Training in the Morris Water Maze Occludes the Synergism Between

Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press

Training in the Morris Water Maze Occludes the Synergism Between ACPD and Arachidonic Acid on Glutamate Release in Synaptosomes Prepared from Rat Hippocampus Bernadette McGahon, 1 Christian Holscher, 2 Liam McGlinchey, 2 Michael J. Rowan, 2 and Marina A. Lynch 1,3 1Department of Physiologyand 2Department of Pharmacologyand Therapeutics Trinity College Dublin 2, Ireland

Abstract In~oducUon It is a widely held view that memory forma- We here that release of glutamate, report tion is dependent on changes in synaptic efficacy inositol phospholipid metabolism, and that allows strengthening (or weakening) of asso- protein kinase C (PKC) activity are ciations between neurons. The hippocampus has increased in synaptosomes prepared from been identified as a brain area that plays a signifi- hippocampi of rats that had been trained in cant role in memory processing (Squire 1992), a spatial learning task. In hippocampi particularly processing of spatial memory (e.g., obtained from animals that were untrained, McNaughton et al. 1986; O'Keefe 1991), and activation of the metabotropic glutamate though it has been extensively studied, the mech- receptor by the specific agonist anisms that underlie acquisition and recall are not trans, l.amino.cyclopentyl, l ,3.dicarboxylate known. Certain cellular changes have been associ- (ACPD) increased release of glutamate but ated with learning; many of these have been iden- only in the presence of a low concentration tiffed as a result of comparison between young of arachidonic acid. A similar interaction animals, which perform well in learning tasks, and between arachidonic acid and ACPD was aged animals, which often exhibit learning impair- observed on inositol phospholipid turnover ments (Barnes 1979), and in studies designed to and on PKC activity. However, the investigate similarities and/or differences between synergistic effect of arachidonic acid and learning and long-term potentiation (LTP). The ACPD on glutamate release was occluded in hypothesis that LTP in the hippocampus is a bio- hippocampal synaptosomes prepared from logical correlate of learning and/or memory is ac- trained rats. Occlusion of the effect on tively challenged, but there is a good deal of indi- inositol phospholipid turnover and PKC rect evidence lending support to this idea. Thus, it activation was also observed. These data has been reported that AP5 blocks induction of suggest that the molecular changes that LTP (see Bliss and Collingridge 1993) and also the underlie spatial learning may include ability of rats to perform in a spatial learning task activation of metabotropic glutamate (Morris et al. 1986), that both LTP and learning receptors in the presence of arachidonic are associated with increased release of glutamate acid and that the interaction between in dentate gyrus (Richter-Levin et al. 1994), and arachidonic acid and ACPD triggers the that protein kinase C (PKC) activity is increased presynaptic changes that accompany following induction of LTP (Akers et al. 1986; An- learning. genstein et al. 1994) and in a learning paradigm (Fordyce et al. 1994). 3Corresponding author. It has been reported recently that when acti-

LEARNING & MEMORY 3:296-304 9 1996 by Cold Spring Harbor Laboratory Press ISSN1072-0502/96 $5.00

L E A R N I N G & M E M O R Y 296 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press TRAINING OCCLUDES SYNERGISM BETWEEN ACPD AND AA vation of metabotropic glutamate receptors is used in these experiments. The platform (diam., blocked, learning in a spatial task is impaired 10 cm) was 4 cm below the water surface during (Richter-Levin et al. 1994), and this observation, training. The water was kept at 23-+2~ and made as well as the finding that learning was accompa- opaque with titanium dioxide. The pool was situ- nied by increased glutamate release in dentate gy- ated in a room with visual cues (e.g., door, win- rus, brings into the context of learning and recall dow, curtain, etc.). All animals were marked with the reports that trans-1-amino-cyclopentyl-1, 3-di- a black water-resistant marker to allow monitor- carboxylate (ACPD) and arachidonic acid (AA) act ing. The animals' movements were recorded with synergistically to increase release of glutamate in a video camera attached to the ceiling, and data cortex (Herrero et al. 1992) and hippocampus was analyzed using a tracking program written by (McGahon and Lynch 1994). AA has not yet been James Mahon (Trinity College, Dublin). During shown to play a role in spatial learning, but it has task aquisition, the program measured latency of been implicated in learning in the chick (HGlscher animals to reach the platform and the distance and Rose 1994), and rat (H/~lscher et al. 1995) covered. The pool was divided into four arbitrary and in maintenance of LTP in the dentate gyrus of quadrants for transfer test analysis. The quadrant the rat (Williams et al. 1989; Clcments et al. 1991; in which the platform had been located (SE) was Lynch and Voss 1994). designated the target quadrant. During the transfer In designing these experiments, we consid- test, the percentage of the distance covered was ered that changes at the level of the synapse, re- assessed for each of the four quadrants. suiting in "strengthening" of these synapses, prob- Animals were placed in the water at one of ably accompany spatial learning and that presyn- four starting positions that alternated in a clock- aptic modifications are likely to contribute to wise manner. Six trials were performed per day, these changes. In this study we have examined the and the intertrial interval was -5 min. The cut-off effect of spatial learning on three parameters in time for a trial, if the animals failed to locate the hippocampal synaptosomes: glutamate release, platform, was 120 sec; in this case the animals inositol phospholipid metabolism, and PKC activa- were manually placed on the platform for 10 sec. tion. In addition we have compared the effect of On day 4, the platform was removed and the ani- AA and ACPD, alone and in combination, on these mals were given a 60-see duration transfer test in three parameters in synaptosomes prepared from the pool. The group that did not learn the task hippocampus obtained from untrained rats and were given six swimming sessions in the pool each hippocampus obtained from rats that had been day without a platform. The duration of the session trained to locate a hidden platform in the Morris was the same as that of the group of animals that water maze. The data presented indicate that spa- learned the task. tial learning increases glutamate release, inositol Animals were sacrificed by cervical disloca- phospholipid metabolism, and PKC activation. tion 1 O-15 min after completion of the behavioral Moreover, although AA and ACPD interact to fur- test. The hippocampus was dissected free; the time ther increase all three measures in synaptosomes taken for this dissection was -1 min. Slices (350 prepared from untrained tissue, this effect is oc- p~m) were prepared using a Mcllwain tissue chop- cluded in hippocampus obtained from trained rats. per, suspended in Krebs solution, thoroughly mixed, and divided into three aliquots for later Materials and Methods assessment of glutamate release, inositol phospholipid metabolism, and PKC activity. Slices were fro- ANIMALS zen in Krebs solution containing 10% DMSO (Haan and Bowen 1981 ) and stored in three sep- Male Wistar rats (250-300 grams) were used arate aliquots in liquid N 2 until required for anal- in these experiments. Animals were housed in ysis. groups of four to six under a 12-hr light schedule. The temperature was controlled between 22~ and 23~ TISSUE PREPARATION Slices were thawed rapidly ( 1.5-2 min) by ag- TRAINING SCHEDULE itation at 37~ and rinsed four times in excess A white fiberglass circular pool (diam., 120 fresh oxygenated Krebs solution. Tissue was ho- cm; height, 48 cm; depth of water, 34 cm) was mogenized in 0.32 M ice-cold sucrose and centri-

L E A R N I N G & M E M O R Y 297 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press McGahon et al. fuged at 5000 rpm for 5 min. The crude synapto- founding effects. In preliminary experiments, re- somal pellet, P2, was prepared by centrifuging the lease of [3H]glutamate was compared in synapto- resulting supernatant at 15,000 rpm for 15 min. P2 somes prepared in sucrose alone and in sucrose to was used for analysis of glutamate release, inositol which bovine serum albumin (BSA) was added to phospholipid metabolism, and PKC activity. All remove polyunsaturated fatty acid that may have biochemical analysis was completed on all samples been released during preparation. No difference that were tested in the behavioral paradigm. between the two groups was observed, and, so, the experiments described here were performed on slices that were prepared in the absence of BSA.

ASSESSMENT OF GLUTAMATE RELEASE

Synaptosomes were resuspended in ice-cold ANALYSIS OF INOSITOL PHOSPHOLIPID TURNOVER Krebs solution (composition in mM: NaCI, 136; KCI, 2.54; KH2PO4, 1.18; MgSO4.7H20 , 1.18; To examine [3H]inositol labeling of phosphoi- NaHCO3, 16; glucose, 10) containing 2 mM CaCl 2 nositides and inositol phosphates, synaptosomes and incubated for 15 min at 37~ in the presence were resuspended in oxygenated ice-cold Krebs of [3H]glutamate (Amersham, UK; specific activity, buffer (composition in mM: NaCI, 136; KCI, 2.54; 20-40 Ci/mmole; final concentration, 5 • 10 -7 KJJ2PO4, 1.18; MgSO4.7H20 , 1.18; NaHCO3, 16; mi). Tissue was aliquoted onto Millipore filters CaCI 2, 1.3; glucose, 10; cytidine, 1; LiCI, 5) and (0.45 ixm) and rinsed under vacuum at least 20 aliquots (50 Ixl) were incubated at 37~ for 35 times by addition of 250 p~l of ice-cold, oxygen- min in a shaking water bath in 120 pA Krebs buffer ated, fresh Krebs solution to the filter papers. Tis- containing [3H]myoinositol (final concentration, sue was then incubated for 5 min at 37~ in 250 Ixl 0.3 ptM; specific activity, 110 mCi/mg; Amersham, oxygenated Krebs solution with no added calcium UK). AA (10 pA; final concentration, 1 ~M) or or with Krebs solution containing 2 mi CaCI 2. The ACPD (10 pd; final concentration, 50 ~ZM) or an filtrate was discarded. This step was repeated, but equivalent volume of control vehicle was added to incubation continued for 10 sec. [In earlier exper- the synaptosomal suspension at that point, and in- iments, we investigated release at different incu- cubation continued for a further 35 min. In some bation times (10, 20, and 60 sec and 5 min). Small experiments, both ACPD (final concentration, 50 time-related differences in the proportions of cal- ~M) and AA (final concentration, 1 p~M) were cium-independent and calcium-dependent release present in the incubation medium. The final assay were observed and because a greater Proportion volume was 250 pA in all experiments. The reac- of calcium-dependent release was observed in the tion was terminated by addition of ice-cold shorter time period, the 10-see incubation period trichloroacetic acid (18 pd; final concentration, was chosen. This finding is consistent with the 5% ), 940 ~zl of chloroform/methanol 2:3 (vol/ literature, which indicates that calcium-dependent vol), 310 pul of chloroform, and 310 &l of water. release is initially very rapid, whereas calcium-in- Organic and aqueous layers were vortex mixed dependent release increases linearly with time and separated by centrifugation. To assess [3H]ino- (e.g., Nicholls 1989).] Filtrate was collected for sitol labeling of inositol phosphates, aliquots (400 scintillation counting. The incubation step was re- p~l) of the aqueous phase were added to 1 ml of peated, but in this case 40 mM KCI was added to Dowex slurry (formate form, 50% in water; depolarize the synaptosomes. In some cases AA Dowex-1, 8% cross-linked, 200 mesh, Sigma) and (final concentration, 1 ~i), ACPD (50 ~M), or washed four times with myoinositol (5 mM). Inosi- both were added during the incubation period to tol phosphates were eluted with 500 pd of 0.1 M examine the effects of these agents on unstimu- NH4OH in 0.1 M formic acid. Aliquots (400 p.l) of lated and KCl-stimulated release. Results from pre- the eluate were added to scintillation fluid for vious experiments indicated that higher concen- scintillation counting and calculation of total la- trations of AA increase release of glutamate in the beling of inositol phosphates by [3H]inositol. To absence as well as in the presence of calcium and assess labeling of phosphoinositides by [3H]inosi- also interfere with glutamate uptake (Lynch and tol, aliquots (320 ~1) of the lower organic phase Voss 1990); a final concentration of 1 pLi AA was were evaporated overnight, scintillant was added, chosen in these experiments to avoid these con- and samples counted for radioactivity. The protein

L E A R N I N G & M E M O R Y 298 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press TRAINING OCCLUDES SYNERGISM BETWEEN ACPD AND AA concentration was estimated by the method of Sigma (UK). All radiochemicals and the PKC assay Bradford (1976), and all values were expresed as kit were obtained from Amersham (UK). cpm/p~g protein.

Results ANALYSIS OF PKC

PKC activity was examined using an enzyme TRAINING assay system (Amersham, UK) that assessed trans- The animals that were trained in the water fer of the radiolabeled phosphate group from ad- maze task learned the task in the acquisition trials. enosine-5-triphosphate ([32p]ATP) to histone, a The period of time required to locate the platform peptide specific for PKC. P2 was prepared and was reduced from 102 sec in trial 1 to 8 sec in trial lysed by incubating at 4~ for 10 min in lysis buffer 18 (Fig. 1). In the transfer task, untrained control [composition: 5 mi Tris-HCl at pH 7.5, containing animals, which swam in the pool during the acqui- 5 mi EDTA, 10 mM EGTA, 0.3% (wt/vol) ~3-mer- sition phase without a platform, did not show a captoethanol, 10 mi benzamidine, and 50 I~l/ml preference for any quadrant (ANOVA over all of phenylmethylsulphonyl fluoride]. The mem- quadrants, F(3,20)=2.4, P>0.05; n=6). In con- brane fraction was pelleted and resuspended in trast, animals that were trained, spent a greater assay buffer [composition: 50 mi Tris-HC1 at pH proportion of time in the SE quadrant (the target 7.5, containing 5 mi EDTA, 10 mi EGTA, 0.3% quadrant) than in any of the other quadrants (wt/vol) [3-mercaptoethanol, 10 mi benzamidine (ANOVA over all quadrants, F(3,20) = 21, and 50 I~l/ml of phenylmethylsulphonyl fluoride]. P<0.001; n=6). A two-tailed unpaired t-test The reaction was started by addition of [32p]ATP showed a significant difference in the distance to the synaptosomal suspension, which was then swum in the SE quadrant by the trained group incubated at 25~ for 15 min. In some experi- compared with the untrained control group ments, the effect of AA (final concentration, 1 p~i), (t= 3.09 with 10 degrees of freedom; P<0.02). ACPD (50 I~i), or both was assessed; in this case Within the trained group, there was a significant the compounds were added prior to the 15-min difference in the distance swum in the target quad- incubation period. The final assay volume was 75 rant (SE) compared with the distance swum in the ill. Phosphorylated peptide was separated by ap- opposite quadrant (NW quadrant; t= 7.1 with 10 plying the reaction mixture to binding paper and degrees of freedom; P<0.001; n = 6). washing with 5% (vol/vol) acetic acid. Binding papers were added to scintillant and 32p counted. Results were expressed as picomoles phosphate A B transferred per minute. 120 50 ~ ~ ~r-~-

STATISTICAL ANALYSIS o 2 9 as l

A one-way analysis of variance was performed r 4(? ~ to determine whether there were significant differences between conditions. When this analysis indicated significance (at the 0.05 level), post hoc 4 6 8 10 12 14 16 18 Tart at Student's Newmann Keuls test analysis was used to Day 1 Day 2 Day 3 NW NE SW SE determine which conditions were significantly dif- Trials Quadrant ferent from each other. In some cases the Student's Figure 1 : (A) Time required to locate a hidden platform t-test for independent means was used to establish in a water maze. The period of time required to locate statistical significance. the platform was reduced from 102 sec in the first trial to 8 sec in the eighteenth trial. (B) Percentage of path swum by rats in a transfer task without a platform. Trained rats MATERIALS swam a greater distance in the target (SE) quadrant compared with the opposite (NW) quadrant (P<0.O01). ACPD was obtained from Tocris (UK); AA and Trained rats also swam a significantly greater distance in other standard chemicals were obtained from the target quadrant than untrained rats (P<0.02).

L E A R N / N G & M E M O R Y 299 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press McGahon et ai.

EFFECTS OF AA AND ACPD ON GLUTAMATE phospholipid turnover (P

L E A R N I N G & M E M O R Y 300 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press TRAINING OCCLUDES SYNERGISM BETWEEN ACPD AND AA

Figure 2: (A) Potassium stimulated calcium-dependent release of [3H]glutamate in synaptosomes prepared from hippocampus obtained Control from trained and untrained rats: effect of ACPD and AA. AA and ACPD, .2 A Release ~ AA in combination (but neither alone), significantly increased KCl-stimu- ]~ ACPD AA/ACPD lated glutamate release (P<0.001 ; one-way ANOVA). Release was sig- ~+ i~~~ nificantly increased in synaptosomes prepared from hippocampus ob- ~ [_~~[~~ tainedfromtrainedrats, comparedwith untrained(+ P<0.05;Stu- ~< .1 dent's t-test for independent means). No effect of AA or ACPD, alone or in combination, affected release hippocampal synaptosomes prepared from trained animals. Results were calculated by subtracting basal release from KCI-stimulated release, and individual values were used to 0 calculate the mean-+s.E.M, for six observations. Release is expressed as 30 B InositolPhospholipid turnover a percentage of the radiolabel present at the start of the incubation period. (B) Inositol phospholipid metabolism in synaptosomes prepared +-----El from hippocampus obtained from trained and untrained rats: effect of ~, ** ACPD and AA. AA (1 I,LM) and ACPD (50 It, M) significantly increased ~" *~ *~~ I1 inositolphospholipid metabolism in hippocampal synaptosomespre-

~• 15 pared from untrained rats (control synaptosomes; *P<0.01; ANOVA); in combination, AA and ACPD increased inositol phospholipid metab- o olism to a greater extent (**P<0.001 compared with control; P<0.05 compared with either AA or ACPD alone). Inositol phospholipid metabolism was significantly increased in hippocampal synaptosomes pre- o pared from trained rats, compared with control (+ P<0.05; Student's t-test for independent means), but neither AA nor ACPD nor both agents 5 C PKC activation had any effect in synaptosomes prepared from trained rats. Values are =_ , +~ means (+S.E.M.) of six observations and are expressed as cpm ( x 10-3)/ I ** I mg protein. (C) PKC activity in synaptosomes prepared from trained and ~-t, 2.s ~[~[~~~~]~.. ~~~~ untrained rats: effect of ACPD and AA. AA (1 p,M) and ACPD (50 p.M) significantly increased PKC activity in hippocampal synaptosomes pre- = pared from untrained rats (*P<0.05; ANOVA); in combination, AA and ACPD increased PKC activity to a greater extent (**P<0.001 compared o with control; P<0.05 compared with either AA or ACPD alone). PKC UNTRAINED TRAINED activity was significantly increased in hippocampal synaptosomes prepared from trained rats, compared with untrained rats (+ P<0.05; Student's t-test for independent means); but no effect of AA, ACPD, or both agents together was observed in hippocampal synaptosomes prepared from trained rats. Values are means (--S.E.M.) of six observations and are expressed as picomoles phosphate incorporatedlmin. untrained rats, was occluded in hippocampi ob- (McGahon and Lynch 1994), and dentate gyrus tained from trained rats. (McGahon and Lynch 1996). This effect was ab- The present data indicate that a low concen- sent in hippocampal synaptosomes prepared from tration of AA did not affect glutamate release, sup- trained rats. This observation suggests that training porting earlier reports (Herrero et al. 1992; Mc- occluded this effect, although it must be consid- Gahon and Lynch 1994) and contrasting with the ered that other factors, for example, stress, may inhibitory effect of higher concentrations (Lynch have contributed to the changes. Because ACPD is and Voss 1990). The finding that ACPD had no a nonspecific activator of mGluRs, it is not possible effect on release indicates that it is unlikely to to speculate on the receptor subtype activated in activate a presynaptic glutamate receptor that these experiments. However, because previous ev- functions as an autoreceptor, although such a re- idence suggests that coupling to cAMP is unlikely ceptor has been observed in neonates (Baskys (McGahon and Lynch 1994) and because PLC ac- 1992). Although no effect of either AA or ACPD tivation is stimulated by ACPD, it might be spec- alone was observed, these agents acted synergisti- ulated that mGluR1 or mGluR5 may play a role caUy to increase glutamate release in synapto- (Pin and Duvoisin 1995). somes prepared from hippocampi of untrained an- The increase in release of glutamate described imals, supporting earlier observations in the cor- here supports the recent observation of Richter- tex (Herrero et al. 1992), whole hippocampus Levin et al. (1995) who reported that training in a

L E A R N I N G & M E M O R Y 301 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press McGahon et al. spatial learning task was associated with increased from trained animals. These parallel effects suggest glutamate release in dentate gyms. In the present that the increase in glutamate release may derive study, tissue was preparcd 15 min after the trans- from increased inositol phospholipid metabolism. fer test, whereas in the study of Richter-Levin, tis- It is well established that IP 3 increases intrac- sue was prepared 4 hr after testing, and, therefore, ellular calcium concentration in several systems, it is possible that recall contributes to a greater including in synaptosomes prepared from dentate extent to the changes described in the present gyrus (Lynch and Voss 1991), and this may con- study. It is of interest that classical conditioning tribute to release; activation of PKC also stimulates has also been associated with increased glutamate transmitter release (Parfitt and Madison 1993; Ter- release in dentate gyrus (Laroche et al. 1987). The rian et al. 1993). One interpretation of these re- fact that increased glutamate release is a feature of sults is that the increase in glutamate release ob- both learning and maintenance of LTP (Lynch et served here is a consequence of increased inositol al. 1989; Canevari et al. 1994) supports the idea phospholipid metabolism. The occlusion of the ef- that LTP may be a biological substrate for learning fects of AA and ACPD on inositol phospholipid and memory. Further support of this nature is pre- metabolism by training in the water maze is similar sented here, because we have found that the syn- to the occlusion observed following induction of ergism between AA and ACPD on glutamate re- LTP (McGahon and Lynch 1996), lending indirect lease in dentate gyrus was occluded by prior in- support to the idea that the mechanisms underly- duction of LTP in this area (McGahon and Lynch ing spatial learning resemble those underlying 1996). maintenance of LTP. In an effort to establish the mechanisms un- Both AA and ACPD alone significantly en- derlying the AA/ACPD-induced increase in gluta- hanced PKC activity in synaptosomes prepared mate release, we examined the effect of these from hippocampus obtained from untrained rats, agents on inositol phospholipid metabolism, an in- and the combination of the two agents induced a crease that leads to formation of inositol trisphos- further increase in whole hippocampus as de- phate (IP3) and diacylglycerol. Both second mes- scribed previoulsy (McGahon and Lynch 1994). In sengers have the ability to increase transmitter re- parallel with the occlusion of the AA/ACPD syner- lease, by increasing intracellular calcium concen- gism on release and inositol phospholipid metab- tration and PKC activity, respectively. The data olism by training, the data indicate that the inter- presented indicate that training in the Morris wa- action between AA and ACPD on PKC is also oc- ter maze led to an increase in inositol phospho- cluded. Training also occluded the separate lipid metabolism in whole hippocampus. Changes actions of AA and ACPD on PKC. The close cou- in inositol phospholipid metabolism in behavioral pling in reponses of phospholipase C and PKC to paradigms have been reported previously; associa- AA and ACPD, both alone and in combination, sug- tive learning is accompanied by an increase in gests that PKC activation may be secondary to basal inositol phospholipid metabolism in the phospholipase C activation. AA has been shown to three subfields of the hippocampus (Laroche et al. activate PKC (Murakami and Routtenberg 1985), 1990), while training in an eight-arm radial maze is but the present results suggest that this may be accompanied by an increase in ibotenate-induced, secondary to an effect on inositol phospholipid but not basal, inositol phospholipid metabolism metabolism. (Nicoletti et al. 1986). We report here that both There is a good deal of evidence suggesting AA and ACPD increased inositol phospholipid me- that activation of PKC plays a role in spatial learn- tabolism, as described previously in the case of AA ing. Thus, physical activity that increases perfor- (Lynch et al. 1988; Lynch and Voss 1990) and as mance in a spatial learning task also increases PKC described in other brain areas in the case of ACPD activity (Fordyce et al. 1994). In addition, PKC (Schoepp et al. 1990; Pin and Duvoisin 1995). In activity in hippocampus of aged mice (C57) with addition to these responses, we found that AA and learning impairments was found to be reduced ACPD interacted to further increase inositol phos- compared with aged mice (F1) in which no age- pholipid metabolism, confirming our previous ob- related learning impairment was observed servation (McGahon and Lynch 1994). In parallel (Fordyce and Wehner 1993). It has also been with our release data, we found that the stimula- shown that performance in the Morris water maze tory effects of AA and ACPD, alone and in combi- was correlated positively with PKC substrate phos- nation, were occluded in synaptosomes prepared phorylation (Wong et al. 1989). It is of interest

L E A R N I N G & M E M O R Y 302 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press TRAINING OCCLUDES SYNERGISM BETWEEN ACPD AND AA that increased activation of PKC following induc- Baskys, A. 1992. Metabolic glutamate receptors and slow tion of LTP has been reported (Akers et al. 1986; excitatory actions of glutamate agonists in the hippocampus. Trends Neurosci. 15: 92-96. Angenstein et al. 1994; Pasinelli et al. 1995), as well as an increase in presynaptic PKC activity in area CA1 (Leahy et al. 1993). Although indirect, Bliss, T.V.P. and G.L. Collingridge. 1993. A synaptic model of memory: Long-term potentiation in the hippocampus. the parallel responses in PKC activity following a Nature 361 : 31-39. learning paradigm and induction of LTP, provide further support for the notion that LTP is a biolog- Bradford, M.M. 1976. A rapid and sensitive method for the ical substrate for learning and/or memory. quantitation of microgram quantities of proteins utilizing the One interpretation of the data presented here principle of protein dye binding. Anal. Biochern. is that spatial learning induces changes in inositol 72" 248-254. phospholipid metabolism leading to PKC activation, and that these changes, in turn, lead to an Canevari, L., G. Richter-Levin, and T.V.P. Bliss. 1994. LTP increase in glutamate release. The inference is that in the dentate gyrus is associated with a persistent NMDA receptor-dependent enhancement of synaptosomal glutamate this sequence of events will occur as a conse- release. Brain Res. 667:115-117. quence of mGluR activation in the presence of AA. However, it is possible that training modifies re- Clements, M.P., T.V.P. Bliss, and M.A. Lynch. 1991. lease of other transmitters and that these may im- Increase in arachidonic acid in a postsynaptic membrane pact on glutamate release. fraction following the induction of long-term potentiation in Our proposal is that learning induces changes the dentate gyrus. Neuroscience 45: 379-389. presynaptically that lead to increased glutamate release and therefore a strengthening of certain Fordyce, D.E. and J.M. Wehner. 1993. Effects of aging on spatial learning and hippocampal protein kinase C in mice. synapses. Our data suggest that the increase in re- Neurobiol. Aging 14:309-317. lease may derive from increased inositol phospholipid metabolism and thence increased PKC acti- Fordyce, D.E., R.V. Bhat, J.M. Baraban, and J.M. Wehner. vation. Because the effects of AA and ACPD, alone 1994. Genetics and activity-dependent regulation of Zif268 and together, are occluded in synaptosomes pre- expression: Association with spatial learning. Hippocampus pared from trained animals, it is tempting to con- 4: 559-568. clude that an interaction between these compounds lies at the heart of the changes induced by Haan, E.A. and D.M. Bowen. 1981. Protection of neocortical prisms from freeze-thaw injury by training and therefore plays a pivotal role in the dimethylsulphhoxide. J. Neurochem. 37: 243-246. process of synaptic strengthening. Herrero, I., M.T. Miras-Portugal, and J. Sanchez-Prieto. 1992. Positive feedback of glutamate exocytosis by metabotropic presynaptic receptor stimulation. Nature 360:163-166. Acknowledgments We wish to acknowledge the generous financial support H61scher, C. and S.P.R. Rose. 1994. Inhibitors of of The Health Research, Ireland, and The Sandoz Foundation phospholipase A 2 produce amnesia for a passive avoidance for Gerontological Research. task in the chick. Behav. Neural Biol. 61: 225-232.

H61scher, C., L. Canevari, and G. Richter-Levin. 1995. References Inhibitors of PLA2 and NO synthase cooperate in producing Akers, R.F., D.M. Lovinger, P.A. Colley, D.J. Linden, and A. amnesia of a spatial task. NeuroReport 6" 730-732. Routtenberg. 1986. Translocation of protein kinase C may mediate hippocampal long-term potentiation. Science Laroche, S., M.L. Errington, M.A. Lynch, and T.V.P. Bliss. 231 : 587-589. 1987. Increase in [3H]-glutamate release from slices of dentate gyrus and hippocampus following classical Angenstein, F., G. Riedel, K. Reymann, and S. Staak. 1994. conditioning in the rat. Behav. Biochern. Res. 25- 23-29. Hippocampal long-term potentiation in vivo induces translocation of protein kinase C. NeuroReport 5: 381-384. Laroche, S., C. Redini Del Negro, M.P. Clements, and M.A. Lynch. 1990. Long-term activation of phosphoinositide Barnes, C.A. 1979. Memory deficits associated with turnover associated with increased release in amino acids in senescence: A neurophysiological and behavioural study in the dentate gyrus and hippocampus following classical the rat. J. Comp. Physiol. Psychol. 93: 74-104. conditioning in the rat. Eur. J. Neurosci. 2" 534-543.

L E A R N / N G & M E M O R Y 303 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press McGahon et aL

Leahy, J.C., Y. Luo, C.S. Kent, K.F. Meiri, and M.L. Vallano. O'Keefe, J. 1991. An allocentric spatial model for the 1993. Demonstration of presynaptic protein kinase C hippocampal cognitive map. Hippocampus 1: 230-235. activation following long-term potentiation in rat hippocampal slices. Neuroscience 52" 563-574 9 Parfitt, K.D. and D.V. Madison. 1993. Phorbol esters enhance synaptic transmission by a presynaptic, Lynch, M.A. and K.L. Voss. 1990. Arachidonic acid calcium-dependent mechanism in rat hippocampus. I. increases inositol phospholipid metabolism and glutamate Physiol. 471: 245-268 9 release in synaptosomes prepared from hippocampal tissue. J. Neurochem. 55: 215-221 9 Pasinelli, P., G.M.J. Ramakers, I.J.A. Urban, J.J.H. Hens, A.B. Oestreicher, P.N.E. deGrann, and W.H. Gispen. 1995. --. 1991. Presynaptic changes in long-term potentiation: Long-term potentiation and synaptic protein phosphorylation. Elevated synaptosomal calcium concentration and basal Behav. Brain Res. 66" 53-59. phosphoinositide turnover in dentate gyrus. J. Neurochem. 56:113-118. Pin, J.-P. and R. Duvoisin. 1995. The metabotropic glutamate receptors: Structure and functions 9 91994. Membrane arachidonic acid concentration Neuropharmacology 34: 1-26. correlates with age and induction of long-term potentiation in the dentate gyrus of the rat. Eur. J. Neurosci. 6: 1008-1014 9 Richter-Levin, G., M.L. Errington, H. Maegawa, and T.V.P. Bliss 91994. Activation of metabotropic glutamate receptors is Lynch, M.A., M.P. Clements, M.L. Errington, and T.V.P. necessary for long-term potentiation in the dentate gyrus and Bliss 91988. Increased hydrolysis of for spatial learning. Neuropharmacology 33: 853-857. phosphatidylinositol-4,5-bisphosphate in long-term potentiation 9Neurosci. Lett. 84: 291-296 9 Richter-Levin, G., L. Canevari, and T.V.P. Bliss 91995. Long-term potentiation and glutamate release in the dentate Lynch, M.A., M.L. Errington, and T.V.P. Bliss. 1989. The gyrus: Links to spatial learning 9Behav. Brain Res. 66: 37-40. increase in [3H]-glutamate release associated with long-term potentiation in the dentate gyrus is blocked by commissural Schoepp, D.D., J. Bockaert, and F. Sladeczek. 1990. stimulation 9Neurosci. Lett. 103: 191-196 9 Pharmacological and functional characteristics of metabotropic excitatory amino acid receptors. Trends McGahon, B. and M.A. Lynch 91994. A study of the Pharmacol. Sci. 11 : 508-515. synergism between metabotropic glutamate receptor activation and arachidonic acid in the rat hippocampus. Squire, L.R. 1992. Memory and the hippocampus: A NeuroReport 5: 2353-2357. synthesis from findings with rats, monkeys and humans. Psychol. Rev. 99: 195-231 9 91996. The synergism between metabotropic glutamate receptor activation and arachidonic acid on Terrian, D.M., D. Kirk-Ways, R.L. Gannon, and D.A. Zetts. glutamate release is occluded by induction of long-term 1993. Transduction of a protein kinase C-generated signal poteniation in dentate gyrus. Neuroscience 72: 847-855 9 into the long-lasting facilitation of glutamate release. Hippocampus 3: 205-220 9 McNaughton, B.L., C.A. Barnes, G. Rao, J. Baldwin, and M. Rasmussen. 1986. Long-term enhancement of hippocampal Williams, J.H., M.L. Errington, M.A. Lynch, and T.V.P. Bliss 9 synaptic transmission and the aquisition of spatial 1989. Arachidonic acid induces a long-term information. J. Neurosci. 6: 563-571. activity-dependent enhancement of synaptic transmission in the hippocampus. Nature 341: 739-742 9 Morris, R.G.M., E. Andersen, G. Lynch, and M. Baudry. Wong, K.L., K. Murakami, and A. Routtenberg. 1989. 1986. Selective impairment of learning and blockade of long-term potentiation by an N-methyI-D-aspartate receptor Dietary cis-fatty acids that increase protein F1 phosphorylation enhance spatial memory. Brain Res. antagonist, AP5. Nature 319: 774-776. 505: 302-305. Murakami, K. and A. Routtenberg. 1985. Direct stimulation of purified protein kinase C by unsaturated fatty acids (oleate, Received June 17, 1996; accepted in revised form August 22, arachidonate) in the absence of phospholipids and calcium 9 1996. FEBS Lett. 192:189-193.

Nicoletti, F., C. Valerio, C. Pellegrino, F. Drago, U. Scapagnini, and P.L. Canonica. 1986. Spatial learning potentiates the stimulation of phosphoinositide hydrolysis by excitatory amino acids in rat hippocampus. J. Neurochem. 51 : 725-729.

Nicholls, D.G. 1989. Release of glutamate, aspartate and aminobutyric acid from isolated nerve terminals 9J. Neurochem. 52: 331-341 9

L E A R N I N G & M E M O R Y 304 Downloaded from learnmem.cshlp.org on September 24, 2021 - Published by Cold Spring Harbor Laboratory Press

Training in the Morris water maze occludes the synergism between ACPD and arachidonic acid on glutamate release in synaptosomes prepared from rat hippocampus.

B McGahon, C Holscher, L McGlinchey, et al.

Learn. Mem. 1996, 3: Access the most recent version at doi:10.1101/lm.3.4.296

References This article cites 40 articles, 2 of which can be accessed free at: http://learnmem.cshlp.org/content/3/4/296.full.html#ref-list-1

License

Email Alerting Receive free email alerts when new articles cite this article - sign up in the box at the Service top right corner of the article or click here.