Neuron,

Vol. 9,1211-1216,

December,

1992, Copyright

0 1992 by Cell Press

Enhancement of Hippocampal Excitatory Synaptic Transmission by Platelet-Activating Factor Gary D. CIark,‘JJ Leo T. Happel,1,4~5~3 Charles F. Zorumski,6 and Nicolas C. Bazan’,‘J ‘Department of Neurology 4Department of Neurosurgery ‘Department of Ophthalmology 2Department of Pediatrics SDepartment of Physiology 3The Neuroscience Center Louisiana State University Medical School New Orleans, Louisiana 70112-2234 6Department of Psychiatry and Department of Anatomy and Neurobiology Washington University Medical School St. Louis, Missouri 63110

Summary The biologically active lipid platelet-activating factor (lQalkyl-2-acetyknglycer~3-phosphorylcholine; PAFI is a mediator of inflammatory and immune responses, and it accumulates in the brain during convulsions or ischemia. We have examined whether PAF may play a second messenger role in the central nervous system by studying effects on synaptic transmission in cultured hippocampal neurons. Carbamyl-PAF, a nonhydrolyzable PAF analog with a similar pharmacologic profile, augmented glutamate-mediated, evoked excitatory synaptic transmission and increased the frequency of spontaneous miniature excitatory synaptic events without increasing their amplitude or altering their time course. This compound had no significant effect on y-aminobutyric acid-mediated inhibitory synaptic responses. Lyso-PAF, the biologically inactive metabolic intermediate, had no effect on synaptic transmission. Moreover, the enhancement of excitatory synaptic transmission by carbamyl-PAF was blocked by a PAF receptor antagonist. These results indicate a specific presynaptic effect of PAF in enhancing excitatory synaptic transmission in cultured rat hippocampal neurons. Introduction Neural signal transduction involves receptor-mediated activation of phospholipases AZ and C. The products generated from the hydrolysis of membrane phospholipids are second messengers that, in turn, modulate a variety of cell functions. The phospholipase C products inositol trisphosphate and diacylglycerol regulate levels of intracellular ionized calcium and protein kinase C, respectively. From arachidonic acid, a cascade is initiated by enzyme-mediated oxygenation, leading to the synthesis of prostaglandins, hydroxyeicosatetraenoic acids, and other biologically active derivatives. Several of these metabolites, including arachidonic acid itself, have been implicated

in long-term synaptic potentiation (Williams and Bliss, 1989; Williams et al., 1989). Pathological conditions including ischemia (Bazan, 1970; Panetta et al., 1987; Birkle et al., 1988), cerebral edema (Le Poncin and Rapin, 1980), or convulsions (Bazan, 1970; Birkle et al., 1988) result in an overactivation of the events that lead to the synthesis and accumulation of membrane-derived second messengers in brain. Phospholipase A2 is activated rapidly under these conditions, predominantly in gray matter as compared with white matter (Bazan, 1971) and in nerve terminals (Birkle and Bazan, 1987). Synaptosomes and not microsomes isolated from the cerebral cortex of rats undergoing bicuculline-induced status epilepticus show phospholipase A2 activation and accumulation of Iipoxygenase products (Birkle and Bazan, 1987). Key events during these pathological conditions are the influx of calcium and the enhanced release of neurotransmitters, yet it is not clear whether some lipid second messenger may be involved in modulating the release of potentially excitotoxic neurotransmitters. One such mediator could be platelet-activating factor (I-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine; PAF), a metabolite that is synthesized in brain during ischemia or convulsions (Kumar et al., 1988). Damage caused by reperfusion of ischemic brain can be attenuated by PAF antagonists (Panetta et al., 1987; Gilboe et al., 1991). PAF enhances ATP release from PC12 cells (Kornecki and Ehrlich, 1988), suggesting that PAF may be involved in vesicular release of neurotransmitters. We have previously presented evidence that a PAF analog augments the release of excitatory neurotransmitter in rat hippocampal culture (Clark et al., 1991, Sot. Neurosci., abstract). Other evidence suggests that PAF may play a role in synaptic physiology, as a PAF antagonist has been reported to prevent the induction of hippocampal long-term potentiation (Del Cerro et al., 1990; Arai and Lynch, 1992). Although cellular membranes contain the phospholipid I-0-alkyl-acyl-sn-glycero-3-phosphocholine as the storage form of PAF, production of PAF virtually does not occur under resting basal conditions (Bazan, 1989). Since this phospholipid is enriched in arachidonoyl acyl chains in the C2 position, thecalcium-dependent route to synthesize PAF first involves the release of arachidonic acid by a phospholipase AZ, followed by an acetylation at C2 (Snyder, 1990). Therefore, several biologically active mediators may be generated by this pathway, the arachidonic cascade metabolites, and PAF. In the retina and in cultured cells, neurotransmitters elicit PAF synthesis involving the de novo route through cytidine S-diphosphocholine in a Ca2+-independent manner (Bussolino et al., 1986). In addition, PAF is generated

by a Ca*+-dependent

phospholipase

NeUKXl 1212

A

before

(Birkle et al., 1988; Marcheselli et al., 1990; Panetta et al., 1987), as well as to have neuroprotective effects during reperfusion of ischemic brain tissue (Panetta et al., 1987; Spinnewyn et al., 1987). Other PAF antagonists have been shown to have similar effects (Lindsberg et al., 1990; Cilboe et al., 1991). These studies suggest that PAF may be an important messenger in the CNS.

wash

c-PAF

Results

Figure 1. C-PAFAugments Excitatory tured Hippocampal Neurons

Synaptic

Responses

in Cul-

(A) Pre- and postsynaptic responses are shown before, during, and after exposure to 1 uM c-PAF. The evoked action potentials are shown at the bottom of each series of traces, and the postsynaptic currents are shown at the top of each series. Action potentialsare superimposed as are the resulting postsynaptic currents. Representative sample traces are shown. In control conditions (left) the median response was -30.2 pA, and there were four failures (30 trials) of synaptic transmission. During c-PAF (center) the median current response was -99.7 pA (p < .OOOl). During wash the median postsynaptic response was -42.1 pA (p < .OOOl [during c-PAF compared with wash], p < .05 [during wash compared with control]). During (30 trials) and after (30 trials) c-PAF there were no synaptic failures recorded. (6) Peak postsynaptic current amplitudes of the synaptic pair shown in (A) are displayed from the synaptic pair. The time of c-PAF perfusion and of wash with bath solution is shown. Synap tic pairs were noted to have augmented postsynaptic responses within seconds of the onset of c-PAF perfusion; however, since the peak effect of c-PAF did not occur for more than 1.5 min, sampling of EPSCs was continued for this period.

A2 that releases I-O-alkyl-glycero-3-phosphorylcholine (lyso-PAF) and arachidonic acid. Lyso-PAF is then acetylated to form PAF (Braquet et al., 1987; Prescott et al., 1990; Bazan et al., 1991). PAF accumulates in both the intracellular and extracellular compartments in neuronal cells in culture (Yue et al., 1990). There are distinct PAF-binding sites in rat cerebral cortex, microsomal sites that may be involved in early gene expression (Marcheselli et al., 1990; Bazan et al., 1991), and synaptic sites with uncertain functions (Marcheselli et al., 1990). A synaptic site PAF antagonist, BN-52021, has been shown to inhibit phospholipase A2 activation induced by ischemia or seizures

Sincecultured hippocampal neuronsform bothglutamate-mediated excitatory and y-aminobutyric acidmediated inhibitory synapses (Yamada et al., 1989; Yoon and Rothman, 1991), we used this preparation to examine synaptic effects of PAF. PAF is rapidly degraded by ubiquitous tissue hydrolases. Therefore, we have used a nonhydrolyzable PAF analog with a similar pharmacologic profile, carbamyl-PAF (c-PAF), in the following experiments. C-PAF was applied by flow tu bes to synapticallyconnetted cultured hippocampal neurons. Since axons and dendrites could not be distinguished in our hippocampal cultures, the synaptic pairs of neurons were selected based upon their proximity to each other (within 200 urn), and approximately one in five successfully penetrated pairs had synaptic connections in the proper orientation (presynaptic cell accessed with a KMeSO., electrode and postsynaptic cell accessed with a CsMeS04 electrode). Five of six excitatory pairs showed a significant increase in the amplitude of the evoked excitatory postsynaptic currents (EPSCs) during perfusion of 1 uM c-PAF (p < 0.05 by Mann-Whitney U test; Figure 1). The percent increase in the median response ranged from 101% to 230%. The sixth excitatory pair showed such a large increase in the frequency of spontaneous synaptic events during the perfusion of c-PAF that the stimulus-coupled currents could not be analyzed. Interestingly, for as long as 7 min (5-7 min)following perfusion with c-PAF, some (four of six) pairs exhibited a persistent significant augmentation, ranging from 38% to 40% (p < .05 by Mann-Whitney U test), of the median postsynaptic responses, suggesting that c-PAF may have longerlived effects on excitatory synaptic transmission. Since longer periods of recordings (>I2 min of simultaneous whole-cell recordings) have been difficult to obtain, it is possible that c-PAF could lead to even longer lasting effects than those that have been observed in this study. Contrary to what would be expected given the increase in postsynaptic responses, perfusion of c-PAF was noted to decrease the amplitude and the area under the evoked presynaptic action potentials. The presynaptic action potential peak amplitudes were significantly smaller, by 8%-12% (N = 6; p < 8001 Mann-Whitney U test), during the perfusion of c-PAF. The areas under the presynaptic action potentials were signficantly smaller, by 5%-13%, in five of six excitatory pairs (p < .OOOl, five of six; see Figure 1).

PAF Enhances Synaptic Transmission 1213

A

c- PAF

Before

50ms

B

Before

Lyso-PAF

C

BN-52021

BN-52021 + PAF

25pA

Figure 2. C-PAF Specifically Enhances mission through a Receptor-Mediated

Excitatory Synaptic Mechanism

Trans-

(A) Outward evoked synaptic currents were completely blocked by 100 RM bicuculline and thus represent T-aminobutyric acidmediated synaptic transmission (Thio et al., 1992). C-PAF (1 PM) was applied to inhibitory pairs of neurons, and the postsynaptic responses were recorded at -30 mV. The left traces are control responses before c-PAF, and the right traces are from the same synapse during perfusion of c-PAF. No significant effect on this pair of neurons was noted. (B) LysoPAF, an inactive form of the PAF molecule, did not enhance excitatory currents. Left traces represent control responses, and right traces represent the responses to the same synapse during 1 RM lyso-PAF. (C) BN-52021 (1 uM), a PAF synaptic-binding site antagonist, was bath perfused and coapplied with 1 PM c-PAF. The left traces represent the EPSCs recorded in the presence of BN-52021, and the right traces represent the postsynaptic responses during the perfusion of c-PAF and BN-52021. No significant effect of c-PAF was seen in the presence of BN-52021.

Lyso-PAF, the inactive form of the PAF molecule, had similar effects on action potentials, with five of six pairs demonstrating a significant decrease in the peak (3%-g%) and five of six pairs demonstrating a significant decrease in the area under the action potential (8%-21%). If the cell body action potential were larger, one may argue that c-PAF has more generalized augmenting effects on neurons and thus on the input into the presynaptic terminal. The presynaptic effect of c-PAF and lyso-PAF upon action potentials is not the expected one for an enhancement of excitatory synap tic transmission and does not explain the c-PAF effect upon EPSCs. C-PAF (1 PM) was also administered to eight pairs of neurons exhibiting inhibitory postsynaptic currents (IPSCs). No significant change in the evoked re-

sponses was seen in three IPSC pairs, and in five a significant decrease in the evoked responses was noted (Figure 2). In the pairs exhibiting a decrease in IPSCs, there was no recovery of the responses, suggesting that an irreversible rundown of IPSCs had occurred. Indeed, in control experiments, IPSC pairs showed a similar irreversible rundown of responses (N = 4). It is thus unlikely that the decrement in IPSCs could explain the observed enhancement of monosynaptic excitatory synaptic transmission. The lack of a consistent effect on IPSCs suggests that c-PAF specifically alters excitatory synaptic transmission. Lyso-PAF, the C2 nonacetylated, biologically inactive PAF-like compound, was used to examine the chemical specificity of c-PAF. No significant enhancement of current amplitude was noted in any excitatory pair exposed to 1 PM lyso-PAF (N = 6; Figure2). It is, therefore, unlikely that the observed c-PAF effect upon excitatory synaptic transmission is due to a nonspecific perturbation of lipid membrane function, since a structurally closely related lipid, lyso-PAF, has no significant effect. Furthermore, c-PAF’s effects are mediated by the PAF receptor found in synaptic regions, because 1 PM BN-52021, a synaptic PAF-binding site antagonist (Marcheselli et al., 1990), blocked the effect of 1 f.rM c-PAF in four pairs of neurons (Figure 2). The enhancing effects of c-PAF are not mediated by an augmentation of postsynaptic glutamate receptor sensitivity. In hippocampal neurons, fast excitatory synaptic responses most likely are mediated by glutamate acting upon a-amino-3-hydroxy5-methyl-isoxazolepropionic acid (AMPA) receptors. These receptors exhibit rapid and profound desensitization during rapid applications of glutamate. In six neurons, currents evoked by 1 mM glutamate in the presence of 2 mM magnesium and no added glycine were not altered by administration of 1 uM c-PAF. Peak and steady-state glutamate currents were 101% + 7% and 103% f 13% of control, respectively (N = 6). Additionally, c-PAF did not alter the time constant of rapid glutamate receptor desensitization, which was 99% f 5% of control responses (mean f SE, N = 6). The average control peak current at a holding potential of -60 mV was 1090 f 170 pA, and the average time constant of desensitization was 23 f 2 ms in these neurons. The failure of c-PAF to alter responses to exogenous glutamate suggests that the effects on evoked EPSCs are mediated presynaptically. This issue was examined by studying effects of c-PAF on spontaneous miniature EPSCs (mEPSCs) recorded in the presence of 1 WM tetrodotoxin and 100 PM bicuculline. Perfusion of 1 uM c-PAF resulted in a 2- to 7-fold increase in the observed frequency of mEPSCs (N = 6) without a significant effect on the amplitude or decay time course. Cumulative probability plots of peak amplitude and of time course (Tw decay) were nearly superimposable before and after c-PAF (Kolmogorov-smirnovstatisticp>.05; Figure3). SincelOOpM bicuculline completely blocks y-aminobutyric acid-mediated in-

NellrOn 1214

B

Before c-PAF

During c-PAF -/------p,-

50ms

D 10 0.8

0.8

Before c-PAF

Before c-PAF . . . . . . During c-PAF

06

0.6

P

P 04

04 02

00

I 0

2

Figure 3. C-PAF increases

4

the Frequency

6

8

but Not the Amplitude

O0 Fr ”

10

or Duration

of Spontaneous

10

20

mEPSCs

Three minutes of spontaneous activity was recorded and digitized before and during c-PAF infusion. (A) Spontaneous synaptic activity of a neuron voltage clamped at -60 mV before the application of c-PAF. (B) During the perfusion of 1 uM c-PAF an increase in the frequency of synaptic activity was seen in all cells (N = 6). In the neuron shown, 1 uM c-PAF increased the frequency of mEPSCs by 629%. (C) Data from another cell illustrating the effect of c-PAF on the time course of mEPSCs. The graph displays the cumulative probability (p) of observing an event less than or equal to a time to one-half decay, T,, of that shown on the abscissa. The curves shown before (49 events) and during (375 events) c-PAF are nearly superimposed (p > .05, by Kolmogorov-Smirnoff). (D) Data from the same cell as in (C) illustrating the effect of c-PAF on the amplitude of mEPSCs. Again the curves shown before and during c-PAF are nearly superimposed (p > .05, by Kolmogorov-SmirnoffL

hibitory synaptic transmission, the observed increase in frequency of spontaneous excitatory events cannot be explained by an effect on IPSCs (Thio et al., 1992). The increase in the frequency of spontaneous excitatory synaptic events is due to an increase in the probability of spontaneous release of excitatory transmitter (Bekkers and Stevens, 1990; Van der Kloot, 1991). Discussion These results indicate that c-PAF specifically augments hippocampal excitatory synaptic transmission in a receptor-mediated, presynaptic manner and support the possibility that PAF may serve a second messenger role in the CNS. Perfusion of c-PAF resulted in an increase in the amplitude of evoked postsynaptic currents while decreasing the size of presynaptic action potentials. The decrease in the size of action potentials could be the result of PAF activation of a potassium current, such as that described in isolated

bullfrog and guinea pig heart cells (Ramos-Franc0 et al., 1992, Biophys. J., abstract; Nakajima et al., 1991). PAF acts through the arachidonic acid pathway, in these studies, to stimulate the G protein-gated muscarinic potassium channel. The data reported here suggest that c-PAF alters excitatory synaptic transmission through a mechanism that does not augment general presynaptic function. C-PAF application had no significant enhancing effect on inhibitory synaptic transmission, suggesting that PAF acts specifically to augment excitatory synaptic transmission. C-PAF did not alter the rapidly desensitizing currents evoked by flow tube applications of glutamate. These results suggest that c-PAF does not exert its action by modulation of postsynaptic glutamate receptors. Concern that c-PAF could serve as a nonspecific membrane perturbing agent was addressed with the use of lyso-PAF, the biologically inactive form of PAF. Lyso-PAF had no effect upon evoked excitatory synap-

PAF Enhances Synaptic Transmission 1215

tic transmission. Furthermore, the c-PAF augmentation of excitatory synaptic transmission was blocked by a synaptic PAF receptor antagonist, BN-52021. Thus c-PAF appears to act at specific synaptic membrane receptors to exert the observed effect upon evoked excitatory synaptic transmission. C-PAF increased the frequency, but not the amplitude or the time course, of spontaneous miniature excitatory synaptic currents. Since these studies were performed in the presenceof bicuculline, it is unlikely that the increase in observed excitatory transmission could be caused by an alteration of inhibitory synaptic currents. This pattern of an increase in the frequency of synaptic events without a significant change in other properties is typical of a presynaptic effect (Bekkers and Stevens, 1990; Van der Kloot, 1991). Presently, it is unknown whether PAF is generated pre- or postsynaptically in the CNS; however, the production of PAF during periods of intense synaptic activation suggests that it could function as a messenger altering presynaptic release of transmitter (Kumar et al., 1988; Prescott et al., 1990; Williams and Bliss, 1989; Williams et al., 1989; Bazan, 1991). Arachidonate is released with PAF, as the former compound is usually in the C2 position of the PAF precursor phospholipid. Therefore, PAF and arachidonate metabolites may comprise a signaling system modulating the release of excitatory amino acids (Williams and Bliss, 1989; Williams et al., 1989). The mechanisms by which PAF augments excitatory synaptic currents are not known for certain, at present. PAF has been shown to activate phosphoinositide turnover, to increase free fatty acids, and to activate protein kinase C in some cells (Yue et al., 1991). If proteins such as synaptotagmin, a synaptic vesicle protein, are involved in the release of excitatory transmitter, then the recent finding of a dramatic increase in synaptotagmin calcium binding in the presence of phospholipids may be important here (Brose et al., 1992). In neural cells and recently in a fraction of cultured hippocampal neurons, PAF has been shown to increase intracellular calcium (Kornecki and Ehrlich, 1988; Yue et al., 1991; Bito et al., 1992). This suggests that excitatory and inhibitory presynaptic terminals differ in their sensitivity to PAF, as a general effect on intracellular calcium would be expected to augment both excitatory and inhibitory synaptic responses. The selective effect on excitatory synaptic transmission makes it important to consider possible roles of PAF in glutamate-mediated synaptic plasticity and excitotoxicity. Thus, PAF may regulate excitatory synaptic transmission, possibly participating in long-term potentiation, in excitotoxicity and in seizure generation. Moreover, an increased formation of PAF could lead to sustained release of excitatory neurotransmitters and, in turn, to neuronal damage.

methods (Huettner and Baughman, 1986; Nakajima et al., 1986; Clark et al., 1990). Matrigel (Collaborative Research), diluted l:lO, was used instead of collagen. Cytosine bc-arabinofuranoside (Ara-C) (10 PM) was added at 24 hr. At 72 hr, the growth medium was completely replaced with a serum-free medium consisting of minimum essential medium (without glutamine, with Earle’s salts) (GIBCO), 400 PM L-glutamine, SO pglml streptomycin, 50 U/ml penicillin, 6 m M glucose, 10 Kg/ml insulin, 10 pg/ ml human transferrin, and 10 rig/ml sodium selenium. At the time of recording, the growth medium was exchanged with a balanced salt solution consisting of 140 m M NaCI, 5 m M KCI, 10 m M HEPES, 10 m M glucose, 5 m M CaCl,, 5 m M MgCI, (pH 7.3). The elevated concentrations of divalent cations were used to depress polysynaptic activity CThio et al., 1992). Additionally, the use of glycine-free solutions containing magnesium allowed study of fast EPSCs mediated by non-NMDA receptors (Forsythe et al., 1988). Presynaptic neurons were studied with patch electrodes (4 MCJ) filled with 150 m M KMeS04, 5 m M NaCI, IO m M HEPES, 5 m M 1,2-bis(o-aminophenoxylethane)-N,N,N’,N’tetra-acetic acid (BAPTA), 0.5 m M CaCIZ, and 2 m M Mg-ATP. Postsynaptic neurons were studied with the same internal solution except that 150 m M CsMeSO, was used to block potassium channels and to improve voltage control of the synapse. Presynaptic neurons were studied in current clamp at - -70 mV. Depolarizing current injection was used to evoke action potentials, and the postsynaptic cell response was recorded at -70 mV under whole-cell voltage clamp (List EPC-7). Neurons were studied as synaptic pairs, and presynaptic neurons were stimulated at 1 Hz to evoke action potentials. The postsynaptic responses were sampled before or during exposure to experimental compounds and after being washed with a bath solution. Presynaptic action potentials and postsynaptic currents were filtered at 3 kHz (& pole Bessel) and digitized at 10 kHz. All drugs were delivered by large-bore flow tubes (580 pm) placed approximately 500 Frn from the synaptic pair. Spontaneous synaptic activity was studied in an external solution consisting of 140 m M NaCI, IS m M KCI, 3 m M CaCI,, 1 m M M&I,, 10 m M glucose, 0.001 m M tetrodotoxin, and 0.1 m M (-)bicuculline methiodide. The elevated potassium was used to increase the probability of spontaneous mEPSCs, and bicuculline was included to block y-aminobutyric acid A-mediated synaptic responses. mEPSCs were filtered at 2 kHz (&pole Bessel) and digitized at 3 kHz. Events were analyzed using routines written in Axobasic (Axon Instruments). BN52021 was a gift from Dr. Pierre Braquet (Institut Henry Beaufour, LePlessis Robinson, France). Data Analysis Synaptic currents are dependent upon the probability of release of vesicles of transmitter, the number of potential release sites, and the amplitude of the postsynaptic response to an individual vesicle release. Thus, there are discrete levels that are, at least theoretically, indefinite and occur independently. Such a function is not normally distributed, and thus tests that assume a normal distribution will not be valid. We have chosen to report results as the median response, because the mean does not usually correlate with any recorded event and does not provide useful information about the distribution of evoked excitatory currents. Median responses are, by definition, the amplitude of the middle response in the rank order of the events under analysis. The Mann-Whitney U test makes no assumptions about the probability distribution. It tests the similarity of the median responses, given the rank order of the other observations in each population being tested. mEPSCs are also not normally distributed. The analysis of mEPSCs was undertaken using cumulative probability plots that make no assumption about the distribution of events. The Kolmogorov-bmirnovstatisticteststheprobabilitythattwocumulative probability plots are similar (Van der Kloot, 1991). Acknowledgments

Experimental

Procedures

Postnatal (postnatal day I-3) hippocampal primary culture

rats were used to prepare whole using modifications of established

The corresponding author is N. C. B. (LSU Neuroscience Center, 2020 Cravier Street, Suite B, New Orleans, LA 701122234). This work was supported by grants NS23002 (N. C. B.), NS04133

Neuron 1216

(G. D. C.), MHOO964, MH45493, and AGO5681 (C. F. Z.) and the American Heart Association-Louisiana (C. D. C.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 USC Section 1734 solely to indicate this fact. Received

July 23, 1992; revised

September

Arai, A., and Lynch, G. (1992). Antagonists of the plateletactivating factor receptor block long-term potentiation in hippocampal slices. Eur. J. Neurosci. 4, 411-419. Bazan, N. G. (1970). Effects of ischemia and electroconvulsive shock on free fatty acid pool in the brain. Biochim. Biophys. Acta 278, I-IO. Bazan, N. G. (1971). Free fatty acid production in cerebral white and grey matter of the squirrel monkey. Lipids 6, 211-212. Bazan, N. G. (1989). Arachidonic acid in the modulation of excitable membrane function and at the onset of brain damage. Ann. NY Acad. Sci. 559, 1-16. Bazan, N. G., Squinto, S. P., Braquet, P., Panetta, T., and Marchesell, V. L. (1991). Platelet-activating factor and polyunsaturated fatty acids in cerebral ischemiaor convulsions: intracellular PAFbinding sites and activation of a Fos/Jun/Apl. Lipids 26, 1-7. Bekkers, J. M., and Stevens, C. F. (1990). Presynaptic mechanism for long-term potentiation in the hippocampus. Nature346,724729. Birkle, D. L., and Bazan, N. G. (1987). Effects of bicucullineinduced status epilepticus on prostaglandins and hydroxyeicosatetraenoic acids in rat brain subcellular fractions. J. Neurothem. 481768-1778. Birkle, D. L., Kurian, P., Braquet, P., and Bazan, N. C. (1988). Platelet-activating factor antagonist BN 52021 decreases accumulation of free polyunsaturated fatty acid in mouse brain during ischemia and electroconvulsive shock. J. Neurochem. 57, 19001905. Bito, H., Nakamura, M., Honda, Z., Izumi, T., Iwatsubo, T., Sey ama, Y., Ogura, A., Kudo, Y., and Shimizu, T. (1992). Plateletactivating factor (PAF) receptor in rat brain: PAF mobilizes intracellular Ca*+ in hippocampal neurons. Neuron 9, 285-294. L., and Vargatig, factor research.

B. B. (1987). Pharmacol.

Brose, N., Petrenko, A. G., Siidhof, T. C., and Jahn, R. (1992). Synaptotagmin: a calcium sensor on the synaptic vesicle surface. Science 256, 1021-1025. Bussolino, F., Gremo, F., Tetta, C., Pescarmona, G., and Camussi, C. (1986). Production of platelet-activating factor by chick retina. J. Biol. Chem. 267, 16502-16508. Clark, G. D., Clifford, D. B., and Zorumski, C. F. (1990). The effect of agonist concentration, membrane voltage and calcium on N-methyl-o-aspartate receptor desensitization. Neuroscience 39, 787-797. Del Cerro, S., Arai, A., and Lynch, G. (1990). Inhibition term potentiation by an antagonist of platelet-activating receptors. Behav. Neural Biol. 54, 213-217.

of longfactor

Forsythe, I. D., Westbrook, C. L., and Mayer, M. L. (1988). Modulation of excitatory synaptic transmission by glycine and zinc in cultures of mouse hippocampal neurons. J. Neurosci. 8, 37333741. Gilboe, D. D., Kinter, D., Fitzpatric, J. H., Emoto, S. E., Esanu, A., Braquet, P. C., and Bazan, N. C. (1991). Recovery of postischemic brain metabolism and function following treatment with a free radical scavenger and platelet-activating factor antagonists. J. Neurochem. 56, 311-319. Huettner, identified Neurosci.

Kumar, R., Harvey, S., Kester, N., Hanahan, D., and Olson, M. (1988). Production and effects of platelet-activating factor in the rat brain. Biochim. Biophys. Acta 963, 375-383. Le Poncin, L. M., and Rapin, J. R. (1980). Effects of Ginkgo biloba on changes induced by quantitative cerebral microembolization in rats. Arch. Int. Pharmacodyn. Ther. 243, 236-244.

18, 1992.

References

Braquet, P., Shen, T. Y., Touqui, Perspectives in platelet-activating Rev. 39, 97-145.

Kornecki, E., and Ehrlich, Y. H. (1988). Neuroregulatory and neuropathological actionsof theether-phospholipid platelet-activating factor. Science 240, 1792-1794.

J. E., and Baughman, R. W. (1986). Primary culture of neurons from the visual cortex of postnatal rats. J. 6. 3044-3060.

Lindsberg, P. J., Yue, T.-L., Fredericks, K. U., Hallenback, J. M., and Feuerstein, G. (1990). Evidence of platelet-activating factor as a novel mediator in experimental stroke in rabbits. Stroke 27, 1452-1457. Marcheselli, V. L., Rossowska, M. J., Domingo, M. T., Braquet, P., and Bazan, N. C. (1990). Distinct platelet-activating factor binding sites in synaptic endings and in intracellular membranes of rat cerebral cortex. J. Biol. Chem. 265, 9140-9145. Nakajima, Y., Nakajima, S., Leornard, R. J., and Yamaguchi, K. (1986). Acetylcholine raises excitability by inhibiting thefast transient potassium current in cultured hippocampal neurons. Proc. Natl. Acad. Sci. USA 83, 3022-3026. Nakijima, T., Sugimoto, T., and Kurachi, Y. (1991). Plateletactivating factor activates cardiac Cx via arachidonic acid metabolites. FEBS Lett. 289, 239-243. Panetta, T., Marcheselli, V. L., Braquet, P., Spinnewyn, B., and Bazan, N. G. (1987). Effects of platelet activating factor (BN 52021) on free fatty acids, diacylglycerols, polyphosphoinositides and blood flow in thegerbil brain: inhibitionof ischemia-reperfusion induced cerebral injury. Biochem. Biophys. Res. Commun. 749, 580-587. Prescott, S. M., Zimmerman, C. A., and McIntyre, T. M. (1990). Platelet-activating factor. J. Biol. Chem. 265, 17381-17384. Snyder, F. (1990). Platelet activating factor and related acetylated lipidsas potent biologicallyactivecellular mediators.Am. J. Physiol. 259, 697-708. Spinnewyn, B., Blavet, N., Clostre, F., Bazan, N. C., and Braquet, P. (1987). Involvement of platelet-activating factor (PAF) in cerebral post-ischemic phase in mongolian gerbils. Prostaglandins 34, 337-349. Thio, L., Clark, G., Clifford, D., and Zorumski, C. (1992). Wheat germ agglutinin enhances EPSC’s in cultured rat postnatal hippocampal neurons by blocking ionotropic quisqualate receptor desensitization. J. Neurophysiol., in press. Van der Kloot, W. (1991). The regulation Neurobiol. 36, 93-130.

of quanta1 size. Prog.

Williams, J. H., and Bliss, T. V. P. (1989). An in vitro study of the effects of lipoxygenase and cyclooxygenase inhibitors of arachidonic acid on the induction and maintenance of long-term potentiation in the hippocampus. Neurosci. Lett. 707, 301-306. Williams, J. H., Errington, M. L., Lynch, M. A., and Bliss, T. V. P. (1989). Arachidonic acid induces a long-term activity-dependent enhancement of synaptic transmission in the hippocampus. Nature 347, 739-742. Yamada, K. W., Dubinsky, J. M., and Rothman, S. M.(1989). Quantitative physiological characterization of a quinoxalinedione non-NMDA receptor antagonist. J. Neurosci. 9, 3230-3236. Yoon, K. W., and Rothman, S. M. (1991J.Adenosine inhibits excitatory but not inhibitory synaptic transmission in the hippocampus. J. Neurosci. 77, 1375-1380. Yue, T.-L., Lysko, P. G., and Feurstein, G. (1990). Production of platelet-activating factor from rat cerebellar granule cells in culture. J. Neurochem. 54, 1809-1811. Yue, T. L., Gleason, M. M., Gu, J. L., Lysko, P. G., Hallenbeck, J., and Feuerstein, C. (1991). Platelet-activating factor (PAF) recep tor-mediated calcium mobilization and phosphoinositide turnover in neurohybrid NC10815 cells: studies with BN-50739, a new PAF antagonist. J. Pharmacol. Exp. Ther. 257, 374-381.