Neuropeptide Y Selectively Inhibits Slow Synaptic Potentials in Rat Dorsal Raphe Nucleus in Vitro by a Presynaptic Action

The Journal of Neuroscience, March 1992, 12(3): 1088-l 093

Neuropeptide Y Selectively Inhibits Slow Synaptic Potentials in Rat Dorsal Raphe Nucleus in vitro by a Presynaptic Action

Samuel B. Kombian and William F. Colmers Department of Pharmacology, The University of Alberta, Edmonton, Alberta, Canada T6G 2H7

Neuropeptide Y (NPY) has been shown to modulate synaptic a contraction of smooth muscle,while subthresholdconcentra- transmission in both peripheral and central tissues via both tions of NPY enhancecontractions causedby NA or other drugs pre- and postsynaptic mechanisms. In this study, we ex- (Ekblad et al., 1984; Wahlestedt and HBkanson, 1987). In the amined the effect of NPY and its analog, peptide YY (PYY), noradrenergicneurons of the brainstem nucleuslocus coeruleus, on slow synaptic potentials in the dorsal raphe nucleus in some of which contain NPY (de Quidt and Emson, 1986), NPY vitro using intracellular recording and single-microelectrode appearsto potentiate postsynaptic responsesto cu,-adrenoceptor voltage-clamp techniques. NPY and PYY inhibited both the activation; it is not known whether there are any effects on slow 5HT,, receptor-mediated IPSP and the a,-adrenocep- synaptic potentials evoked in this nucleus (Illes and Regenold, tor-mediated slow EPSP while not affecting the fast, amino 1990). NPY’s actions therefore vary with the system, presum- acid-mediated synaptic responses. PYY also inhibited phar- ably to elicit an appropriate, coordinated responseto the higher macologically isolated slow synaptic responses. NPY/PYY levels of stimulation thought to cause preferential release of appear to mediate the observed inhibitions via a presynaptic peptides (Bartfai et al., 1988). mechanism, as the postsynaptic conductances mediated by The dorsal raphe (DR) nucleus is a brainstem nucleus com- activation of 5-HT,, receptors or a,-adrenoceptors were un- posedmainly of 5-HT-secreting neurons that project rostrally affected by the peptides. NPY/PYY act via a different mech- to supply the serotonergic innervation of the forebrain (And& anism than presynaptic 5-HT,, receptors. NPY/PYY probably et al., 1966; Azmitia and Segal, 1978; Fuxe and Jonsson, 1984; act via presynaptic Y, receptors, as the C-terminal fragment Mulligan and Tork, 1987, 1988; Aitken and Tork, 1988). These NPY 13-36 and the Y,-selective agonist CS-NPY are effec- cellsreceive a variety of synaptic inputs, including fast responses tive. Since NPY and its receptors are present in the dorsal mediated by the amino acids glutamate and GABA (Kalen et raphe nucleus, this peptide may act as an endogenous mod- al., 1985, 1986; Pan and Williams, 1989a),and slowerpotentials ulator of the state of activity of neurons in this region and mediated by 5-HT (Chazal and Rahlston, 1987; Williams et al., may thus have a role in the modulation of neuronal output 1988) and NA (Aghajanian and Wang, 1977; Yoshimura et al., from this nucleus. 1985). The pharmacology of thesesynaptic responseshas been well characterized (Baraban and Aghajanian, 1980; Yoshimura Neuropeptide Y (NPY) has been shown to have several neu- andHigashi, 1985; Yoshimuraetal., 1985; Williamsetal., 1988; romodulatory actions at pre- and postsynaptic sitesin a number Pan and Williams, 1989a,b; Bobker and Williams, 1990). In of tissues(Pemow et al., 1986; see other referencesbelow). In addition, DR neurons are relatively compact electrically, per- the rat hippocampusin vitro, NPY inhibits the fast (glutamater- mitting acceptablesingle-electrode voltage clamp of slow mem- gic) synaptic potentials evoked in CA1 pyramidal neurons by brane conductanceswith conventional sharp electrodes.NPY stimulation of stratum radiatum. This action appearsto be pure- is present in moderately high concentrations in DR (O’Dono- ly presynaptic in nature, as no evidence for postsynaptic actions ghue et al., 1985), and NPY binding sitesare concentrated there of the peptide could be seen(Colmers et al., 1987, 1988). At (Saria et al., 1985; Martel et al., 1990). The DR is therefore a sympathetic neuroeffector junctions, NPY inhibits release of good system in which to test the hypothesis that NPY acts at noradrenaline (NA), and of itself, from presynaptic sympathetic presynaptic sitesto regulate synaptic transmission. The results terminals (Lundbergand Tatemoto, 1982;Lundberg and Stj&ne, indicate that NPY and related peptides selectively inhibit slow 1984; Lundberg et al., 1984; Lundberg and Hakfelt, 1986; Wahl- synaptic responsesmediated by 5-HT and NA, by an action at estedtet al., 1990). Here, however, it also has direct and indirect a presynaptic site, while not affecting the rapid synaptic re- postsynaptic actions; high concentrations of NPY directly cause sponsesmediated by amino acid release.The peptide could, by its combined presynaptic actions, induce a change in state of the DR neurons similar to that induced in hippocampal pyra- Received July 1, 1991; rexised Sept. 25, 1991; accepted Oct. 23, 1991. We are grateful to Dr. Alain Fournier of the INRS Sant6, Pointe-Claire, Quest%, midal neuronsby the postsynaptic action of NA. for his generous gifts of PYY and NPY 13-36; to Dr. John Krstenansky (Merrell Dow R&arch Ir&tute, Cincinnati, OH) for his generous gift of CZ-NPY, and to Materials and Methods Sandoz Canada Ltd. for the gift of cyanopindolol. We thank Dr. J. T. Williams for helpful discussions throughout the course of the experiments. This work was In vitro slicepreparation. Methodsused were essentially similar to those supported by the Medical Research Council of Canada. W.F.C. is an Alberta reportedearlier (Williams et al., 1988).Briefly, maleSprague-Dawley Heritage for Medical Research Scholar; S.B.K. is a recipient of AHFMR and MRC rats(200-350 gm) werelightly anesthetizedwith halothaneand killed (Canada) studentships. by a heavyblow to the spinalcord. The brain wasthen quickly removed Correspondence should be addressed to Dr. William F. Colmers, Department and placedin ice-cold,carbogenated (95% 0,, 5OhCO,), artificial ce- of Pharmacology, University of Alberta, Edmonton, Alberta, Canada T6G 2H7. rebrospinalfluid (ACSF, pH 7.35). Coronalsections (350 pm thick) Copyright 0 1992 Society for Neuroscience 0270-6474/92/121086-08$05.00/O werethen cut with a vibratomefrom a blockof tissuecontaining caudal The Journal of Neuroscience, March 1992, 12(3) 1087 regions of the DR nucleus in cold (4°C) carbogenated ACSF. The slices were immediately placed in a solution of carbogenated ACSF and in- cubated at 34°C for a minimum of 1 hr. A slice containing caudal regions Slow EPSP of the DR was then transferred to a recording chamber where it was submerged and continuously superfused at a rate of 2.5-3 ml/min with carbogenated ACSF, which- had been prewarmed to 34 + 0.2”C. The comnosition of the ACSF was fin mM) NaCl. 126: KCl. 2.5: NaH,PO,. 1.2; MgCl,, 1.2; CaCl,, 2.4; &glucose, 11: NaHCO, 25: To evoke synaptic potentials, a bipolar tungsten stimulating electrode, connected to a stimulus isolating unit, was placed in, or adjacent to, the DR. Intracellular recording. Neurons were impaled with glass microelec- trodes filled with 2 M KC1 (50-120 Mfi), connected to the headstage of IPSP an Axoclamp 2A amplifier. Only neurons with the characteristic com- I / plex synaptic responses, and that responded to the application of the w 5-HT, agonist 5-carboxamidotryptamine (5-CT) with an inwardly rectifying potassium conductance, were studied. Synaptic potentials were b recorded in bridge current-clamp mode upon stimulation via the bipolar I Control electrode. Stimuli (1 MO V, l-2 msec) were presented either singly or in pairs (interstimulus interval, 20 msec). In the presence of neurotrans- mitter agonists, current ofthe appropriate sign was applied to the neuron via the balanced bridge circuit to keep it at the original membrane potential (l’,,,). The fast, amino acid-dependent, depolarizing synaptic potentials (DSPs; Pan and Williams, 1989a) were blocked with a cocktail containing 10 PM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), 50 PM m-2-amino-5-phosphonovaleric acid (APV), and 50 PM picrotoxin. Slow EPSPs (sEPSPs) were blocked with prazosin (100 nM), and IPSPs were blocked with cyanopindolol (1 PM), or reduced with m-trifluoro- methylphenylpiperazine (TFMPP, 1 PM). For voltage-clamp experiments, the discontinuous single-electrode voltage-clamp technique (switching frequency, 3.4-4.2 kHz) was used; electrodes were coated with Sylgard (Dow-Coming) to reduce capacitance. Headstage voltage was continuously monitored to ensure proper voltage clamp; headstage voltage pulses had to be square, and electrode settling after the current passing phase had to be complete prior to voltage sampling for the clamp to be considered acceptable. Neurons were routinely held near their resting membrane potential at -60 mV (I’*). For ramp experi- kd Recovery ments, the voltage was slowly (2 set) ramped from -60 mV to - 125 mV and then held for 2 set to stabilize the membrane before ramping gradually (-20 set) to -40 mV. In a few experiments, the voltage was ramped from - 115 mV to -40 mV. Ramps were digital averages of two successive responses taken 30 set apart. In some experiments, volt- Control age steps of 1 set duration were applied to potentials both negative and positive to Vn. Data were corrected for any electrode offset observed at the end of an experiment. Digital averages of three successive synaptic responses, taken at 60 set intervals, were used for figures and for data analysis. The maximum amplitude of each synaptic response was taken as the magnitude of that response. W 10 mV All data were stored on computer for on-line analysis using the ~CLAMP Control program (Axon Instruments) or digitally on videotape for playback and I off-line analysis using a Nicolet 4094 digital oscilloscope. Current and moms 4s voltage data were recorded continuously during an experiment with a pen recorder (Gould Brush 2200). For some figures, current records Figure 1. NPY and PYY inhibit only slow synaptic potentials in the were digitally subtracted using AXUM software (TriMetrix, Seattle, WA). rat DR nucleus. a, Complex synaptic potential evoked in a DR nucleus Data are expressed as percentage inhibition of control values; averages neuron by focal electrical stimulation. See text for detailed description. are presented as means f SEM. Preparations served as their own con- Digital sampling rate was slowed eightfold at the center of the trace to trols; all data are from preparations that showed significant recovery capture the decay of the sEPSP. b, Superimposed traces of synaptic upon washout. Statistical comparisons were performed using a Student’s potentials in control and after application of 1 PM NPY, the peptide t test. Statistical differences were considered significant at p IS 0.05. inhibited both of the slow components. c, NPY trace from b is super- Drugs were dissolved in warmed, carbogenated ACSF just prior to imposed on trace taken after 30 min wash, demonstrating recovery. d, use and applied via bath perfusion. NPY, peptide YY (PYY), and NPY PYY (1 PM) inhibits the slow synaptic potentials of the cell in b and c 13-36 were obtained from INRS, Pointe-Claire, Quebec, and dissolved (recovery not illustrated). The peptides were applied by superfusion for in ACSF in a final volume of 10 ml mior to use. CZ-NPY was a gift of 3-4 min; maximal peptide effect was observed 5-l 0 min following the Dr. J. L. Krstenansky (Metrell Dow). -Cyanopindolol was a gift of Sandoz start of washout. (Canada). TFMPP was obtained from RBI (Natick, MA). All other drugs were obtained from Sigma (St. Louis, MO).

Results 1989a,b), as well as to antagonistsof the GABA, receptors(the NPY/PYY reduceslow but not fast synaptic potentials intracellular electrodesused in theseexperiments contained KCl, Focal electrical stimulation of DR elicits a complex synaptic which shifted the chloride equilibrium potential positive to rest, potential in DR serotonergic neuronsthat consistsof fast and thus producing a depolarizing response;Williams et al., 1988). slow components as shown in Figure la. A fast DSP is first The slow componentsconsist of a potassium-dependentIPSP, elicited that is sensitive to antagonistsof quisqualate/kainate mediated by S-HT,, autoreceptors (Yoshimura and Higashi, and NMDA excitatory amino acid receptors(Pan and Williams, 1985; Pan and Williams, 1989b; Bobker and Williams, 1990), 1088 Kombian and Colmers * NPY Inhibits Slow Postsynaptic Potentials in Dorsal Raphe

a I

b I 8 ms (I I

Figure 2. NPY and PYY have no effect on amino acid-mediated synaptic responses in the rat DR nucleus. Superimposed are the DSP evoked in control and in the presence of NPY (1 PM). NPY has no effect on the DSP, while it inhibits the IPSP.

followed by a sEPSP, mediated by a,-adrenoceptors (Yoshimura et al., 1985; Pan and Williams, 1989). Bath application of 1 PM NPY caused a reduction both in the IPSP (by 33.3 f 4.4%; n = 10, p < O.OOOl), and the sEPSP (by 40.2 f 7.8%; n = 11, p < 0.00 1; Fig. lb). This inhibition was slow (5-10 min) in onset and reversed upon prolonged washout (Fig. 1c). All data reported are from preparations demonstrating I , substantial (> 90%) recovery from peptide effect. The NPY an- l alog, PYY (1 PM), which shares 70% sequence homology with, 500 ms 4 s and a high degree of analogy to, NPY (Tatemoto, 1982), also Figure 3. PYY inhibits isolatedslow synapticpotentials in the DR reversibly inhibited the IPSP by 45.4 f 4.0% (n = 9, p < 0.00 l), nucleus.(I, Superimposedare the pharmacologicallyisolated IPSP in and the sEPSP by 31.2 f 6.0% (n = 6, p < 0.005) (Fig. Id). control (a cocktailof the aminoacid recentor antaaonists CNOX/APV/ picrotoxin, andprazosin, to blockthe sEPSP)and ifter additionof PW Because the effects of NPY and PYY on slow synaptic responses (1 FM). b, The sEPSP was isolated pharmacologically with the amino were statistically indistinguishable, they were used interchang- acid receptor antagonist cocktail and cyanopindolol, to block the IPSP. ably in all following experiments. By contrast, neither NPY nor The isolated sEPSP is also inhibited by PYY (1 PM). c, PYY inhibits PYY had an effect on the fast synaptic component (n = 11, p the IPSP and sEPSP in the presence of 1 PM TFMPP, a 5-HT,, receptor antagonist. > 0.5; Fig. 2).

NPYIPYY inhibit isolatedslow synaptic responses we examined the action of the peptide in the presenceof a high To determine if the slow componentsof the complex synaptic concentration of the 5-HT,,-selective agonist TFMPP (1 PM). responsewere modulated individually, we isolatedthe slow syn- TFMPP caused an inhibition of evoked, 5-HT-mediated re- aptic responsesby pharmacologicalmeans and then studiedthe sponses(Bobker and Williams, 1990a,b), reducing the IPSP by effect of NPY/PYY. The DSP was blocked with a cocktail con- 61.7 f 9.8% (n = 5). The IPSP remaining after TFMPP treat- taining supramaximal concentrations of antagonists to quis- ment was reversibly reduced by PYY (by 36.3 + 6.1%; n = 4, qualate/kainate and NMDA receptorsand a GABA, (chloride) p < 0.01; Fig. 3~). As well, PYY reduced the sEPSP in the channel blocker (Pan and Williams, 1989a; seeMaterials and presenceof TFMPP (and the amino acid antagonist cocktail) Methods). Additionally, the sEPSP was completely eliminated by 21.5 f 4.2% (n = 3, p < 0.025; not illustrated). in the presenceof 100 nM prazosin, a selective a,-adrenoceptor antagonist (Pan and Williams, 1989b, Bobker and Williams, NPYIPYY have no effect on postsynaptic conductances 1990b). PYY reduced the resulting isolated IPSP by 22.0 & To determine the locus of the peptides’ action on synaptic re- 4.2% (n = 3, p < 0.025; Fig. 3a), indicating that the IPSP sponses,we applied single-microelectrodevoltage clamp to DR inhibition was direct. neurons(Finkel and Redman, 1984; Williams et al., 1988; Pan Becausethe 5-HT,, antagonist spiperone also blocks cy,-ad- and Williams, 1989)to test whether the peptidesaffected voltage renoceptors(Pan and Williams, 1989a),we explored other phar- and ligand-induced conductances.Application of 1 PM NPY or macological means of isolating the sEPSP. Blockade of both PYY alone did not affect the resting membrane conductance of 5-HT,, and 5-HT,, receptorswith 1 MM cyanopindolol (Bobker DR neurons, nor did it affect steady-state voltage-dependent and Williams, 1990b)in the presenceof the amino acid antag- conductanceselicited by a slow (= 20 set), steady voltage ramp onist cocktail completely isolatedthe sEPSP.In the presenceof from - 115 to -40 mV. At the same time, the slow synaptic cyanopindolol, PYY reversibly inhibited the sEPSP by 35.0 f potentials were inhibited (Fig. 4). 8.6% (n = 3, p < 0.025; Fig. 3b). Having seenno effect on resting membraneconductances, we To determine whether inhibition of the IPSP by NPY/PYY next determined whether the peptides modulated ligand-in- receptorsoccurred via the same mechanismas the presynaptic duced conductances,as has been postulated for NPY in locus 5-HT,, receptor that inhibits releaseof 5-HT in DR (Bobker coeruleusneurons (Illes and Regenold, 1990). Selective agonists and Williams, 1990a;J. T. Williams, personalcommunication), were applied to activate either the postsynaptic 5-HT receptors The Journal of Neuroscience, March 1992, 12(3) 1089

10m”l 1 zdn I I I

-60 mV voltage

PYY effect

1 I -100 -70 -100 -70 Voltage (mv) Voltage (mV)

Figure 4. PYY at 1 MM affects neither resting nor voltage-activated membrane conductances. a, Chart record of the resting membrane current in a DR neuron held near rest in voltage clamp. A slow voltage ramp protocol (from - 115 to -40 mv) was applied; voltage range was corrected for electrode offset. Application of 1 PM PYY does not alter the resting membrane current. b,, Superimposed are the steady-state Z-V relationship in the absence (control, 0) and in the presence of 1 PM PYY (r). Inset b2, Superimposed are the synaptic potentials evoked in control (0) and in the presence of 1 PM PYY 0, showing the inhibition of both the IPSP and sEPSP in the DR neuron in a. Inset b,, Digital subtraction of the steady- state Z-V relationships in b,. or (Y,-adrenoceptors; we then examined whether coapplication cells were held near rest in voltage clamp ( V, = - 65 mv), 5-CT of the peptide altered the ligand-induced changesin membrane elicited an outward current, which rectified inwardly (Fig. 5aJ conductance, while at the same time examining the effects of as previously reported (Williams et al., 1988). Voltage stepcom- the peptides on synaptic potentials. mands (1 set) were also applied to different potentials (Fig. 5~~); Application of 100 n.~ 5-CT, a full agonist at the postsynaptic no significant differenceswere observed in current values mea- 5-HT,, receptors (Hoyer, 1988; Williams et al., 1988; Pan et sured at the end of a 1 set voltage step and during the slow al., 1989; Andrade and Chaput, 199 l), hyperpolarized DR neu- voltage ramp (Fig. 5a,), indicating that the ramp measuredsteady- rons in current clamp (Williams et al., 1988). This was accom- state membrane conductances(Williams et al., 1988). Appli- panied by a marked reduction of the IPSP, the sEPSPwas short- cation of PYY (1 PM) in the presenceof 5-CT had no effect on ened but not greatly reduced in amplitude (Fig. 5aJ. When the the time-dependent (Fig. 5b,,,) or steady-state membrane con- 1090 Kombian and Colmers l NPY Inhibits Slow Postsynaptic Potentials in Dorsal Raphe

3 40 mV

-Lr -130mv

3 0.0

9 5-CT effect 2 -0.2 -

i? u3 6.4 -

-0.6 I I I -100 -70 -40 -100 -70 -40 Voltage (mV) Voltage (mV)

k- i-! 1s 4

I I , I I -0.6 L I -0.50 _- Ifi *n -130 -100 -70 40 -130 -100 --IV -L)u Voltage (mV) Voltage (mV) Figure 5. PYY at 1 PM does not alter the 5-HT,, receptor-mediated membrane conductance while inhibiting the sEPSP in DR neurons. a,, Application of 100 ILM S-CT causes a hyperpolarization or outward current in DR neurons, with a marked reduction of the IPSP. Superimposed are the steady-state current-voltage (Z-F’) relationship in a DR neuron obtained by application of a slow depolarizing voltage ramp from - 125 to -40 mV (the voltage range has been corrected for electrode offset). Symbols (control, 0; S-CT, r) represent the values of membrane currents measured at the end of 1 set voltage-clamp steps to different potentials (inseta,). Inset a,, Synaptic potentials, evoked at the same membrane potential in control and in the presence of 5-CT. The IPSP is reduced by 5-CT, the sEPSP is curtailed. The neuron was depolarized with current to the same potential as control in 5-CT to compare synaptic potentials. Inset a,, Digital subtraction of the ramp currents in u,, showing the inwardly rectifying conductance induced by 5-CT. b,, Steady-state I-Vrelationship of the neuron in a in the presence of 5-CT alone (7) superimposed on the Z-V relationship in the presence of 5-CT with 1 PM PYY (B). Insetbp, The synaptic potentials evoked in 5-CT alone and in 5-CT with PYY showing the inhibition of the sEPSP. b,, digital subtraction of the slow ramp Z-V relationships in b,. The Journal of Neuroscience, March 1992, f2(3) 1091

-0.50 -0.5 I I I -130 -100 -70 -40 -130 -100 -70 40 Voltage (mV) Voltage (mV) bl 0.25 -m mV

- 0.00 1 pMPEr 3 1 3 3 0.1 t: PE + 1 pM PYYm 3 i; -0.25 9 -0.1

5 4.3 5 CJ

-0.5 I I I -0.50 - -130 -100 -70 -40 -130 -100 -70 -40 Voltage (mV) Voltage (mV) Figure 6. PYY at 1 PM doesnot alter cr,-adrenoceptor-mediatedmembrane conductance changes, while inhibiting the IPSP in a DR neuron.a, The q-agonistPE (1 PM)evokes a steady-stateinward current in voltage-clampedDR neurons.a,, The steady-stateZ-Y curvesin the absenceof PE (control,0) and in the presenceof 1 PM PE (v) are shownsuperimposed. Symbols represent the membranecurrent values from voltagesteps as in Figure 4. Inset a,, the synapticpotentials, evoked at the samemembrane potential in control and in the presenceof PE. The neuronwas hyperpolarizedto the control potentialfor comparisonof synapticpotentials. The sEPSPis occludedby the applicationof PE. Inset a,, Digital subtractionof the steady-stateZ-V curvesin a,. Zr,Steady state Z-V curves in PE (V), and in PE with 1 PM PYY m. Insets asin a. AlthoughPYY hasno effect on the membranecurrents, the IPSP evokedin currentclamp is inhibited. ductances(Fig. 5b,,,) over the voltage rangeexamined. However, ylephrine (PE, 1 PM; Baraban and Aghajanian, 1989) induced the sEPSPrecorded in current clamp was reversibly inhibited a steady-state inward current near rest (Fig. 6a,,& as well as by 37.95 f 7.2% (n = 5, p < 0.005; Fig. 5bJ. appearingto reduce an outward current in the depolarizing re- In analogousexperiments, the a,-adrenoceptor agonist phen- gion of the current-voltage relationship (Fig. ~cz,,~;Pan and Wil- 1092 Kombian and Colmers - NPY Inhibits Slow Postsynaptic Potentials in Dorsal Raphe

control p < 0.01; Fig. 7b). The inhibition of both slow synaptic potentials by C2-NPY was reversible (not illustrated).

Discussion In this study, we have demonstratedthat NPY and PYY inhibit both hyperpolarizing and depolarizing slow synaptic potentials in DR, while not measurably affecting the fast synaptic potentials. This observation is in contrast to their previously reported NPY, -+f control effect in the hippocampus (Colmers et al., 1987, 1991). These actions are presynaptic since there was no effect on resting, voltage-activated, or agonist-induced membrane conductances. As well, there was no effect on the fast, amino acid-mediated synaptic responses,consistent with a selective presynaptic action at serotonergic and noradrenergic terminals alone. b I The actions of NPY/PYY are mimicked consistently, though control less potently, by the carboxy-terminal NPY 13-36 fragment. li The Y, receptor-selective ligand C2-NPY is an agonist as well. These data suggestthat the actions of NPY/PYY may be me- /” diated by the activation of a presynaptic Y, receptor, as has c2-NPY + already been demonstratedin the hippocampus (Colmerset al., 4 \ I99 1) and elsewhere(Wahlestedt et al., 1986). In most systems, c2-NPY -A7 10mv the agonist fragment is less potent than the intact peptide; in control 1 rat hippocampus NPY 13-36 is about 50% as potent as NPY 5ooms4s (Colmers et al., 199 1). Interestingly, the peptidergic inhibition Figure 7. Y, receptor agonists inhibit slow synaptic potentials in the of the IPSP was reduced after blockade of cu,-adrenoceptors, DR. a, Synaptic potentials evoked in a DR neuron in control and after suggestingthat a portion of the IPSP was facilitated by norad- application of 1 PM NPY 13-36 are shown superimposed. b, Synaptic renergic excitation of neighboring DR neurons. potentials in a different DR neuron in control and after application of Although this study did not addressin detail the mechanism C2-NPY. Membrane potential was the same in control and peptide for each cell. Responses recovered upon washout. of presynaptic action by NPY/PYY, it seemsclear that the peptidergic inhibition of the IPSP appearsto be independent of the presynaptic 5-HT,, receptor and its mechanismof action, liams, 1989), correspondingto a depolarization in current clamp; as PYY inhibited the IPSP remaining after a supramaximal the sEPSPwas also occluded (Fig. 6a2). Application of PYY (1 concentration of the 5-HT,, agonistTFMPP wasapplied. There FM) did not significantly alter the steady-state membrane con- is evidence in hippocampus (Colmers et al., 1988) and in cul- ductance in the presenceof PE (Fig. Sb,.,), but the isolatedIPSP, tured dorsal root ganglion neurons(Bleakman et al., 1991) that recorded in current clamp, was inhibited by 3 1.O + 9.9% (n = Y, receptors inhibit Ca2+ influx. However, the mechanismof 4, p < 0.005; Fig. 6bJ. Y, presynaptic action in DR will need to be elucidated in detail, perhapsaided by the likelihood that there are two presynaptic Synaptic effectsof NPYIPYY are mimicked by Y, receptor mechanismsby which the IPSP can be inhibited. agonist The present results indicate that NPY and PYY modulate Two receptors for NPY/PYY have been identified, which can slow synaptic transmissionin DR by a presynaptic mechanism. be differentiated on the basis of agonist fragments of NPY or This meansthat at central, as well as peripheral, synapses(Wahl- PYY. The Y, receptors require the entire, intact peptide, while estedt et al., 1986) NPY can inhibit the release of NA. The the Y, receptorscan be activated by amidated C-terminal pep- observed inhibition by NPY of a 5-HT-induced synaptic re- tide fragmentsas short as NPY 13-36 (Wahlestedt et al., 1986). sponse(probably through inhibition of release)is a novel find- Becausepresynaptic inhibition mediated by NPY at peripheral ing. Data from microdialysis of hypothalamic nuclei with NPY (Wahlestedt et al., 1986)and central mammalian synapses(Col- show a reduction in levels of NA, 5-HT, and their metabolites merset al., 1991) has been shown to be mediated by Y, recep- (Shimizu and Bray, 1989), suggestinga presynaptic inhibitory tors, we examined the effect of NPY 13-36 on slow synaptic action, as we have observed here. Although NPY has been re- responsesin DR to determine whether this effect was also me- ported to modulate the NA response of the locus coeruleus diated by a Y, receptor. A 1 PM concentration of NPY 13-36 neurons via a postsynaptic mechanism (Illes and Regenold, reversibly inhibited IPSP and sEPSP by 15.88 + 1% (n = 3, p 1990), we observed no such interaction in the DR. < 0.005) and 26.58 f 10.82% (n = 2, p < 0.25), respectively The inhibition by NPY of both slow inhibitory and excitatory (Fig. 7a). Although the inhibition of the sEPSPwas not statis- synaptic inputs to DR neurons appearscontradictory. There is, tically significant, NPY 13-36 clearly demonstrateda reversible however, a potential explanation for this apparent contradic- inhibition of this response. tion. Noradrenergic tone is required for the tonic activity of DR To examine further the question of receptor subtype, we ex- neurons observed in vivo (Baraban and Aghajanian, 1980). Re- amined whether the recently developed Y, receptor ligand [Cys2, leaseof NPY onto its receptors in DR would be expected to 8-aminooctanoicacid5J4, D-C~@NPY (C2-NPY, MDL 29,6 16; reduce this activity. At the same time, NPY would be expected McLean et al., 1990) affected slow synaptic potentials in DR. to reduce the 5-HT,, receptor-mediated inhibition of the DR Application of C2-NPY (1 PM) inhibited the IPSP by 33.2 f 5-HT cells. Becausethe action of NPY is entirely presynaptic, 0.5% (n = 2, p < O.Ol), and the sEPSPby 47.1 f 0.8% (n = 2, other mediating or modulating influences on the postsynaptic The Journal of Neuroscience, March 1992. U(3) 1093

membrane will remain unaltered. We suggest that the release grade wheat germ agglutinin and D-[‘HI-aspartate tracing. Brain Res of NPY could therefore induce a state of “quiet readiness” in 360:285-297. Kalen P, Pritzel M, Nieoullon A, Wiklund L (1986) Further evidence DR neurons, where the tonic level of activity is lowered, but for excitatory amino acid transmission in the lateral habemular pro- that the response to suprathreshold stimuli (such as from the jection to the rostra1 raphe nuclei: lesion-induced decrease of high habenular excitatory amino acid inputs; Ralen et al., 1985, 1986) affinity glutamate uptake. Neurosci Lett 68:35-40. Of will be enhanced, because the activity-dependent autoinhibi- Lundberg I~ JM. , Hijkfelt T (1986)~ I Multinle co-existence DeDtideS_ _ and classical transmitters in peripheral autonomic and sensory neurons- tion, caused by the release of SHT, is attenuated. A similar functional and pharmacological implication. In: Progress in brain change of state is induced by NA in hippocampal pyramidal research, Vol 68 (Hiikfelt T, Fuxe K, Pemow B, eds), pp 241-262. neurons, albeit by a postsynaptic mechanism (Madison and Ni- Amsterdam: Elsevier. ~011, 1982). Lundberg JM, Stjtie L (1984) Neuropeptide Y (NPY) depresses the secretion of ‘H-noradrenaline and the contractile response evoked by field stimulation in rat vas deferens. Acta Phvsiol Stand 120:477- 479. References Lundberg JM, Tatemoto K (1982) Pancreatic polypeptide family (APP, Aghajanian GK, Wang RY (1977) Habemular and other midbrain BPP. NPY. and PYY) in relation to cY-adrenocentor-resistant svm- raphe afferents demonstrated by a modified retrograde tracing tech- pathetic vasoconsttiction. Acta Physiol Stand 1 l&393-403. . nique. Brain Res 122~229-242. Lundberg JM, Hua XY, France-Cereceda A (1984) Effects of neuro- Aitken A, Tork I (1988) Early development of serotonin-containing peptide Y (NPY) on mechanical activity and neurotransmission in neurons and pathways as seen in wholemount of fetal rat brains. J the heart, vas deferens and urinary bladder of the guinea-pig. Acta Comp Nemo1 274132-47. Physiol Stand 121:325-332. Anden NE, Dahlstrom A, Fuxe K, Larsson K, Olson L, Ungerstedt U Madison DV, Nicoll RA ( 1982) Noradrenaline blocks accommodation (1966) Ascending monoamine neurons to the telencephalon and di- of pyramidal cell discharge in hippocampus. Nature 299:636-638. encephalon. ActaPhysiol Stand 67:3 13-326. Mattel J, Foumier A, St Pierre S, Quirion R (1990) Quantitative Andrade R. Chanut Y (199 1) The electronhvsioloev of serotonin re- autoradiographic distribution of [lz51] Bolton-Hunter neuropeptide Y ceptor subtypes. In: Serotohin receptor subiypes:%asic and clinical receptor binding sites in rat brain. Comparison with [1251]-peptide YY aspects (Peroutka SJ, ed), pp 103-124. New York: Wiley-Liss. receptor sites. Neuroscience 36:255-283. Azmitia EC, Segal M (1978) An autoradiographic analysis of the dif- McLean LR, Buck SH, Krstenansky JL (1990) Examination of the ferential ascending projections of the dorsal and median raphe nuclei role of the amphipathic a-helix in the interaction of neuropeptide Y in the rat. J Comp Neurol 179:641-668. and active cyclic analogues with cell membrane receptors and di- Baraban JM, Aghajanian GK (1980) Suppression of firing activity of myristoylphosphatidylcholine. Biochemistry 29:20 16-2022. 5-HT neurons in the dorsal raphe by alpha-adrenoceptor antagonists. Mulligan KA, Tork I (1987) Serotonergic axons form basketlike ter- Neuropharmacology 19:355-363. minals in cerebral cortex. Neurosci Lett 8 1:7-l 2. Bartfai T, Iverfeldt K, Fisone G, Serf&ii P (1988) Regulation of the Mulligan KA, Tork I (1988) Serotonergic innervation of the cat ce- release of coexisting neurotransmitters. Ann Rev Pharmacol Toxic01 rebral cortex. J Comp Neurol270:86-110. 28:285-3 10. O’Donoghue TL, Chronwall BM, Pruss RM, Mezey E, Kiss JZ, Eiden Bobker DH, Williams JT (199Oa) Serotonin-mediated inhibitory post- LE, Massari VJ, Tessel RE, Pickel VM, DiMaggio DA, Hotchkiss AJ, synaptic potentials in guinea pig prepositus hypoglossi and feedback Crowley WR, Zukowska-Grojec Z (1985) Neuropeptide Y and pep- inhibition by serotonin. J Physiol (Land) 422:447462. tide YY neuronal and endocrine systems. Peptides 6:755-768. Bobker DH, Williams JT (1990b) Ion conductances affected by 5-HT Pan ZZ, Williams JT (1989a) GABA- and glutamate-mediated syn- receptor subtypes in mammalian neurons. Trends Neurosci 13: 169- aptic potentials in rat dorsal raphe nucleus in vitro. J Neurophysiol 173. 61:719-726. Chazal G, Ralston JH III (1987) Serotonin-containing structures in Pan ZZ, Williams JT (1989b) Differential actions of cocaine and am- the nucZeus ruphe dorsalis of the cat: an ultrastructural analysis of phetamine on dorsal raphe neurons in vitro. J Pharmacol Exp Ther dendrites, presynaptic dendrites and axon terminals. J Comp Neurol 25 1:56-62. 259:327-329. Pan ZZ, Colmers WF, Williams JT (1989) 5-HT mediated synaptic Colmers WF, Lukowiak KD, Pittman QJ (1987) Presynaptic action potentials in the dorsal raphe nucleus: interactions with excitatory of neuropeptide Y in CA1 of rat hippocampal slice. J Physiol (Lond) amino acid and GABA neurotransmission. J Neurophysiol 62:48 l- 383:285-299. 486. Colmers WF, Lukowiak KD, Pittman QJ (1988) Neuropeptide K Pemow J, Saria A, Lundberg JM (1986) Mechanisms underlying pre- action in the rat hippocampal slice: site and mechanism ofpresynaptic and post-junctional effects of neuropeptide Y in sympathetic vascular inhibition. J Neurosci 8:3827-3837. control. Acta Physiol Stand 126:239-249. Colmers WF, Klapstein GJ, Foumier A, St-Pierre S, Treheme KA Saria A, Theodorsson-Norheim E, Lundberg JM (1985) Evidence for (199 1) Presynaptic inhibition by neuropeptide Y in rat hippocampal specific neuropeptide Y-binding sites in rat brain synaptosomes. Eur slice in vitro is mediated by a Y, receptor. Br J Pharmacol 120:41- J Pharmacol 107:105-107. 44. Shimizu H, Bray GA (1989) Effects of neuropeptide Y on noradren- de Quidt ME, Emson PC (1986) Distribution of neuropeptide Y-like aline and serotonin metabolism in rat hypothalamus in vivo. Brain immunoreactivity in the rat central nervous system. II. Immunohis- Res Bull 221945-950. tochemical analysis. Neuroscience 18:545-6 18. Tatemoto K (1982) Neuropeptide Y: complete amino acid sequence Ekblad E, Edvinson L, Wahlestedt C, Uddman R, HIkanson R, Sundler of the brain peptide. Proc Natl Acad Sci USA 79:5485-5489. F ( 1984) Neuropeptide Y co-exists and co-operates with noradren- Wahlestedt C, Yanaihara N, H&tnson R (1986) Evidence for pre- aline in perivascular nerve fibers. Regul Pept 8:225-235. and postjunctional receptors for NPY and related peptides. Regul Finkel AS, Redman S (1984) Theory and operation of single micro- Pept 13:307-318. electrode voltage clamp. J Neurosci Methods 11: 101-127. Wahlestedt C, Grundemar L, Htianson R, Heilig M. Shen GH, Zu- Fuxe K, Jonsson G (1974) Further mapping of central 5-hydroxy- kowska-Grojec Z, Reis DJ (1990) Neuropepcde k receptor’sub- tryptamine neurons: studies with the neurotoxic dihydroxytrypt- types. Y, and Y,. Ann NY Acad Sci 611:7-26. amine. In: Advances in biochemical psychopharmacology, Vol 10, Wiiliams JT, Colmers WF, Pan ZZ (1988) Voltage and ligand-acti- Serotonin: new vistas, histochemistry and pharmacology. (Costa E, vated inwardly rectifying currents in dorsal raphe in vitro. J Neurosci Gessa GL, Sandler M, eds), pp 1-12. New York: Raven. 8:3499-3506. Hoyer D (1988) Functional correlates of serotonin 5-HT, recognition Yoshimura M, Higashi H (1985) 5-Hydroxytryptamine mediates in- sites. J Recept Res 8:59-8 1. hibitory postsynaptic potentials in rat dorsal raphe neurons. Neurosci Illes P, Regenold JT (1990) Interaction between neuropeptide Y and Lett 53167-74. noradrenaline on central catecholamine neurons. Nature 344:62-63. Yoshimura M, Higashi H, Nishi S (1985) Noradrenaline mediates Kalen P, Karlson M, Wiklund L (1985) Possible excitatory amino acid slow excitatory synaptic potential in rat dorsal raphe neurons in vitro. afferents to nucleus raphe dorsalis of the rat investigated with retro- Neurosci Lett 6 1:305-3 10.