Modulation of paired-pulse activation in the hippocampal dentate gyrus by cholecystokinin, baclofen and bicuculline

Neuropeptides (1993) 24,263-270 0 Longman GroupUK Ltd 1993

Modulation of Paired-pulse Activation in the Hippocampal Dentate Gyrus by Cholecystokinin, Baclofen and Bicuculline E. RICH-BENNETT, D. DAHL and B. B. LECOMPTE III The University of Texas at Dallas, Human Development, (Reprint requests to DD)

GR4.1, Dallas, Texas 75083-0888 USA

Abstract-Paired-pulse stimulation was used to evaluate the effects of the sulfated octapeptide of cholecystokinin (CCK&S), the y-aminobutyric acide (GABAe) agonist (-1 baclofen, and the GABA,, antagonist (-1 bicuculline on hippocampal dentate gyrus (DG) granule cell excitability. Evoked action potentials (EAPs) and excitatory postsynaptic potentials (EPSPs) were recorded in response to orthodromic stimulation of the medial (MPP) or lateral (LPP) perforant pathway. Paired-pulse indices were determined using interpulse intervals (IPls) across the range of E-1000 ms. As reported by others, three phases of pairedpulse effects were revealed under control (drug-free ACSF) conditions: early paired-pulse inhibition (PPI), intermediate paired-pulse facilitation (PPF) and late PPI. With EAPs, CCK&S enhanced only the intermediate PPF on both pathways, with no effect on the early or late PPls. The effects of (-) baclofen were similar to CCK8-S. (-) Bicuculline attenuated the early and late PPI as well as the PPF. No differences were measured on the MPP- or LPP-evoked EPSPs in any of the drug conditions. These results indicate a similarity of CCK8-S- with GABAB-mediated modulation on neuronal activation in the DG. CCK&S disinhibition of DG granule cells may play a role in the induction of long-lasting synaptic modifications.

Introduction Cholecystokinin (CCK) is an ubiquitous neuropeptide found throughout the mammalian central nervous system, and is colocalized with GAB A in many cortical and hippocampal neurons.15J3 Although CCK has an excitatory effect on hippocampal neurons,5J2 the mechanism underlying this action is

Date received 7 December 1992 Date accepted 10 December 1992

unlmown. CCK modulation of dopamine release in other areas of the CNS,zo26 and theories regarding the function of other neuropeptide/neurotransmitter colocalization,‘6 would suggest that CCK may modulate GABA release. One electrophysiological method for the analysis of neuronal excitation and inhibition is the pairedpulse technique. ~3 To evaluate the presumptive effects of CCK on GABA modulation of neuronal excitability, this technique was employed in conjunction with perfusion of CCK&S, the GABAB

263

264

NJ3JRoPEFmDEs

Fig. 1 Placement of stimulating and recording electrodes within the dentate gyms. Stimulatiug electrodes were placed in either the

LPP (Sl) or MPP (S2). Recording electrodes were placed in the molecular layer corresponding to the projection field of the indicated PP component (Rl and R2) to record evoked EPSPs and in the granule cell layer (R3) to record EAPs (adapted from 10).

agonist (-) baclofen, and the GABA* antagonist (-) bicuculline. As previously suggested, (-) baclofen, although a GABA* receptor agonist, may inhibit the release of GABA via GABAe autoreceptor activation.‘J3J4 If GABA release is also modulated by CCK8-S, then one would expect similar effects of CClWS and (-) baclofen on paired-pulse activation. Methods Hippocampal slices (375400 pm) were cut from the brains of male rata (Sprague-Dawley, 100-125 g). Brains were removed from decapitated heads by stripping the overlying skull and separating the brain from the surrounding meninges. Brains were then set in a Petri dish with ice-cold (5-7’C) ACSF, the cerebellum was removed and the brain hemisected by a mid-saggital cut. The thalamus and brainstem were removed using curved forceps and the hippocampal formation was reflected and dissected free from the attached entorhinal cortex. Slices were cut on a McIlwain tissue chopper from the mid-dorsal region of hippocampi that were carefully oriented to the blade to obtain maximum exposure of the DG molecular and cell layers. Brain slices were set on a nylon net in a modified Andersen-type chamber at the interface between circulating, buffered ACSF, and a warmed, humidified atmosphere consisting of 95% 02/5% COS. Buffered ACSF (pH = 7.2) had the following composition (inrnM): NaCl125, KC1 3.0, MgS04 1.3, CaCh 2.4, NaHC03 26, NaH2P04 1.25, and dextrose 10. Temperature of the slice chamber

was maintained at 3 1OC. Once on the net, the slices were left to equilibrate for 1 h with continuous ACSF perfusion before the experiment was started. Stimulation electrodes consisted of a 2-4” length of teflon-coated stainless steel or platinum-iridium wire (ca. 80 pm in diameter), mounted in a drawn glass capillary. Extracellular recording electrodes were drawn from glass micropipettes, filled with 2.0 M NaCl, and had tip resistances in the range of 5-10 MQ. The placement of stimulating and recording electrodes is outlined in Figure 1. Briefly, the stimulating electrode was placed in either the MPP or LPP, the entorhinal cortical afferents to the DG. A recording electrode was placed in the molecular layer responding to the respective perforant pathway projection field (to record EPSPs), and an additional electrode was placed in the granule cell layer A

Sl

I

I

I 410

Fig. 2

S2B

=A

SEC*

45-loo0

MS IPb

The paired-pulse paradigm. (A) A single reference stimulus (Sl) was followed 10 s later by paired stimuli (S2* and SUB) with IPIs in the range of 5 to 1000 ms. Au index of PPI or PPF was determined by the ratio S2$S 1. (B) PPI (15 ms IPI) of EAP and EPSP are illustrated. The solid lines are the response to S 1, broken lines the response to S2.

MODULATION

OF PAIRED-PULSE

ACTIVATION

INDG

BY CHOLECYSTOKININ,

(to record EAPs). Electrophysiological profiles were obtained by the method of Dahl and Sarvey9 in order to insure MPP or LPP separation. Responses were

BACLOFEN

AND BICUCULLINE

265

amplified using a DC amplifier with a high Erequency cut-off at 2 KHz. EPSPs were analyzed by measurement of the maximum initial negative slope

LiATERALlpERpoRANTPATH 1 /.&I CC!K8-S

A 4.5

T

*

4 -3.5 -3 --

t

2.6 --

s2lSl 1 -1.5 -1 --

--I 15

al

t

I 100

MEDIALsPERFORANTPATH 1 /.&I CCK8-S B

* t *

Fig. 3 Paired-pulse response series of EAPs in control (drug-free) and CCK perfusion conditions. EAPs are from orthdromic (A) LPP stimulation and (B) h4PP stimulation. CCK&S was without effect on the early (IPI < 40 ms) or late (IPI > 200 ms) PPIs, but significantly enhanced the PPF (IPI 4@-200 ms).

266

NEUROPEFTIDES

LATERAL PERFORANI’ PATH 1oonMBAcLomN

A

n

CONTROL

0 BACLQFEN

7

*

--

6 --

4

--

3 --

2 -1 --

15

60

B

n 6

6

4

!a313 2

1

0

! *

*

CONTROL

0 BACJLOF’EN

ld+td+L 15

40

Fig. 4 Paired-pulse response series of EAPs in control (drug-free) and (-) baclofen conditions. (-) EAPs are from orthodromic (A) LPP stimulation and (B) MPP stimulation, Baclofen was without effect on the early (IPI < 40 ms) or late (IPI > 200 ms) PPIs, but significantly enhanced PPF (IPI 40-200 ms).

MODULATION

OF PAIRED-PULSE

ACTIVATION

INDG BY CHOLECYSTOKINLN,

BACLOFBN AND BICUCULLINE

267

and EAPs by the maximum peak amplitude.9 Responses were measured and stored using a PCbased data acquisition system to be analyzed off-line. In the paired-pulse paradigm (Fig. 2), a control stimulus (S 1) was followed 10 s later by the paired stimuli (S2A and S2B) separated by IPIs in the range of 5-1000 ms. A total of 5 paired-pulse responses were averaged for each IPI. The ratio S2B/Sl was calculated to estimate the degree of PPI or PPF. These ratios were then used to plot a paired-pulse response series. After obtaining satisfactory, pathway-isolated responses, a 15 min, single stimulus baseline period was monitored, to insure stable neuronal responses. In addition, paired-pulse activation was applied at 10 ms IPI and 60 ms IPI, to ensure the induction of PPI and PPF, respectively. Additional evidence for pathway isolation was evident at 60 ms IPI. As previously reported,19 paired-pulse stimulation of the LPP reveals a PPF of the EPSP, whereas MPP stimulation reveals either no change or a PPI. After the baseline period, a control drug-tree paired-pulse response series was obtained. The slices were then perfused with a drug solution for at least 30 min in order to ensure adequate drug exposure. While drug perfusion continued, a second paired-pulse response series was obtained, then the experiment was terminated. A two-tailed Students t test was used to compare the control and drug conditions, with differences considered to be significant at p < 0.05. The drugs used in this series of experiments were (-) bicuculline (Sigma Chemicals, Inc.), (-) baclofen hydrochloride (Research Biochemicals, Inc.), and the sulfated octapeptide of cholecystokinin (CCKSS, BaChem Bioscience, Inc.).

CCKS-S, (-) baclofen, and (-) bicuculline are presented in Figures 3,4, & 5, respectively. For clarity of presentation, not all IPIs used are shown; specific IPIs (15,60,100,140,180,200, and 1000 ms IPIs) were selected as representative of the three phases of paired-pulse activation. Columns show the mean and vertical bars reflect the S.E.M. for both the drugfree control and drug-perfusion paired-pulse series. Asterisks indicate statistical significance at the p < 0.05 level. The effects of 1uM CCK8-S on paired-pulse activation of DG granule cells are shown in Figure 3. Paired-pulse indices in response to LPP stimulation are in the top panel, MPP stimulation in the bottom panel. In the presence of CCK8 S a &gnificam@; lt0.05, 5 df for LPP, 4 df for M&Pi enhancement of the paired-pulse indices at 60, 140, 180, and 200 ms IPI for LPP evoked responses, and at 60, 100, 140, 180, and 200 ms IPI for MPP evoked responses. CCK8-S appears to affect only the intermediate PPF, with no significant change for both the early and late PPIs. Similar results were obtained with 100 nM (-) baclofen, as shown in Figure 4: an enhancement of the intermediate PPF, with no significant effect on the early or late PPIs. Enhancement of the pairedpulse indices were significant 0, < 0.05,4 df) at 60, 100,140, and 280 180 ms IPI for both LPP and MPP evoked responses. Figure 5 illustrates the results with 500 nM to 1 pM to (-) bicuculline. As with the CCK8-S and (-) baclofen figures, the top panel (A) represents the LPP stimulation and the bottom panel (B) MPP stimulation. (-) Bicuculline significantly attenuated (at all IPIs shown) both the early and late PPIs as well as the intermediate PPF (p < 0.05,4 df) for both the LPP and MPP.

Results

Discussion

As described by others,4 paired-pulse activation revealed three phases of DG excitability: an early inhibition at 5-40 ms IPI, an intermediate facilitation at 40-200 ms IPI, and a late inhibition at 200-1000 ms IPI. With evoked EPSPs, although drug perfusion resulted in a general increase in the EPSP slope, the paired-pulse response series revealed no significant difference between control and drug conditions (not shown). The EAP paired-pulse response series for

The results with CCK-8S, when compared to the results with (-) baclofen, indicate a similar effect of CCK8-S- and GABAB-mediated modulation of postsynaptic responses in the DG. CCK8-S and (-) baclofen clearly enhance the PPF by GABAB or CCK receptor activation. In contrast, (-) bicuculline attenuated paired-pulse effects at all interpulse intervals shown, i.e., bothPPIs andthe PPF were reduced. These results indicate a possible complexity of CCK&S effects at both presynaptic and postsynap-

268

NEUROPEFTIDES

LATERALPERFJORANTPATH 500 nM - 1 pM BIcucuLLINE

A

4-

3.6 --

3 --

2.6 --

s2lsl 2 -1.6 --

1

0.6

0

1000

200

B

MEDIALPERFfORANTPATH 500x&f- 1 pMBICUCtJLLJNE

n

CONTROL

0 BICUCUUINE

* +

-

Fig. 5 Paired-pulse response series of EAPs in control and (-) bicuculline perfusion conditions. (-) EAPs are from orthodromic (A) LPP stimulation and(B) MPP stimulation. Bicuculline attenuated the early (IPI e 40 ms) and late (IPI > 200 ms) PPIs, and also attenuated the PPF (IPI 40-200 ma).

MODULATION

OF PAIRED-PULSE

ACTIVATION INDG BY CHOLECYSTOKININ,

tic sites. This activation could be presynaptic, postsynaptic, or a combination of the two. Evidence exists7,13J4for the possibility of a disinhibitory action of (-) baclofen via a GABAs autoreceptor modulation of GABA release. Our fmdings with (-) baclofen provide additional support for this hypothesis. The similarity of CCKS-S and baclofen effects indicates CCK8-S may also modulate GABA release. In other areas of the central nervous system, evidence exists for a presynaptic modulation ofdopamine by CCK8S,20J6 which provides indirect support for this hypothesis. Evidence for a possible postsynaptic action of CCK8-S comes from the finding that CCK8-S can decrease an outward K+ current.6n22This action of CCK8-S would increase granule cell excitability, resulting in an increase in the PPF. However, a postsynaptic increase in excitability should also affect the early and late PPIs, which was not seen with CCK8-S. Preliminary results with the GABAB antagonists phaclofen and 2-hydroxysaclofen (10 PM each, not shown), did reveal an effect on the early PPI and an enhancement of the PPF, although the PPF enhancement was not as closely correlated with the CCKS-S results as was (-) baclofen. The results with (-) bicuculline indicate that paired-pulse activation is modulated at least in part by GABA* receptor activation. The attenuation of the PPI with bicuculline reflects a postsynaptic reduction of inhibition. The mechanism underlying the attenuation of the PPF is not as clear. The intermediate PPF could be mediated directly by GABA* activation, and evidence exists for a GABA*, Cl-mediated depolarization of membrane potential in hippocampal neurons2,3 which could support this explanation. An alternative explanation could be that with the doses of bicuculline used further facilitation of the test response was prevented because the cells were already maximally activated. However the attenuating effect of (-) bicuculline was also seen at lower doses (250 nM, not shown). These and similarz5results in the DG, as well as previously reported results in hippocampal field CA31v with (-) bicuculline indicate that paired-pulse activation is modulated at least in part by GABA* receptor activation. Modulation of inhibitory activation in DG granule cells may be an underlying factor in many central nervous system functions. For example, the modulation of excitability may be a factor in deter-

BACLOFEN AND BICUCULLINE

269

mining high frequency stimulation effects on synaptic transmission. Disinhibition has also been shown to play a role in the induction of p-adrenergic longlasting potentiation.7 Both GABAA and GABAs as well as CCKLo have been shown receptors, 11~21~27 to play a modulatory role in the induction of longterm potentiation (LTP). Since LTP is a presumptive biological substrate for memory processing24 in the hippocampal formation, CCK is likely to be an important influence on the mechanisms of attention and memory that are dependent on the perforant pathway input to the DGs8 Acknowledgements We would like to thank Dr. Larry Caulker for his valuable comments. This work was supported by a grant from The Whitehall Foundation.

References 1. Adamec, R. E., McNaughton, B., Racine, R. and Livingston, K. E. (1981). Effects of diazepam on hippocampal excitability in the rat: action in the dentate area. Epilepsia. 22: 205-215. 2. Alger, B. E. and Nicoll, R. A. (1982). Pharmacological evidence for two kinds of GABA receptor on rat hippocampal pyramidal cells studied in vitro. J. Physiol. (Lond.). 328: 125-141. 3. Andersen, P., Dingledine, R., Gjentad, L., Langmoen, I. A. and Mosfeldt Laursen, A. (1980). Two different responses of hippocampal pyramidal cells to application of gammaamino butyric acid. J. Physiol. (Lond.). 305: 279-296. 4. Austin, K. B., Bronzino, J. D. and Morgane, P. J. (1989). Paired-pulse facilitation and inhibition in the dentate gyms is dependent on behavioral state. Exp. Brain Res. 77: 594-604. 5. Brooks, P. A. and Kelly, J. S. (1985). Cholecystokinin as a potent excitant of neurons of the dentate gyms of rats. Ann. N.Y. Acad. Sci. 448: 361-374. 6. Buckett, K. J. and Saint, D. A. (1989). Cholecystokinin modulates voltage dependent Kls+ currents in cultured rat hippocampal neurones. Neurosci. Lett. 107: 162-166. 7. Burgard, E. C. and Sarvey, J. M. (1991). Long-laatingpotentiation and epileptiform activity produced by GAB& receptor activation in the dentate gyrus of rat hippocampal slice. J. Neurosci. 1 l(5): 1198-1209. 8. Dahl, D. (1988). Central administration of cholecystokinin potentiates evoked potential amplitude in the hippocampal dentate gyms. Neuropept. 11: 147-15 1. 9. Dahl, D. and Sarvey, J. M. (1989). Norepinephrine induces pathway-specific long-lasting potentiation and depression in the hippocampal dentate gyms. Proc. Natl. Acad. Sci. USA. 86: 4776-4780. 10. Dahl, D., Rich-Bennett, E. and LeCompte III, B. B. (1992). The interaction of cholecystokinin and affzrent tetanization in the induction of synaptic modifications in field CA1 of the rat hippocampal slice. Sot. Neurosci. Abstr. 274.28. 11. Davies, C. H., Starkey, S. J., Pozza, M. F. and Collingridge, G. L. (1991). GABAe autoreceptors regulate the induction of LTP. Nature (Lond.). 349: 609-611.

270

NEUROPEPTIDES

12. Dodd, J. and Kelly, J. S. (1981). The actions of cholecys-

13.

14.

15.

16.

17.

18.

19.

20.

tokinin and related peptides on pyramidal neurones of the mammalian hippocampus. Brain Res. 205: 337-350. Harrison, N. L. (1990). Gn the presynaptic action ofbaclofen at inhibitory synapses between cultured rat hippocampal neurones. J. Physiol. (Land.). 422: 433446. Harrison, N. L., Lange, G. D. and Barker, J. L. (1988). (-) Baclofen activates presynaptic GABAB receptors on GABAergic inhibitory neurones from embryonic rat hippocampus. Neurosci. Lett. 85: 105-109. Hendry, S. H. C., Jones, E. G., De Felipe, J., Schmechel, D., Brandon, C. and Emson, P. C. (1984) Neuropeptide-containing neurons of the cerebral cortex are also GABAergic. hoc. Natl. Acad. Sci. USA. 81: 6526-6530. H&felt, T., Millhorn, D., Seroogy, K. et al. (1987). Coexistence of peptides with classical neurotransmitters. Experientia. 43: 768-780. Kehl, S. J. and McLemtan, H. (1983). Evidence for a bicuculline-insensitive long-lasting inhibition in the CA3 region of the rat hippocampal slice. Brain Res. 279: 278-281. Lemo, T. (1971). Patterns of activation in a monosynaptic cortical pathway: the perforant path input to the dentate area of the hippocampal formation. Exp. Brain Res. 12: 1845. McNaughton, B. L. (1980). Evidence fortwophysiologically distinct uerforant oathwavs to the fascia dentata. Brain Res. 199: l-is. Meyer, D. K. and Krause, J. (1983). Dopamiue modulates

21.

22.

23.

24.

25.

26.

27.

cholecystokinin release in neostiatum. Nature (Lond.). 301: 338-340. Mott,D.D. andLewis,D. V.(1991). Facilitationoftheinduction of long-term potentiation by GABAs receptors. Science (Wash. D. C.). 252: 1718-1720. Saint, D. A. and Bucket& K. J. (1991). Modulation of the transient potassium current in rat hippocampal neurones by cholecystokmin. Neuropept. 20: 15 1-157. Somogyi, P., Hodgson, A. J., Smith, A. D., Nunzi, M. G., Gorio, A. and Wu, J.-Y. (1984). Different populations of GABAergic neurons in the visual cortex and hippocampus of cat contain somatostatin- or cholecystokinin-immunoreactive material. J. Neurosci. 4(10): 2590-2603. Teyler, T. J. and DiScenna, P. (1984). Long-term potentiation as a candidate mnemonic device. Brain Res. Rev. 7: 15-28. TutT, L. P., Racine, R. J. and Adamec, R. (1983). The effects of kindling on GABA-mediated inhibition in the dentate gyms of the rat. I. Paired-pulse depression. Brain Res. 277: 79-90. Voigt, M., Wang, R. Y. and Westfall, T. C. (1986). Cholecystokinin octapeptides alter the release of endogenous dopamine from the rat nucleus accumbens in vitro. J. Pharmacol. Exp. Ther. 237: 147-153. Wigstrom, H. and Gustaffson, J. B. (1983). Facilitated induction of long-lasting potentiation during blockade of inhibition. Nature (Land.). 301: 603-605.