- Email: [email protected]
Metabolism of arachidonic acid and prostaglandins in the torpedo electric organ: Modulation by the presynaptic muscarinic acetylcholine receptor
Neuroscience Vol. 13, No. 4, pp. 1359-1364, Printed in Great Britain
1984
0306-4522184 $3.00 + 0.00 Pergamon Press Ltd “8 1984IBRO
METABOLISM OF ARACHIDONIC ACID AND PROSTAGLANDINS IN THE TORPEDO ELECTRIC ORGAN: MODULATION BY THE PRESYNAPTIC MUSCARINIC ACETYLCHOLINE RECEPTOR I. PINCHASI, M. BURSTEIN and D. M. MICHAELSON* Department of Biochemistry, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv 69978, Israel Abstract-We have found that Torpedo electric organ readily synthesizes prostaglandin E, from both exogenous and endogenous arachidonate and that activation of the presynaptic muscarinic acetylcholine receptor increases the rate of prostaglandin E, synthesis by inducing the release of tissue arachidonate from its phospholipid pools. The incorporation of radiolabeled arachidonate into tissue phospholipids is slow and Ca *+ independent However, the electric organ slices readily oxidize the externally added, radiolabeled arachidonate via the cycle-oxygenase pathway, with prostaglandin E, being the major product (22 + 4% of the initial radioactivity). This process is not affected by either Ca2+ or mepacrine. Torpedo electric organ slices also synthesize prostaglandin E$ from endogenous substrates, and release it into the medium. This process, however, is enhanced by Ca2+ and inhibited by mepacrine. Activation of the Torpedo muscarinic acetylcholine receptor by the agonist oxotremorine results in a dose-dependent atropine-sensitive increase in the synthesis of prostaglandin E, from endogenous tissue substrates and in the concomitant release of arachidonate into the medium. By contrast, oxotremorine has no effect on either the formation of [‘%]prostaglandin E, from exogenous arachidonate, the incorporation of radiolabeled arachidonate into tissue phospholipids or its liberation from prelabeled slices. These results suggest that activation of the muscarinic acetylcholine receptor induces lipolysis which results in the liberation of endogenous arachidonate and its subsequent conversion to prostaglandin E,.
The cholinergic nerve terminals of the Torpedo electric organ contain presynaptic muscarinic acetylcholine receptors (mAChR) which regulate acetylcholine (ACh) release by feedback inhibition.4.5*8.9*” In a previous study we demonstrated that activation of these receptors stimulates the formation of a prostaglandin E2 (PGE,)-like substance by isolated Torpedo nerve terminals (synaptosomes) and presented data which suggest that this substance mediates the regulation of ACh release by mAChR.14 In a variety of mammalian preparations, activation of mAChR promotes a net breakdown of phos‘J*S This phospholipid contains a phatidylinositol. high proportion of arachidonic acid, which is the precursor of prostaglandins and leukotrienes. Recently, Deutsch and Kelly3 demonstrated that arachidonic acid is a prominent constituent of several phospholipids contained in the electric organ of Narcine and represents 6.5 mol% of the fatty acid content of total lipid extracts of this tissue. It seems reasonable to assume then, that the metabolism of arachidonate in the electric organ may play a role in the regulation of its cellular activities, and in particular, in neurotransmitter release. *To whom correspondence should be addressed. Abbreuiulions: ACh, acetylcholine; EGTA, ethyleneglycolhis@-aminoethylether)N,N’-tetra-acetic acid; mAChR, muscarinic acetylcholine receptor; PGE,, prostaglandin E 2.
In the present work we studied the metabolism of arachidonate in Torpedo electric organ slices in order to elucidate its role in mAChR activity and to investigate the biochemical mechanism underlying this phenomenon. EXPERIMENTALPROCEDURES Fish Torpedo ocelluta were caught live off the coast of Israel between December and May and kept in sea-water aquaria. The fish were sacrificed within a month of capture and their electric organs were excised at 4°C and cut into slices of the desired weight. The slices were then preincubated in modified elasmobranch buffer (vide infera) for about 60 min at 25°C to allow for recovery after excision, after which they were immediately transferred into a fresh medium for experimental use. Radiochemical assay of prostaglandin synthesis
Electric organ slices (l-1.5 g) were incubated at 25°C for the designated time intervals with [Vlarachidonic acid (0.5 x lo5 to 1 x lo5 cpm/ml) in 2 ml of the specified buffer. When the effects of indomethacin, mepacrine or muscarinic ligands were to be tested, they were added to washed slices 60 min before the addition of the “C-labeled arachidonate. The basic buffer employed was modified elasmobranch buffer containing (in mM): NaC1, 250; KCI, 4.8; MgCl,, 2.4; ethyleneglycolbis(B-aminoethylether)N,N’-tetra-acetic acid (EGTA), 0.1; D-glucose, 10; sucrose, 200; Hepes, 20; pH 7.2. When testing the effect of Ca’+ we either raised the EGTA concentration to 1 mM (Ca2+ = 0) or added excess CaCI, in the presence of 1 mM EGTA to yield the required final concentration of free Ca”.
1359
I360
I. Pinchasi rr al.
Following incubation the medium was acidified to pH 3.5 with I M citric acid and the prostaglandins were extracted twice with 20ml chloroform. The dried extract was dissolved in 0.03-0.05 ml of chloroform:methanol (2: I) and chromatographed on Silica Gel 60 plates (Merck), using either the AIX solvent system of Hamberg and Samuelsson” (ethyl acetate:water:iso-octane:acetic acid, I I: IO:5:2, upper phase) or the chloroform:methanol:acetic acid (90:5:5) solvent system. The samples were run in parallel to unlabeled prostaglandins FZu, El, Dz and A, and arachidonic acid. The spots were located by exposure of the plates to iodine vapor. Strips, 1cm in length, were scraped off into scintillation vials and counted in a Packard Tricarb liquid scintillator in 4.5 ml of Lumac’s Hydro-luma scintillation liquid. Radiochemical assay ofarachidonate incorporation into tissue phospholipids
Electric organ slices (0.35 + 0.05 g) were incubated at 25°C for the designated time intervals with [3H]arachidonate (I x IO’-1.5 x 10’cpm/ml) in 2ml of the specified buffer, as described in the previous section. The muscarinic ligands to be tested were added to washed slices 60min prior to the addition of [‘Hlarachidonate. Following incubation, the slices were rinsed in cold buffer and imediately homogenized in 2 ml of IOOmM acetate which contained 5 mM EGTA. The homogenate was extracted according to Bligh and Dyer’ and chromatographed on Silica Gel 60 plates with a concentrating zone (Merck) in a solvent system consisting of chloroform:methanol:acetic acid:water (lOO:20: 12: 5). In parallel, unlabeled phosphatidylserine, phosphatidylinositol, phosphatidylcholine, phosphatidylethanolamine and arachidonic acid were subjected to chromatography. The spots were located by exposure to iodine vapor, scraped off and counted as described above. Radioimmunoassay of prostagiandin E2
Washed electric organs slices (0.25 + 0.05 g) were incubated at 25°C for the designated time intervals in 2 ml of the specified buffer. When the effects of muscarinic ligands were tested they were added 90 min (atropine) and 60 min (oxotremorine) before the slices were transferred to a fresh medium. Indomethacin and mepacrine were also added 90 min before the final incubation. Medium samples were withdrawn at the indicated times and kept on ice until assayed. The content of PGE, in the samples was determined by means of a specific lz51 radioimmunoassay (New England Nuclear). The antiserum is highly specific, having a crossreactivity of only 3.7% with prostaglandin E, and even less (i 0.4%) with other members of the prostaglandin family. The assay is sensitive enough to enable the determination of 0.5-25 pg PGE,.
Protein
Protein was determined according to Bradford’ using bovine serum albumin as standard. Materials [3H]Arachidonate (86 Ci/mmol) and [“‘Clarachidonate (- 55 mCi/mmol) were from New England Nuclear (NEN). Oxotremorine was from Aldrich. Atropine, arachidonic acid, prostaglandins A,, D,, E, and F,a and all the phospholipids were from Sigma. Indomethacin was kindly donated by Dr. S. Cherkez of Teva, Israel. All other reagents were of analytical grade. RESULTS Metabolism
of radiolabeled
arachidonate
The ability of Torpedo electric organ to metabolize arachidonic acid was tested by incubating tissue slices with radiolabeled arachidonate, followed by extraction of both the medium and the tissue and separation of the products by thin layer chromatography. Analysis by the AIX solvent system (see Experimental Procedures) of the [I4Clarachidonate metabolites released by tissue slices into the medium during a 15-min period of incubation in modified elasmobranch buffer revealed the formation of a major product, which comigrated with a PGEz standard and which comprised 22 + 4% (n = 3) of the initial radioactivity. In addition, two minor peaks were detected (3.4 + l.So/, and 4.5 i 2%) which comigrated with prostaglandin F, r and DZ standards, respectively (Fig. 1). The identity of these radioactive products was further ascertained by separation of the medium extract in a different solvent system and by testing the
-
Determination of free fatty acids
Washed electric organ slices (1 g) were incubated at 25°C in 1 ml of modified elasmobranch buffer, which contained 1 mM free Ca2+ and 75 pg/ml indomethacin. Oxotremorine and atropine were added 60 and 90 mitt, respectively, before the slices were transferred to a fresh medium. Following a final incubation of 90 min, a sample (0.60.8 ml) was withdrawn and diluted ten-fold with Dole reagent (1 N H,SO,:n-heptane:iso-propanol, I : 10:40), which contained 5 pg of pentadecanoic acid (C 15:O) as internal standard. The mixture was incubated at 60°C for 10 min followed by extraction with n-heptane and back-extraction with ethanolic KOH (0.05 N KOH in 50% ethanol). The water phase, containing the K-salts of the fatty acids, was acidified to pH - 3 with 1 N H,SO,, extracted with heptane and dried under N,. Methylation of the free fatty acids was performed according to Schlenk and Gellerman.‘8 The final sample was dissolved in hexane and analyzed on a Packard model 417 gas chromatograph.
IA
1
L
1
IJ-43 I
I
I
I
I
Distance
I
I
I
I
I
10
5
from
origin
km)
Fig. 1. Separation of [“‘Clarachidonic acid metabolites, produced by Torpedo electric organ and released into the medium, by thin layer chromatography utilizing the AIX solvent system. Electric organ slices (l-l .5 g) were incubated at 25°C for 15 min in modified elasmobranch buffer containing [‘4C]arachidonate (0.5 x 10s-I x 10’cpm/ml). The metabolites released by the tissue into the medium were extracted and separated by the AIX solvent system. AA. arachidonic acid; A,, Dz, Ez, F,a; prostaglandins.
Muscarinic modulation of arachidonate metabolism in Torpedo
1361
affected by elevation of the EGTA concentration (to 1 mM), introduction of free Ca*+ (up to 2 mM) or addition of mepacrine (10-100 PM). The incorporation of labeled arachidonate into tissue phospholipids was examined under conditions similar to those employed above. After a 15-min period of incubation with [3H]arachidonate only 10% of the radioactivity was recovered in the tissue, all of it in the form of free arachidonate. With a longer period of incubation (1 h) about 30% of the added radioactivity was recovered in the tissue, of which about 10% was incorporated into the major tissue phospholipid classes: phosphatidylinositol, phos-
10
5
Distance
from
origin
(cm)
Fig. 2. Separation of [‘4C]arachidonic acid metabolites, produced by Torpedo electric organ and released into the medium, by thin layer chromatography utilizing the chloroform:methanol:acetic acid (90: 5: 5) solvent system. Electric organ slices were incubated as described in the legend to Fig. 1, in the absence (bold lines) or presence (broken lines) of
phatidylserine, phosphatidylcholine and phosphatidylethanolamine (Table 1). The remaining 90;,‘, was in the form of free arachidonate. This pattern was not significantly altered when much longer (6 h) incubation periods were employed (Table 1). The presence of Ca*+ (1 mM) had no effect on the incor-
poration
of labeled arachidonate I
into tissue phos-
I
,
I
60
70
5
indomethacin (30 pgg/ml).The medium was then extracted and separated by the chlorofonn:methanol:acetic acid (90:5 :5) solvent system. AA, arachidonic acid; D, , E, , F, a ; prostaglandins. effects of cycle-oxygenase inhibitors on their synthesis. As shown in Fig. 2, the profile of oxygenated arachidonate derivatives revealed by the chloroform:methanol:acetic acid (90:5:5) solvent system was similar to that obtained with the AIX solvent system (Fig. I), and consisted of a major peak (22 + 3%; n = 4) comigrating with the PGE, standard and two minor peaks (3.2 + 1.2% and 5 + 1.5%) comigrating with the prostaglandin F, tl and D, standards, respectively. Examination of the tissue blocks under the same conditions revealed that they were virtually devoid of [‘4C]PGE2 (less than 10% of the amount recovered in the medium). Inclusion of the cycle-oxygenase inhibitor indomethacin (30 pg/ml) in the incubation medium inhibited virtually all (> 95%) of the [‘4C]arachidonate oxidation (Fig. 2). Similar results were obtained with aspirin (300pg/ml; not shown). The extent and rate of conversion of [‘4C]arachidonate into “C-labeled products were not Table 1. Incorporation
10
20
30
40
50
Fig. 3. Prostaglandin E, synthesis by electric organ slices from endogenous arachidonate and the effects thereon of Ca2+ and indomethacin. Slices (0.25 + 0.05 g) were incubated at 25°C in modified elasmobranch buffer in the absence (0) or presence (H) of indomethacin (30pgg/ml), 1mM free Cal+ (0) or in the presence of 1 mM EGTA (A). At the designated time intervals aliquots were removed from the medium and assayed for their PGE, content by a specific radioimmunoassay. Results presented are the mean of three experiments and the SD was less than 15”d.
of [‘Hlarachidonic acid into Torpedo electric organ phospholipids Phospholipid
Conditions 15min lh 6h
Ca2+ = Ca*+ = Ca’+ = Ca” = Ca2+ = Ca2+ =
0 1 mM 0 1mM 0 1 mM
60
Time (mid
PI
PS
PC
N.D. N.D. 2.3 + 1.2 (4) 2.6 k 1.4(5) 4.5 * 1.3 (3) 4.8 k 2.4 (3)
N.D. N.D. 1.3+ 0.9 (4) 1.6 & 0.6 (5) 4.6 + 3.4 (3) 4.2 + 1.65 (3)
N.D. N.D. 2.0 f 0.2 (4) 3.2 + 0.8 (5) 5.3 _+2.6 (3) 6.2 + 2.2 (3)
PE N.D. N.D. 3.9 & 1.65 (4) 5.0 + 2.0 (5) 6.4 + 2.0 (3) 5.7 + 2.1 (3)
Results presented are average f SD of the per cent of the radioactivity recovered in the tissue which is incorporated into the corresponding phospholipid. Numbers in brackets designate the number of experiments Performed, each in duplicate. N.D., not detectable; PC, PE, PI, PS; phosphatidylcholine, ethanolamine, inositol and serine respectively.
1. Pinchasr el al.
I362
1 1
! I
15 Time
30
10
(mm)
20
30
40
Oxotremorme
50
[pbl]
Fig. 4. Effect of muscarinic ligands on the synthesis of PGE, from electric organ precursors. Washed slices (0.25 k 0.05 g) were incubated at 25°C in modified elasmobranch buffer containing 1 mM free Ca2+. in the absence and presence of muscarinic ligands, as described in the Experimental Procedures. (A) The time dependence of the effect of oxotremorine and atropine on PGE, synthesis. 0 corresponds to control. 0, A, 4 correspond to 10, 20 and 50 pM oxotremorine. 0 corresponds to 50 PM oxotremorine and 10 PM atropine. (8) Dose/response curve for the effect of oxotremorine on PGE, synthesis. Data presented is from (A) and corresponds to the amount of PGE, synthesized following a 30 min incubation period in the absence (100%) and presence of the indicated concentrations of oxotremorine. Results presented are the mean of three experiments and the SD was less than 15%.
pholipids at any of the incubation (Table 1).
periods tested
Synthesis of prostaglandin E, from endogenous arach idonate
Incubation of tissue slices in modified elasmobranch buffer resulted in a massive, indomethacinsensitive synthesis of PGEz and in its liberation into the medium (Fig. 3). In contrast to the conversion of [I4Clarachidonate to radiolabeled PGE,, the synthesis of PGE, from endogenous arachidonate was markedly affected by the concentration of free Ca2+. Elevation of the EGTA concentration to 1 mM reduced PGE, liberation to about 50% of control levels;
Table 2. Effect of muscarinic ligands on the release of fatty acids from electric organ slices
Control Fatty acid (nmol/gh) ______ Cl, 5.1 + 1.4 CL, 4.6 ri: 1.7 C IXI 2.6 k 0.85 C ikz 0.37 5 0.09 C 204 I. 1 + 0.28 C,?., 4.3 +_0.28
Effect of muscarinic ligands (“/, of control) Oxotremorine Oxotremorine + atropine 125 k 30 106 + 15 116k20 13Ok29 171+ II 115+ 15
128 + 20 114 + 10 114* 10 120f20 96 + 20 108 + 10
Control corresponds to the electric organ tissue slice suspended in modified elasmobranch buffer which contained 1 mM Ca*+ and no muscarinic ligands. The release of free fatty acids in the presence of oxotremorine (50 fl M) and oxotremorine (50 p M) + atropine (10 p M) is expressed as per cent of the amount of the corresponding fatty acid which was released in the absence of muscarinic ligands. Results presented are average + SD of three experiments.
conversely, introduction of free Ca2+ (1 mM) resulted in a two-fold increase in the liberation of PGE, (Fig. 3). Furthermore, mepacrine (50 FM) inhibited PGE2 release by 50% (not shown). Effect of metabolism
muscarinic
ligands
on
arachidonate
Addition of either the muscarinic agonist oxotremorine (l-100pM) or the antagonist atropine (O.l-10pM) had no effect on the oxidation of [‘4C]arachidonate to [14C]PGE2 (105 f 8% and 100 f 5% of control, respectively; n = 3). On the other hand, the production of PGE, from endogenous substrate was strongly affected; oxotremorine caused a dose-dependent increase in PGE, synthesis of up to 230% of control levels, which was reversed by atropine (Fig. 4). These results can be interpreted as implying that activation of the presynaptic mAChR induces the liberation of arachidonic acid from tissue phospholipids rather than directly stimulates its oxidation to PGE,. This hypothesis was further investigated by measuring directly the effects of muscarinic ligands on the liberation of endogenous and radiolabeled arachidonate from tissue blocks. Incubation of tissue slices in modified elasmobranch buffer resulted in a basal release of several long-chain fatty acids into the medium, at a rate ranging between 0.5 and 5 nmol/g h (Table 2). Addition of oxotremorine (50/*M) increased the release of arachidonate to 17O”/0of control levels. This effect was reversed by atropine (10 PM) which by itself had no effect (Table 2). In contrast to its effect on endogenous arachidonate, oxotremorine had no effect on either the incorpo-
Muscarinic modulation of arachidonate metabolism in Torpedo ration of [‘Hlarachidonate into tissue phospholipids or on the rate of its liberation from prelabeled slices. DISCUSSION
The metabolism of exogenous, radiolabeled arachidonate and of endogenous arachidonate were studied in the Torpedo electric organ, and the effects thereon of activation of the presynaptic mAChR were investigated. Incubation of tissue blocks with labeled arachidonate result in the formation and release of a major product, which contains -25% of the added radioactivity. On the basis of its comigration with a PGE, standard in two solvent systems, as well as the complete inhibition of its synthesis by indomethacin and aspirin (Figs 1 and 2), this product was identified as PGEr Since virtually no labeled PGE, is found in the tissue under these conditions, it seems that [14C]PGE, is released into the medium immediately following its synthesis. In addition to [‘4C]PGE2, the electric organ slices synthesize small amounts of indomethacin-sensitive substances, which comigrate with prostaglandins F, c( and D, (Figs 1 and 2). This profile of oxidized [‘4C]arachidonate metabolites is qualitatively similar reported for Torpedo to that previously synaptosomes14 except that the extent of the conversion of [‘4C]arachidonate by the synaptosomes (- 3%) is much lower than that observed with the tissue slices (3079, and that the synaptosomes synthesize an additional major product (product II), which comigrates with prostaglandin D2 but is resistent to indomethacin.‘4 A similar oxidation profile was observed with tissue homogenates before fractionation. Interestingly, when homogenization is performed under conditions which minimalize cycloinactivation oxygenase (N,-saturated buffer containing I:/, fatty acid-free bovine serum albumin), the amount of [‘4C]PGE, formed by synaptosomes is increased five-fold and that of product II is drastically reduced. However, under these conditions the synaptosomes are depleted of their ACh content and lose their ability to release ACh when depolarized in the presence of Ca*+ (unpublished observations). Thus the synaptosomes are not the most suitable preparation for the parallel investigation of ACh release and arachidonate metabolism. Synthesis of PGE2 from endogenous arachidonate was demonstrated by employing a highly specific radioimmunoassay. The results obtained (0.2 + 0.03 nmol/gh, n = 4) are very similar to those previously found using the rat stomach bioassay (0.25 f 0.02 nmol/gh, n = 3) (unpublished observations). As expected, the synthesis of PGE, by tissue blocks is totally inhibited by indomethacin and aspirin (Fig. 3). It thus appears that Torpedo electric organ oxidizes both exogenous and endogenous arachidonic acid via the cycle-oxygenase pathway, with PGE, being the major product.
1363
The mechanism of synthesis of PGE, from radiolabeled arachidonate was compared to that from endogenous arachidonate. Two major differences were found: (1) the formation of PGE, from endogenous arachidonate is stimulated by Ca*+ while that of labeled PGE, is not; (2) mepacrine has no effect on the synthesis of [‘4C]PGE2 but inhibits the conversion of endogenous arachidonate into PGE,. These results are interpreted as suggesting that while exogenous arachidonate is directly converted into PGE,, the endogenous precursor has to be liberated from a phospholipid pool prior to oxidation. This contention is also supported by the finding that following a brief incubation of the tissue with labeled arachidonate, there is no detectable incorporation of the label into tissue phospholipids, whereas labeled PGE, is massively formed (Table 1 and Figs 1 and 2). Much longer incubation periods (up to 6 h) are needed in order to achieve detectable incorporation of radioactivity into tissue phospholipids. Under these conditions the major phospholipids become uniformly labeled, in a Ca2+-independent manner (Table 1). Incubation of tissue slices with the muscarinic agonist oxotremorine causes a marked dosedependent and atropine-reversible increase in the liberation of PGE2 into the medium (Fig. 4). The concentration of oxotremorine inducing a halfmaximal effect (10 PM) is similar to its previously reported binding constant to Torpedo mAChR.8,9 These results are similar to those previously observed with Torpedo synaptosomes, using the rat stomach bioassay. I4 Comparison of the specific rates of the muscarinic stimulation of PGE, formation with 50 p M oxotremorine shows that the synaptosomes synthesize - 30 pmol of PGE,/mg protein/min, while the tissue slices synthesize -0.50 pmol of PGE,/mg proteimmin. This marked enhancement in the specific “responsiveness” of synaptosomes to muscarinic activation relative to that of intact tissue, is consistent with their relative enrichment (by more than ten-fold) in muscarinic receptors, as determined by ligand binding.* Muscarinic stimulation can increase PGE, formation either by activating the oxidation of arachidonic acid or by increasing the arachidonate pool available for oxidation through triggering its liberation from phospholipids. In order to distinguish between these possibilities we examined the effect of oxotremorine on the conversion of [‘4C]arachidonate to [‘4C]PGE2 and on the release of endogenous arachidonate. Our results show that neither oxotremorine nor atropine has a significant effect on the conversion of [‘4C]arachidonate to [‘4C]PGE,. In contrast, oxotremorine induces an atropine-sensitive increase in the release of endogenous arachidonate, into the medium (Table 2). The fact that the release of C22:6, which is a major constituent of Narcine electric organ phospholipids3 is not affected by OXotremorine implies that activation of the presynaptic mAChR induces selective lipolysis.
1364
I. Pinchasi et (II.
The basal incorporation of labeled arachidonate into tissue phospholipids is very low even after prolonged incubation (Table 1). This observation and the finding that oxotremorine has no effect on either the incorporation of [3H]arachidonate into Torpedo phospholipids or on its liberation from prelabeled tissue, imply that the turnover of phospholipids in Torpedo is very slow and that externally added arachidonate is not readily incorporated into the receptor-linked phospholipid pool. The nature of the lipolytic event induced by muscarinic stimulation and the mechanism of coupling between mAChR activation and the lipolytic enzymes are not known. In many systems mAChR coupling to phosphatidylinositol turnover is thought to be a possible mechanism by which external stimuli are transduced into the ~~41.‘~“~‘~ Since phosphatidylinositol is rich in arachidonate esterified to the 2-Sn position, it is possible that a phospholipaseA, is activated by muscarinic stimulation, resulting in
the release of arachidonate and in its subsequent conversion to PGE,. Such a mechanism has indeed been shown in non-neuronal systems.‘“,” Our finding that mepacrine inhibits the formation of PGE, from endogenous substrates supports this contention. We have recently shown that Torpedo nerve terminals contain a guanosine 5’-triphosphate-binding protein, which is adenosine 5’-pyrophosphateribosylated by cholera toxin,‘O and which regulates the binding of muscarinic agonists to the receptor.4,‘Y It is tempting to suggest that this protein may be involved in the coupling between the mAChR and the enzymatic apparatus responsible for the generation of PGE, Acknowledgements-We are grateful to Prof. A. Raz of the Department of Biochemistry, Tel-Aviv University, for his valuable advice and practical help throughout this study. This work was supported in part by grants from the
Muscular Dystrophy Association and the U.S.-Israel Binational Science Foundation (BSF Grant No. 2699).
REFERENCES I. Bligh E. G. and Dyer W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 91 l-917.
2. Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt. Biochem. 72, 248-254. 3. Deutsch J. W. and Kelly R. B. (1981) Lipids of synaptic vesicles: relevance to the mechanism of membrane fusion. Biochemistry 20, 278-285.
4. Dowdall M. J., Golds P. R. and Strange P. G. (1982) Properties of Torpedo electric organ muscarinic receptors. J. Physiol., Paris 78, 379-384.
5. Dunant Y. and Walker A. I. (1982) Cholinergic inhibition of acetylcholine release in the electric organ of Torpedo. Eur. J. Pharmac. 78, 201-212.
6. Hamberg M. and Samuelsson G. (1966) Prostaglandins in human seminal plasma: prostaglandins and related factors, J. biol. Chem. 241, 257-263.
7. Hawthorne J. N. and Pickard M. R. (1979) Phospholipids in synaptic function. J. Neurochem. 32, 5-14. 8. Kloog Y., Michaelson D. M. and Sokolovsky M. (1978) Identification of muscarinic receptors in the Torpedo electric organ: evidence for their presynaptic localization. FEBS Left. 95, 331-334. 9. Kloog Y., Michaelson D. M. and Sokolovsky M. (1980) Characterization of the presynaptic muscarinic receptor in synaptosomes of Torpedo electric organ by means of kinetic and equilibrium binding studies. Bruin Res. 194, 97- 115. 10. Lester H. A., Steer M. L. and Michaelson D. M. (1982) ADP-ribosylation of membrane proteins in cholinergic nerve terminals. J. Neurochem. 38, 1080-1086. II. Michaelson D. M., Avissar S., Kloog Y. and Sokolovsky M. (1979) Mechanism of acetylcholine release: possible involvement of presynaptic muscarinic receptors in regulation of acetylcholine release and protein phosphorylation. Proc. natn. Acad. Sci. U.S.A. 76, 6336-6340.
12. Michell R. H. (1979) Inositol phospholipids in membrane function. Trends Biochem. Sci. 3, 128-13 1. 13. Michell R. H. (1982) Phosphatidyl inositol turnover in signal transduction. Neurosci. Res. Prog. Bull. u), 338-350. 14. Pinchasi I., Shanietzki B., Schwartzman M., Raz A. and Michaelson D. M. (1982) Is presynaptic muscarinic inhibition of acetylcholine release mediated by second messengers? The role of cyclic AMP and prostaglandins. In Presynaptic Receptors, Mechanisms and Functions (ed. DeBelleroche J.), pp. 114129. Ellis Horwood, Chichester. 15. Pinchasi I., Burstein M. and Michaelson D. M. (1985) Prostaglandins mediate the muscarinic inhibition of acetylcholine release from Torpedo nerve terminals. In Molecular Basis of Neme Actiaity (eds Changeux J. P., Hucho F., Maelicke A. and Neumann E.). Walter de Gruyter. In press. 16. Putney Jr. J. W. (1981) Recent hypotheses regarding the phosphatidyl inositol effect. Life Sci. 9, 1183-l 194. 17. Rubin R. P. (1982) Calcium-phospholipid interactions in secretory cells: a new perspective on stimulus-secretion coupling. Fedn Proc. Fedn Am. Sots exp. Biol. 41, 2181-2187. 18. Schlenk H. and Gellerman J. L. (1960) Esterification of fatty acids with diazomethane on a small scale. Analyt. Chem. 32, 1412-1416. 19. Sokolovsky M., Avissar S., Egozi Y., Gurwitz D.. Kloog Y. and Michaelson D. M. (1981) Agonist and antagonist interaction with muscarinic pre- and postsynaptic receptors: biochemical characterization. In Neurotransmitters and Their Receptors (eds Littauer U., Dudai Y., Silman I., Teichberg V. and Vogel Z.), pp. 257-270. John Wiley, New York. (Accepted 14 May 1984) Note added in proof. Recent experiments in which we measured concomitantly the effects of oxotremorine on ACh release and on the metabolism of arachidonate in tissue slices revealed that the pattern of PGs synthesized following muscarinic activated depends on the time of its measurement following exision of the electric organs. These findings are presented elsewhere.”
1984
0306-4522184 $3.00 + 0.00 Pergamon Press Ltd “8 1984IBRO
METABOLISM OF ARACHIDONIC ACID AND PROSTAGLANDINS IN THE TORPEDO ELECTRIC ORGAN: MODULATION BY THE PRESYNAPTIC MUSCARINIC ACETYLCHOLINE RECEPTOR I. PINCHASI, M. BURSTEIN and D. M. MICHAELSON* Department of Biochemistry, The George S. Wise Faculty of Life Sciences, Tel Aviv University, Ramat Aviv 69978, Israel Abstract-We have found that Torpedo electric organ readily synthesizes prostaglandin E, from both exogenous and endogenous arachidonate and that activation of the presynaptic muscarinic acetylcholine receptor increases the rate of prostaglandin E, synthesis by inducing the release of tissue arachidonate from its phospholipid pools. The incorporation of radiolabeled arachidonate into tissue phospholipids is slow and Ca *+ independent However, the electric organ slices readily oxidize the externally added, radiolabeled arachidonate via the cycle-oxygenase pathway, with prostaglandin E, being the major product (22 + 4% of the initial radioactivity). This process is not affected by either Ca2+ or mepacrine. Torpedo electric organ slices also synthesize prostaglandin E$ from endogenous substrates, and release it into the medium. This process, however, is enhanced by Ca2+ and inhibited by mepacrine. Activation of the Torpedo muscarinic acetylcholine receptor by the agonist oxotremorine results in a dose-dependent atropine-sensitive increase in the synthesis of prostaglandin E, from endogenous tissue substrates and in the concomitant release of arachidonate into the medium. By contrast, oxotremorine has no effect on either the formation of [‘%]prostaglandin E, from exogenous arachidonate, the incorporation of radiolabeled arachidonate into tissue phospholipids or its liberation from prelabeled slices. These results suggest that activation of the muscarinic acetylcholine receptor induces lipolysis which results in the liberation of endogenous arachidonate and its subsequent conversion to prostaglandin E,.
The cholinergic nerve terminals of the Torpedo electric organ contain presynaptic muscarinic acetylcholine receptors (mAChR) which regulate acetylcholine (ACh) release by feedback inhibition.4.5*8.9*” In a previous study we demonstrated that activation of these receptors stimulates the formation of a prostaglandin E2 (PGE,)-like substance by isolated Torpedo nerve terminals (synaptosomes) and presented data which suggest that this substance mediates the regulation of ACh release by mAChR.14 In a variety of mammalian preparations, activation of mAChR promotes a net breakdown of phos‘J*S This phospholipid contains a phatidylinositol. high proportion of arachidonic acid, which is the precursor of prostaglandins and leukotrienes. Recently, Deutsch and Kelly3 demonstrated that arachidonic acid is a prominent constituent of several phospholipids contained in the electric organ of Narcine and represents 6.5 mol% of the fatty acid content of total lipid extracts of this tissue. It seems reasonable to assume then, that the metabolism of arachidonate in the electric organ may play a role in the regulation of its cellular activities, and in particular, in neurotransmitter release. *To whom correspondence should be addressed. Abbreuiulions: ACh, acetylcholine; EGTA, ethyleneglycolhis@-aminoethylether)N,N’-tetra-acetic acid; mAChR, muscarinic acetylcholine receptor; PGE,, prostaglandin E 2.
In the present work we studied the metabolism of arachidonate in Torpedo electric organ slices in order to elucidate its role in mAChR activity and to investigate the biochemical mechanism underlying this phenomenon. EXPERIMENTALPROCEDURES Fish Torpedo ocelluta were caught live off the coast of Israel between December and May and kept in sea-water aquaria. The fish were sacrificed within a month of capture and their electric organs were excised at 4°C and cut into slices of the desired weight. The slices were then preincubated in modified elasmobranch buffer (vide infera) for about 60 min at 25°C to allow for recovery after excision, after which they were immediately transferred into a fresh medium for experimental use. Radiochemical assay of prostaglandin synthesis
Electric organ slices (l-1.5 g) were incubated at 25°C for the designated time intervals with [Vlarachidonic acid (0.5 x lo5 to 1 x lo5 cpm/ml) in 2 ml of the specified buffer. When the effects of indomethacin, mepacrine or muscarinic ligands were to be tested, they were added to washed slices 60 min before the addition of the “C-labeled arachidonate. The basic buffer employed was modified elasmobranch buffer containing (in mM): NaC1, 250; KCI, 4.8; MgCl,, 2.4; ethyleneglycolbis(B-aminoethylether)N,N’-tetra-acetic acid (EGTA), 0.1; D-glucose, 10; sucrose, 200; Hepes, 20; pH 7.2. When testing the effect of Ca’+ we either raised the EGTA concentration to 1 mM (Ca2+ = 0) or added excess CaCI, in the presence of 1 mM EGTA to yield the required final concentration of free Ca”.
1359
I360
I. Pinchasi rr al.
Following incubation the medium was acidified to pH 3.5 with I M citric acid and the prostaglandins were extracted twice with 20ml chloroform. The dried extract was dissolved in 0.03-0.05 ml of chloroform:methanol (2: I) and chromatographed on Silica Gel 60 plates (Merck), using either the AIX solvent system of Hamberg and Samuelsson” (ethyl acetate:water:iso-octane:acetic acid, I I: IO:5:2, upper phase) or the chloroform:methanol:acetic acid (90:5:5) solvent system. The samples were run in parallel to unlabeled prostaglandins FZu, El, Dz and A, and arachidonic acid. The spots were located by exposure of the plates to iodine vapor. Strips, 1cm in length, were scraped off into scintillation vials and counted in a Packard Tricarb liquid scintillator in 4.5 ml of Lumac’s Hydro-luma scintillation liquid. Radiochemical assay ofarachidonate incorporation into tissue phospholipids
Electric organ slices (0.35 + 0.05 g) were incubated at 25°C for the designated time intervals with [3H]arachidonate (I x IO’-1.5 x 10’cpm/ml) in 2ml of the specified buffer, as described in the previous section. The muscarinic ligands to be tested were added to washed slices 60min prior to the addition of [‘Hlarachidonate. Following incubation, the slices were rinsed in cold buffer and imediately homogenized in 2 ml of IOOmM acetate which contained 5 mM EGTA. The homogenate was extracted according to Bligh and Dyer’ and chromatographed on Silica Gel 60 plates with a concentrating zone (Merck) in a solvent system consisting of chloroform:methanol:acetic acid:water (lOO:20: 12: 5). In parallel, unlabeled phosphatidylserine, phosphatidylinositol, phosphatidylcholine, phosphatidylethanolamine and arachidonic acid were subjected to chromatography. The spots were located by exposure to iodine vapor, scraped off and counted as described above. Radioimmunoassay of prostagiandin E2
Washed electric organs slices (0.25 + 0.05 g) were incubated at 25°C for the designated time intervals in 2 ml of the specified buffer. When the effects of muscarinic ligands were tested they were added 90 min (atropine) and 60 min (oxotremorine) before the slices were transferred to a fresh medium. Indomethacin and mepacrine were also added 90 min before the final incubation. Medium samples were withdrawn at the indicated times and kept on ice until assayed. The content of PGE, in the samples was determined by means of a specific lz51 radioimmunoassay (New England Nuclear). The antiserum is highly specific, having a crossreactivity of only 3.7% with prostaglandin E, and even less (i 0.4%) with other members of the prostaglandin family. The assay is sensitive enough to enable the determination of 0.5-25 pg PGE,.
Protein
Protein was determined according to Bradford’ using bovine serum albumin as standard. Materials [3H]Arachidonate (86 Ci/mmol) and [“‘Clarachidonate (- 55 mCi/mmol) were from New England Nuclear (NEN). Oxotremorine was from Aldrich. Atropine, arachidonic acid, prostaglandins A,, D,, E, and F,a and all the phospholipids were from Sigma. Indomethacin was kindly donated by Dr. S. Cherkez of Teva, Israel. All other reagents were of analytical grade. RESULTS Metabolism
of radiolabeled
arachidonate
The ability of Torpedo electric organ to metabolize arachidonic acid was tested by incubating tissue slices with radiolabeled arachidonate, followed by extraction of both the medium and the tissue and separation of the products by thin layer chromatography. Analysis by the AIX solvent system (see Experimental Procedures) of the [I4Clarachidonate metabolites released by tissue slices into the medium during a 15-min period of incubation in modified elasmobranch buffer revealed the formation of a major product, which comigrated with a PGEz standard and which comprised 22 + 4% (n = 3) of the initial radioactivity. In addition, two minor peaks were detected (3.4 + l.So/, and 4.5 i 2%) which comigrated with prostaglandin F, r and DZ standards, respectively (Fig. 1). The identity of these radioactive products was further ascertained by separation of the medium extract in a different solvent system and by testing the
-
Determination of free fatty acids
Washed electric organ slices (1 g) were incubated at 25°C in 1 ml of modified elasmobranch buffer, which contained 1 mM free Ca2+ and 75 pg/ml indomethacin. Oxotremorine and atropine were added 60 and 90 mitt, respectively, before the slices were transferred to a fresh medium. Following a final incubation of 90 min, a sample (0.60.8 ml) was withdrawn and diluted ten-fold with Dole reagent (1 N H,SO,:n-heptane:iso-propanol, I : 10:40), which contained 5 pg of pentadecanoic acid (C 15:O) as internal standard. The mixture was incubated at 60°C for 10 min followed by extraction with n-heptane and back-extraction with ethanolic KOH (0.05 N KOH in 50% ethanol). The water phase, containing the K-salts of the fatty acids, was acidified to pH - 3 with 1 N H,SO,, extracted with heptane and dried under N,. Methylation of the free fatty acids was performed according to Schlenk and Gellerman.‘8 The final sample was dissolved in hexane and analyzed on a Packard model 417 gas chromatograph.
IA
1
L
1
IJ-43 I
I
I
I
I
Distance
I
I
I
I
I
10
5
from
origin
km)
Fig. 1. Separation of [“‘Clarachidonic acid metabolites, produced by Torpedo electric organ and released into the medium, by thin layer chromatography utilizing the AIX solvent system. Electric organ slices (l-l .5 g) were incubated at 25°C for 15 min in modified elasmobranch buffer containing [‘4C]arachidonate (0.5 x 10s-I x 10’cpm/ml). The metabolites released by the tissue into the medium were extracted and separated by the AIX solvent system. AA. arachidonic acid; A,, Dz, Ez, F,a; prostaglandins.
Muscarinic modulation of arachidonate metabolism in Torpedo
1361
affected by elevation of the EGTA concentration (to 1 mM), introduction of free Ca*+ (up to 2 mM) or addition of mepacrine (10-100 PM). The incorporation of labeled arachidonate into tissue phospholipids was examined under conditions similar to those employed above. After a 15-min period of incubation with [3H]arachidonate only 10% of the radioactivity was recovered in the tissue, all of it in the form of free arachidonate. With a longer period of incubation (1 h) about 30% of the added radioactivity was recovered in the tissue, of which about 10% was incorporated into the major tissue phospholipid classes: phosphatidylinositol, phos-
10
5
Distance
from
origin
(cm)
Fig. 2. Separation of [‘4C]arachidonic acid metabolites, produced by Torpedo electric organ and released into the medium, by thin layer chromatography utilizing the chloroform:methanol:acetic acid (90: 5: 5) solvent system. Electric organ slices were incubated as described in the legend to Fig. 1, in the absence (bold lines) or presence (broken lines) of
phatidylserine, phosphatidylcholine and phosphatidylethanolamine (Table 1). The remaining 90;,‘, was in the form of free arachidonate. This pattern was not significantly altered when much longer (6 h) incubation periods were employed (Table 1). The presence of Ca*+ (1 mM) had no effect on the incor-
poration
of labeled arachidonate I
into tissue phos-
I
,
I
60
70
5
indomethacin (30 pgg/ml).The medium was then extracted and separated by the chlorofonn:methanol:acetic acid (90:5 :5) solvent system. AA, arachidonic acid; D, , E, , F, a ; prostaglandins. effects of cycle-oxygenase inhibitors on their synthesis. As shown in Fig. 2, the profile of oxygenated arachidonate derivatives revealed by the chloroform:methanol:acetic acid (90:5:5) solvent system was similar to that obtained with the AIX solvent system (Fig. I), and consisted of a major peak (22 + 3%; n = 4) comigrating with the PGE, standard and two minor peaks (3.2 + 1.2% and 5 + 1.5%) comigrating with the prostaglandin F, tl and D, standards, respectively. Examination of the tissue blocks under the same conditions revealed that they were virtually devoid of [‘4C]PGE2 (less than 10% of the amount recovered in the medium). Inclusion of the cycle-oxygenase inhibitor indomethacin (30 pg/ml) in the incubation medium inhibited virtually all (> 95%) of the [‘4C]arachidonate oxidation (Fig. 2). Similar results were obtained with aspirin (300pg/ml; not shown). The extent and rate of conversion of [‘4C]arachidonate into “C-labeled products were not Table 1. Incorporation
10
20
30
40
50
Fig. 3. Prostaglandin E, synthesis by electric organ slices from endogenous arachidonate and the effects thereon of Ca2+ and indomethacin. Slices (0.25 + 0.05 g) were incubated at 25°C in modified elasmobranch buffer in the absence (0) or presence (H) of indomethacin (30pgg/ml), 1mM free Cal+ (0) or in the presence of 1 mM EGTA (A). At the designated time intervals aliquots were removed from the medium and assayed for their PGE, content by a specific radioimmunoassay. Results presented are the mean of three experiments and the SD was less than 15”d.
of [‘Hlarachidonic acid into Torpedo electric organ phospholipids Phospholipid
Conditions 15min lh 6h
Ca2+ = Ca*+ = Ca’+ = Ca” = Ca2+ = Ca2+ =
0 1 mM 0 1mM 0 1 mM
60
Time (mid
PI
PS
PC
N.D. N.D. 2.3 + 1.2 (4) 2.6 k 1.4(5) 4.5 * 1.3 (3) 4.8 k 2.4 (3)
N.D. N.D. 1.3+ 0.9 (4) 1.6 & 0.6 (5) 4.6 + 3.4 (3) 4.2 + 1.65 (3)
N.D. N.D. 2.0 f 0.2 (4) 3.2 + 0.8 (5) 5.3 _+2.6 (3) 6.2 + 2.2 (3)
PE N.D. N.D. 3.9 & 1.65 (4) 5.0 + 2.0 (5) 6.4 + 2.0 (3) 5.7 + 2.1 (3)
Results presented are average f SD of the per cent of the radioactivity recovered in the tissue which is incorporated into the corresponding phospholipid. Numbers in brackets designate the number of experiments Performed, each in duplicate. N.D., not detectable; PC, PE, PI, PS; phosphatidylcholine, ethanolamine, inositol and serine respectively.
1. Pinchasr el al.
I362
1 1
! I
15 Time
30
10
(mm)
20
30
40
Oxotremorme
50
[pbl]
Fig. 4. Effect of muscarinic ligands on the synthesis of PGE, from electric organ precursors. Washed slices (0.25 k 0.05 g) were incubated at 25°C in modified elasmobranch buffer containing 1 mM free Ca2+. in the absence and presence of muscarinic ligands, as described in the Experimental Procedures. (A) The time dependence of the effect of oxotremorine and atropine on PGE, synthesis. 0 corresponds to control. 0, A, 4 correspond to 10, 20 and 50 pM oxotremorine. 0 corresponds to 50 PM oxotremorine and 10 PM atropine. (8) Dose/response curve for the effect of oxotremorine on PGE, synthesis. Data presented is from (A) and corresponds to the amount of PGE, synthesized following a 30 min incubation period in the absence (100%) and presence of the indicated concentrations of oxotremorine. Results presented are the mean of three experiments and the SD was less than 15%.
pholipids at any of the incubation (Table 1).
periods tested
Synthesis of prostaglandin E, from endogenous arach idonate
Incubation of tissue slices in modified elasmobranch buffer resulted in a massive, indomethacinsensitive synthesis of PGEz and in its liberation into the medium (Fig. 3). In contrast to the conversion of [I4Clarachidonate to radiolabeled PGE,, the synthesis of PGE, from endogenous arachidonate was markedly affected by the concentration of free Ca2+. Elevation of the EGTA concentration to 1 mM reduced PGE, liberation to about 50% of control levels;
Table 2. Effect of muscarinic ligands on the release of fatty acids from electric organ slices
Control Fatty acid (nmol/gh) ______ Cl, 5.1 + 1.4 CL, 4.6 ri: 1.7 C IXI 2.6 k 0.85 C ikz 0.37 5 0.09 C 204 I. 1 + 0.28 C,?., 4.3 +_0.28
Effect of muscarinic ligands (“/, of control) Oxotremorine Oxotremorine + atropine 125 k 30 106 + 15 116k20 13Ok29 171+ II 115+ 15
128 + 20 114 + 10 114* 10 120f20 96 + 20 108 + 10
Control corresponds to the electric organ tissue slice suspended in modified elasmobranch buffer which contained 1 mM Ca*+ and no muscarinic ligands. The release of free fatty acids in the presence of oxotremorine (50 fl M) and oxotremorine (50 p M) + atropine (10 p M) is expressed as per cent of the amount of the corresponding fatty acid which was released in the absence of muscarinic ligands. Results presented are average + SD of three experiments.
conversely, introduction of free Ca2+ (1 mM) resulted in a two-fold increase in the liberation of PGE, (Fig. 3). Furthermore, mepacrine (50 FM) inhibited PGE2 release by 50% (not shown). Effect of metabolism
muscarinic
ligands
on
arachidonate
Addition of either the muscarinic agonist oxotremorine (l-100pM) or the antagonist atropine (O.l-10pM) had no effect on the oxidation of [‘4C]arachidonate to [14C]PGE2 (105 f 8% and 100 f 5% of control, respectively; n = 3). On the other hand, the production of PGE, from endogenous substrate was strongly affected; oxotremorine caused a dose-dependent increase in PGE, synthesis of up to 230% of control levels, which was reversed by atropine (Fig. 4). These results can be interpreted as implying that activation of the presynaptic mAChR induces the liberation of arachidonic acid from tissue phospholipids rather than directly stimulates its oxidation to PGE,. This hypothesis was further investigated by measuring directly the effects of muscarinic ligands on the liberation of endogenous and radiolabeled arachidonate from tissue blocks. Incubation of tissue slices in modified elasmobranch buffer resulted in a basal release of several long-chain fatty acids into the medium, at a rate ranging between 0.5 and 5 nmol/g h (Table 2). Addition of oxotremorine (50/*M) increased the release of arachidonate to 17O”/0of control levels. This effect was reversed by atropine (10 PM) which by itself had no effect (Table 2). In contrast to its effect on endogenous arachidonate, oxotremorine had no effect on either the incorpo-
Muscarinic modulation of arachidonate metabolism in Torpedo ration of [‘Hlarachidonate into tissue phospholipids or on the rate of its liberation from prelabeled slices. DISCUSSION
The metabolism of exogenous, radiolabeled arachidonate and of endogenous arachidonate were studied in the Torpedo electric organ, and the effects thereon of activation of the presynaptic mAChR were investigated. Incubation of tissue blocks with labeled arachidonate result in the formation and release of a major product, which contains -25% of the added radioactivity. On the basis of its comigration with a PGE, standard in two solvent systems, as well as the complete inhibition of its synthesis by indomethacin and aspirin (Figs 1 and 2), this product was identified as PGEr Since virtually no labeled PGE, is found in the tissue under these conditions, it seems that [14C]PGE, is released into the medium immediately following its synthesis. In addition to [‘4C]PGE2, the electric organ slices synthesize small amounts of indomethacin-sensitive substances, which comigrate with prostaglandins F, c( and D, (Figs 1 and 2). This profile of oxidized [‘4C]arachidonate metabolites is qualitatively similar reported for Torpedo to that previously synaptosomes14 except that the extent of the conversion of [‘4C]arachidonate by the synaptosomes (- 3%) is much lower than that observed with the tissue slices (3079, and that the synaptosomes synthesize an additional major product (product II), which comigrates with prostaglandin D2 but is resistent to indomethacin.‘4 A similar oxidation profile was observed with tissue homogenates before fractionation. Interestingly, when homogenization is performed under conditions which minimalize cycloinactivation oxygenase (N,-saturated buffer containing I:/, fatty acid-free bovine serum albumin), the amount of [‘4C]PGE, formed by synaptosomes is increased five-fold and that of product II is drastically reduced. However, under these conditions the synaptosomes are depleted of their ACh content and lose their ability to release ACh when depolarized in the presence of Ca*+ (unpublished observations). Thus the synaptosomes are not the most suitable preparation for the parallel investigation of ACh release and arachidonate metabolism. Synthesis of PGE2 from endogenous arachidonate was demonstrated by employing a highly specific radioimmunoassay. The results obtained (0.2 + 0.03 nmol/gh, n = 4) are very similar to those previously found using the rat stomach bioassay (0.25 f 0.02 nmol/gh, n = 3) (unpublished observations). As expected, the synthesis of PGE, by tissue blocks is totally inhibited by indomethacin and aspirin (Fig. 3). It thus appears that Torpedo electric organ oxidizes both exogenous and endogenous arachidonic acid via the cycle-oxygenase pathway, with PGE, being the major product.
1363
The mechanism of synthesis of PGE, from radiolabeled arachidonate was compared to that from endogenous arachidonate. Two major differences were found: (1) the formation of PGE, from endogenous arachidonate is stimulated by Ca*+ while that of labeled PGE, is not; (2) mepacrine has no effect on the synthesis of [‘4C]PGE2 but inhibits the conversion of endogenous arachidonate into PGE,. These results are interpreted as suggesting that while exogenous arachidonate is directly converted into PGE,, the endogenous precursor has to be liberated from a phospholipid pool prior to oxidation. This contention is also supported by the finding that following a brief incubation of the tissue with labeled arachidonate, there is no detectable incorporation of the label into tissue phospholipids, whereas labeled PGE, is massively formed (Table 1 and Figs 1 and 2). Much longer incubation periods (up to 6 h) are needed in order to achieve detectable incorporation of radioactivity into tissue phospholipids. Under these conditions the major phospholipids become uniformly labeled, in a Ca2+-independent manner (Table 1). Incubation of tissue slices with the muscarinic agonist oxotremorine causes a marked dosedependent and atropine-reversible increase in the liberation of PGE2 into the medium (Fig. 4). The concentration of oxotremorine inducing a halfmaximal effect (10 PM) is similar to its previously reported binding constant to Torpedo mAChR.8,9 These results are similar to those previously observed with Torpedo synaptosomes, using the rat stomach bioassay. I4 Comparison of the specific rates of the muscarinic stimulation of PGE, formation with 50 p M oxotremorine shows that the synaptosomes synthesize - 30 pmol of PGE,/mg protein/min, while the tissue slices synthesize -0.50 pmol of PGE,/mg proteimmin. This marked enhancement in the specific “responsiveness” of synaptosomes to muscarinic activation relative to that of intact tissue, is consistent with their relative enrichment (by more than ten-fold) in muscarinic receptors, as determined by ligand binding.* Muscarinic stimulation can increase PGE, formation either by activating the oxidation of arachidonic acid or by increasing the arachidonate pool available for oxidation through triggering its liberation from phospholipids. In order to distinguish between these possibilities we examined the effect of oxotremorine on the conversion of [‘4C]arachidonate to [‘4C]PGE2 and on the release of endogenous arachidonate. Our results show that neither oxotremorine nor atropine has a significant effect on the conversion of [‘4C]arachidonate to [‘4C]PGE,. In contrast, oxotremorine induces an atropine-sensitive increase in the release of endogenous arachidonate, into the medium (Table 2). The fact that the release of C22:6, which is a major constituent of Narcine electric organ phospholipids3 is not affected by OXotremorine implies that activation of the presynaptic mAChR induces selective lipolysis.
1364
I. Pinchasi et (II.
The basal incorporation of labeled arachidonate into tissue phospholipids is very low even after prolonged incubation (Table 1). This observation and the finding that oxotremorine has no effect on either the incorporation of [3H]arachidonate into Torpedo phospholipids or on its liberation from prelabeled tissue, imply that the turnover of phospholipids in Torpedo is very slow and that externally added arachidonate is not readily incorporated into the receptor-linked phospholipid pool. The nature of the lipolytic event induced by muscarinic stimulation and the mechanism of coupling between mAChR activation and the lipolytic enzymes are not known. In many systems mAChR coupling to phosphatidylinositol turnover is thought to be a possible mechanism by which external stimuli are transduced into the ~~41.‘~“~‘~ Since phosphatidylinositol is rich in arachidonate esterified to the 2-Sn position, it is possible that a phospholipaseA, is activated by muscarinic stimulation, resulting in
the release of arachidonate and in its subsequent conversion to PGE,. Such a mechanism has indeed been shown in non-neuronal systems.‘“,” Our finding that mepacrine inhibits the formation of PGE, from endogenous substrates supports this contention. We have recently shown that Torpedo nerve terminals contain a guanosine 5’-triphosphate-binding protein, which is adenosine 5’-pyrophosphateribosylated by cholera toxin,‘O and which regulates the binding of muscarinic agonists to the receptor.4,‘Y It is tempting to suggest that this protein may be involved in the coupling between the mAChR and the enzymatic apparatus responsible for the generation of PGE, Acknowledgements-We are grateful to Prof. A. Raz of the Department of Biochemistry, Tel-Aviv University, for his valuable advice and practical help throughout this study. This work was supported in part by grants from the
Muscular Dystrophy Association and the U.S.-Israel Binational Science Foundation (BSF Grant No. 2699).
REFERENCES I. Bligh E. G. and Dyer W. J. (1959) A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 91 l-917.
2. Bradford M. M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analyt. Biochem. 72, 248-254. 3. Deutsch J. W. and Kelly R. B. (1981) Lipids of synaptic vesicles: relevance to the mechanism of membrane fusion. Biochemistry 20, 278-285.
4. Dowdall M. J., Golds P. R. and Strange P. G. (1982) Properties of Torpedo electric organ muscarinic receptors. J. Physiol., Paris 78, 379-384.
5. Dunant Y. and Walker A. I. (1982) Cholinergic inhibition of acetylcholine release in the electric organ of Torpedo. Eur. J. Pharmac. 78, 201-212.
6. Hamberg M. and Samuelsson G. (1966) Prostaglandins in human seminal plasma: prostaglandins and related factors, J. biol. Chem. 241, 257-263.
7. Hawthorne J. N. and Pickard M. R. (1979) Phospholipids in synaptic function. J. Neurochem. 32, 5-14. 8. Kloog Y., Michaelson D. M. and Sokolovsky M. (1978) Identification of muscarinic receptors in the Torpedo electric organ: evidence for their presynaptic localization. FEBS Left. 95, 331-334. 9. Kloog Y., Michaelson D. M. and Sokolovsky M. (1980) Characterization of the presynaptic muscarinic receptor in synaptosomes of Torpedo electric organ by means of kinetic and equilibrium binding studies. Bruin Res. 194, 97- 115. 10. Lester H. A., Steer M. L. and Michaelson D. M. (1982) ADP-ribosylation of membrane proteins in cholinergic nerve terminals. J. Neurochem. 38, 1080-1086. II. Michaelson D. M., Avissar S., Kloog Y. and Sokolovsky M. (1979) Mechanism of acetylcholine release: possible involvement of presynaptic muscarinic receptors in regulation of acetylcholine release and protein phosphorylation. Proc. natn. Acad. Sci. U.S.A. 76, 6336-6340.
12. Michell R. H. (1979) Inositol phospholipids in membrane function. Trends Biochem. Sci. 3, 128-13 1. 13. Michell R. H. (1982) Phosphatidyl inositol turnover in signal transduction. Neurosci. Res. Prog. Bull. u), 338-350. 14. Pinchasi I., Shanietzki B., Schwartzman M., Raz A. and Michaelson D. M. (1982) Is presynaptic muscarinic inhibition of acetylcholine release mediated by second messengers? The role of cyclic AMP and prostaglandins. In Presynaptic Receptors, Mechanisms and Functions (ed. DeBelleroche J.), pp. 114129. Ellis Horwood, Chichester. 15. Pinchasi I., Burstein M. and Michaelson D. M. (1985) Prostaglandins mediate the muscarinic inhibition of acetylcholine release from Torpedo nerve terminals. In Molecular Basis of Neme Actiaity (eds Changeux J. P., Hucho F., Maelicke A. and Neumann E.). Walter de Gruyter. In press. 16. Putney Jr. J. W. (1981) Recent hypotheses regarding the phosphatidyl inositol effect. Life Sci. 9, 1183-l 194. 17. Rubin R. P. (1982) Calcium-phospholipid interactions in secretory cells: a new perspective on stimulus-secretion coupling. Fedn Proc. Fedn Am. Sots exp. Biol. 41, 2181-2187. 18. Schlenk H. and Gellerman J. L. (1960) Esterification of fatty acids with diazomethane on a small scale. Analyt. Chem. 32, 1412-1416. 19. Sokolovsky M., Avissar S., Egozi Y., Gurwitz D.. Kloog Y. and Michaelson D. M. (1981) Agonist and antagonist interaction with muscarinic pre- and postsynaptic receptors: biochemical characterization. In Neurotransmitters and Their Receptors (eds Littauer U., Dudai Y., Silman I., Teichberg V. and Vogel Z.), pp. 257-270. John Wiley, New York. (Accepted 14 May 1984) Note added in proof. Recent experiments in which we measured concomitantly the effects of oxotremorine on ACh release and on the metabolism of arachidonate in tissue slices revealed that the pattern of PGs synthesized following muscarinic activated depends on the time of its measurement following exision of the electric organs. These findings are presented elsewhere.”