Enantiospecific synthesis of bioactive hydroxyeicosatetraenoic acids (HETEs) in Hydra magnipapillata

BiochimicaL et Biophysics &ta

ELSEVIER

Biochimica et Biophysics Acta 1213 (1994) 215-223

Enantiospecific

synthesis of bioactive hydroxyeicosatetraenoic ( HETEs) in Hydra magnipapillata 1

acids

Thomas Leitz a, Werner Muller a, Luciano De Petrocellis b, Vincenzo Di Marzo ‘j* a Zoologisches Institut der Urkersitat, Fachrichtung Physiologic, Im Neuenheimer Feld 230, Heidelberg, Germany b Istituto di Cibernetica de1 C.N.R., Arco Felice, Naples, Italy

’ Istituto per la Chimica di Molecole di Interesse Biologico de1 C.N.R., Ka Toiano 6, 80072, Arco Felice, Naples, Italy (Received 10 November 1993)

Abstract Recent arachidonic

reports have shown the occurrence of regiospecific and enantioselective lipoxygenase-mediated metabolism of acid (AA) in cytosolic extracts of marine and freshwater hydroids. Here we report that cytosolic extracts of Hydra magnipapillata are unique among hydrozoans for their capability of converting AA into two major metabolites which showed

chromatographic, mass spectrometric and nuclear magnetic resonance properties identical to those of 11-R- and 12-S-hydroxyeicosatetraenoic acid (ll-R-HETE and 1ZS-HETE). The production of neither compound was affected by co-incubation of H. magnipapillata

extracts with the cytochrome P-450 inhibitor proadifen. The 5- and 12-lipoxygenase inhibitor nordihydroguaiaretic acid (NDGA), while inhibiting 12-S-HETE formation at high concentrations, did not influence ll-R-HETE production, thus suggesting the co-localisation, unprecedented in hydroids, of two separate enantioselective lipoxygenase-like activities. The possible role of the two metabolites in the control of hydroid body pattern was investigated. At low micromolar concentrations, both enantiomers of ll-HETE inhibited diacylglycerol-induced ectopic head formation (EHF), while 12-S-HETE, and its likely precursor 12-S-hydroperoxy-eicosatetraenoic acid (12~S-HPETE), enhanced bud formation, thus providing the first example of endogenous metabolites controlling, respectively, hydroid ‘head activation potential’ and asexual reproduction.

Key words: Hydroxyeicosatetraenoic

acid; HETE;

Hydroid

development;

1. Introduction Lipoxygenases (LOS), which catalyse the regio- and stereo-specific addition of 0, to polyunsaturated fatty acids (PUFAs), are enzymes ubiquitous in both plants and animals [l]. When the substrate is arachidonic acid (AA) typical products of LO-mediated activity are hydroperoxy-eicosatetraenoic acids (HPETEs) which can be reduced either spontaneously or enzymatically to hydroxyeicosatetraenoic acids (HETEs) [2]. These metabolites exhibit a wide range of biological actions which are often enantiospecific and span from chemotaxis and control of cell proliferation to regulation of enzyme activity, hormone secretion and receptor-mediated CAMP formation (for a review see [3]). LO-mediated formation of either R or S enantiomers of 5-, 8-,

* Corresponding author. Fax: +39 81 8041770. ’ Dedicated to the memory of Dr. Gianpaolo Nitti. 0005-2760/94/$07.00 0 1994 Elsevier SSDI 0005.2760(94)00050-9

Science

B.V. All rights reserved

Invertebrate

eicosanoid;

Lipoxygenase

9-, ll-, 12- and 15-HETEs has been described to occur in both vertebrate and invertebrate tissues, with a diversity of important biological functions, including intra- and/or extra-cellular mediators of cell development and neural transmission [4,.5], that were previously unconsidered. In particular, in invertebrates, 12LO metabolites are used as second messengers for the action of neurotransmitters in the ganglia of the mollust Aply& californica [6]; the S-LO product 8-RHEPE controls enantiospecifically starfish oocyte maturation [7], and egg hatching in barnacles [8], while 11-R- and 12-R-LO-like activities have been detected in sea urchin eggs [9]. Recently, we have presented evidence for the presence of LO-like activities in cytosolic extracts of both marine and freshwater hydroids. Hydra oligactis and Halocordyle disticha were shown to contain 11-LO enzymatic activity, while a compound co-eluting with 9-HETE was detected upon incubation of Aglaophenia pluma extracts with AA [lo]. The possible involvement of LO-induced forma-

216

T. Leitz et al. /Biochimica

et Biophysics Acta 1213 (1994) 215-223

tion of 8-R-HETE in Hydructiniu echinata metamorphic events was also suggested [I 11. An abundant regioselective and enantioselective ll-R-LO, catalysing AA conversion into ll-R-HPETE followed by formation of II-R-HETE was found and partially characterised in Hydra uulgaris [12]. This enzyme also recognised other PUFAs including the very ones present in hydroid fatty acid pools, e.g. linoleic, y- and a-linolenic, eicosapentaenoic and eicosatrienoic acids, leading to the regiospecific and enantioselective formation of hydroxy-derivatives which, in the case of 1l-R-HETE and 9-~-hydro~octadecadienoic acid (9-S-HODE), were found to induce Hydra tentacle regeneration in an enantiospecific fashion [13]. Evidence for the participation of AA and HETEs in the regulation of Hydra magnipupii~ata body pattern has recently been presented [14] and substantiated by similar independent findings in H. uulguris [15-171. H mugnipapillutu, wild-type strain 105, exhibits striking morphological changes, consisting in the appearance of supemumera~ fectopic) tentacles and complete heads along the body column, upon repeated pulse-type stimulation with AA and the protein kinase C activator dioctanoylglycerol (DAG) [14,18,19]. This phenomenon, which, together with bud fo~ation and tentacle regeneration, is dependent on cell proliferation but is primarily an expression of an increased developmental potential known as ‘head activation potential’ [20], has not yet been fully investigated from a biochemical point of view. The LO-inhibitor nordihydroguaiaretic acid (NDGA) was found to inhibit AA- and DAG-induced formation of ectopic heads, but a complete characterisation of the HETEs actually biosynthesised during and involved in ‘head activation potential’ is still lacking. In the present report we present evidence for the presence of two separate regio- and enantiospecific LO activities catalysing AA conversion into two HETEs in cytosolic extracts of H. mugnipapilfufu. Bioassays, also conducted in this study, assign to these metabolites a potential role in the control of body pattern, cell proliferation and asexual reproduction.

2. Materials and methods 2.1. Animals Hydra mugn~papi~lutu, strain wt 105, were grown under standard conditions described previously [21].

2.2. Biosynthesis and purification of HETEs from cytosoiic extracts 1000 polyps of H. mugnipupillata starved for 3 days were collected and washed twice with culture medium, once with ice-cold homogenisation buffer (20 mM Tris-

HCl, 100 mM NaCl, 2 mM EDTA, 2 mM DTT (pH 7.4)) and then homogenised in 8 ml buffer in a 15 ml glass homogeniser with 30 strokes. The homogenate was centrifuged for 30 min at 4°C and 16000 X g. The protein concentration in the supernatant was determined by the method described in [22] and modified as described in 1231. In macroassays, 9 ml of supematant (1.6 mg/ml total protein content) were incubated with 5 mg AA. To facilitate identification of metabolites, microassays were performed incubating 50 ~1 of the same supernatant with 30 pg AA and 20000 dpm of [14ClAA (Amersham, 2 GBq/mmol). CaCl, was added in all assays to a 15 mM final concentration. Control incubations were conducted under the same conditions but with boiled supernatant. After a 2 h incubation at 20°C the reaction was stopped by the addition of ethanol to 15% and by lowering the pH to 2 by addition of 3% formic acid. The mixture was briefly centrifuged and the supernatant loaded onto Sep-pak plus cartridges (Waters Ass.), which had been primed with 10 ml ethanol and 20 ml water. The cartridges were washed with 20 ml water, 20 ml 15% ethanol and 20 ml petroleum ether (b.p. 40-60°C) and eicosanoids eluted with 9 ml ethyl acetate which was then evaporated in a Speed Vat (Savant Ins., USA). HETEs were purified by means of HPLC on a Spherisorb Silica column (5 pm, 250 X 4 mm, Latek, Heidelberg, Germany) using a Jasco HPLC gradient system equipped with two 880 PUi pumps, a Jasco 870 UV detector, a Spectra-Physics Data Jet Integrator and a Gitson fraction collector. Buffer A was hexane/acetic acid 99: 1 and buffer B was hexane/2-propanol/acetic acid 98: 1: 1. The column was equilibrated with 50% buffer B and eluted with a 40 min gradient to 100% buffer B at a flow rate of 1.5 ml/min ill]. 1.5 ml fractions were collected. In microassays, the eiuates were directly collected into scintillation vials, mixed with 3 ml of either Quickszint 1 or Quicksafe A (Zinsser, Germany) and p emission counted. In macroassays, the metabolites were detected by their UV absorption at 234 nm and tentatively identified/quantitated by their retention times/peak areas compared to those of known amounts of authentic HETE standards (Sigma, Munich) chromatographed in different runs. The solvent of fractions containing HETEs was evaporated and the metabolites subjected to NMR analysis, or derivatised in order to carry out mass spectrometric analysis and chiral phase HPLC. 2.3. Effect of inhibitors on HETE formation in cytosolic extracts In a separate set of experiments 315 ~1 of the supernatant, containing 250 pg total proteins, were incubated with 100000 dpm radiolabelled AA and 1.6 PM unlabelled AA and with or without either the

T. Leitz et al. /Biochimica

et Biophysics Acta 1213 (1994) 215-223

cytochrome P-450 inhibitor proadifen (75 PM) or the 5- and 12-LO inhibitor NDGA (8.3 and 20 PM). After 2 h incubation at 20°C samples were treated as described above and analysed by HPLC, counting p emission of each 1.5 ml fraction. Each incubation was carried out in triplicate. 2.4. Structure characterisation of HETEs ‘H-NMR analyses gave clear cut results only for the most abundant metabolite, i.e. lZHETE, using a 500 MHz Bruker instrument (Germany), by dissolving the HPLC fraction, after prolonged evaporation of non-deuterated solvents, in 0.5 ml CD,OD. After NMR, solvents were evaporated and samples dissolved in methanol to be submitted to methylation with diazomethane for 15 min at room temperature. In order to determine HETE stereochemical composition, aliquots of the methylated samples were submitted to chiral phase HPLC. This was carried out using a Chiralcel OB column (Daicel Chem. Ind., 250 X 4.6 mm) eluted with hexane/Zpropanol 99: 1 at a flow rate of 1.5 ml/min and monitoring the absorbance at 234 nm [12,24]. Under these conditions S enantiomers elute at least 3.5 min earlier than R enantiomers. In order to carry out acetylation of the hydroxy group, the remaining methylated samples were dried down, dissolved in anhydrous pyridine (200 ~1) and treated with acetic anhydride (50 ~1) for 40 min at 60°C. The reaction was stopped by adding methanol and solvents were evaporated under nitrogen. Acetylated HETE methyl esters were analysed by gas chromatography-electron impact mass spectrometry (GC-EIMS) conducted using a TRIO-2 apparatus equipped with a quadrupole mass analyser (VG Analytical) and a fused silica SE-30 capillary column (25 m X 0.32 mm) eluted using a programmed temperature gradient from 80 to 280°C at lO”C/min. 2.5. Bioassays Different types of treatment of H. magnipapillata with optically active HETEs were tested in order to study the effect on budding (both in intact and in regenerating hydra), ectopic head formation (in intact polyps) and tentacle regeneration (in excised gastric segments). HETEs were administered (a) in a single incubation after cutting, (b) in 2-h pulses at several consecutive days before cutting, and (cl in daily 2-h pulses without decapitation. Statistical analyses of means were conducted by using either the Student’s t-test or the Mann-Whitney U test. Ratios of budding activities (buds produced per individual and unit time) in experimental and reference groups were checked for statistical significance using the Fisher-Yates x2 test.

217

2.5.1. Budding in intact polyps and ectopic head formation

In daily treatments, budding polyps were collected and divided into groups of at least 30 specimens each and treated with: (a) R-or S-HETEs only at three different doses; (b) HETE pure enantiomers plus DAG (50 PM); (c) HETE pure enantiomers plus DAG plus NDGA (5 or 10 PM); (d) DAG only (50 PM); (e) DAG + NDGA only; (f) incubation medium only (control). In a different set of experiments 12-S-HPETE was also tested. Treatment was done daily, beginning 4 h after feeding, over a period of 12 days. Ethanolic solutions of the substances to be tested were put in 4.5 ml polystyrol dishes and the alcohol evaporated with a stream of nitrogen immediately before use; incubation with optically pure HETEs was carried out for 16 h, whereas treatments with DAG or DAG + NDGA were conducted for only 2 h. A 30 min pre-incubation with NDGA was also carried out. Buds released over a period of 9-10 days (counted from day 3 from the beginning of treatment) were taken into account as a measure of budding activity. Animals with tentacles in the gastric region formed up to 12 days after the beginning of treatment were used as a measure of ectopic head formation. 2.5.2. Tentacle regeneration in excised animals These tests were conducted only with optically pure ll-HETEs in order to provide a comparison with analogous data previously shown for these compounds in H. vulgaris [12]. The same procedure described previously was carried out [12].

3. Results 3.1. H. magnipapillata cytosolic extracts convert AA into two major metabolites

When incubated with AA, membrane-free homogenates of H. magnipapillata converted it into two UV (234 nm) sensitive metabolites which were purified and tentatively identified by means of their HPLC retention times as ll- and 1ZHETE (Fig. 1). Incubation of [14C]AA yielded two radioactive peaks co-eluting with the above UV visible peaks (Fig. l), thus substantiating their biosynthetic origin as AA metabolites. No AA oxidising activity was detected in boiled extracts. The amounts of the two metabolites were calculated from the areas of their UV peak on the assumption, justified by data shown below, that they were ll- and lZHETE, and it was found that AA conversion into each compound was respectively 2.6 and 1.1% per mg total protein per ml homogenate. These values are in agreement with data previously found for AA conversion into ll-HETE in H. vulgaris

T. Leitz et al. / Biochimica et Biophysics Acta 1213 (1994) 215-223

218

[12], and indicate that, like the latter but differently from other hydroid species [lo], also H. mugnipupillata represents an abundant source of AA metabolising enzymatic activity. When incubations were conducted in the presence of AA cascade inhibitors, no significant effect was found with the cytochrome P-450 inhibitor proadifen (75 PM) on the amounts of both metabolites, while the LO inhibitor NDGA exerted an inhibition on the levels of the peak corresponding to 12-HETE, but only at a high concentration (Table 1). 3.2. Full structure characterisation of AA metabolites The ‘H-NMR spectra (region from 6 2.5 to 7.0) of synthetic and putative endogenous lZHETE, shown in Fig. 2, were perfectly superimposable, the only differences being due to traces of non-deuterated solvents (from the purification procedure) in the natural metabolite. In particular, the signal at 6 2.98 (dd, 2H, C-7), due to a methylene group between a double bond and a diene, allowed to distinguish between 1ZHETE and ll- or 9-HETE, while the absence of a signal at 6 4.34, substituted by a signal at 6 4.19 (dd, lH, C-121, ruled out the possibility of the compound being a hydroperoxide [12]. However, NMR analysis does not distinguish between 12- and 8-HETE, although the HPLC retention time of the latter is quite different from that of 12-HETE (Fig. 1). GC-EIMS analysis of the acetylated methyl ester of this metabolite was carried out in order to conclusively determine its chemical structure. This type of derivatisation was preferred to trimethylsilylation, since the latter usually produces a higher degree of contamination (from the silylating agents) which was to be avoided due to the very low amounts of metabolites. The major GC peak co-eluted with

L

12-HETE acetylated methyl ester standard. The fragmentation pattern observed, with peaks at m/z = 317 (loss of *O-CO-Me), 316 (loss of acetic acid), which generated respectively the fragments at m/z = 206 and 205 (cleavage between C-12 and C-13), and peaks at m/z = 193 (cleavage between C-12 and C-11), 265 (cleavage between C-12 and C-131, the latter generating a fragment at m /z = 223 (loss of chetene CH,=C=O from the 265 ion), established the site of oxidation on C-12. A peak at m/z = 235 (loss of 141 from the molecular ion, due to the cleavage between C-7 and C-S), was also found. 11-HETE-like metabolite was less abundant than 12-HETE and its full structural elucidation was carried out uniquely by means of the GC-EIMS technique. Also in this case the major GC peak co-eluted with ll-HETE acetylated methyl ester standard. The mass spectrum of the acetylated methyl ester is shown in Fig. 3. Again fragments at m/z = 316 and 195 corresponded to loss of acetic acid and cleavage between C-10 and C-11 respectively. The second fragment also generated the peaks at m/z = 153 and 135 upon loss of either a chetene or acetic acid. A fragment at m/z = 167, due to cleavage between C-10 and C-9, was also observed. Amplification of peaks in the mass range 200-340 a.m.u. (only partially visible in Fig. 3) showed also the presence of fragments at m/z = 317, 285, 229 and 215, due to fragmentation on the carboxyend of the molecule (loss of *O-CO-Me from the whole molecule and of -OMe, -(CH,),-COOMe and -(CH,l,-COOMe from m /z = 3161, and at m/z = 259 and 245 due to fragmentation on the aliphatic end of the molecule (loss of butanyl and pentenyl groups from m/z = 316). This fragmentation pattern is completely compatible with the structure of ll-HETE for the second AA metabolite in H. magnipapillata cytosolic extracts. Enantiomeric composition of natural ll- and 12HETE was determined by means of chiral phase HPLC of the methyl esters, following a procedure previously described [12,23]. UV chromatograms of two typical runs are shown in Fig. 4. ll-HETE was composed of 87.2% R enantiomer and, surprisingly, 12-HETE was mostly (95.5%) present as the S enantiomer, thus suggesting that the two enzymatic activities leading to the synthesis of these two metabolites act in an enantioselective fashion. 3.3. Biological activity of 11- and 12-HETE papilla ta

in H. magni-

5

Fig. 1. Typical UV (234 nm) and radioactivity profiles of normal phase HPLC purification of a sample from Sep-pak extraction of H. magnipapilluta cytosolic extracts incubated with AA. Chromatographic conditions are described in Section 2. Arrows show 8-, 9-, lland 12-HETE standard elution times.

The effect of various combinations of DAG, NDGA, and 1ZHETE and of 12-S-HPETE on budding rate (BR) and ectopic head formation (EHF) in intact hydra upon daily treatment is shown in Tables 2 and 3. Both enantiomers of ll-HETE exerted a R and S ll-

T. Leitz et al. /Biochimica Table 1 The effect of inhibitors magnipapillata extracts Inhibitor

Dose

et Biophysics Acta 1213 (1994) 215-223

219

100

on ll-HETE

Enzyme

and 1ZHETE

inhibited

12-HETE

production

by H.

ll-HETE

(/.LM) Proadifen

75

NDGA

20

Cytochrome P-450 mono-oxygenases Lipoxygenase(s)

86.8 + 27.1

98.9* 11.6

51.0+ 10.9

108.5 + 16.3

%FS

ll-HETE and 12-HETE production was measured as described in Section 2. Data are expressed as percent of control cpm+S.E. (n = 3). Control values were 90+ 17 and 136 +40, respectively, for ll- and 12-HETE (S.E., n = 3). Only amounts exhibiting the highest effect are shown.

0

m/z 50

150

100

250

200

Fig. 3. EIMS spectrum of acetylated ll-HETE magnipapillata extracts incubated with AA. cussed in Section 3.

strong inhibition of bud formation only at a 20 PM concentration, but were more effective in inhibiting DAG-induced EHF already at a 2 FM dose. The LO inhibitor NDGA (5 PM) also inhibited EHF, and its effect was additive most significantly to that of 11-S

300

350

methyl ester from H. Fragmentation is dis-

HETE. Up to an optimal low micromolar concentration, both 12-S-HETE and 12-SHPETE significantly increased BR, while exerting the opposite effect at

B

b A

lwm

8.76

em

8.s

em

1.71

s.50

5.25

5.ca

4.75

4.50

4.25

rm

w5

3.54

3.a

3.m

27s

2.60

Fig. 2. ‘H-NMR spectrum (500 MHz, CDsOD, region from 6 2.5 to 7.0) of lZHETE-like metabolite from H. magnipapillata extracts with AA (A), in comparison with that of authentic synthetic 12-S-HETE (25 @g) (B). Dots indicate signals due to HETEs.

incubated

220

T. Leitz et al. /Biochimica

Table 2 Effect of daily treatment

with II-HETEs

on budding

Substance applied

(PM)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)

2 20 2 20 50 50 + 2 50 + 2 50 + 5 50+5+2 50+5+2

Dose

Control 1l-S-HETE 1 I-S-HETE 1 I-R-HETE 1 I-R-HETE DAG DAG + 11-S DAG + 11-R DAG + NDGA DAG + NDGA DAG + NDGA

35 polyps in Y days, (see text) errors are

were and and not

+ 11-S + 11-R

et Biophysics Acta 1213 (1994) 215-223

rates and DAG-induced BR

head formation % Animals

(EHF)

in intact

with EHF

(P vs. (1)) 5.4 4.9 1.9 4.8 1.0 4.0 5.2 5.2 3.7 4.5 3.9

animals Significance (PI

0 0 0 0 0 67 14 11 49 24 35

n.s. < 0.01 ns. < 0.001 n.s. n.s. n.s. < 0.05 n.s. n.s.

vs. (6) < 0.001 vs. (6) < 0.001 vs. (9) < 0.05 vs. (9) < 0.05

incubated daily with the various mixtures of drugs, previously deposited onto Petri dishes, for 2 h. BR is buds released per animal provides an estimate of cell proliferative activity. The formation of ectopic heads is directly linked to ‘head activation potential’ was measured after 12 days from incubation with drugs. DAG, dioctanoylglycerol; NDGA, nordihydroguaiaretic acid. Standard shown for the sake of clarity and were never higher than 15%. n.s., not significant at the 5% level of significance.

high doses. 12-S-HPETE counteracted NDGA inhibition of BR, and enhanced BR also in gastric segments excised from daily-treated animals (Table 3), thus suggesting an effect on ‘head activation potential’. 12-RHETE was ineffective at concentrations up to 2.0 PM (not shown). Data on the effect of 12-S-HETE and 12-S-HPETE on EHF are not yet available, for the assay demands quantities of HETEs which can be procured only in a long-term project. Administered to excised gastric segments, 11-R- and ll-S-HETE (12.5 PM) exerted a slight inhibition of the average tentacle number regenerated in 7 days (respectively 6.98 + 0.12, not significant, and 6.68 & 0.15, P < 0.01, vs. 7.31 + 0.12 in control, means * S.D., n = 40, 50 and 50). Higher doses of the two compounds were toxic to decapitated hydra. 12-S-HETE and 12-STable 3 Effect of daily repeated

treatment

Substance applied (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

ectopic

Significance

Control 1 12-S-HPETE 12.S-HPETE 12-SHPETE DAG DAG + 12-S DAG + NDGA DAG + NDGA DAG + NDGA Control 2 12-S-HETE 12-S-HETE

with 12-S-HPETE Dose

or 12-S-HETE Animals

HPETE had no significant influence on ATN (data not shown).

4. Discussion We presented here evidence for the enzymatic conversion of AA into 11-R- and 12-S-HETE in H. magnipupilluta cytosolic extracts. This finding is in agreement with previous data suggesting the presence of lipoxygenase-like enzymes in marine and freshwater hydroids [lo-141 and represents an unusual example, in invertebrates, of (an) enzyme(s) catalysing the regiospecific and enantioselective addition of 0, to AA with simultaneous formation of two different metabolites. Theoretically, apart from LO action, these two compounds

on budding BRa

+ 12-S + 12-S-

1.6 16

Significance

BRb

71 35 35 35 31 31 36 38 35 47 48 49

3.1 4.2 4.0 2.3 2.3 2.8 1.2 1.6 1.5 4.5 5.6 3.7

vs. (1) vs. (1) vs. (1) vs. (1) n.s. vs. (5) vs. (5) vs. (5)

Significance (PI

(P)

(PM) 0.25 2.5 12.5 50 50 + 2.5 50 + 5 50 + 5 + 2.5 50 + 5 + 12.5

rate (BR)

< < < <

0.01 0.01 0.01 0.01

< 0.01 n.s. n.s.

vs. (10) < 0.001 vs. (10) < 0.001

1.01 0.96 1.54 1.31 1.71 2.21 1.03 1.12 1.43

vs. (1) n.s. vs. (1) vs. (1) vs. (5) vs. (7) vs. (7)

< 0.01 < 0.01 < 0.01 < 0.01 n.s. n.s.

N.D. N.D.

Animals were treated daily and the buds released per animal in 10 days scored. After 10 days the treatment was finished and the upper gastric column excluding the original budding zone was excised. Buds appearing in newly established budding zones within 5 days from cutting were scored. Buds appearing in intact animals indicate proliferative activity, while buds appearing in regenerating segments indicate the level of ‘head activation potential’ (see text). Two separate sets of experiments were conducted for 12-S-HPETE and 12-S-HETE, and relative controls are shown. Means were compared using the Fisher-Yates x2 test. Standard errors are not shown for the sake of clarity and were never higher than 15%. n.s., not significant at the 5% level of significance; N.D., not determined. Abbreviations as in Table 2. a before cutting after 10 days and b 5 days after cutting.

T. Leitz et al. /Biochimica

0100

dbo

lO.kl

15100 Remrnb”

et Biophysics Acta 1213 (1994) 215-223

20100 omn,min,

Fig. 4. Typical UV (234) profiles of chiral phase HPLC analyses of methylated ll-HETE (upper trace) and methylated 1ZHETE (lower trace) from H. magnipapillata extracts incubated with AA. Arrows show the retention times of synthetic R- and S-enantiomers of each HETE. Chromatographic conditions are described in Section 2.

might derive also from: (1) enantiospecific cytochrome P-450-mediated epoxydation of AA on the All double bond of AA with formation of the epoxydic intermediate 11,12-epoxy-eicosatrienoic acid followed by non regiospecific reduction at C-11 or C-12 and simultaneous production of 12- and ll-HETE respectively, through a mechanism unprecedented in nature; (2) enantiospecific, non-regiospecific cytochrome P-450mediated hydroxylation of either C-11 or C-12 of AA. However, two observations seem to indicate the unlikelihood of both hypotheses. First, cytochrome P-450 catalysed oxidations have seldomly been reported to occur in an enantioselective and regiospecific fashion [25]. More importantly, co-incubation of cytosolic extracts with the cytochrome P-450 inhibitor proadifen, conducted in this study, did not prevent the production of either metabolite. The LO inhibitor NDGA, at a dose (20 PM) corresponding to the IC,, for the inhibition of mammalian 12-S-LO, but much lower than that reported to inhibit cylooxygenase (100 PM), exerted a selective inhibition on 12-HETE production, thus sug-

221

gesting that the enzymatic activity leading to 1ZHETE formation is not only an LO-like enzyme, but is distinct from that catalysing the formation of ll-HETE. The latter is probably analogous to the abundant ll-R-LO previously described in another Hydra species, H. vulguris [12], which was also unaffected by proadifen and NDGA. This enzymatic activity was not influenced by 5,8,11-eicosatriynoic acid, another LO inhibitor, nor by inhibitors of cyclooxygenases, such as indomethacin, which, therefore, were not tested in this study. The finding of ll-HETE and 1ZHETE synthesising activities in H. magnipupillutu is partly in agreement with preliminary data concerning the presence of 5-, 9-, ll-, 12- and 15-HETE, as measured by GC-EIMS, in the same species [13]. In the present study the formation of 5-, 15- and 9-HETE was not detected. This discrepancy is probably due to the different procedure used here, where disruption of cellular compartments upon homogenisation might prevent some enzyme from exerting its catalytic action. On the other hand, this procedure was necessary: (a) in order to obtain a scale up of the amounts of the metabolites produced so as to determine their enantiomeric composition and to characterise their chemical nature with major confidence; and (b) to characterise in a preliminary way the type(s) of enzymes involved in HETE biosynthesis. The possible biological role of 11-R- and 12-S-HETE produced by H. mugnipupillutu homogenates was assessed bearing in mind: (1) the well known enantiospecific properties of HETEs as regulators of cell motility and proliferation in mammals [3] - both these cellular processes are of significance also in hydroid tentacle regeneration, ectopic head formation (EHF) and asexual reproduction through budding, e.g. in the complex of phenomena which critically depend on a quantitative cellular parameter termed ‘head activation potential’ [19,20]; (2) the previously reported enhancement by ll-R-HETE, and not by ll-S-HETE, of average tentacle number in regenerating polyps of H. vulgar& [12]; (3) the previously reported potentiation or inhibition of DAG-induced EHF, by AA or the LO inhibitor NDGA respectively, in H. magnipupillutu [14]. The possible influence on ‘head activation potential’ was studied primarily by assessing the effect of HETEs on DAG-induced EHF in intact hydra and on the budding activity of excised gastric segments during regeneration. In addition, the effect of ll-HETEs on tentacle regeneration in excised gastric fragments was scored. Finally, a possible influence of HETEs on cell proliferation could be inferred from data on the budding rate in intact animals: inhibition or activation of budding may reflect respectively an inhibition or activation of cell proliferation. From the data shown in Tables 2 and 3, it appears that the two eicosanoids exert in intact animals a bipolar control on the formation of buds (and, there-

222

T. Leitz et al. / Biochimica et Biophysics Acta 1213 (1994) 215-223

fore, on cell proliferation), albeit at different concentrations and with a different specificity. While the non-enantioselective inhibitory effect of 11-HETE seems to be shared by many compounds when using high concentrations (including 12-S-HETE and -HPETE, Table 3) of similar lipophilic nature [17], and may have no physiological significance, the specific facilitatory action exhibited by low micromolar doses of 12-S-HETE, and its likely precursor 12-S-HPETE, represents the first example of endogenous putative signal molecules displaying such a stimulator-y action in Hydra. Thus, the inhibition exerted by NDGA on bud formation (this study and [14]) may in part be accounted for by its inhibitory action on the 12-S-LO-like activity described here, although the latter was observed at a concentration (20 PM) higher than that necessary for inhibition of budding (5 PM), probably because of differences between the efficacy of the inhibitor in vivo and in vitro. Indeed, 12-S-HPETE counteracted NDGA inhibition at the same concentration (2.5 PM) exerting maximal enhancement of budding, With respect to ‘head activation potential’, both 1l-HETE enantiomers exhibited a suppressing effect when DAG-induced EHF was used as a criterion. This effect was observed in the low micromolar range (2 PM) and was additive to the suppressing effect of NDGA. Using a second criterion, i.e. inhibition of budding rate in regenerating gastric segments, this effect was confirmed (not shown). ll-HETFs are the first endogenous metabolites suppressing stimulus-induced increase of ‘head activation potential’ so far identified. Interestingly, NDGA exerted an analogous effect, and while low doses of ll-HETEs and NDGA acted additively, NDGA did not significantly potentiate the inhibition of DAG-induced EHF exerted by maximal doses of ll-HETEs. This suggests that ll-HETE inhibitory effect is exerted through a mechanism similar to that of NDGA, i.e. via inhibition of certain LOS, although more efficaciously. Other unidentified LO product(s), possibly derived from PUFA substrates different from AA (see [131), might be produced during repeated treatment of H. mag~ipapiIlata with DAG and induce EHF, while their synthesis would be counteracted by NDGA and, endogenously, by ll-HETE. This hypothesis is corroborated by previous studies on non-enantiospecific HETE-inhibition of LO activity, for example, in platelets [26], and fits well with the observation, described here, that NDGA does not inhibit 1I-HETE biosynthesis. ll-HETEs also influenced ATN (see Section 31, although the two enantiomers seemed to alter tentacle regeneration to significantly different degrees in H. magnipapillata (this study) and H. vulgaris [12]. In conclusion, the present report has described the co-existence of two abundant enantiospecific LO-like

activities in cytosolic extracts of Ii magnipapillata, a finding unusual in invertebrates, the only other example known being the co-localisation of R-11- and 12LOS in sea urchin eggs [9]. The main products of the catalytic action of these enzymes on AA, i.e. 11-R and I2-S-HETE, were suggested to play a modulatory role on cell proliferation and on the mechanisms controlling hydroid body pattern. Further studies will be aimed at fully characterising both enzymes as well as HETE molecular mode of action in the developmental biology of these basic multicellular organisms.

Acknowledgements

The authors wish to thank Dr. G. Cimino for his support, Dr. C. Gianfrani and Dr. A. Milone for their valuable assistance in some experiments and Mr. R. Turco for artwork. NMR and GC-ElMS analyses were performed at the ‘Servizio NMR de1 CNR’ and ‘Servizio di Spettrometria di Massa dell’universita’ di Napoli e de1 CNR’, and staff at both centres are acknowledged.

References

111Schewe, T. and Kuhn, K. (1991) Trends Biochem. Sci. 16, 369-373.

121Yamamoto, S. (1991) Free Radical Biol. Med. 10, 149-159. j31 Spector, A.A., Gordon, J.A. and Moore, S.A. (1988) Prog. Lipid

Res. 27,2X-329. [41 Chang, WC., Ning, CC., Lin, M.T. and Huang, J.D. (19931 in Eicosanoids and Other Bioactive Lipids in Cancer, Inflammation and Radiation Injury (Nigam, S., Honn, K.V., Marnett, L.J. and Walden, T.L. Jr., eds.1, pp. 463-466, Kluwer Academic. ISI Shim& T. and Wolfe, L.S. (1990) J. Neurochem. 55, l-15. [61 Piomelli, D., Shapiro, E., Zipkin, R., Schwartz, J.H. and Feinmark, S. (1989) Proc. Natl. Acad. Sci. USA 86, 8550-8554. [71 Meijer, L., Brash, A.R., Bryant, R.W., Ng, K., Mclouf, J. and Sprecher, II. (1986) J. Biol. Chem. 261, 17040-17047. WI Hill, E.M. and Holland, D.L. (1992) Proc. R. Sot. Lnnd. B 247, 41-46. 191 Hawkins, D.J. and Brash, A.R. (1987) J. Biol. Chem. 262, 7629-7634. i101 Di Marzo, V., Gianfrani, C., De Petrocellis, L., Milone, A., Villani, G. and Cimino, G. (1993) Comp. Biochem. Physiol., 106B, 901-906. [ill Leitz, T., Beck, H., Stephan, M., Lehmann, W.-D., De Petrocellis, L. and Di Marzo, V. (1994) J. Exp. Zool., in press. 1121Di Marzo, V., De Petrocellis, L., Gianfrani, C. and Cimino, G. (19931 Biochem. J. 295, 23-29. [131 Di Marzo, V., Gianfrani, C., De Petrocellis, L., Milone, A. and Cimino, G. (1994) Biochem. J. 300, 501-507. [141 Muller, W.A., Leitz, T., Stephan, M. and Lehmann, W.-D. (1993) Roux’s Arch. Dev. Biol. 202, 70-76. ]151 Di Marzo, V., De Petrocellis, L. and Cimino, G. (19931 in Eicosanoids and Other Bioactive Lipids in Cancer, Inflammation and Radiation Injury (Nigam, S., Honn, K.V., Marnett, L.J. and Walden, T.L. Jr., eds.), pp. 209-212, Kluwer Academic. [16] De Petrocellis, L., Di Marzo, V. and Cimino, G. (19931 Experientia 49, 57-64.

T. Leitz et al. / Biochimica et Biophysics Acta 1213 (1994) 215-223 [17] De Petrocellis, L., Di Marzo, V., Gianfrani, C. and Minei, R. (1993) Comp. Biochem. Physiol. C 105, 219-224. [181 Muller, W.A. (1989) Development 105, 309-316. [19] Muller, W.A. (1990) Differentiation 42, 131-143. [20] McWilliams, H.K. (1983) Dev Biol. 96, 239-257. [21] Sugiyama, T. and Fujisawa, T. (1977) Dev. Growth Diff. 19, 187-200. [22] Bradford, M.M. (1976) Anal. Biochem. 72, 248-254.

223

[23] Read, S.M. and Northcote, D.H. (1981) Anal. Biochem. 116, 53-64. [24] Brash, A.R. and Hawkins, D.J. (1990) Methods Enzymol. 187, 187-195. 1251 Fitzpatrick, F.A. and Murphy, R.C. (1989) Pharmacol. Rev. 40, 229-241. [261 Vanderbroeck, J.Y., Bryant, R.W. and Bailey, J.M. (1982) Biochem. Pharmacol. 31, 3463-3467.