay for tissue myeloperoxidase To rule out any infiltration of the excised epidermis with PMNs during the suction, the presence of PMNs in the suction blister skin was determined by a myeloperoxidase assay as previously described by Bradley et al. [25]. Briefly, a representative part of the excised epidermis was chopped into small pieces and placed in a glass tube containing 400 ~1 of 0.5% HTAB in 50 mM potassium phosphate buffer pH 6.0. The tissue was then homogenized with a Kinematica AG polytron (10 s x 3 on ice) and then sonicated (10 s x 3 on ice). The homogenized epi-

1). 2.6. Co-incubation of suction blister lifed epidermis and PMNs To determine the effect of blister-obtained epidermis/PMNs interaction, small pieces of chopped epidermis (0.021-0.054 g of wet weight) alone, chopped epidermis (0.025-0.068 g wet weight) mixed with PMNs (10 x 106) and PMNs (10 x

0

2

1

Number

of

PMNs

(~10~)

Fig. 1. Standard curve of MPO activity determined with varying numbers of PMNs.

194

106) alone were pre-incubated in KGM containing CaCl, (0.87 mM) for 10 min at 37°C. Then the mixtures containing chopped epidermis and/or cells (PMNs) were stimulated with A 23187 (5 PM) dissolved in ethanol (final concentration < 1%) and incubated for an additional 5 min. Incubations were terminated by the addition of 2 vol of ice-cold MeOH. 2.7. Preparation of LTA, and incubation with suction blister lifted epidermis

Free LTA4 used in these incubations was prepared from LTA4-methyl ester as previously described [7,13], and the concentration estimated by UV-absorption at 280 nm. To determine whether isolated epidermis was able to transform LTA4 into LTB4, small pieces of chopped epidermis were preincubated for 10 min at 37°C in 1 ml of KGM containing 0.87 mM CaCl, and 1 mg of human serum albumin (HSA) per ml. LTAl (10 PM) was added and incubated for 10 min, at 37°C. In the control experiments, LTA, was added to chopped epidermis that had been inactivated (boiled for 20 min), or to the medium without tissue. Incubations were terminated by the addition of 2 vol of ice cold MeOH. 2.8. Sub-fractionation of epidermis and cultured keratinocytes

To ascertain the subcellular localization of LTA4-hydrolase in epidermis, the isolated human epidermis was homogenized in Tris-HCl buffer (pH 7.5) with a Kinematica AG polytron (3 x 10 s on ice). The homogenate was filtered through glass wool to remove debris and then subjected to centrifugation at 12 000 x g for 15 min. The 12 000 x g pellet obtained was washed twice, before resuspension in buffer. Aliquots from the 12 000 x g supernatant was removed for protein determination and for incubation. The remaining portion of the 12 000 x g supernatant was further centrifuged at 105 000 x g for 60 min. The high speed 105 000 x g pellet was washed twice and resuspended in buffer. Aliquots were taken from the resuspended pellet and the 105 000 x g supernatant for protein determination. The protein contents of the subcellular fractions were determined according

L. Ivrrsen et al. /J. Dermntol.Sci. 7 f I9941 191-201

to Bradford [26] with bovine serum albumin as standard. Pellets were resuspended in buffer (500 ccl), incubated with LTA4 (10 PM) for 1 min and assayed for LTB4 formation. Aliquots (500 ~1) from the supernatant fractions were similarly incubated with LTA4. Cultured keratinocytes were released from the culture dish by scraping and the 105 000 x g supernatant fraction was prepared as described for the epidermis. 2.9. Inhibition of LTA4-hydrolase activity The effects of two inhibitors (bestatin, captopril) of LTA.,-hydrolase activity [27,28] were tested regarding LTA,-hydrolase activity in human epidermis. Aliquots (0.5 ml) of the 105 000 x g supernatant were pre-incubated with or without captopril (100 PM) or bestatin (O-70 PM) for 15 or 60 min at 37°C before the addition of LTA4 (10 PM). The incubations were then carried out for 1 min at 37°C. All incubations were terminated by the addition of 2 vol of ice cold MeOH as indicated previously. 2.10. Extraction of lipids After termination of incubations, reaction mixtures were kept at -20°C for 20 min and then centrifuged (1500 x g) for 10 min at 4°C to remove precipitated proteins. The supernatant was collected and the lipids extracted as previously described by Fogh et al. [24], application to octadecylsilyl (ODS) silica columns (Cis-SEP-PAK cartridges). The metabolites were eluted sequentially with 2 ml of water, 2 ml of 25% MeOH, 2 ml of petroleum ether and 3 ml of MeOH. The methanol fraction was collected and taken to dryness under a stream of N2 and re-suspended in 70% MeOH. 2.11. Reversed-phase high performance liquid chromatography (RP-HPLC) Extracted lipids were separated by RP-HPLC as previously described [ 131. The RP-HPLC system from Waters was equipped with a Hypersil C,, column (5 pm, 100 x 4.6 mm). The elution program was a modified isocratic method using methanol/water/acetic acid, 70:30:0.01 (by vol) as the mobile phase. For the first 10 min the flow rate was

0.8 mlimin and thereafter the flow rate was increased to 1.4 ml/min. The detector was set at 270 nm from 0- 15 min and at 235 nm from 15-30 min. Eluents were collected by a fraction collector programmed to obtain 1 min fractions. The identity of LTB, was ascertained as previously described [ 131 by characteristic ultraviolet absorbance and by chromatographic comparison with authentic LTB4 standard, and in selected samples by radioimmunoassay (RIA). 6-transLTB4, 12-epi-6-trans-LTB4, 5(S),6IR)-DiHETE. 12-HETE and 15-HETE were identified only by chromatographic comparison with authentic standards. Quantitation was based on integration of optical density, and by RIA in the experiments with co-incubations. 2.12. RIA for LTB, Fractions co-chromatographing with authentic LTB4 were collected and RIA analysis of selected samples was performed as previously described [ 131. 2.13. Protein anal?‘sis ( Western blots) The recombinant LTA,-hydrolase used as standard was prepared as previously described [29]. Standard LTA,-hydrolase (100 ng), the cytoplasmic fraction of human epidermis and cultured human keratinocytes ( 18 pg protein) were analysed on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), (9% polyacrylamide gel). Before loading of the sample to the gel, 5 ~1 of 2% sodium dodecyl sulphate/lO% glycerol/5% 2mercaptoethano110.002”~ bromophenol blue was added per 20 pl sample, and then the samples were boiled for 2 min. After gel electrophoresis the proteins were transferred from the gel to a hybondECL nitrocellulose membrane by using a SemiDry Electrobiotter (Ancos Denmark ApS, Denmark). Complete transfer of all proteins in the range of 50-80 kDa was achieved by 5 h at 20 V as shown by the lack of staining of the gels by Coomassie blue. The membrane was blocked with 3% bovine serum albumin and then incubated with the affinity-purified antibody for 1 h. The second antibody also incubated for 1 h was aflinitypurified goat anti-rabbit immunoglobulins conjugated with horseradish peroxidase (I :2000 dilu-

tion). Finally the membrane was incubated for 1 min with the ECL Western blotting detection reagents and immediately after exposed to a hyperlilm-ECL to visualize the immunoreactive bands. This was done according to the manufacturer’s instructions. 2.14. Statistical analysis Results are expressed as the mean f standard error of mean (S.E.M.). Statistical significance was assessed by Wilcoxon’s ranksum test. 3. Results 3. I. Myeloperoxidase activity The MPO activity was measured to determine whether the epidermis obtained by suction blister formation was contaminated with PMNs. A standard curve of MPO activity determined with varying numbers of PMNs (1.5 x 10’ to 2 x lo5

200

.

Epidermis

Fig. 2. Co-incubation

of chopped

PMNS

epidermis

I.

Epidsrmis+PMNs

and PMNs; forma-

tion of LTB., by chopped epidermis (0.021-0.054 g wet weight) alone, PMNs alone (IO x IO6 cells) and co-incubation of chopped epidermis (0.025-0.068 g wet weight) plus PMNs (IO x lo6 cells); pre-incubation in KGM containing CaC12 (0.87 mM) was carried out for IO min at 37°C prior to addition of A 23187 (5 PM) and further incubation for additional 5 mitt; LTB, formation was determined by RIA as described in 2. I?; results are expressed as mean + S.E.M. from 6 duplicate experiments.

196

L. Iversen et al. /J. Dermatol. Sci. 7 (1994) 191-201

Run time

:

Run time

0

Run time

SD inl”

Run time

Fig. 3. Representative reverse-phase HPLC chromatograms of epidermis incubated in 1 ml PBS containing 0.87 mM CaClz with LTA, (10 PM) for 20 min at 37OC; UV-absorbance was monitored at 270 nm from O-15 min and at 235 nm from 15-30 min; (A) epidermis incubated with LTA,; (B) epidermis incubated without LTA4; (C) control; inactivated epidermis incubated with LTA,; (D) control; LTA, incubated without epidermis.

cells) revealed a close relationship between the number of PMNs and the MPO-activity (Fig. 1). These data indicate that MPO activity was detectable with only 1.5 x lo3 PMNs. In contrast, epidermis prepared by suction blister under our conditions showed no detectable MPO activity (data not shown) indicating a PMN contamination of < 1.5 x lo3 cells per sample. In separate experiments the challenge of 1.5 x lo3 PMNs in a volume of 0.5 ml with ionophore, or with LTA4 (10 PM), resulted in negligible generation of LTB4

as determined by HPLC. Therefore, contaminating PMNs cannot contribute substantially to the conversion of LTA4 to LTB4 in the epidermis preparations. 3.2. Ionophore-induced biosynthesis of LTB,

by epidermis PMNs and by co-incubation of epidermis and PMNs

The effect of calcium ionophore (A 23 187) on LTB4 biosynthesis by chopped epidermis, by PMNs (10 x lo6 cells) and by co-incubation of

197

L. lversen er al. /J. Dermarol. Sri. 7 (1994) 191-201

chopped epidermis and PMNs are shown in Fig. 2. Negligible amounts of LTB4 were found in the A 23187 challenged epidermal specimens as determined by RIA (detection limit at 4 pg). In contrast, similar stimulation of PMNs (10 x lo6 cells) alone produced 119 f 21 ng LTB,. Coincubation of suction blister lifted epidermis and PMNs resulted in an increase of LTB4 formation (54%). when compared to PMNs incubated alone. 3.3. Transformation qfLTA4 into LTB, by human epidermis To determine whether human epidermis possesses the capacity to transform LTA, into LTB,, homogenized epidermis prepared from blister lifted skin was incubated with or without LTA, (10 PM in a total of 1 ml) at 37°C for 20 min. HPLC-chromatography of the extracted lipids revealed a peak that co-migrated with authentic LTB, (Fig. 3A), whereas the incubation of epidermis alone revealed no detectable peak of LTB4 (Fig. 3B). Incubations of LTA4 with either inactivated epidermis (boiled for 20 min) or medium showed the two non-enzymatically formed products 6-trans-LTB4 and 12-epi-6-trans-LTB4, but no LTB4 formation (Figs. 3C,D). Further characterization of the chromatographically identified LTB., was provided by the characteristic

Table I Subcellular epidermis

distribution

of LTA,-hydrolase

activity

in human

Fraction

Specific activity tng LTBdmg protein/mitt)

Homogenate I2 000 X g ( I5 min) supernatant 12 000 X g (I5 mitt) pellet 105 000 x g (60 mint supernatant I05 000 x g (60 min) pellet

28.8 45.1 ND 48.5 ND

l

4.40 12.0

l

I I.8

l

Epidermal pieces obtained from suction blister skin were homogenized in 3.7 ml of Tris-HCI buffer (pH 7.5) and then subjected to differential centrifugation as described in the text: pellets were resuspended in the same volume of buffer as the corresponding supernatants: aliquots (0.5 ml) were incubated with LTA, (IO pM) for I min; LTB, formation was determined by HPLC and the identity ascertained by characteristic UVabsorption scan; results are expressed as mean f S.E.M. of 3 experiments assayed in duplicate; ND. not detectable.

120 110

0.0

0.2

0.4

mg

0.a

0.8

cytosollc

1.0

1.2

1.4

1,s

protein

Fig. 4. Effect of increasing amounts of crude high speed cytosolic fraction on transformation of LTA, into LTB,; increasing amounts of crude I05 000 x g supernatant protein were equilibrated for IO min at 37°C and then incubated with LTA, (IO pM) for an additional I min (incubation volume 0.5 ml); LTB, formation was determined by HPLC; results are expressed as mean f S.E.M. from 5 experiments.

triplet UV-absorbance pattern (maximum absorption at 270 nm) [13], and by RIA. Furthermore, it appears that LTA, is transformed both enzymatically and nonenzymatically into isomers of 5,6-DiHETE (Figs. 3A,C). The two arachidonic acid metabolites, 12-HETE and 15-HETE. were also detected in the chromatograms. Notably, quite substantial amounts of these monohydroxy without addition of acids were obtained, arachidonic acid to the incubations. 3.4. Subcellular distribution qf LTB4 qwthetic uctivity in epidermis To delineate the cellular localization of the LTA,-hydrolase activity found in the epidermis, subcellular fractions of homogenized epidermis were incubated with LTA4. Table 1 shows that most of the LTA,-hydrolase activity was detected in the high speed (105 000 x g) supernatant fraction. There was no detectable activity in the high speed particulate fraction. A relationship between the 105 000 x g cytosolic protein and LTB4 formation is shown in Fig. 4. Specifically, our data

198

L. hersen et al./J. Dermatol. Sci. 7 (1994) 191-201 A

B

C

origin +

25

N-5

0

5

IO phi

15

20

LTA4

25

30

35

40

+

Fig. 6. Western blot analyses of LTA4-hydrolase in epidermis and keratinocytes; the cytoplasmic fraction from human epidermis and cultured human keratinocytes or purified LTA, hydrolasc were subjected to SDS-PAGE (9% polyacrylamide gel);

0.04

after electroblotting to nitrocellulose membranes an affinitypurified anti-LTA, hydrolase antibody was used for immunodetection; (A) the cytoplasmic fraction of human epidermis (18 pg protein); (B) the cytoplasmic fraction of cultured human keratinocytes (18 pg protein); (C) purified LTA,-hydrolase (100 ng).

p 0.03 + c 2

Front

(substrate)

0.02

0.01

-0.2

-0.1

0

l/N

Fig. 5. (A) Effect of LTA4-hydrolase activity; were added to previously aliquots of the 105 000 x

0.1

0.2

0.3

(t&M)

increasing substrate (LTAJ on increasing concentrations of LTA, equilibrated (10 min at 37°C) 0.5 ml g supernatant cytosolic fraction (0.5

mg protein) and incubated for I min; LTB4 formed was determined by HPLC, the identity was ascertained by characteristic UV-absorption scan, results are expressed as mean f S.E.M. from 5 experiments; (B) Lineweaver-Burk plot of data in Fig. 5A, omitting the measurement corresponding to LTA, concentration 2.5 pM.

revealed that increasing amounts of cytosolic protein (in 0.5 ml), incubated with a fixed amount of LTA4 (10 PM), resulted in increased LTB4 formation, suggesting that an active enzyme protein is present in the epidermal cytosol. Other experiments with variable amounts of substrate (LTA,)

and a fixed amount of cytosolic protein (0.5 mg in 0.5 ml) showed that the reaction obeyed MichaelisMenten kinetics with an apparent K, of 6 PM (VIII,, 300 pmol LTBd mg proteimmin) (Fig. 5). in the 10.5 000 X of homogenized epidermis and cultured keratinocytes by Western blot

3.5. Analysis of LTA,-hydrolase g supernatant

To confirm the existence of LTA4-hydrolase in the cytoplasmic fraction of the epidermis and cultured human keratinocytes, Western blot analysis of this fraction was carried out. Samples of the 105 000 x g fraction of homogenized human epidermis and cultured keratinocytes were analyzed in parallel with enzyme purified from human leukocyte, and a single intense and clear LTA4hydrolase band with the same retention as the authentic standard appeared for the epidermis as well as for the cultured keratinocytes (Fig. 6).

L. lversrn et ~1. _;J. Dermatol.

Sci. 7 (19941 191-201

3.6. Ejjkct of inhibitors of LTA4 hjldrolase on epidermal biosynthesis of LTB, The inhibitors bestatin (70 PM) or captopril (100 PM) were added to 0.S ml aliquots (1 .O mg protein) of the 105 000 x g epidermal supernatant. 15 or 60 min before the addition of LTA4 (10 PM). The data in Fig. 7 revealed that both inhibitors exerted significant inhibition of LTA, transformation into LTB4. After pre-incubation for 15 min the inhibition was 76% with captopril and 96% with bestatin. When pre-incubated for 60 min, captopril and bestatin inhibited LTB4 formation by 69% and 98X, respectively. When cytosolic fractions were pre-incubated with different concentrations of bestatin for 60 min. the inhibition of LTB4 formation was found to be dose dependent with an ICsO of 10.5 PM (Fig. 8).

199

0

10

20

30

40

pM

BESTATIN

50

60

70

Fig. 8. Pre-incubation of crude 105 000 x g supernatant of epidermis with different concentrations of bestatin followed by incubation with LTA,; 0.5 ml aliquots (0.5 mg protein) of epidermal 105 000 x g cytosolic fraction were pre-incubated with varying concentrations of bestatin (O-70 PM) at 37°C for 60 min: LTA, I 10 pM) was added and incubation was carried out for an addttional 1 min; LTB4 formation was determined by HPLC, and the results are expressed as mean f S.E.M. from 6 experiments.

. : pco.05

15

Ill,”

Prr-Incubation

80

mln

time

Fig. 7. Inhibition of epidermal LTA,-hydrolase activity by bestatin and captopril; 0.5 ml aliquots of epidermal 105 000 x g cytosolic fractions (1.0 mg protein) were preincubated with bestatin (70 pM) or captopril (100 PM) at 37°C for 15 and 60 min: LTA, ( IO pM) was added and incubated for an additional 1 min: control experiments contained only cytosolic protein with no bestatin or captopril: LTB, formed was determined by HPLC. and the results are expressed as mean f S.E.M. from 6 experiments.

4. Discussion

In the present study we have shown that LTA,hydrolase activity is present in normal human epidermis. The activity is localized in the 105 000 x g supernatant (cytoplasmic) fraction which is consistent with the localization of LTAdhydrolase in cell types such as erythrocytes and leukocytes [30,31]. The transformation of LTA4 into LTB4 was proportional to the amount of cytosolic protein, and the reaction conformed to Michaelis-Menten kinetics with an apparent K, of 6 PM. This is in reasonable agreement with previous determinations of K, for LTA,-hydrolase [I% Although the levels of LTB4 have been found to be raised significantly in inflammatory skin diseases such as psoriasis [21-231 and atopic der-

L. iversen et al. /J.

200

matitis [32], this is the first report demonstrating LTA4-hydrolase activity in human skin. Inhibition of epidermal LTB4 formation was observed, when the 105 000 x g fraction was preincubated with bestatin and captopril, both of which previously have been shown to be inhibitors of LTA4-hydrolase [27,28]. In our study an IC5s of 10.5 PM was found for bestatin. This degree of inhibition is similar to that described in a recent report in which bestatin was shown to inhibit LTB4 formation by LTA4-hydrolase in a human leukocyte cytosol fraction [33]. The LTA4-hydrolase activity in the cell free system was inhibited to a higher degree by bestatin than in whole cells [13]. Although the explanation of this difference is not immediately clear, it is likely that this may be due, at least in part, to a reduced penetration into whole cells. The existence of the LTA,-hydrolase enzyme in the epidermis and in human keratinocytes was confirmed by Western blot analysis of the 105 000 x g supernatant. The molecular mass of the epidermal and keratinocyte LTArhydrolase was the same as for leukocyte LTA4-hydrolase (close to 70 kDa). The role of LTArhydrolase in transcellular LTB4 formation has recently gained much interest. It has been shown that cellular interaction in LTB, formation can take place among PMNs and other cell types such as erythrocytes [12] and endothelial cells (111, both of which express negligible 5-LO activity. We and other investigators have also recently reported that cultured human keratinocytes can interact with PMNs resulting in the generation of LTB4 [13,14], and in this study we report that human epidermis can transform PMN derived LTA4 into LTB4. Future studies aim at purification and full epidermal LTAdof the characterization hydrolase, and elucidation of possible variation of the epidermal LTA4-hydrolase activity, in inflammatory skin diseases characterized by high levels of LTB4.

2

6

191-201

Ruzicka T, Printz MP: Arachidonic acid metabolism in skin: a review. Rev Physiol Biochem Pharmacol 100: 12 I- 160, 1984. Nijkamp FP, Sitsen JMA: Leukotrienes, allergy and inflammation. Pharmaceutisch Weekblad Sci Ed 4: 165-171, 1982. Ford-Hutchinson AW: Leukotrienes as mediators of inflammation. IS1 Atlas of Science: Pharmacology 1: 25-28, 1987. Lewis RA, Austen KF, Soberman RJ: Leukotrienes and other products of the 5lipoxygenase pathway: biochemistry and relation to pathobiology in human diseases. N Engl J Med 323: 645-655, 1990. Marcus AJ: Eicosanoids: transcellular metabolism. Inflammation: Basic Principles and Clincal Correlates, 129- 137,

Medina JF, Haeggstrom J, Kumlin M, Ridmark 0: Leukotriene A,: metabolism in different rat tissues. Biochim Biophys Acta 961: 203-212, 1988. Haeggstrom J, Meijer J, Ridmark 0: Leukotriene A4: enzymatic conversion into 5,6-dihydroxy-7,9,11,14-eicosatetraenoic acid by mouse liver cytosolic epoxide hydrolase. J Biol Chem 261: 6332-6337, 1986. 10

11

Jakobsson PJ, Odlander 9, Claesson HE: Effect of monocyte-lymphocyte interaction on the synthesis of leukotriene 94. Eur J Biochem 196: 395-400, 1991. Claesson HE, Haeggstrom J: Human endothelial cells

I2

stimulate leukotriene synthesis and convert leukotriene A4 into leukotrienes 94, Cd, D4 and E,. Eur J Biochem 173: 93-100, 1988. McGee JE, Fitzpatrick FA: Erythrocyte-neutrophil interac-

13

14

15

16

17 Borgeat P, Samuelsson 9: Arachidonic acid metabolism in polymorphonuclear leukocytes: effect of ionophore A 23187. Proc Nat1 Acad Sci USA 76: 2148-2152, 1979.

Sci. 7 (1994)

1988. Maycock AL, Anderson MS, De Sousa DM, Kuehl FA Jr: Leukotriene A,: preparation and enzymatic conversion in a cell-free system to leukotriene 9,. J Biol Chem 257: 1391 l-13914, 1982.

6. References I

Dermatol.

tions: formation of leukotriene 94 by transcellular biosynthesis. Proc Nat1 Acad Sci USA 83: 1349-1353, 1986. Iversen L, Fogh K, Ziboh VA, Kristensen P, Schmedes A, Kragballe K: Leukotriene B4 formation during human neutrophil keratinocyte interactions: evidence for transformation of leukotriene A4 by putative keratinocyte leukotriene Ad hydrolase. J Invest Dermatol 100: 293-298, 1993. Sola J, Godessart N, Vila L, Puig L, de Moragas JM: Epidermal cell-polymorphonuclear leukocyte co-operation in the formation of leukotriene 9, by transcellular biosynthesis. J Invest Dermatol 98: 333-339, 1992. Ridmark 0, Haeggstrom J: Properties of leukotriene Ad-hydrolase. Adv Prostaglandin Thromboxane Leukotriene Res 20: 35-45, 1990. Minami M, Ohno S. Kawasaki H, Ridmark 0. Samuelsson 9, Jornvall H, Shimizu T, Seyama Y, Suzuki K: Molecular cloning of a cDNA coding for human leukotriene A, hydrolase: complete primary structure of an enzyme involved in eicosanoid synthesis. J Biol Chem 262: 13873- 13876, 1987. Funk CD, Ridmark 0, Fu JY, Matsumoto T, Jornvall H, Shimizu T, Samuelsson 9: Molecular cloning and amino acid sequence of leukotriene Ad hydrolase. Proc Nat1 Acad Sci USA 84: 6677-6681, 1987.

L. lversen et al. /J. Dermarol. Sci. 7 (1994)

18

19

20

21

22

23

24

25

191-201

201

Izumi T, Shimizu T, Seyama Y, Ohishi N, Takaku F: Tissue distribution of leukotriene A, hydrolase activity in guinea pig. Biochem Biophys Res Commun 135: 139- 145, 1986. Ohishi N, Minami M, Kobayashi J, Seyama Y, Hata J, Yotsumoto H, Takaku F. Shimizu T: Immunological quantitation and immunohistochemical localization of leukotriene A, hydrolase in guinea pig tissues. J Biol Chem 265: 7520-7525. 1990. Fu JY, Haeggstrom J, Collins P, Meijer J, Ridmark 0: Leukotriene A, hydrolase: analysis of some human tissues by radioimmunoassay. Biochim Biophys Acta 1006: 121-126, 1989. Brain S. Camp R, Dowd P, Black AK, Greaves M: The release of leukotriene B1-like material in biologically active amounts from the lesional skin of patients with psoriasis. J Invest Dermatol 83: 70-73, 1984. Grabbe J, Czarnetzki BM, Rosenbach T, Mardin M: Identification of chemotactic lipoxygenase products of arachidonate metabolism in psoriatic skin. J Invest Dermatol 82: 477-479, 1984. Fogh K. Herlin T. Kragballe K: Eicosanoids in acute and chronic psoriatic lesions: leukotriene Bq. but not 12-hydroxyeicosatetraenoic acid is present in biologically active amounts in acute guttate lesions. J Invest Dermatol 92: 837-841, 1989. Fogh K, Herlin T. Kragballe K: In vitro inhibition of leukotriene B, formation by exogeneous 5-lipoxygenase inhibitors is associated with enhanced generation of I5-hydroxyeicosatetraenoic acid ( 15-HETE) by human neutrophils. Arch Dermatol Res 280: 430-436, 1988. Bradley PP, Priebat DA, Christensen RD. Rothstein G: Measurement

of cutaneous

inflammation:

estimation

of

26

27

28

29

30

31

32

33

neutrophil content with an enzyme marker. J Invest Dermatol 78: 206-209, 1982. Bradford M: A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254. 1976. Orning L, Krivi G. Bild G, Gierse J. Aykent S. Fitzpatrick FA: Inhibition of leukotriene A4 hydrolase/aminopeptidase by captopril. J Biol Chem 266: 16507-16511. 1991. Orning L, Krivi G, Fitzpatrick FA: Leukotriene A, hydrolase: inhibition by bestatin and intrinsic aminopeptidase activity establish its functional resemblance to metallohydrolase enzymes. J Biol Chem 266: 1375- 1378. 1991. Minami M. Minami Y. Emori Y. Kawasaki H. Ohno S. Suzuki K. Ohishi N, Shimizu T. Seyama Y: Expression of human leukotriene A, hydrolase cDNA in E.whwichia co/i. FEBS Lett 229: 279-282, 1988. McGee J. Fitzpatrick F: Enzymatic hydration of leukotriene A,: purification and characterization of a novel epoxide hydrolase from human erythrocytes. J Biol Chem 260: 12832-12837, 1985. Rsdmark 0. Shimizu T, Jornvall H, Samuelsson B: Leukotriene A4 hydrolase in human leukocytes: purification and properties. J Biol Chem 259: 12339-12345. 1984. Fogh K. Herlin T. Kragballe K: Eicosanoids in skin of patients with atopic dermatitis: prostaglandin E, and leukotriene B4 are present in biologically active concentrations. J Allergy Clin Immunol 83: 450-455, 1989. Evans JF, Kargman S: Bestatin inhibits covalent coupling of t3H)LTA, to human leukocyte LTA, hydrolase. FEBS Lett 297: 139-142. 1992.