Noradrenergic depression of synaptic responses in hippocampus of rat: Evidence for mediation by alpha1-receptors

~euro~~a?macoIogy Vol. 27, No. 4, pp. 391-398, Printed in Great Britain

1988

0~28~3~8/88

$3.00 + 0.00

Pergamon Press pie

NORADRENERGIC DEPRESSION OF SYNAPTIC RESPONSES IN HIPPOCAMPUS OF RAT: EVIDENCE MEDIATION BY ALPHA, -RECEPTORS

FOR

MICHELLE MYNLIEFF’ and T. V. DUNWIDDIE’~~ ‘Department of Pharmacology, University of Colorado Medical Center, Denver, CO 80262, U.S.A. and *Medical Research Service, Veterans Administration Medical Center, Denver, CO 80220, U.S.A. (Accepted 28 October

1987)

hulas-Norepinephrine (NE) has been shown to have a biphasic effect on evoked potentials in the CA1 region of the hippocampus of the rat in vitro, with a beta receptor mediating an increase and an alpha receptor eliciting a decrease in the amplitude of the population spike. The purpose of this study was to use selective alpha-adrenergic agonists and antagonists to determine the subtype of receptor mediating the depressant response of NE. The present investigations demonstrated that the selective alpha, agonist, phenylephrine (2-50 PM) elicited a dose-dependent depression of the amplitude of the population spike. Clonidine, a relatively selective alpha,-agonist, also depressed the amplitude of the population spike, but only at concentrations (1OpM) that were inconsistent with a selective action upon alpha,-receptors. Another alpha,-agonist, alpha-methylnorepinephrine (100-400 nM) did not depress the amplitude of the population spike. The depressant effect of NE was antagonized by the nonselective alpha antagonist, phentolamine (0.5-50 PM) and the alpha, -selective antagonist, prazosin (1 ,uM), but not by the alpha,-selective antagonist, idazoxan (l-10 FM). Phentolamine and prazosin antagonized the response to phenyleph~ne but not to clonidine. The depressant effect of NE was not antagonized by the antagonist of serotonin and dopamine, spiperone (100 nM); conversely, the effect of 8-hydroxy-Z-(di-~-propylamine) tetralin (50 pM), a S-HT,, receptor-selective agonist, which afso depresses the amplitude of the population spike, was not antagonized by phentolamine (5pM). These data indicate that, in the in tiitro slice preparation, NE acts upon an alpha, -adrenergic receptor to depress the amplitude of the population spike, and not an alpha,-adrenergic receptor, a S-HT,, receptor or a dopamine receptor. The data also support the conclusion that serotonin acts independently to reduce the response of the hippocampal population spike. Key words: hippocampus, norepinephrine,

Norepinephrine mitter which

synaptic transmission, alpha receptors.

(NE) is an important

neurotransdistributed in the central nervous system. Many aspects of its neurotransmitter is widely

action still remain unclear. In the hippocampus, ei~trophysiological responses to norepinephrine have been studied by Segal and Bloom (1974a, 1974b) using in situ recording techniques. Either local application of NE, or stimulation of the locus coeruleus were found to depress the spontaneous firing rates of cells in the CA, region, and this effect was blocked by sotalol, a beta antagonist. However, the responses were frequently more complex (e.g. biphasic inhibitory/excitatory effects) and the identity of the receptors mediating the responses was not clearly established. The in vitro hippocampal slice preparation has been used more recently to characterize further the effects of NE on different subtypes of receptors (Mueller, Hoffer and Dunwiddie, 1981; Mueller, Palmer, Hoffer and Dunwiddie, 1982). In these studies, the effect of NE on the amplitude of an evoked excitatory response, the population spike, in the CA, region was examined. It was found that superfusion of NE evoked a biphasic response in this preparation. At small doses NE increased the amplitude of the population spike; this effect was antagon-

ized by beta-selective

receptor

antagonists.

At larger

doses, NE decreased the amplitude of the population spike, and this response was antagonized by the alpha-receptor selective antagonist, phentolamine. Intracellular studies in the in vitro slice preparation have demonstrated similar effects, in that NE both depolarized and hyperpolarized the membrane potential. Norepinephrine can also reduce accommodation of cell firing in response to a depolarizing current pulse, most probably by reducing the amplitude and duration of the slow calcium-activated potassiumafter-hyperpolarization that follows dependent depolarization-induced action potentials (Madison and Nicoll, 1982; Madison and Nicoll, 1986). This latter response was attributed to an action of NE on beta,-receptors, but the characte~zation of the putative alpha-re~ptor-mediated portion of these responses (hy~~olarization and decrease in the amplitude of the population spike) has been unsuccessful. There are at least two reasons why it is important to characterize the alpha,- compared to alpha,-mediation of these responses. First, other studies have suggested that alpha,-responses are excitatory, whereas alpha,-receptors are generally inhibitory (Rogawski, 1985; Szabadi, 1979); this being 391

MICHELLE MYNLIEFF and

392

the case, it is important to determine whether the same holds true in the hippocampus. In addition, the cellular mechanisms by which alpha,-, and alpha,responses are brought about are quite different (activation of phospholipase C and subsequent increases in inositol trisphosphate and diacylglycerol, compared to activation of a potassium channel and perhaps inhibition of adenylate cyclase). Identification of the specific receptor involved in these responses is clearly a prerequisite to the subsequent characterization of the cellular mechanisms of action. Once the presence and identity of functional receptors has been established, this information can be used to characterize further the effects of NE in vivo. The goal of the present study was to examine the depression of evoked potentials in the hippocampus by noradrenaline and to use selective agonists and antagonists to identify the receptor responsible for this depression. In these experiments, the authors have used the in vitro hippocampal slice preparation, which offers numerous advantages over other preparations in the study of receptor pharmacology. Because many adrenergic drugs are only somewhat receptor selective, and at larger concentrations can often interact with other receptors, the ability to perfuse slices with known concentrations of drug is critical to identifying subtypes of receptors. By applying the drug directly to the slice one can also circumvent any peripheral metabolism or effects of the drug, which are particularly important in the study of noradrenergic drugs that can affect the cardiovascular system.

METHODS

T. V.

DUNWIDDIE

Electrical stimulation and recording The stimulating electrode was a twisted nichrome wire which was placed in the stratum radiatum under visual guidance near the border of CA,-CA,. The synaptic response was elicited by a monophasic 0.1 msec pulse of 5-30 V, delivered to the slice once every 60 set and the evoked potential, measured in these experiments, was the population spike recorded from the stratum radiatum, which reflects the summated synchronous firing of the pyramidal neurons. The stimulation voltage was set to evoke a submaximal population spike of l-3 mV. The recording eiectrode consisted of a 2-3 MR glass microelectrode, filled with 3 M NaCl, which was placed under visual guidance in the CA, pyramidal cell layer. The slices were maintained without perfusion until they were tested, at which point fresh, oxygenated preheated medium was pumped through the chamber at 2ml/min. Drugs were added to the flow of medium with a calibrated Sage Model 355 or Raze1 syringe pump. Previous studies in this laboratory have shown that the addition of distilled water to the perfusion medium had no effect on evoked responses when the amount added induced only a 0. l-l % change in flow rate (Mueller et al., 1981). In all of the experiments, a stable response was obtained for at least 10min before adding any drugs. The perfusion of any particular drug was continued until a maximal response was seen. Antagonist drugs were present in the medium for a minimum of 10 min before the addition of an agonist. If the antagonist was ineffective in blocking the response to the agonist, a longer preincubation period (60 min or more) with the antagonist was used to ensure that the drug had reached equilibrium.

Subjects

Collection of data and analysis

Male Sprague-Dawley rats (150-250 g) were obtained from Sasco, St. Louis, Missouri for all experiments. They were housed in groups of 2-5 under a 12-hr light/dark cycle and were maintained on laboratory rat chow and tap water ad libitum.

The data were entered into a NOVA 3/12 computer in digital form for subsequential analysis by computer. The data are presented as the mean f the standard error of the mean. Statistical significance was determined with the Students t-test. Drugs

Preparation and maintenance of hippocampal slices The rats were decapitated and the hippocampus was dissected free from the rest of the brain. Coronal slices of the hippocampus were prepared as described previously (Dunwiddie and Lynch, 1978). Slices of 400 pm thickness were made on a Sorvall tissue chopper and immediately placed in ice-cold artificial cerebral spinal fluid consisting of 124mM NaCl, 3.3 mM KCl, 1.2mM KH,PO,, 2.4mM MgSO,, 2.5 mM CaCl,, 25.7 mM NaHCO, and 10 mM glucose, which was pregassed with 95% O2 and 5% CO,. The slices were transferred within 5 min to the recording chamber (33-34°C). The slices were allowed to equilibrate for a minimum of 1 hr, during which time the level of the medium was maintained at, or just below, the upper surface of the slice.

Clonidine, phenylephrine, norepinephrine (Sigma Chemical Co., St. Louis, Missouri), idazoxan (Reckitt & Colman, Kingston-upon-Hull, England), timolol (Merck Sharp & Dohme, Rahway, New Jersey), phentolamine (Ciba-Geigy Corp., Summit, New Jersey), I-hydroxy-Z(di-n-propylamine)tetralin (RBI, Wayland, Massachusetts) and alpha-methylnorepinephrine (Sterling-Winthrop Res. Inst. Rensselaer, New York) were dissolved in distilled, degassed water at 100-1000 times the desired concentration. Spiperone (Janssen, Beers, Belgium) was dissolved in a drop of concentrated HCI and then brought up to volume in distilled water. The pH of this solution was adjusted to 6.5-7.5 with NaOH. Prazosin (Pfizer, Inc., Groton, Connecticut) was added to distilled water and sonicated for a minimum of 1 hr.

Alpha-adrenergic

responses in hippocampus

RESULTS

As shown in Figure 1, perfusion of hippocampal slices with 25 PM NE produced a slight depression in the amplitude of the population spike with a maximal reduction of 14 k 5.1% (n = 19). If 1 PM timolol was included in the superfusion medium to antagonize effects mediated by beta-adrenergic receptors, 25 PM NE produced a much larger depression of the amplitude of the population spike (46 f 7.6%, n = 16). After the addition of 50 PM phentolamine to reduce the alpha-adrenergic component of the response, 25 p M NE elicited an increase in the amplitude of the population spike (99 f 25%, n = 6). These data suggest that NE decreased the evoked response through alpha-adrenergic receptors and increased the evoked response through the beta-adrenergic receptors. These results are consistent with those of previous studies using this preparation (Mueller et al., 1981; Mueller et al., 1982). Because NE has a low but significant affinity for both serotonin and dopamine receptors, the present authors first considered the possibility that the depressant response might reflect the actions of NE

393

upon non-adrenergic receptors, particularly since it has been reported that both dopamine and serotonin can affect the amplitude of the population spike (Marciani, Calabresi, Stanzione and Bernardi, 1984; Beck, Clarke and Goldfarb, 1985; Beck and Goldfarb, 1985; Peroutka, Mauk and Kocsis, 1987). Two different types of experiments were conducted to determine whether this was the case. Pretreatment with 100 nM spiperone, which is 1250 times the Kd for dopamine D, receptors (Zahniser and Dubocovich, 1983) 100 times the Kd for 5-HT, sites and IO times the K,, for 5-HT,, sites (Beck et al., 1985), had no significant effect upon the response to 25 p M NE plus 1 PM timolol (Fig. 2). Conversely, pretreatment with 5 PM phentolamine, which significantly reduced the depressant response to NE (cf. Fig. 4) had no effect on the response to 50 PM 8-hydroxy-2-(di-n-propylamine)tetralin, a selective 5-HT,, agonist. In order to characterize further the receptors mediating the depressant component of the response to NE, the effects of agonists, selective for subtypes of alpha-receptors were examined: 1 PM timolol was included in all experiments to block any effects

i‘* I-:..-*. 1 I :-r-r-~_._r-:-r-:-:‘: I -60

NOREPINEPHRINE

-80 0

5

10

15 Tm

20

25

30

(MINUTES)

Fig. 1. The top portion of the figure represents the time course of the effect of NE on the amplitude of the population spike. Perfusion with 25pM NE (circles; n = 19) produced a slight depression in the amplitude of the population spike, 25 PM NE + 50 PM phentolamine elicited a large increase in the amplitude of the population spike (triangles; n = 6) while 25j~M NE + 1PM timolol induced a depression in the amplitude of the population spike (diamonds; n = 16). The effect of NE on the population spike is expressed as a percentage change in the amplitude from the pre-NE baseline. The perfusion of NE was carried out from lo-22 min as indicated by the horizontal line in the lower portion of the graph, whereas phentolamine and timolol when present were perfused from the beginning of the experiment. Each point represents the mean change in amplitude f the standard error of the mean. The bottom half of the figure shows responses of individual slices treated as outlined above. In A, B, and C, 1 is the control population spike, 2 is the population spike after a maximal response to drug has been reached, and 3 is the population spike after the drug has been washed out. A is an example of the effect of 25 PM NE alone; B, the effect of 25 PM NE with 50 PM phentolamine as pretreatment and C, the effect of 25 PM NE with 1PM timolol as pretreatment. The calibration bar in the right hand corner is 1 mV by 2 msec.

MICHELLE MYNLIEFF and T. V. DUNWIDDIE

394

NJ3

NE

+ SPJP

8-OH-DPAT &OH-DPAT + PHA

Fig. 2. Effect of spiperone @PIP) on the response to NE and the effect of phentolamine on the response to 8-hydroxy-2-(di-n-propylamine)tetralin (8-OH-DPAT). The response to 25 PM NE plus 1 PM timolol was measured in the absence and presence of 100 nM spiperone (preincubated for 10 min); the percentage depression of the population spike after perfusion for 15 min with 50 PM 8-hydroxy-2-(di-n-propylamine)tetralin was measured in the absence and presence of 5 PM phentolamine (PHA); neither difference was statistically significant. Note that the mean response to I-OH-DPAT was also significantly larger than the response to any of the alpha-agonists studied (cf. Fig. 3). The data are expressed as the mean k SEM response from 4-9 slices.

mediated

by

beta-adrenoceptors.

Phenylephrine,

a

selective alpha, -agonst (Drew, 1976; Starke, Endo and Taube, 1975) consistently reduced the amplitude of the population spike, with a threshold concentration of approximately 2 PM (Fig. 3). Clonidine, an alpha,-selective agonist (UPrichard, Greenberg and Snyder, 1977) was approximately equipotent with phenylephrine and produced approximately the same maximal depression of the population spike as did NE and phenylephrine, with 10 p M clonidine causing

a 41 + 6.1% (n = 17) depression after 10 min. Previous studies in this laboratory have established a threshold of 1 PM for clonidine in this preparation, in the absence of timolol (Mueller et al., 1981) and a similar potency was observed here as well. Another alpha,-adrenergic agonist, alpha-methylnorepinephrine (UPrichard et al., 1977) was tested at 100 nM (n = 6) and 400 nM (n = 3) on the preparation and was found to have no depressant effect on the amplitude of the population spike.

E" z 0 0.5

1.0

10.0

II I.0

a DOSE OFPHENYLEPHRINE(@d) Fig. 3. Dose-response curve for phenylephrine in the presence of 1 PM timolol. The effect of phenylephrine is expressed as the mean k SEM percentage depression of the amplitude of the population spike after 10 min perfusion, relative to the pre-drug control period (N = 4-12 slices/point). The narrow range of the dose-response curve is typical of most agents that affect the response of the population spike (opiates, beta-adrenergic agonists, adenosine).

Alpha-adrenergic

responses

in hippocampus

395

L I

**

EL -liL -

NE

NE

NE

f

+

PHA

PR

NE + IDA

Fig. 4. Effect of alpha-adrenergic antagonists on the response to 25 PM NE + I pM timolol. The percentage depression of the population spike after perfusion for 10 min with NE was measured in the absence and presence of the alpha-selective antagonists. Phentolamine (PHA 5 PM; P < 0.005) and prazosin (PR 1 PM; P < 0.01) significantly reduced responses to NE, but 1 pM idazoxan (IDA) was ineffective. In each experiment the effect of NE plus timolol, in conjunction with an alpha antagonist, was compared to the effect of NE plus timolol on control slices, obtained from the same rats. The data are expressed as the mean response from 10 -23 slices + SEM.

In subsequent experiments, the effects of selective alpha-adrenergic receptor antagonists were examined, again in the presence of 1 PM timolol to eliminate the complicating effects of the betaadrenergic response. Agonist responses in the presence of each antagonist were compared to the effect of the agonist alone on control hippocampal slices from the same group of rats. This protocol was followed because recovery from some adrenergic agonists, such as clonidine, is poor, perhaps because of poor wash out of drug, thereby making it more difficult to apply repeated doses of a drug to a single slice. Phentolamine, which is a relatively non-selective alpha-adrenergic antagonist, antagonized the effects of 25 p M NE at a concentration of 50 p M. Considerably smaller concentrations were also effective, but only when the slices were preincubated for at least an hour with the antagonist; for example, 0.5 PM phentolamine antagonized the response to NE by 38% (P < O.l), and 5pM, by 72% (P <0.005, Fig. 4). Prazosin (1 PM), a selective alpha, -antagonist (Aghajanian, 1985; North and Yoshimura, 1984), antagonized the effect of 25 PM NE by 59% (P < 0.01). Idazoxan, a selective alpha,-antagonist (Doxey, Roach and Smith, 1983; Freedman and Aghajanian, 1984), did not reduce the depressant response of the hippocampal slice to NE plus timolol, at concentrations of 1 and lOpM, even with a preincubation period of 1 hr. At lOpM, idazoxan had direct action of its own: the evoked response was depressed within 10 min of application, providing direct evidence that the compound was able to penetrate the slice. Pretreatment with phentolamine (5 /J M) antagonized the response to phenylephrine

(1OpM) plus timolol by 65% (P < 0.05, but not the response to clonidine (10 PM, Prazosin (1 PM) also reduced the response to phenylephrine plus timolol by 59% (P < 0.05,

n = 8) n = 6). 10 p M n = 8).

DISCUSSION

The present experiments confirm previous work which demonstrated a biphasic effect of NE on responses of the evoked population spike in the CA, region of the hippocampus, and showed that the excitatory and inhibitory responses were mediated by different types of adrenergic receptors (Mueller et al., 1981; Mueller et al., 1982). The alpha-mediated depression of the evoked population spike could be blocked by smaller concentrations of phentolamine than have previously been reported, but only when the slices were pretreated with the antagonist for longer than 1Omin. This suggests that the antagonism of depressant responses, exerted by phentolamine, is likely to reflect actions at alpha-receptors but that, under normal conditions, the diffusion of phentolamine into the slice is relatively slow. Previous experiments with phenylephrine were somewhat inconclusive, since phenylephrine (50-1OOpM) had given variable results. In the present study, the combination of timolol and phenylephrine resulted in a consistent dosedependent depression of the population spike, suggesting that phenylephrine, like NE, may activate both alpha- and beta-receptors. Phenylephrine has been reported to have very little activity at alpha,-receptors (Drew, 1976; Starke et al., 1975),

396

MICHELLE MYNLIEFF and

which would make it unlikely that, at the concentration used, phenylephrine was acting upon alpha,-receptors. The EC,, for phenylephrine in these experiments was approximately 2-3 PM which corresponds well with the affinity of phenylephrine for the alpha, -receptor (K, = 2.6 PM, U’Prichard et al., 1977). In terms of the antagonist studies, prazosin has been reported to be a highly selective alpha, -receptor antagonist, which in other in vitro electrophysiological experiments appears to block putative alpha, adrenoceptor-mediated responses at concentrations between 7555000 nM (Aghajanian, 1985; North and Yoshimura, 1984). In the present experiments, I PM prazosin reduced, but did not entirely block, the depressant response to NE or phenylephrine plus timolol. On the other hand, idazoxan, an alpha,selective antagonist (Doxey et al., 1983) that inhibits alpha,-responses in the locus coeruleus (Freedman and Aghajanian, 1984) and substantia gelatinosa (North and Yoshimura, l984), if anything enhanced the depressant responses to NE in the hippocampus. These data, in combination with those concerning selective agonists, suggest that NE caused a depression of the amplitude of the population spike through an interaction with the alpha, -subtype of adrenergic receptors. Although clonidine, a selective alpha,-agonist, also produced a depression of the amplitude of the population spike, the ECS, value (-2 FM, Mueller et al., 1981) was approximately 1000 times the Kd found in binding studies for the alpha,-receptor (U’Prichard et al., 1977). The previously reported threshold for the depression of the population spike (I PM) was also much greater than the threshold concentration required to hyperpolarize neurons in the locus coeruleus in vitro (3 nM), an effect probably mediated by alpha,-receptors (Williams, Henderson and North, 1985). It is likely that, in micromolar concentrations, clonidine can act upon receptors other than the alpha,-receptor. In this context, it is notable that alpha-methylnorepinephrine, which is also relatively selective for the alpha,-receptor (UPrichard et al., 1977), did not depress the amplitude of the population spike at concentrations consistent with effects upon alpha,-receptors. The observation that pretreatment with 5 PM phentolamine significantly antagonized the responses to both NE and phenylephrine but not to clonidine suggests that the response to clonidine may not involve either subtype of alpha receptor. The fact that alpha,-receptors are sparse in the hippocampus (Young and Kuhar, 1980) also would make it less likely that NE was acting upon alpha,-receptors. A possibility that must always be considered in such experiments is that NE is acting through some non-adrenergic type of receptor to inhibit the response of the population spike. It has recently been reported (Malenka and Nicoll, 1986) that dopamine, when perfused in sufficient amounts, can elicit betaadrenergic responses in the hippocampus, and the

T. V.

DUNWIDDIE

present authors have made similar observations (Dunwiddie, unpublished). Likewise, a sufficiently large concentration of exogenous NE might act upon either dopaminergic, or perhaps serotonergic, receptors to modify the evoked response, since NE has a measurable affinity for both types of receptors (Bennett and Snyder, 1976; Lovell and Freedman, 1976; Burt, Creese and Snyder, 1976), Serotonin, in particular, has been shown to decrease the amplitude of the population spike in the CA, region of the hippocampus, an effect which is thought to be mediated by the 5HT,, subtype of receptor (Beck et al., 1985, Peroutka et al., 1987). However, pretreatment with spiperone in doses that block responses to serotonin in the in vitro hippocampal slice (Beck et aI., 1985) did not affect the response to NE, suggesting that NE was not acting on serotonin receptors to produce this response. This concentration of spiperone would also appear to preclude effects on dopamine receptors as well (Zahniser and Dubocovich, 1983). Pretreatment with 5 PM phentolamine, which markedly attenuated responses to NE, in turn, was also unable to affect the response to 8-hydroxy-2-(di-n -propylamine)tetralin, indicating that the serotonergic response reported by Beck et al. (1985), Beck and Goldfarb (1985) and Peroutka ef al. (1987) does not involve an alpha-receptor. An important question raised by this study is why an alpha, -adrenergic response is apparently inhibitory, whereas in other regions of the brain alpha,adrenergic receptors have most frequently been linked to an excitatory response (Rogawski, 1985; Szabadi, 1979). One possibility is that the depressant effect of NE is an indirect action, mediated by an alpha, -excitation of hippocampal interneurons. Previous studies have shown that the depression of the population spike is not due to a reduction in synaptic transmission, since the excitatory post-synaptic potential is unaffected by the perfusion or NE (Mueller et al., 1981). Thus, the depression must reflect some post-synaptic change which is either intrinsic to the pyramidal neurons, or is secondary to changes in afferent activity. If the alpha,-receptors are located on the inhibitory interneurons, then excitation of these interneurons would result in a decrease in the number of pyramidal cells firing in response to the same synaptic input, which would be consistent with the present observations. Direct intracellular recordings from inhibitory interneurons, under conditions where synaptic transmission is suppressed, might confirm this possibility. An alternative possibility is that, although in other tissues the alpha,-response is excitatory, in hippocampal neurons it may be inhibitory. Many previous studies have used local application techniques, where the agonist was applied directly to the soma. In the present experiments the agonist was perfused, allowing it to reach receptor sites other than those on the soma. The possibility exists that the population of receptors on the soma differ in their response when

Alpha-adrenergic

responses

compared to the population of receptors located on the dendrites. Another explanation may be related to the fact that in slices of brain, activation of alpha, -receptors stimulates phospholipase C-mediated breakdown of polyphosphatidyl-4,%bisphosphate into inositol trisphosphate and diacylglycerol (Marx, 1987). The inositol trisphosphate, in turn, causes release of calcium from internal storage sites and diacylglycerol stimulates the activity of protein kinase C. It has recently been shown that in the dorsal raphe of the rat, activation of alpha,-receptors prolongs the calcium-activated potassium currents (Freedman and Aghajanian, persona1 communication) an effect which may be related to the release of intracellular calcium by inositol trisphosphate. Because of the indirect nature of this alpha, -receptor-mediated response, it is certainly possible that regions of the brain may differ in terms of the inhibitory and excitatory effects, depending upon the way in which intracellular second messengers, such as calcium released from cytoplasmic stores by inositol trisphosphate and the activation of protein kinase C by diacylglycerol, affect cellular activity. In conclusion, these experiments suggest that the depressant effect of NE on evoked potentials in the CA, region of the hippocampus is mediated by alpha, -adrenergic receptors. This could reflect either direct actions upon pyramidal neurons, or the activation of alpha, -receptors on hippocampal interneurons that might indirectly reduce the excitability of the pyramidal cells. In either case, activation of noradrenergic afferents to the hippocampus would be expected to have rather complex effects that would reflect not only the reduction in hippocampal excitability that can be attributed to activation of alpha, -receptors, but opposing increases in excitability linked to activation of beta, -adrenoceptors. Acknowledgements-This research was supported by DA 02702 and by the Veterans Administration Medical Research Service.

REFERENCES Aghajanian G. K. (1985) Modulation of a transient current in serotonergic neurones by alpha-adrenoceptors. Nature 315: 501-503. Beck S. G., Clarke W. P. and Goldfarb J. (1985) Spiperone differentiates multiple 5-hydroxytryptamine responses in rat hippocampal slices in vitro. Eur. J. Pharmac. 116: 1955197. Beck S. G. and Goldfarb J. (1985) Serotonin produces a reversible concentration dependent decrease of population spikes in rat hippocampal slices. Life Sci. 36: 557-563. Bennett J. P. and Snyder S. H. (1976) Serotonin and lysergic acid diethylamide binding in rat brain membranes: relationship to postsynaptic serotonin receptors. Molec. Pharmac. 12: 373-389.

Burt D. R., Creese I. and Snyder S. H. (1976) Properties of ( 3H)haIoperidol and (3H)dopamine binding associated with dopamine receptors in calf brain membranes. Molec. Pharmac. 12: 800-812.

in hippocampus

397

Doxey J. C., Roach A. G. and Smith C. F. C. (1983) Studies on RX 781094: a selective, potent and specific antagonist of alpha2-adrenocentors. Br. J. Pharmac. 78: 489-505. Drew G. M. (1976) Effects of alpha-adrenoceptor agonists and antagonists on pre- and postsynaptically located alpha-adrenocentors. Eur. J. Pharmac. 36: 313-320. Dunwiddie T. V.-and Lynch G. (1978) Long-term potentiation and depression of synaptic responses in the rat hippocampus: localization and frequency dependency. J. Physiol., Lond. 276: 353-367. Freedman J. E. and Aghajanian G. K. (1984) Idazoxan (RX 781094) selectively antagonized alpha2-adrenoceptors on rat central neurons. Eur. J. Phar&c. 105: 265-272. Lovell R. A. and Freedman D. X. (1976) Stereosoecific receptor sites for d-lysergic acid diethylamide in rat brain: effects of neurotransmitters, amine antagonists, and other psychotropic drugs. Molec. Pharmac. 12: 62&630. Madison D. V. and Nicoll R. A. (1982) Noradrenaline blocks accommodation of pyramidal cell discharge in the hippocampus. Nature 299: 636638. Madison D. V. and Nicoll R. A. (1986) Actions of noradrenaline recorded intracellularly in rat hippocampal CA, pyramidal neurones, in vitro. J. Physiol. 372: 221-244. Malenka R. C. and Nicoll R. A. (1986) Dopamine decreases the calcium-activated afterhyperpolarization in hippocampal CA, pyramidal cells. Bruin Res. 379: 210-215. Marciani M. G., Calabresi P., Stanzione P. and Bernardi G. (1984) Dopaminergic and noradrenergic responses in the hippocampal slice preparation. Neuropharmacology 23: 303-307. Marx J. L. (1987) Polyphosphoinositide research updated. Science 235: 974976. Mueller A. L., Hoffer B. J. and Dunwiddie T. V. (1981) Noradrenergic responses in rat hippocampus: evidence for mediation by alpha and beta receptors in the in vitro slice. Brain Res. 214: 113-126. Mueller A. L., Palmer M. R., Hoffer B. J. and Dunwiddie T. V. (1982) Hippocampal noradrenergic responses in uivo and in vitro: characterization of alpha and beta components. Naunyn-Schmiedebergs Arch. Pharmac. 318: 259-266. North R. A. and Yoshimura M. (1984) The actions of noradrenaline on neurones of the rat substantia gelatinbosa in vitro. J. Physiol. 349: 43-55. Peroutka S. J., Mauk M. D. and Kocsis J. D. (1987) Modulation of hippocampal neuronal activity by 5-hydroxytryptamine and 5_hydroxytryptamine,. selective drugs. Neuropharmacology 26: 139-146. Rogawski M. A. (1985) Norepinephrine. Neurofrunsmifter Actions in lhe Vertebrale Nervous System, pp. 241-284. Plenum Press, New York. Segal M. and Bloom F. E. (1974a) the action of norepinephrine in the rat hippocampus. 1. Ionotophoretic studies. Brain Res. 72: 79-97. _ Segal M. and Bloom F. E. (1974b) The action of norepinephrine in the rat hippocampus. II. Activation of the input pathway. Brain Res. 72: 99-114. Starke K.. Endo T. and Taube H. D. (1975) Relative nreand postsynaptic potencies of alpha-ahenoceptor agonists in the rabbit pulmonary artery. Naunyn-Schmiedebergs Arch. Phurmac. 291: 55-78. Szabadi E. (1979) Adrenoceptors on central neurones: microelectrophoretic studies. Neuropharmacology 18: 831-843. UPrichard D. C., Greenberg D. A. and Snyder S. H. (1977) Binding characteristics of a radiolabeled agonist and antagonist at central nervous system alpha noradrenergic receptors, Molec Pharmac. 13: 454473. Williams J. T., Henderson G. and North R. A. (1985) Characterization of alpha,-adrenoceptors which increase potassium conductance in rat locus coeruleus neurones. Neuroscience 14: 955101.

398

MICHELLEMYNLIEFFand T. V. DUNWIDDIE

Young W. S. III and Kuhar M. J. (1980) Noradrenergic alpha1 and alpha,-receptors. Light microscopic autoradiographic localization. Proc. natn. Acad. Sci. U.S.A. 77: 169fS1700.

Zahniser N. R. and Dubocovich M. L. (1983) Comparison of dopamine receptor sites labeled by (3H)-s-sulpiride and (‘H)-spiperone in striatum. J. Pharmac. exp. Thu. 227: 592-599.