TiPS - September 1990 [Vol. 11] 367

28 Takeda, N. et aL (1986) Acta Otolaryngol. Venerol. (Suppl. 115), 1-43 34 Pipkom, U. et al. (1987) Allergy 101,416-421 32 Granerus, G., Olafsson, I. H. and (Copenhagen) 42, 496-501 29 Tung, A. S. et al. (1985) Biochem. Roupe, G. (1985) Agents Actions 16, Pharmacol. 34, 3509-3515 244-248 30 August, T. F. el al. (1985) J. Pharm. Sci. 33 Neitaanmaki, H., Fraki, ]. E., Harvima, IPDllS1T: dimethyl-2-[4-(3-ethoxy-2- 74, 871-875 R. J. and Forstrom, L. (1989) Arch. hydroxypropoxy)phenylcarbamoyllethyl 31 Olafsson, I. H. (I985) Acta Derm. Dermatol. Res. 281, 99-104 sulfonium-p-toluene sulfonate

polyacrylamide gel electrophor- metabolites of esis) and has no apparent cofactor requirement 4. A cDNA encoding in neuronal leukocyte 12-1ipoxygenase has been isolated and sequenced 5. Like all other , 12- transmembrane signalling lipoxygenase catalyses the intro- duction of molecular oxygen into Daniele Piomelli and Paul Greengard a 1,4-(cis,cis)-pentadiene moiety, converting arachidonic acid into the hydroperoxide, (12s)-hydro- Studies of invertebrate and vertebrate nervous tissue have demonstrated that peroxyeicosatetraenoic acid (12- free arachidonic acid and its lipoxygenase metabolites are produced in a HPETE) - a reaction that is both receptor-dependent fashion. The intracellular actions of these compounds regiospecific and stereospecific. include the regulation of activity of membrane ion channels and protein Other cis-polyunsaturated fatty kinases. In this article Daniele Piomelli and Paul Greengard review the acids, such as , lino- evidence that these lipophilic molecules constitute a novel class of intracellular lenic acid and docosahexaenoic second messenger, possibly involved in the modulation of neurotransmitter acid are also good substrates for release. the leukocyte 12-1ipoxygenase.

Many neurotransmitters exert Recent work has suggested that Lipoxygenases in the brain their actions on neurons by stimu- the membrane polyunsaturated In a recent study the levels of lating the formation of intracellu- fatty acid arachidonic acid and its 12-1ipoxygenase in several areas of lar second messengers, which metabolites produced through the porcine brain were estimated by include such diverse molecules as lipoxygenase pathways may con- immunoassay, using a mono- cyclic nucleotides, Ca 2+, inositol stitute a novel class of neuronal clonal antibody raised against 1,4,5-trisphosphate and 1,2-diacyl- second messengers. These lipids, leukocyte 12-1ipoxygenase°. Detect- glycerol. In most cases second which affect the activities of neur- able levels of the enzyme were messengers stimulate the activity onal ion channels and protein found in the cytosolic fractions of specific protein kinases, which kinases, might participate in the of olfactory bulb, hypothalamus, phosphorylate and hence modu- regulation of neurotransmitter cerebellum and pineal body; the late the activity of a large number release. highest |evels were detected in of enzymes, membrane ion chan- Arachidonic acid is found in parenchymal cells of the anterior nels and structural proteins (for esterified form in pituitary lobe. review, see Ref. 1). However, phospholipids from which it can Even though 12-HPETE is second messengers can also acti- be liberated through multiple reduced very rapidly to the alco- vate phosphatases either directly enzymatic pathways (see Box 1). It hol (12s)-hydroxyeicosatetraenoic (as in the case of the Ca2+-cal - can then diffuse out of the cell, acid (12-HETE), which has long modulin-dependent phosphatase be reincorporated into phospho- been known to be a major metab- calcineurin), or indirectly through lipids, or undergo metabolism. olite of arachidonic acid in the phosphorylation of phosphatase Three pathways of arachidonic acid brain, the metabolism of 12- inhibitors 2. In some cases, second metabolism have been described HPETE is not limited to its reduc- messengers can act without the in animal tissues: the cyclooxy- tion (Fig. 1). The fatty acid hydro- participation of enzyme inter- genase pathway, the lipoxygenase peroxide can give rise to a group mediates, by binding to mem- pathway and the cytochrome P-450 of biologically active hydroxy- brane ion channels (or to regu- or 'epoxygenase' pathway (Box 1). epoxides, the 'hepoxilins', two of latory proteins closely associated In nervous tissue the major enzy- which - hepoxilin As and hepox- with them) and changing their matic route involved in the metab- ilin I]3 - have been identified in properties (for instance in the case olism of arachidonic acid is the 12- nervous tissue from both ver- of Ca2+-activated K + channels; for lipoxygenase pathway (Box 1 and tebrates and invertebrates 7,s (Fig. review, see Ref. 3). Fig. 1). There is also good evidence 1). The hepoxilins are short-lived for the occurrence in nervous tissue compounds and undergo further of the 5-1ipoxygenase pathway. metabolism by the action of a 12-Lipoxygenase, first isolated hepoxilin hydrolase, which has D. P;omelli is a Postdoctoral Associate and from blood platelets and now been purified from rat liver and P. Gree:rga:'d is Profi'ssor and ttead at tfle shown to be present in brain. This Laboratory o[ Molecular and Celhdar Neuro- purified to near homogeneity science, The Rockefeller University, New from leukocytes, is a cytosolic pro- enzyme catalyses the specific York, NY 10021, USA. tein of 72 kDa (based on SDS- cleavage of the hepoxilin epoxide

~) 1990, Elsevier Science Publishers Lid. (UK) 0165 - b147/901.$02.0i) 368 TiPS - September 1990 [Vol. 11] - Box 1 Pathways of arachidonic

Liberation of arachidonic acid from membrane phospho- 1). The two steps required for reincorporation - the lipids mainly occurs by two pathways (Fig. 1). Hydrolytic reactions of arachidonoyl-CoA synthetase and cleavage at the sn-2 position of the glycerophospholipid arachidonoyl-lysophospholipid transferase - are thought backbone, catalysed by Ca2+-activated A2, to make up the major mechanism responsible for keeping yields free fatty acid and lysophospholipidL Alternatively, the concentration of free arachidonic acid low within cells. formation of free arachidonic acid can be initiated by the activation of a phospholipase C, which cleaves the phos- The three pathways of arachidonic acid metabolism pholipid at the phosphate ester bond, producing 1,2- described for other animal tissues - cycloo×ygenase, diacylgJycerol. This intermediate is broken down by iipoxygenase and cytochrome P-450 (Ref. 4) - bdve also diacylglycerol-and monoacylglycerol lipases to yield free been found in the brain (for review, see ReL 5). The fatty acid and glycerol-'. cyclooxygenase pathway leads to the formation of prosta- After liberation, free arachidonic acid has several fates. It glandins, prostacyclin (PGI2) and (TXA2). can diffuse out of the cell. Alternatively, it can be either Cyclooxygenase is inhibited by some non-steroidal anti- metabolized (see below) or converted to arachidonoyl- inflammatory drugs, such as indometacin, by the anti- coenzyme A and reincorporated into phospholipidss (Fig. oxidant nordihydroguaiaretic acid and by the false sub- strate eicosatetraynoic acid4. ~phospholipid~ Lipoxygenases, which are inhibited nonspecifically by nordihydroguaiaretic acid, eicosatetraynoic acid and baicalein, form hydroperoxyeicosatetraenoic acids polar~ I "X (HPETE) as their primary products. HPETE can undergo a complex metabolism (see main text for details). 5-HPETE, formed by 5-1ipoxygenase, is converted to the short-lived [k, lyso- )~ epoxide , LTA4, by a dehydrase activity intrin- •,l, I '~ phospholipid (J I sic to 5-1ipoxygenase6. LTA4 can be transformed either into OA.G AA-CoA LTB4, by an LTA4-hydrolase7, or into LTC4 by a -S-transferase (GSH-S-transferase)8. 5-Lipoxy- genase activity is specifically inhibited by 5,6-methano- . methyl ester and 5,6-dehydroarachidonic acid. In addition, fo,-mation of 5-1ipoxygenase products can / \ / be inhibited in intact cells by MK886, which prevents I CoA-S" 5-1ipoxygenase activaUon by binding to a newly described MAG ('~ ~ AA 5-1ipoxygenase activatiz:g protein (FLAP)9. Metabolism of 15-HPETE show.~ considerable similarities to that of 12-HPETE (see Refs 10 and 11). Cytochrome P-450 catalyses the conversion of arachi- donic acid into an array of epoxyeicosatrienoic acids (EET), extracellular metabolism which are hydrolysed to the corresponding diols by the diffusion action of an (inhibited by trichloro- F-~j. I. Pathways of arach~onic acid turnover. AA, arachidonic propene oxide). In addition, cytochrome P-450 has also ac/d; PLA=, phoso,hoiipase A2; PLC, phospholipase C; DAG, been shown to produce hydroxyeicosatetraenoic acids 1,2-diacylglycerol; MAG, monoacylglyceroL (HETE) .2.

ring, yielding a family of isomeric arachidonic acid to oxygenated 12-1ipoxygenase, requires both trihydroxy acids termed "trioxi- derivatives (mainly epoxylated Ca 2+ and ATP for maximal ac- lins '9. In addition to hepoxilins and hydroxylated eicosatetraenoic tivity. Upon binding of Ca 2+ the and trioxilins, the nervous tissue acids), some of which are formed enzyme translocates to the cell of the marine mollusk Aplysia cali- in nervous tissue ~2. In addition to membrane, where its lipophilic fornica converts 12-HPETE into this direct role in arachidonic acid substrate, arachidonic acid, is the keto acid, 12-ketoeicosa- metabolism (for review, see Ref. more readily accessible. Trans- tetraenoic acid (12-KETE) (Fig. 1; 13), cytochrome P-450 can convert location is followed by a rapid Ref. 10). fatty acid hydroperoxides into process of enzyme inactivation TM. The enzymes involved in the hydroxyepoxides and keto acids in As would be predicted, agents metabolism of 12-HPETE are still vitro 14. A cytochrome P-450-1ike that increase intracellular Ca 2+ not characterized. Heme-containing activity may participate in the concentration, such as the proteins and hematin can catalyse biotransformation of 12-HPETE in ionophore calcimycin (A23187), the conversion of lipid hydro- intact neurons of Aplysia 15. stimulate formation of LTB4 and of peroxides to hydroxyepoxides and Another lipoxygenase present the peptide-containing leuko- keto acids H, but this does not rule in the brain is 5-1ipoxygenase, trienes LTC4, LTD4 and LTE4 in rat out the possibility that specific which initiates the biosynthesis of cortical tissue 17,18. A role for the enzymes act in intact cells. An the , potent chemo- leukotrienes in the regulation of intriguing candidate is cytochrome tactic agents for leukocytes and K + channels in mammalian atrial P-450- this NADPH-dependent mediators of anaphylaxis. Purified myocytes has been proposed (for monooxygenase can metabolize mammalian 5-1ipoxygenase, unlike review, see Ref. 19). TiPS - September 1990 iVol. 11] 369

acid turnover and metabolism COOH

arachidonic acid

cyclooxygenase lipoxygenases cytochmme P-450 • indometacin, proadifen • NDGA, E'IrYA clotdmazole $ 4, $ (, NDGA, ETYA $ $ prostaglandins PGI 2 TXA2 I EET HETE (PGE2 , PGD2 . etc.) 12- 5, lip°xygenase I lipoxygenase lipoxygenase'°-1 oxide hydro/ase MLTA | ~ TCPO : DHAA / MK886 L Fig. 2. Metabolism of arachidonic ,& Diols acid. Enzyme inhibitors are given in I:>-HPETE 5-HPETE IS-HPETE bold type. NDGA, nordihydroguai- aretic acid; ETYA, eicosatetraynoic $ acid; MLTA, 5,6-methanoleukotriene LTA 4 ,44 methyl ester;DHAA, 5,6.dehydro- arachidonic acid; TCPO, trichloropro- I pene oxide. LTA 4 hydrolase~ GSH-S-transferase

4, LTB 4 LTC4 $ LTDd $ References LTE4 1 Waite, M. (1987) The (Handbook of Lipid 7 Samue]sson, B. and Funk, C. D. (1989) I. Biol. Chem. 264, Research, Vol. 5) (Hanahan, D. l., ed.), Plenum Press 19469-19472 2 Berridge, M. I. and Irvine, R. F. (1989) Nature 341, 197-205 8 Samuelsson, B. (1983) Science 220, 568-573 3 Sun, G. Y. and MacQuarrie, R. A. (1989) Ann. NY Acad. Sci. 9 Miller, D. K. et al. (1990) Nature 343, 278-281 559 37-55 10 Hamberg, M., Herman, C. A. and Herman, P. R. (1986) 4 Needleman, P., Turk, J., Jakschik, B. A., Morrison, A. R. and Biochim. Biophys. Acta 877, 447-457 Lefkowith, ]. B. (1986) Annu. Rev. Biochem. 55, 69-102 11 Weiss, R. H., Arnold, J. L. and Estabrook, R. W. (1987) Arch. 5 Shimizu, T. and Wolfe, L. S. (1990) ]. Neurochem. 55, 1-15 Biochem. Biophys. 252, 334-338 6 Samuelsson, B., Dahl~n, S-E., Lindgren, J-A., Rouzer, C. A. 12 Fitzpatrick, F. A. and Murphy, R. C. (1989) PharmacoL Rev. and Serhan, C. N. (1987) Science 237, 1171-1176 40, 229-241

Receptor-dependent formation of stimulates arachidonic acid liber- Modulation of K + channels by lipoxygenase products ation, as a result of receptor- lipoxygenase metabolites Neurotransmitters cause forma- dependent activation of phospho- The results obtained in Aplysia tion of 12-1ipoxygenase metabo- lipase A2 (Ref. 23). suggested that lipoxygenase metab- lites in nervous tissue. NMDA In Aplysia, formation of 12- olites of arachidonic acid may receptor stimulation, apparently HETE by neural ganglia is stimu- participate in the modulation of by increasing Ca 2+ entry and acti- lated by the application of hista- K + channel activity and in pre- vating phospholipase A2, stimu- mine or the tetrapeptide FMRF synaptic inhibition. This idea was lates liberation of arachidonic acid amide 24,25. These two modulatory substantiated by electrophysio- and the generation of 12-HETE in substances can alter neuronal logical experiments. In two hist- a variety of nervous tissue membrane potential by opening aminoceptive neurons (L10 and preparations 2°-22. The absolute K + channels, producing pre- L14), extracellular application of configuration of the HETE formed synaptic inhibition of neurotrans- 12-HPETE produced changes in (12s) gives definitive evidence of mitter release. It is interesting, membrane potential that were its origin from a 12-1ipoxygenase- therefore, that electrical stimula- similar to those produced by the catalysed reaction 21. tion of L32 neurons, an identified neurotransmitter s. Similarly, in l',Iorepinephrine, acting via 0¢- group of Aplysia cells, results in sensory cells, arachidonic acid and adrenoceptors, also elicits forma- the generation of both 12-HETE 12-HPETE mimicked the actions of tion of 12-HETE in rat cortex 2t. In and hepoxilin A3 as well as in the FMRF amide, increasing the prob- hippocampal neurons in culture, inhibition of neurotransmitter ability of opening of a subclass of activation of 5-HT2 receptors release from LI0 neurons a,24. K + channels, the 5-HT-inactivated 370 TiPS - September 1990 [Vol. 11l

~ OOH arachidonic actd NDGA 12-1ipoxygenase baicalein i ¢~-V'~ COOH

12-HPETE

I l ~ glutathione proadifen ~ cytochrome P-450-like? peroxidase I 1 .ol oo. l oo. Ho-K/-~

12-H ETE

12-KETE hepoxilin B a hepoxilin A 3 HO I hepOxilia hydrOlase I hepOxilin hydrOlase /~,~OOH HO.h,f--'~/""~OOH .o,J _ - - Fig. I. Metabolism of arachidonic acid via the 12-1ipoxy- O~,~._ . , HOt`'- ~V genase pathway. Enzyme inhibitors are given in bold H type. NDGA, nordihydroguaiarebcacid. trioxilin A 3 trioxilin B 3

IK(5-HT~ (S-K +) channels 2s. By con- the IK(S-HT) channel, or does it act (SKF525A) significantly reduced trast, a 5-!ipoxygenase-derived by modulating the activity of pro- the response of sensory neurons product, 5-HPETE, and the hy- tein kinases or protein phospha- to 12-HPETE, suggesting that droxy acids, 12-HETE and 5-HETE, tases? In cell-free patches of Aplysia metabolism is carried out by a were inactive. In addition, the sense,~'y cell membranes, in the cytochrome P-450-1ike enzyme 15. actions of histamine and FMRF absence of cytosolic components, The identity of the derivative of amide on membrane potential 12-HPETE could not affect the 12-HPETE that activates IK(5_HT ) could be prevented by treatment opening of IK(5-HT~ channels is. Its channels in Aplysia sensory cells is with the phospholipase A2 in- biological activity was restored, unknown. It is clear, however, hibitor p-bromophenacylbromide, however, when the patches were that two of the candidates, hepox- or with the lipoxygenase inhibitor bathed in a solution containing ilin A 3 and 12-KETE, have differ- nordihydroguaiaretic acid, but hematin (which catalyses the con- ent electrophysiological actions on not with the cyclooxygenase in- version of the lipid hydroperoxide identified Aplysia neurons: hepox- hibitor indometacin2S.2~. (For a to various products, including ilin A 3 hyperpolarizes the mem- discussion of the drawbacks of the hepoxilins and 12-KETE). Thus brane of L14 cells, an effect that phospholipase A2 inhibitors pres- biotransformation of 12-HPETE is appears to result from increased ently available, see Ref. 27.) On necessary for biological activity. conductance to K + ionsS; by con- the basis of these findings it was Since the experiments were car- trast, 12-KETE produces a biphasic proposed that a 12-1ipoxygenase ried out in the absence of ATP and response (depolarization-hyper- metabolite of arachidonic acid, GTP, they also indicate that K + polarization) in these neurons 1°. possibly 12-HPETE, may act as an channel modulation by 12-HPETE Hepoxilin A 3 also hyperpolarizes intracellular second messenger in does not require the participation neurons of the CA1 area in super- the control of K + channel activity of a G protein or of a protein fused rat hippocampal slices2s. by histamine and FMRF amide in kinase; furthermore, it cannot be Recently, Buttner and co- Aplysia neurons. due to a phosphoprotein phos- workers reported that, although These findings raise several phatase, since it is reversible. 12-HPETE is ineffective when questions. Is 12-HPETE itself the It is still unclear what enzyme, if applied to the intracellular side of active intracellular messenger, or any, catalyses the metabolism of the neuronal membrane, it can does it require further metabol- 12-HPETE and what metabolite is produce a large increase in IK(5-HT) ism? If it is metabolized, which responsible for the biological re- channel activity when applied to of its products is the active sponse. Belardetti and colleagues the outside; 12-HPETE may there- one? Does this 12-1ipoxygenase- found that the ~eversible cyto- fore act as a local intercellular sig- derived messenger act directly on chrome P-450 inhibitor proadifen nalling molecule 29. While its tran- TiPS - September 1990 [Vol. 11] 371 sient chemical nature seems to crine role played by the catechol- donic acid regulates Ca 2+ currents argue against this possibility, amines in mammalian brain in hippocampal neurons32, chick there are other examples of ex- In rat hippocampal CAt sympathetic3-~ and frog sympath- tremely short-lived metabolites of neurons, somatostatin, a putative etic neurons34. arachidonic acid that appear out- transmitter in the hippocampus, side the cell of origin and act on augments a non-inactivating, Inhibition of neurotransmitter membrane receptors of neighbor- voltage-gated K ÷ current, termed release ing cells. For example, the cyclo- the M current (IM). Augmentation Phosphorylation of specific pro- oxygenase product thromboxane of IM results in a slow hyperpolar- teins in the presynaptic nerve ter- A2, a powerful platelet-aggregat- ization of the membrane and minal participates in modulating ing and vasoconstricting autacoid, decreased neuronal excitability3°. neurotransmitter release. The state is released from blood platelets Lipoxygenase metabolites of of phosphorylation of the synaptic and acts both on the platelets arachidonic acid may be involved vesicle-associated protein, synap- themselves and on vascular smooth in the response to somatostatin. sin I, is thought to regulate the muscle (in both cases via a mem- Schweitzer ef al. 31 found that phos- availability of synaptic vesicles for brane receptor) 19. It is not under- pholipase A2 inhibitors (mepacrine exocytosis (for review, see Ref. stood how these short-lived metab- and p-bromophenacylbromide) 35). In its dephosphorylated state, olites leave cells. Perhaps an anion and specific 5-1ipoxygenase in- synapsin I may cross-link synaptic carrier of the type sensitive to the hibitors (5,6-methanoleukotriene vesicles to the surrounding cyto- uricosuric agent probenecid may A4 methyl ester and 5,6-dehydro- skeletal lattice. According to this be involved. arachidonic acid) inhibit the re- model, when synapsin I is phos- Whatever the mechanism for sponse to somatostatin in CA1 phorylated on its 'tail'-region by the release of 12-HPETE, its dif- cells. In addition, stimulation of IM Ca2+--calmodulin-flependent pro- fusion in a circumscribed area by somatostatin can be mimicked tein kinase 1I, its interaction both around the neuron of origin might by the application of arachidonic with synaptic vesicles and with bring about a local spreading of acid or LTC4 (a 5-1ipoxygenase cytoskeletal elements is reduced, neuronal inhibition. This could metabolite)31. resulting in dissociation of the represent a means of modulating These findings suggest that vesicles from the cytoskeleton. or synchronizing the activity of a lipoxygenase-derived This would in turn increase the small ensemble of neighboring have a widespread role in reducing number of vesicles available for neurons involved in a single neuronal excitability through the exocytosis. Therefore, reducing physiological function. Such a role regulation of K ÷ channel activity. the state of phosphorylation of would be analogous to the para- There is also evidence that arachi- synapsin I may be a way to reduce - Box 2

Arachidonic acid in hippocampal long-term potentiation

Long-term potentiation (LTP) is a mammalian model of duce LTP (Ref. 7). These findings suggest that free fatty synaptic plasticity and information storage, which can acids, liberated by the action of a phospholipase, may be produced in the hippocampus by high frequency play a role in the maintenance phase of LTP. stimulation of the perforant pathway (a group of fibers The cellular site of action of this lipid(s) is still unclear connecting the entorhinal cortex to the dentate gyrus)L and both presynaptic6 and glial8 sites have been LTP is generally believed to consist of two phases - proposed. Likewise, little is known of their mechanism induction and maintenance. Induction is initiated by of action. An involvement of protein ki,-,ase C has been the postsynaptic entry of Ca2+, which occurs through suggested 7 on the grounds of the ability of unsaturated cation channels associated with excitatory amino acid fatty acids to activate this protein kinase in a Ca 2+- and receptors of the NMDA type, and consequent activation phospholipid-independent fashion9. of a Ca2+-dependent protein kinase2-4. Maintenance appears to be produced at least partly by presynaptic mechanismss. References Williams and collaborators° reported that perfusion 1 Landfield,P. W. and Deadwyler, S. A., eds (1988) Long-Term with arachidonic acid produced a transient inhibition of Potentiation: From Biophysics to ~,e!:nvior, Alan R. Liss synaptic transmission in the dentate gyros of the 2 Hu, G. Y. et al. (1987) Nature 328, 426--429 anesthetized rat. They also found, however, that by 3 Malenka,R. C. et aL (1988) Nature 342, 554-557 combining arachidonic acid perfosion with a weak 4 Malinow, R., Schulman, H. and Tsien, R. W. (1989) Science stimulation of the perforant path (unable to produce 245, 862-866 LTP per se) a long-term enhancement in synaptic 5 Lynch, M. A., Errington, M. L. and Bliss, T. V. P. (1989) strength resulted, which was indistinguishable from Neuroscience 30, 693-701 LTP. This effect was not inhibited by the mixed 6 Williams, I. H., Errington, M. L., Lynch, IH. A. and Bliss, T. V. P. (1989) Nature 341,739-742 lipoxygenase/cyclooxygenase blocker nordihydro- 7 Linden, D. I., Shen, F-S., Murakami, K. and Routtenberg, A. guaiaretic acid indicating that it was produced by the (1987) J. Neurosci. 7, 3783-3792 free fatty acid, rather than by a metabolite. In agreement 8 Barbour, g., SzatkowsKt, M., Jngiedew, N. and Attweli, D. with this interpretation, oleic acid, a ~onounsaturated (1989) Nature 342, 918-920 fatty acid which is not a substrate for metabolism by 9 Naor, Z., Shearman, M. S., Kishimoto, A. and Nishizuka, Y. lipoxygenase or cyclooxygenase, was also able to pro- (1988) Mol. E,rdocrinol. 2, 1043--1048 372 TiPS - September 1990 [Vol. 11]

dependent kinase II. This inhibi- arachidonic acid tory effect of arachidonic acid could also be demonstrated in a preparation of intact synaptic nerve endings (synaptosomes) 37. 12-HPETE The effects of arachidonic acid and its lnetabolites on K + chan- nels and on Ca2+-calmodulin - dependent protein kinase II can be integrated in a model for the role played by these Epids in metabolite JfCa2~-calmodulin-dependent protein kinase II activity presynaptic inhibition (Fig. 2). According to this model, free arachidonic acid generated in a receptor-dependent fashion is metabolized by 12-1ipoxygenase t K" channel phosphorylation to form 12-HPETE, A metabolite activity of synapsin I of 12-HPETE (possibly a hepox- ilin) may act by modulating the activity of K + channels, while I 12-HPETE itself may inhibit V Ca2+-calmodulin-dependent pro- Ca2" entry number of synaptic vesicles tein kinase II activity and reduce available for Ca2+-stimulated secretic,~ Ca2+-stimulated phosphorylation of synapsin I and other synaptic- v~icle-associated proteins. These two independent actions might be synergistic in decreasing synaptic neurotransmitter release strength.

Fig. 2. Mechanism by which 12-1ipoxygenase metabolites of arachidonic acid may inhibit neurotransmitter release. 12-HPETE and its metabolites may decrease Ca 2÷ entry and concomitantly reduce the number of synaptic vesicles available for Ca 2 ~- Acknowledgements stimulated release, These actions would be expected to have a synergistic effect in The authors thank Drs Robert inhibiting neurotransmitter release. A. Nichols, Jack A. Grebb and Andrew J. Czernik for critical reading of the manuscript and Drs Leonard S. Wolfe and George R. synaptic strength independently reducing Ca 2+ influx during the Siggins for sharing findings before of moddlation. action potential, also decreases the publication. During some of these Inhibition of CaZ+-dependent sensitivity of the secretory appar- studies DP was supported by a protein phosphorylation may atus to an elevated concentration NARSAD Young Investigator underlie some of the inhibitory of intracellular Ca 2+, possibly by Award. actions exerted by FMRF amide in affecting protein phosphorylation. molluscan neurons. In identified Arachidonic acid and its metab- neurons ot the snail Helisoma, olites may regulate neurotrans- References FMRF amide reduced the release mitter release partly by a direct 1 Nestler, E. J. and Greengard, P. (1984) of acet3,1choline by a mechanism action on neuronal K + channels Protein Phosphorylation in the Nervous involving reduction of a voltage- System, I. Wiley & Sons and partly by regulating Ca 2+- 2 Shenolikar, S. and Nairn, A. C. Adv. dependent Ca 2+ current, and a dependent protein phosphoryl- Prof. Phosphorylation Second Messenger direct action on the secretory ation within the synaptic terminal. Res. (in press) apparatus 36. To demonstrate the Lipoxygenase-derived eicosanoids 3 Kaczmarek, L. K. and Levitan, I. B., eds latter effect, Man-Son-Hing and are potent and selective inhibitors (1987) Neuromodulation: The Biochemical Control of Neuronal Excitability, Oxford co-workers clamped the intracel- of purified Ca2+-calmodulin - University Press lular Ca 2+ concentration by using dependent protein kinase II (Ref. 4 Yokoyama, C. et al. (1986) ]. Biol. Cllem. the Ca 2+ chelator nitr 5, whose 37). 12-HPETE inhibited this en- 261, 16714-16721 affinity for Ca 2+ is decreased by zyme with a half-maximal effect 5 Yoshimoto, T. et al. (1990) Proc. Natl UV light. Illumination of Helisoma Acad. Sci. USA 87, 2142-2146 at a concentration of 0.7 ~t~a. By 6 Ueda, N. el al. (1990) ]. Biol. Chem. 265, neurons preloaded with nitr 5 contrast, it had no effect on the 2311-2316 produced a constant, elevated activities of protein kinase C, 7 Pace-Asciak, C. R. (1988) Biochem. Bio- internal Ca 2+ concentration and cAMP-dependent protein kinase, phys. Res. Con.nun. 151, 493-498 enhanced acetylcho~ine release. the Ca2+-calmodulin-dependent 8 Piomelli, D., Shapiro, E., Zipkin, R., When FMRF amide was applied Schwartz, J. H. and Feinmark, S. J. protein kinases I and lII, or the (1989) Proc. Natl Acad. Sci. USA 86, under these conditions, it still Ca2+-calmodulin-activated phos- 1721-1725 produced inhibition of neuro- phatase, calcineurin. Arachidonic 9 Pace-Asciak, C. R. and Lee, W-S. (1989) transmitter release, but without acid was less potent than 12-HPETE I. Biol. Chem. 264, 9310-9313 altering the Ca 2+ concentration. 10 Piomelli, D., Feinmark, S. J., Shapiro, E. by a factor of about 40 as an and Schwartz, J. H. (1989) J. Biol. Chem. Thus FMRF amide, in addition to inhibitor of Ca2+-calmodulin- 263, 16591-16596 TiPS - September 1990 [Vol. 11] 373

11 Pace-Asciak, C. R. (1984) Biochim. Bio- and Rostworoski, K. (1990) /. Neural 33 Bug, W., Role, L., Siegelbaum, S. A. and phys. Acta 793, 485--488 Transm. 29, 29-37 Simmons, L. (1989) Soc. Neurosci. Abstr. 12 Ojeda, S. R., Urbanski, H. F., Junier, 22 Lazarewicz, 1. W., Wroblewski, J. T., 15, 177 M-P. and Capdevila, j. (1989) Amt. NY Palmer, M. E. and Costa, E. (1988) 34 Bley, K. R. and Tsien, R. W. (1990) Acad. Sci. 559, 192-207 Neuropharmacology 27, 765-769 Neuron 2, 379-391 13 Fitzpatrick, F. A. and Murphy, R. C. 23 Felder, C. C., Kanterman, R. Y., Ma, 35 De Camilli, P., Benfenati, F., Valtorta, F. (1989) Pharmacol. Rev. 40, 229-241 A. L. and Axelrod, J. (1990) Proc. Natl and Greengard, P. Annu. Rev. Cell Biol. 14 Weiss, R. H., Arnold, J- L. and Acad. Sci. USA 87, 2187-2191 (in press) Estabrook, R. W. (1987) Arch. Biochem. 24 Piomelli, D., Shapiro, E., Feinmark, S. I- 36 Man-Son-Hing, H., Zoran, M. l., Biophys. 252, 334-338 and Schwartz, i. H. (1987) I. Neurosci. 7, Lukowiak, K. and Haydon, P. G. (1989) 15 Belardetti, F., Campbell, W. B., Falck, 3675-3686 Nature 341,237-239 l- R., Demontis, G. and Rosolowski, M. 25 Piomelli, D. et al. (1987) Nature 328, 37 Piomelli, D. et al. (1989) Proc. Natl Acad. (1989) Neutral 3, 497-505 38-43 Sci. USA 86, 8550-8554 16 Samuelsson, B. and Funk, C. D. (1989) 26 Shapir,-,, E. ct al. (1988) Cold Spri~g I. Biol. Chem. 264, 19469-19472 Harbour Syrup. Quant. Biol. 53, 425--433 17 Lindgren, J-A., H6kfelt, T., Dahl6n, 27 Dennis, E. A. (1987) Drug Dev. Res. 10, FMRFamide: Phe-Met-Arg-Phe-NH2 S-E., Patrono, C. and Samuelsson, B. 205-220 hepoxilin A~: 8-hydroxy-11,12-epoxy- (1984) Proc. Natl Acad. Sci. USA 81, 28 Carlen, P. L. et al. (1989) Brain Res. 497, eicosatrienoic acid 6212-6216 171-176 hepoxilin B3: 10-hydroxy-ll,12-epoxy- 18 Shimizu, T., Takusagawa, Y., Izumi, T., 29 Buttner, N., Siegelbaum, S. A. and eicosatrienoic acid Ohishi, N. and Seyama, Y. (1987) Volterra, A. (1989) Nature 342, 553-555 12-HETE: (12s)-hydroxyeicosatetraenoic I. Neurochem. 48, 1541-1546 30 Moore, S. D., Madamba, S. G., loels, M. acid 19 Shimizu, T. and Wolfe, L. S. (1990) and Siggins, G. R. (1988) Science 239, 12-HPETE: (12s)-hydroperoxyeicosa- ]. Neurochem. 55, 1-15 278-280 tetraenoic acid 20 Dumuis, A., Sebben, M., Haynes, L., 31 Schweitzer, P., Madamba, S. and Siggins, 12-KETE: 12-ketoeicosatetraenoic acid Pin, I-P. and Bockaert, J. (1988) Nature G. R. (1990) Nature 346, 464--467 MK886: 3-[1-(p-chlorophenyl)-5-isopropyl- 336, 68--70 32 Keyser, D. O. and Alger, B. E. (1989) 3-tert-butylthio-lH-indol-2-yl]-2,2- 21 Wolfe, L. S., Pellerin, L., Drapeau, C. Soc. Neurosci. Abstr. 15, 178 dimethylpropanoic acid

adjuvant activity, manifested by Immunomodulatory activity of a potent stimulation of the pro- duction of antibodies against anti- gen administered simultaneously2, small peptides MDP stimulates nonspecific resist- ance against bacterial, viral and Vassil St Georgiev parasitic infections, and against tumors by stimulating cytotoxic macrophages and natural killer The activity of the immune system can be modulated by a wide variety of cells 2, and increases the duration natural and synthetic peptides. Here, Vassil St Georgiev summarizes the of slow-wave sleep 2. However, it actions of some of the immunostimulatory and immunosuppressant small also has other, largely undesirable pelJtides that have shown most promise as therapeutic agents. Some are biological effects such as pyro- already in use as vaccine adjuvants or to prevent graft rejection. There are genicity, transient leukopenia, now indications that these peptides may also be of benefit in conditions in sensitization to endotoxin and which the immune system is compromised, in autoimmune disease and in induction of arthritis, granuloma cancer. and uve~tis 2. A large number of natural and lmmunostimulatory peptides synthetic peptides have been Many peptides have been iso- muramyl moiety evaluated for their immuno- lated that have nonspecific stimu- 6 modulatory effects. These derive latory effects on the immune 4 C H2OH O from (or are modifications of) system. Work is under way to sources as diverse as bacterial cell develop these peptides as adju- wall peptidoglycans, microbial vants in vaccines and as thera- HN' ', and fungal metabolites, peptide peutic agents for conditions in- ',, R1 hormones, fragments of immuno- cluding immune deficiencies and R2 R4 globulins and other plasma pro- cancer. teins and colostrum and milk O CR3 L-alanine 0II ~-isoglutamine proteins. Although in many cases • Muramyl peptides moiety moiety little is known of the mechanism Freund's complete adjuvant - of action, the clinical potential of killed mycobacterial cells emul- MDP many of these peptides has clearly sified in mineral oil - has long been demonstrated. Some of the been known to be a powerful less well-characterized peptides adjuvant of bumoral and cell- In order to determine which may also prove to be useful; their mediated immune reactions. The structural features of the molecule properties are summarized in smallest biologically active unit of of MDP are important for bio- Table I. bacterial cell wall peptidoglycan logical activity, congeners with capable of replacing whole Myco- changes in the muramyi, c-alanyl V. St Georgiev is Vice-President of Chenlical bacteria in Freund's complete and o-isoglutamyl moieties have Sciences, Division of Life Sciences, Orion Research and Technologies Corporation, PO adjuvant is muramyl dipeptide been tested in vitro and in vivo i Box 463, Tampa, FL 33601-0463, USA. (MDP) t. In addition to its immuno- (Table IIL 19qo, ElsevierScience Publishers Ltd. (UK) I)~65- 6147/90/$02.00