Neuroscience Letters, 134 (1992) 223-228 223 © 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 03.50

NSL 08308

Opioid peptides reduce synaptic transmission in the nucleus accumbens

X i a o r u Yuan* , Samuel M a d a m b a and G eo rg e R o b e r t Siggins

Department of Neuropharmacology, The Scripps Research Institute, La Jolla, CA (U.S.A.)

(Received 3 July 1991; Revised version received 20 September 1991; Accepted 30 September 1991)

Key words: Electrophysiology; Brain slice preparation; Opiate addiction; Intracellular recording; Synaptic potential; Membrane property

Behavioral studies implicate the nucleus accumbens (NAcc) as a brain area pivotal for the rewarding effects of opiates like heroine and morphine. Therefore, we studied the effect of a variety of opioids on membrane properties and responses to synaptic stimulation in a slice preparation of the NAcc using intracellular recording. Superfusion of opioid peptides did not affect the membrane potential or input resistance of NAcc neurons, but significantly reduced both depolarizing and hyperpolarizing synaptic potentials. Naloxone superfusion significantly reversed the depressant effects of the/~ and 8 receptor agonists (but not those of the x agonist) on synaptic transmission, suggesting involvement of opiate receptors. These results imply that the predominant effect of opiates in NAcc is a reduction of synaptic transmission.

Despite decades of research, the brain mechanisms re- sponsible for human addiction to drugs are still unknown. Recent behavioral studies in rats with local injection of opioids and opiate antagonists have shown that the NAcc may be a key area supporting self-admin- istration of heroin [7, 19]. Thus, opiate addiction may involve a direct action of the opiate in this brain region. Morphological studies have revealed a profuse network of opioid peptide-containing fibers in the nucleus accum- bens (NAcc) [20], as well as several different kinds of opiate-receptor subtypes [1, 6, 8, 12, 13, 18]. Further- more, this area may regulate extrapyramidal motor function and the motor activity associated with motiva- tion [10]. In vivo electrophysiological studies indicate that the major effect of opiates and opioid peptides in the NAcc is depression of neuronal excitability [2, 3, 9, 21], and that there may be more than one neuron type in this region [2]. To examine the mechanism(s) underlying this depressant effect, and possibly illuminate the cellular processes underlying opiate self-administration, we per- formed intracellular and pharmacological studies on rat NAcc neurons in a slice preparation. We now report that behaviorally relevant concentrations of opioid peptides, acting through opiate receptors, reduce both excitatory

"Current address: Department of Physiology, Nanjing Medical Col- lege, 140 Hanzhong Rd., Nanjing, Jiangsu 210029, People's Republic of China. Correspondence: G.R. Siggins, Department of Neuropharmacology, The Scripps Research Institute, 10666 N. Torrey Pines Rd., La Jolla, CA, U.S.A.

and inhibitory postsynaptic potentials in accumbens neurons, without apparent direct effect on resting electri- cal properties of the postsynaptic membranes.

Sprague-Dawley rats of 110-170 g weight were the source of the brain tissue. We cut transverse (coronal) slices of NAcc (350-400 pm thick) on a vibrating cutter (Vibroslice; Frederick Haer) and completely submerged and continuously superfused them with warm (30- 3 I°C), gassed (95% 02, 5% CO2) artificial cerebrospinal fluid (ACSF) at a constant rate (2-4 ml/min). The com- position of the ACSF was (mM): NaCI 130, KC1 3.5, NaH2PO 4 1.25, MgSO4.7H20 1.5, CaC12 2.0, NaHCO a 24, and glucose 10. We performed intracellular current- clamp recording from NAcc neurons via micropipettes filled usually (106 cells) with 3 M potassium acetate (impedence: 120-150 MO), or less often (13 cells) with 3 M KCI (60-90 MI2), led to an Axoclamp 2A amplifier (Axon Instruments). We took recordings from the NAcc at levels 2.2-0.7 mm from bregrna [11] and surrounding, but especially ventromedial, to the anterior commissure (see Fig. 1A). We elicited postsynaptic potentials (PSPs) and action potentials by stimulating the slice with a bi- polar tungsten electrode (0.1 mm tip diameter, 0.2 mm apart) placed ventral to the NAcc in fiber pathways seen emanating from the area of the olfactory tubercle (Fig. 1A). The resulting PSPs may result from direct activa- tion of tubercle-NAcc fibers, although activation of oth- er fibers within NAcc cannot be ruled out. Continuous DC recordings were stored on a polygraph, and high- speed data such as current-voltage and stimulus-re- sponse ( ' input-output ' ) curves were digitized, stored and

224

A f .

CX

~ * t , l

~ g m a 1.2 mm

B C > E

v

.,,._, t -

O Q_

"O

-40

-60

-80

-100

-120

-140 -1200

Control / * D-Pen

/ -800 -400 0 400 800 Current S t e p (pA)

D Control DAGO 11' DAGO + Nal 15'

,_l] L_ t I I -

RMP-85mV lrll1 , . . . . ,, , JlH

. . . . . I ' - - - I ' l l r

omv, LIII ' 30 soc

Fig. 1. A: usual recording and stimulus sites in the NAcc (stipled area). Most recordings (checkered area) were taken from the nucleus 'core' and most stimulation (double arrows) was in the area of the olfactory tubercle. CC, corpus callosum; AC, anterior commisure; Cd, caudate-putamen. Taken from Paxinos and Watson [11]. B: typical voltage response of a NAcc neuron to current injection. Note the strong inward (anomolous) rectifi- cation, evidenced by smaller potential changes with increasingly hyperpolarizing current steps and the time-dependent 'sag' in the most hyperpolar- ized responses. Calibration bars= 20 mV and 50 ms. KC1 micropipette. Resting membrane potential (RMP; dashed line)=-81 mV. C: lack of effect of the 5-selective agonist o-PEN enkephalin (1/tM) on current-voltage curves (measured at steady-state) of another accumbens cell. Note also the slight inward rectification in this cell as shown in the non-linearity of the curve in the hyperpolarizing range. Arrow indicates RMP. Potas- sium acetate pipette. D: polygraph record of membrane potential in another NAcc neuron showing lack of effect on RMP of superfusion of the /l-selective agonist DAGO enkephalin (1 /~M). Upward deflections early in each panel are PSP responses to pathway stimulation; later deflections are voltage responses and spikes (attenuated by the polygraph) of current voltage curve analyses. The current voltage relationship was also not

altered by DAGO enkephalin. Nal, naloxone l/zM. Potassium acetate-filled micropipette.

ana lyzed on an AT- type compu te r using the P -C la mp

(Axon Ins t ruments ) p rograms . We added opio ids and

na loxone to the superfusate in k n o w n concen t ra t ions for

8-25 min. Differences between t r ea tment condi t ions

were tested for s tat is t ical significance by one- fac tor

A N O V A for repeated measures , with pos t -hoc analysis

(Fisher) .

We s tudied a to ta l o f 119 neurons , loca ted within the

NAcc as indicated in Fig. 1A, at depths within the slice

o f 14-324 /zm (average: 116 /ira). These neurons dis-

p layed large resting m e m b r a n e potent ia l s (RMPs) aver-

aging - 8 1 4- 1 mV ( m e a n + S . E . M . ; ranging f rom - 6 8

to - 9 4 mV; n = 119) and evoked spikes averaging 99 mV

(range: 80-120 mV; n = 8 1 ) . Input resistance was 5 8 + 5

Mr2 (mean + S.E.M.; n = 75). N o n e o f the cells was spon-

taneous ly active. In cu r r en t -vo l t age curves, many o f

these neurons d isp layed cons iderable a n o m a l o u s rectifi-

ca t ion [16, 17], result ing in a t ime-dependen t posi t ive-

going t ra jec tory ( ' sag ' ) o f the e lect rotonic po ten t ia l

genera ted by hyperpo la r iz ing current (Fig. 1B, lower 3

traces). Other cells showed a more l inear cu r r en t -vo l t age

curve. In addi t ion , many NAcc neurons d isp layed a

t ime-dependent ' r a m p i n g ' response to depolar iz ing cur-

rent steps (Yuan, M a d a m b a , Deisz, Zieglg/ insberger and

Siggins, in prepara t ion) , resembling the effect of a K +

conduc tance te rmed the D-cur ren t [l 5].

We tested op io id agonis ts in 85 cells. The op io ids tested were the 5-selective agonis t [D-Pen2,5]-enkephalin

(D-PEN), the/z-selective agonis t [D-Ala 2 ,NMe-Phea ,Gly-

ol ] -enkephal in ( D A G O ) , and the K-selective agonis t U-

50,488h (U-50). Superfusion of D-PEN, DAGO or U-50 (all at 1 /tM) for 8-20 min had no significant effect on the neuronal membrane potential or input resistance in these ceils (Fig. 1C,D). In fact, the current-voltage curves obtained during opiate superfusion were usually superimposable with control curves, even in the anoma- lous rectifying portion of the hyperpolarizing range (Fig. 1C). In a group of 11 cells studied with KC1 pipettes to avoid artifacts of high electrode resistance (as with potassium acetate), the opiates altered neither the trajec- tory of the 'sag' due to anomalous rectification (see above), the 'ramping' of the presumptive D-current, nor the spike amplitudes. In addition, the opiates did not consistently alter the number of spikes triggered by depolarizing steps or the afterhyperpolarizations (AHPs) following these steps (Yuan et al., in preparation).

225

By contrast, the opioid peptides markedly depressed postsynaptic potentials (PSPs) evoked by electrical stim- ulation of the sites shown in Fig. 1A. In most cells, such stimulation produced only depolarizing PSPs, even when potassium acetate-filled micropipettes were used and the membrane was depolarized by current to enhance or re- veal possible hyperpolarizing IPSPs. The depolarizing PSPs were generated by a range of stimulus strengths to obtain 'input-output' curves (Fig. 2B-D). All three of the opioids tested at 1 pM reduced the PSPs over the whole range of stimulus strengths, although in some individual cells the responses to lower intensity stimuli seemed more affected (see Fig. 2A). PSP amplitudes were averaged at stimuli for threshold, half-maximal and maximal (just below spiking) responses across all cells tested for each peptide (Fig. 2D). On average, DAGO

A Control t k

12 V ~

9 v I

B

> E

v 6 0)

"13

Q . E 4 < (3_ 09 13.. 2

40

D A G O 1 7' Na lox + D A G O 1 8'

t T 20 mV ]

C D 0ms 20

~ Control

{)-Pen ICt 15

10

5

60 80 100

Stimulus Intensity (V)

~3- - ConlroI US0,488h 8'

~ # ~ U50 + Nalox T

4 - - - - 5 10 15

Stimulus Intensity (V)

120[ .,,

100

0 | m m

°tt' : "

80 - , , , i i ~ 60 *

~ 40

~, 2o

0 ~ D-Pen DAGO U-50

Fig. 2. A: the/~-specific agonist DAGO enkephalin (1 pM) reduces the amplitude of evoked PSPs in an accumbens neuron, with reversal of the effect by naloxone 1/~M (Nalox). Shown are responses to three different stimulus strengths (9, 12 and 15 V), applied ventrally to the accumbens (at time marked by arrows). Note that responses to the weaker stimuli are decreased most by DAGO in this cell. The resting membrane potential ( - 88 mV) was not significantly altered by DAGO. Potassium acetate electrode used. B,C: opioid and naloxone effects on depolarizing PSPs evoked by a range of stimulus strengths to give input-output curves for two representative cells. B: NAce cell (RMP = - 7 3 mV); PSPs are reduced by D-PEN enkephalin (I /iM, 10 min superfusion), with reversal of this effect by naloxone 1 /~M (28 min). C: another cell (RMP= - 8 3 mV): PSPs are reduced by 1 pM U-50,488h superfused for 8 rain, and this effect is reversed by 1/IM naloxone (7 min). Potassium acetate electrodes used in both these cells. D: PSP data taken from all NAce cells tested, averaged for responses at threshold (solid columns), half-maximal (white columns) and maximal (hatched columns) stimuli and plotted as percent of the averaged control (dashed line). Asterisks indicate statistical significance at P < 0.001 (ANOVA, Fisher post-hoc test). Short bars above each column indicate average PSPs for each stimulus during naloxone (I /~M) plus

opioid agonist superfusion, n = 16 for D-PEN, 18 for DAGO and 17 for U50.

226

was the most effective in reducing PSPs at all 3 stimulus levels (Fig. 2D). DAGO (n = 18) reduced threshold PSPs by 47%, half-maximal PSPs by 35% and maximal PSPs by 25%. PSP differences elicited by all three opioids at 1 /~M, except the D-PEN threshold effect and the U50 half- maximal effect, were highly significant by ANOVA and the Fisher PLSD comparison test (Fig. 2D).

We superfused naloxone (1 /~M) together with the opioid agonist to test for reversibility by a specific opiate antagonist (Fig. 2D). Naloxone significantly reversed the D-PEN effects on PSPs averaged for half-maximal and maximal stimuli and the DAGO effects on threshold and half-maximal responses. However, naloxone did not sig- nificantly reverse the U50 effect on average PSP re- sponses at any stimulus strength, although some individ- ual cells did show partial naloxone antagonism (Fig. 2C). This relative inability of naloxone (1/~M) to reverse U50 effects may indicate an action of U50 not involving opiate receptors, or a relative ineffectiveness of naloxone in blocking x receptors. Further tests using other antag- onists or higher naloxone concentrations will be required to distinguish between these possibilities.

Included in the average PSP data are 3 cells that showed opioid-induced increases of the depolarizing PSPs (2 cells with D-PEN, 1 with D A G O ) . In contrast to the opioid-induced PSP decreases, these augmenting effects were not significantly reversed by naloxone. In addition, 5 neurons showed clear hyperpolarizing PSPs (probable IPSPs) when potassium acetate pipettes were used or the cells were depolarized by intracellular cur- rent injection. The opioids reduced the IPSPs (Fig. 3) in all 4 cells tested with the opioids (o-PEN in one cell and DAGO in three), with reversal of the effects by nalox- one.

With respect to the membrane properties of NAcc neurons, our results agree with those of Uchimura, et al. [17], who found pronounced inward or anomalous recti-

fication (a non-linearity in the current-voltage curve resulting from more ionic current passing through mem- brane channels at hyperpolarized than at resting mem- brane potentials). However, we also noted some cells showing more linear properties, suggesting the presence of more than one electrophysiological state or neuron type in the NAcc. We are currently using imaging meth- ods combined with electrophysiological characterization to assess these possibilities.

We conclude from our findings that the predominent effect of opioid agonists on neurons in NAcc is depres- sion of postsynaptic potentials, with little direct postsy- naptic effect on membrane potential or input resistance. We intentionally tested concentrations of the opioids at the high end of the range of behaviorally relevant con- centrations (0.5-1/~M; see ref. 24) to minimize the possi- bility that insufficient opioid would reach the appropri- ate receptors due to catabolism or problems of penetra- tion through the slice. It might be argued that the lack of opioid effect on membrane potential in our popula- tion of cells results from the proximity of the large rest- ing membrane potentials to the reversal potential for K + or CI- (two candidate ions for a hyperpolarizing effect); however, there were many cells with RMPs in the - 6 8 to - 7 5 mV range that also showed no opioid effect on resting potential. In addition, the lack of effect of the opioids on several measures of membrane properties (e.g., the anomalous rectifier, post-train AHPs), and over the large range of membrane potentials tested in the current-voltage curves, suggests a relative lack of effect on voltage-dependent conductances in this range. How- ever, more extensive, voltage-clamp studies (e.g., of Ca 2+ and Na + conductances) will be required to deter- mine if the opiates alter voltage-dependent conductances at more depolarized potentials.

The finding that a few cells showed an increase in PSPs (at some stimulus strengths) may suggest the presence of

Control D-Pen Enk 15' Wash 16'

5 mV 10 mV

RMP = -69 mV

50 V

RMP = -68 mV RMP = -68 mV

I

T T Fig. 3. IPSP reduction by the &-selective agonist D-PEN enkephalin (1 #M). Responses to two stimulus strengths (40 and 50 V, applied at arrows) are shown. Note the greater gain (calibration bar = 5 mV) in the upper row of records. Dotted line indicates resting membrane potential (RMP). This cell is slightly more depolarized than the average NAcc cell, allowing easy observation of the IPSP. Note the presence of an early and a late IPSP; both are reduced by D-PEN, suggesting a disinhibitory action. In a previous test on this cell (not shown), the IPSP reduction by D-PEN was

reversed by naloxone I/~M. Potassium acetate micropipette.

more than one opiate receptor subtype. It may be rele- vant that, whereas all three different types of opioids (more selective for /t, ~ or x receptors) most often reduced PSPs, the/t-selective agonist DAGO was the most effective at 1 pM. However, our preliminary dose- response studies indicate that both DAGO and D-PEN significantly reduce PSPs at 50--100 nM, whereas U50 is ineffective at 100--300 nM. These concentrations of DAGO and o-PEN are comparable to the 500 nM plasma concentrations obtained in behavioral studies of opiate analgesia, tolerance and dependence (see refs. 22, 24). Further studies will be needed to distinguish which receptor subtypes are involved in the PSP effect. The pronounced reduction of the IPSPs seen in a few cells is also of interest. These initial results suggest that opiates can have disinhibitory actions in NAcc, as first reported for the excitatory opiate effects seen in hippocampus [23]. This mechanism could account for the small but sig- nificant percentage of cells in NAcc shown to be excited by opiates in vivo [2].

As for the opioid-induced reduction of depolarizing PSPs, at least two possible mechanisms may be in play: (1) presynaptic modulation; that is, a presynaptic action causing decreased release of synaptic transmitter, as has been observed classically for opioids in gut and other autonomic systems [4, 5]; this idea is consistent with re- cent ultrastructural studies showing axoaxonic termina- tions of opioid peptide-containing elements in the NAcc [14]; (2) postsynaptic modulation [22]; that is, a reduc- tion of responses of specific receptors or ionophores for other transmitters (for example, glutamate). However, we cannot rule out the possibility of a direct postsynaptic membrane effect (e.g., a hyperpolarization or resistance decrease) at remote (e.g., dendritic) sites that could re- duce the PSPs indirectly. Studies with local application of glutamate are now in progress to test these possibili- ties.

Regardless of the exact mechanisms involved, these re- suits have important implications with respect to the cel- lular substrate of opiate addiction. Hence, hypotheses of NAcc involvement in the reinforcing properties of her- oin [7, 19] should take into account the likelihood that either exogenous or endogenous opioids will predomi- nantly reduce synaptic input into NAcc, without directly affecting postsynaptic excitability and intrinsic spike dis- charge. Thus, the combined electrophysiological, behav- ioral [7, 19] and anatomical [14] data suggest that rats will work to suppress synaptic transmission in their NAcc. Of course, the effects of systemic opiates may also involve actions on other brain regions projecting to NAcc [2, 3]. Clarification of the synaptic and intercellu- lar events occurring 'downstream' from this process also will require further study. It is thought that a major

227

afferent ouput from the NAcc projects to the substantia innominata (ventral pallidum). AS this pathway is con- sidered to be largely GABAergic, and therefore inhibi- tory, opioid inhibition of synaptic transmission in NAcc should constitute a form of disinhibition for s. innomi- nata neurons. Thus, we predict that endogenous or exo- genous opioids should increase the discharge of these ventral pallidal neurons.

We thank Drs. F. Bloom, S. Henriksen, G. Koob and M. Zeise for comments on the manuscript. The NAcc slice was developed in the laboratory of Drs. Deisz and Zieglg/insberger and supported by grants from ADAMHA (DA-03665, AA-06420, MH-44346 and MH-47680) and the Alexander-Von-Humboldt Stiftung.

1 Goodman, R.R. and Pasternak, G.W., Visualization ofmul opiate receptors in rat brain by using a computerized autoradiographic subtraction technique, Proc. Natl. Acad. Sci. U.S.A., 82 (1985) 6667~671.

2 Hakan, R.L. and Henriksen, S.J., Systemic opiate administration has heterogeneous effects on activity recorded from nucleus neu- rons in vivo, Neurosci. Lett., 83 (1987) 307-312.

3 Hakan, R.L. and Henriksen, S.J., Opiate influences on nucleus accumbens neuronal electrophysiology: dopamine and non°dopa- mine mechanisms, J. Neurosci., 9 (1989) 3538 3546.

4 Henderson, G. and Hughes, J., Modulation of frequency-depend- ent noradrenaline release by calcium, angiotensin and morphine, Br. J. Pharmacol., 52 (1974) 455~,56P.

5 Henderson, G., Hughes, J. and Kosterlitz, H.W., Modification of catecholamine release by narcotic analgesics and opioid peptides. In D.M. Paton (Ed.), The Release of Catecholamines from Adre- nergic Neurones, Pergamon, Oxford, 1979, pp. 217-228.

6 Herkenham, M., Edley, S.M. and Stuart, J., Cell clusters in the nu- cleus accumbens of the rat, and the mosaic relationship of opiate receptors, acetylcholinesterase and subcortical afferent termina- tions, Neuroscience, 11 (1984) 561-593.

7 Koob, G.F. and Bloom, F.E., Cellular and molecular mechanisms of drug dependence, Science, 242 (1988) 715 723.

8 Mansour, A., Kachaturian, H., Lewis, M.E., Akil, H. and Watson, S.J., Anatomy of CNS opioid receptors, Trends Neurosci., 11 (1988) 308-314.

9 McCarthy, P.S., Walker, R.J. and Woodruff, G.N., Depressant ac- tion of enkephalins on neurons in the nucleus accumbens, J. Phys- iol., 267 (1977) 40.

10 Mogenson, G.J., Limbic-motor integration, Prog. Physiol. Psy- chol., 12 (1987) 117-170.

11 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordi- nates, 2nd edn., Academic Press, London, 1986.

12 Petit, F., Hamon, M., Fournie-Zaluski, M.C., Roques, B.P. and Glowinski, J., Further evidence for a role of delta-opiate receptors in the presynaptic regulation of newly synthesized dopamine re- lease, Eur. J. Pharmacol., 126 (1986) 1 9.

13 Quirion, R., Zajac, J.M., Morgat, J.L. and Roques, B.P., Autora- diographic distribution of mu and delta opiate receptors in rat brain using highly selective ligands, Life Sci., 33 Suppl I (1983) 227 230.

14 Sesack, S.R. and Pickel, V.M., Ultrastructural evidence for interac- tions between opioid and dopaminergic neurons in the rat mesolim- bic system. In J.M. van Ree, A.H. Mulder, V.M. Wiegant and T.B.

228

van Wimersma Greidanus (Eds.), New Leads in Opioid Research, Elsevier, Amsterdam, 1990, pp. 241-243.

15 Storm, J.F., Temporal integration by a slowly inactivating K ÷ cur- rent in hippocampal neurons, Nature, 33 (1988) 379-381.

16 Surmeier, D.J., Stefani, A., Foehring, R.C. and Kitai, S.T., Devel- opmental regulation of a slowly-inactivating potassium conduc- tance in rat neostriatal neurons, Neurosci. Lett., 122 (1991) 41~,6.

17 Uchimura, N., Cherubini, E. and North, R.A., Inward rectification in rat nucleus accumbens neurons, J. Neurophysiok, 62 (1989) 128(~1286.

18 Unterwald, E.M., Koob, G~F. and Zukin, R.S., Characterization and neuroanatomical localization of multiple opioid receptors in rat brain following selective nucleus accumbens lesions, Soc. Neu- rosci. Abstr., 12 (1986) 48.11.

19 Vaccarino, F.J., Bloom, F.E. and Koob, G.F., Blockade of nucleus accumbens opiate receptors attenuates the intravenous heroin re- ward in the rat, Psychopharmacology, 86 (1985) 37~,2.

20 Wamsley, J.K., Young, W.S. and Kuhar, M.J., Immunohistochem- ical localization of enkephalin in rat forebrain, Brain Res., 190 (1980) 153-174.

21 Woodruff, G.N., McCarthy, P.S. and Walker, RA., Studies on the pharmacology of neurons in the nucleus accumbens of the rat, Brain Res., 115 (1976) 233-242.

22 Zieglg~insberger, W. and Bayerl, J., The activity of opiates in the spinal cord of cat, Brain Res., 115 (1976) I 11-128.

23 Zieglg/insberger, W., French, E.D., Siggins, G.R. and Bloom, F.E., Opioid peptides may excite hippocampal pyramidal neurons by inhibiting adjacent inhibitory interneurons, Science, 205 (1979) 415~417.

24 Zieglg~insberger, W. and North, R. A., Opiate withdrawal signs in single myenteric neurones, Brain Res., 144 (1978) 208-211.