commissural synapses

Neuroscience Letters, 100 (1989) 141-146

141

Elsevier Scientific Publishers Ireland Ltd. NSL 06049

Differential effects of phospholipase inhibitors in long-term potentiation in the rat hippocampal mossy fiber synapses and Schaffer/commissural synapses Daisuke Okada, Shunichi Yamagishi and Hiroyuki Sugiyama National Institutefor PhysiologicalSciences, Department of Cellular Physiology, Okazaki (Japan)

(Received 29 December 1988; Revised version received 14 January 1989; Accepted 17 January 1989) Key words: Long-term potentiation; Hippocampus; Phospholipase A2; Phospholipase C; Arachidonate;

Inositol phospholipid Bath application of the inhibitors of phospholipases, nordihydroguaiaretic acid (NDGA) and p-bromophenacyl bromide (BPB), to the rat hippocampal slices suppressed long-term potentiation (LTP) in Schaffer/commissural-CAl pyramidal synapses. On the other hand, neither of the two inhibitors suppressed LTP in mossy fiber-CA3 pyramidal cell synapses. BPB did not suppress phosphatidylinositol-specific phospholipase C (PI-PLC) activity of the slices. These results suggested that the mechanisms of LTP were quite different in the CAI and CA3 subfields of rat hippocampus: in CA1, the involvement of an arachidonate metabolism was strongly suggested, whereas in CA3, an arachidonic acid cascade may not be necessary for LTP.

Recent studies on long-term potentiation (LTP) in the h i p p o c a m p u s have revealed that m a n y aspects o f L T P in the C A 1 (Schaffer collateral/commissural f i b e ~ C A 1 pyramidal cell synapses) and CA3 (mossy fiber-CA3 pyramidal cell synapses) subfields differ from each other. Firstly, the receptors involved in L T P are different. The activation o f N-methyi-D-aspartate ( N M D A ) type glutamate receptor is required for L T P in C A I [1 1], whereas an N M D A receptor antagonist, 2 - a m i n o - 5 - p h o s p h o n o v a lerate, did not suppress L T P in the CA3 subfield [5]. Secondly, both pre- and postsynaptic mechanisms have been observed in CA1 LTP, but no postsynaptic changes during L T P have been reported in CA3 subfield [2]. In C A 1 LTP, depolarization o f a postsynaptic cell is required [4, 5], and the enhanced transmitter release from a presynaptic cell was reported [12]. Some investigators postulated that hydroperoxyeicosatetetraenoate ( H P E T E ) , one o f the intermediate o f the arachidonate cascade, was a retrograde messenger necessary for the induction o f L T P in CA1 [15]. Thirdly, we Correspondence: H. Sugiyama, Department of Cellular Physiology, National Institute for Physiological

Sciences, Myodaiji, Okazaki 444, Japan. 0304-3940/89/$ 03.50 © 1989 Elsevier Scientific Publishers Ireland Ltd.

142

recently reported that pertussis toxin suppressed LTP in CA3 but not in CA I showing the involvement of a certain pertussis toxin sensitive G protein in kTP in CA3 subfield [8]. Williams and Bliss [15] reported that Ca 2 +-induced LTP in CA I and dentate gyrus were suppressed by high concentration of nordihydroguaiaretic acid (NDGA), an inhibitor of phospholipase A2 which therefore could have suppressed the production of HPETEs. The effect of N D G A on LTP in CA3 had not been reported. The present study is performed to make clear the involvement of arachidonate cascade in LTP in CA3 subfield by the use of the following inhibitors of enzymes functioning in the cascade: N D G A which may inhibit at high concentrations both lipoxygenase and phospholipase A2 [3], and p-bromophenacyl bromide (BPB) which may inhibit phospholipase A2 and phosphatidylinositol-specific phospholipase C (PI-PLC) [7]. Rat hippocampal slices (400/~m thick) were prepared and maintained at 32°C as reported previously [8]. Field potentials were measured to evaluate tetanic stimulation-induced LTP in CAi and CA3 subfields (see Methods in the legend to Fig. 1). Fig. 1 shows the effects of N D G A and BPB on LTP in CA1 subfield. In the presence of 100 pM N D G A , CAI population excitatory postsynaptic potential (pEPSP) showed a slight and reversible decrease (by about 20% of the pre-tetanic control; Fig. IC) as described in ref. 15, whereas BPB (up to 25/~M) did not alter CAl pEPSP as shown in Fig. 1D. Tetanic stimulations induced LTP in control slices (Figs. 1A,B and 3) but no LTP in 100/~M N D G A (Figs. IC and 3) or 25/tM BPB (Figs. 1A, D and 3). In some experiments with N D G A , tetanic stimulation strength was slightly increased so that it evoked the same amplitude of pEPSP as before N D G A application, and the results were similar (no LTP). When N D G A was washed out for 30 min, the second tetanic stimulation elicited an intermediate extent of potentiation (44_+ 10% in amplitude and 38_+7% in slope, n=6). On the other hand, once slices were perfused with BPB, the second tetanic stimulation given after 30 rain of BPB wash-out did not potentiate the pEPSP. Thus the inhibitory effect of BPB was irreversible. Fig. 3 summarizes the results of the repeated experiments indicating that 100/~M N D G A and 25/~M BPB clearly suppressed LTP in CAI subfield. When a lower concentration of BPB (10/~M) was used, an intermediate extent of LTP was observed as shown in Fig. 3. Thus the effect of BPB on LTP was dose-dependent in the CA I subfield. In Fig. 2, the effects of the two inhibitors in CA3 subfield are shown. In contrast to CA l, pEPSP in CA3 subfield was not affected by N D G A , and was slightly enhanced by BPB (mean enhancement _+S.E.M., 13 4- 6% for amplitude, 12 -+ 9% for initial slope; Fig. 2D). LTP was observed to a similar extent to the controls (Figs. 2A, B and 3) when tetanic stimulation was given in the presence of 100/tM N D G A (Figs. 2C and 3) or 25/IM BPB (Figs. 2A,D and 3). As summarized in Fig. 3, LTP in mossy fiber synapses was not suppressed by either 25/iM BPB or 100/tM NDGA. A preliminary experiment showed that it was not suppressed even by 250/~M BPB (data not shown). The mean extent of the potentiation with BPB in CA3 was a little greater than the control values as shown in Fig. 3. This may be explained by the enhancing effect of BPB on pEPSP. When potentiation in BPB was evaluated by

143

C

~;200 A

CONTROL

t

Z 0

BPB

100

_1

"

I-Z bLl I--" O 13-

0 NDGA

-3'o -2'0 -1.0 T[

B

ME

;

1'o

(min

)

2.0

3'0

D

~3001

1=~ i

~200

200

t

o 100

100 Z ta

0 BPB

0 !

i

-10

i

i

i

i

0

10

20

30

T IME (mi n)

Q-

a

-30

i

-20

I

-10 TIME

I

0 (m

I

10

I

I

20

30

i n )

Fig. 1. The effects of NDGA and BPB on LTP in CAI subfield. The pEPSP was evoked by an orthodromic stimulus on Schaffer/commissural fibers with a tungsten bipolar electrode (80 gsec duration) and recorded from the dendritic field of CA1 pyramidal cell (stratum radiatum) with a glass microelectrode (3 8 Mohm). A: the pEPSPs in pre-tetanic control period (1, see Methods), and 30 min after tetanic stimulation (2) are shown for control and 25/zM BPB-applied slices. The arrowhead represents the time of the stimulation. Calibration: the horizontal bar represents 2 ms and the vertical one 1 mV. Time course of the LTP of the control (B), 100/~M NDGA-applied (C), and 25/tM BPB-applied (D) slices expressed by the potentiation (see Methods) of pEPSP amplitude (open circle) and initial slope (filled circle). The bar indicates the application of the inhibitors. The arrow indicates tetanic stimulation. The abscissa represents the time after tetanic stimulation in min. Methods. The amplitude and the initial slope of pEPSP were recorded at each 20 s and the data were digitized with an Autonics $210 digitizer, stored in a floppy disk and analysed off line with a microcomputer. For NDGA, 5 consecutive pEPSPs (20 s intervals) were recorded and averaged at each 5 min. Before application of the inhibitors, a stable 15 min of control sampling was made, averaged, and defined as the pre-tetanic control pEPSP. Tetanic stimulation was given at 100 Hz for 1 s at the same strength as in pre-tetanic control sampling. Mean values of pEPSP amplitudes and initial slopes between 25 min and 30 min after tetanic stimulation were divided by those of the pre-tetanic control. The potentiation represents the percent increase of this ratio. Either 100 mM NDGA (Sigma) or 25 mM BPB (Sigma) was dissolved in dimethylsulfoxide on each day and added to Krebs solution [8] to a final concentration immediately before use. To avoid light-induced radical formation, BPB was dealt under dim light. PI-PLC activity was assayed by scintillation counting of [3H]inositol produced by hydrolysis of [3H]phosphatidylinositol (5000 cpm/sample, 16.6 Ci/mmol). Lipid vesicles composed of radioactive and cold phosphatidylinositol (10/1g/sample) suspended in Hepes buffer (I00/tl/sample, 50 mM, pH 7.3 containing 100 mM NaCI and 1 mM CaC12) was mixed with 10-50 gg of soluble protein (100/~l/sample) obtained by homogenization and centrifugation of the hippocampal slices. The reaction was performed at 3T'C for 10 min and the assay was linear with respect to protein amounts less than 50/~g.

144

C ,4 CONTROL

BPB

~200 z uJ 0 a_

0

NDGA

-3'0 -2'0

10 0 TIME ( m i

D

B

~

20

30

j , ~ ~*-~, ~ ", \

~30(3_.

~2°° F

10 n )

~2oo

Y. ~oo

.... ~

;. loo

[11 i

[li

a -10

0 r | M E

10 20 ( m i n )

30

BPg -30

20

~-10 I ] M [

0 10 ( m i n )

20

30

Fig. 2. The effects of BPB on LTP in CA3 subfield. Granular cell layer of dentate gyrus was stimulated and pEPSP was recorded from CA3 stratum lucidum. A: the pEPSPs in pre-tetanic control period (1), and 30 min after tetanic stimulation (2) recorded from control and 25/iM BPB-applied slices are shown. Calibration: the horizontal bar denotes 2 ms and the vertical one 0.5 inV. Time course of the LTP of the control (B), 100/tM NDGA-applied (C), and 25/~M BPB-applied (D) slices are shown in the same manner as in Fig. l, c o m p a r i n g with p E P S P at 15 min after application o f BPB, the mean potentiation ± S.E.M. was 7 0 + 8 % for amplitude and 101 _+25% for slope. Thus, N D G A and BPB suppressed L T P in CA1 but not in CA3. N D G A is reported to inhibit phospholipase A2 and lipoxygenase [3, 7, 15]. BPB is reported to inhibit both phospholipase A2 and P I - P L C by covalent modification o f the enzymes [7, 14]. Therefore we measured the P I - P L C activity o f BPB-applied slices by the method o f H o f m a n n and Majerus [6] to know which enzyme or cascade is blocked by BPB. P I - P L C activity observed in the h o m o g e n a t e o f the hippocampal slices after treatment with 2 5 / t M BPB for 30 min was c o m p a r e d with that o f control slices. The BPB-treated slices showed the same P I - P L C activity as control slices did ( 5 . 4 + 1.3 and 5.7_+0.9 /~g PI hydrolyzed/min/mg protein (mean _+ S.E.M., n = 4 ) for BPBtreated and control slices, respectively). Even when BPB concentration was increased to 250/~M, essentially the same activity was observed (84% o f the control). Thus, hippocampal P I - P L C activity was not suppressed by bath-applied BPB, suggesting that the inhibitory effect o f BPB on L T P in CA1 subfield was due to its blockade o f phosp h o l i p a s e A 2. These results suggested that both N D G A and BPB suppressed L T P in the hippocampal C A I subfield by inhibiting phospholipase A2 activity, and neither inhibitor

145

CA3

CA1

I

I

CONTROL 100gM NDGA 2511M BPB

t,,i

IOBM

I ~

BPB I

I

100

,

I

,

50

I

0

POTENTIATION

i

0

I

I

50

I00

(%)

Fig. 3. The differential effects of NDGA and BPB on LTP in the CA1 and CA3 subfields. The bars show the S.E.M. The open and hatched columns represent the percent increases above the pre-tetanic controls in pEPSP amplitude and initial slope, respectively. In CA3, n= 17 (control), 7 (NDGA) and 9 (BPB). In CA1, n = 11 (control), 13 (NDGA), 8 (10 pM BPB) and 9 (25 pM BPB).

inhibited mossy fiber LTP in the CA3 subfield. It is likely that LTP in CA1, but not in CA3, requires the activation of arachidonate cascade. If phospholipid metabolism is involved in mossy fiber LTP, our results described above suggest that it would be turnover reactions other than phopholipase A2 reaction, most likely phospholipase C-mediated PI turnover [1, 9]. One of the candidates of neurotransmitter receptors responsible for such reactions is the metabotropic type of glutamate receptors [13], which activates phospholipase C through pertussis toxinsensitive G protein in Xenopus oocytes [13], and in fact, CA3 LTP is suppressed by pertussis toxin [8]. 1 Bar, P.R., Wiegant, F,, Lopes da Silva, F.H. and Gispen, W.H., Tetanic stimulation affects the metabolism of phosphoinositides in hippocampal slices, Brain Res., 321 (1984) 381 385. 2 Barrionuevo, G., Kelso, S.R., Johnston, D. and Brown, T.H., Conductance mechanism responsible for long-term potentiation in monosynaptic and isolated excitatory synaptic inputs to hippocampus, J. Neurophysiol., 55 (1986) 54(F550. 3 Billah, M.M., Bryant, R.W. and Siegel, M.I., Lipoxygenase products of arachidonic acid modulate biosynthesis of platelet activating factor (I-O-alkyl-2-acetyl-sn-glycero-3-phosphocholine) by human neutrophils via phospholipase A2, J. Biol. Chem., 260 (1985) 6899~906. 4 Gustafsson, B., Wigstrom, H., Abraham, W.C. and Huang, Y.Y., Long-term potentiation in the hippocampus using depolarizing currer~ pulses as the conditioning stimulus to single volley synaptic potentials, J. Neurosci., 7 (1987) 774-780. 5 Harris, E.W. and Cotman, C.W., Long-term potentiation of guinea pig mossy fiber responses is not blocked by N-methyl-t)-aspartate antagonists, Neurosci. Lett., 70 (1986) 132-137. 6 Hofmann, S.L. and Majerus, P.W., Modulation of phosphatidylinositol-specific phospholipase C activity by phospholipid interactions, diglycerides, and calcium ions, J. Biol. Chem., 257 (1982) 14359 14364. 7 Hofmann, S.A., Prescott, S.M. and Majerus, P.W., The effect of mepacrine and p-bromophenacyl bromide on arachidonic acid release in human platelets, Arch. Biochem. Biophys., 215 (1982) 237-244.

146

9 10 11 12 13 14 15

Ito, I., Okada. D. and Sugiyama, H., Pertussis toxin suppresses long-term potentiation in the rat hippocampal mossy tiber synapses, Neurosci. Left., 90 (1988) 181 185. Lynch, M.A., Clements, M.P., Errington, M.L. and Bliss, T.V.P., Increased hydrolysis of phosphatidylinositol-4,5-bisphosphate in long-term potentiation, Neurosci. Lett., 84 (1988) 291 296~ Malinow, R. and Miller, J.P., Postsynaptic hyperpolarization during conditioning reversibly blocks induction of long-term potentiation, Nature (Lond.), 320 (1986) 529 530. Nicoll, R.A., Kauer, J.A. and Malenka, R.C., The current excitement in long-term potentiation, Neuron, 1 (1988)97 103. Skrede, K.K. and Malthe-Sorenssen, D., Increased resting and evoked release of transmitter following repetitive electrical tetanization in hippocampus, Brain Res., 208 (1981) 436 441. Sugiyama, H., lto, I. and Hirono, C., A new type of glutamate receptor linked to inositol phospholipids metabolism, Nature (Lond.), 325 (1987) 531 533. Volwerk, J.J., Pieterson, W.A. and de Haas. G.H., Histidine at the active site of phospholipase A:, Biochemistry, 13 (1974) 1446 1454. Williams, J.H. and Bliss, T.V.P., Induction but not maintenance of calcium-induced long-term potentiation in dentate gyrus and area CA I of the hippocampal slices is blocked by nordihydroguaiaretic acid, Neurosci. Lett., 88 (1988) 81-85.