Actions of Neurotensin: A Review of the Electrophysiological Studies" WEI-XING SHI AND BENJAMIN S. BUNNEY Departments of Psychiatry and Pharmacology Yale University School of Medicine New Haven, Connecticut 06510

INTRODUCTION Neurotensin (NT) is widely and heterogeneously distributed in both the central and the peripheral nervous system and is believed to act as a neurotransmitter or neuromodulator. High-affinity NT receptors, recently suggested to belong to the family of G protein-coupled receptors, I are localized throughout the nervous system. Their activation modulates a variety of cellular biochemical processes. Centrally administered NT profoundly affects various animal behaviors. This review will focus on electrophysiological actions of NT on single cells in both the central and peripheral nervous system. The effects of NT on the cells in various areas studied are summarized in TABLE1. It should be emphasized that these effects were all evoked by exogenously applied NT. They may reflect the true actions of NT at its synaptic sites. However, it will not be surprising if the effects of synaptically released NT turn out to be more complicated. Due to the unavailability of a NT receptor antagonist, the question of whether the observed effect is mediated by an action on NT receptors can only be tested by using various NT fragments that have been shown to bind or not to bind NT receptors. If the effects of these fragments correlate with their binding affinity, it is likely that the observed effect is NT receptor-mediated. In many studies, these fragments were not tested, and it is therefore not clear whether the observed effects in those areas were mediated by NT receptors. Since the target areas for NT terminals are much broader than those listed, the complete range of NT action remains to be defined. Among the areas studied, the effects of NT on midbrain dopamine (DA) neurons in both the substantia nigra and the ventral tegmental area have been most extensively examined. NT is shown to directly affect DA cell firing and modulate DA autoreceptormediated effects. Diverse effects of NT were also reported in other areas. In this review, we will first discuss how NT produces some of its effects on midbrain DA cells and then describe the effects of NT in other brain areas from cortex to caudal areas and in the periphery. 0 This work was supported by U.S. Public Health Service Grants MH28849 and MH25642, the Stanley Foundation for Research on Serious Mental Diseases and the Abraham Ribicoff Research Facilities at the Connecticut Mental Health Center, State of Connecticut.

129

ANNALS NEW YORK ACADEMY OF SCIENCES

130

TABLE1. Electrophysiological Effects of Neurotensin Target Area Cortex Prefrontal Sensory-motor Nucleus accumbens Bed nucleus of stria terminals Hypothalamus Preoptic-anterior

Arcuate nucleus

Midbrain DA neurons

Locus ceruleus Nucleus tractus solitarius Periaqueductal gray Cerebellum Purkinje cell

Spinal cord Motoneurons Interneurons Dorsal cells Preganglionic cells Retina Amacrine cells Ganglion cells Mesenteric ganglion Myenteric ganglion NG 108-15

Oocyte with rat mRNAs

Effect

Mediated Direct by NT or Receptors" Indirect

Excitation Excitation Inhibition Excitation

Likely ? ? ?

Direct ?

Excitation Inhibition

Likely ?

Direct Indirect?

Modulation Excitation Inhibition

? ? ?

Both Indirect?

Modulation

Likely

Excitation

Likely

Likely direct Direct

Inhibition Inhibition Excitation

? ? ?

? ? ?

Excitation

?

Direct

Excitation Inhibition

? ?

Direct? Indirect

Excitation Excitation Excitation Modulation Excitation Inhibition

? ? ? ? ? ?

Indirect? ? ?

Excitation Excitation Excitation

? ? Likely

? ?

Both

Excitation Inhibition Excitation

? ? ?

Direct ? ?

Inhibition 1 a,

? ?

?

?

Possible Mechanisms and Other Comments 1

K + conductance

DPI ?

Direct

?

? ? ?

DPI at high dose Blocked by zero Ca2+; due to DPI in vivo? t Glutamate-induced excitation DPI at high doses? Blocked by low Ca2+ and high Mg2+ 1 DA-induced inhibition; m a y be due to t CAMP 1 Membrane conductances At high concentrations, DPI? DPI? DPI at high concentrations? No change or slight 1 membrane conductances

NE release from locus NE afferents

References 36 38 39 42

43-44 43-44 43 46 46 12, 17, 18, 20, 23-25 17, 20, 23, 31-33 20, 23, 33 47, 48 57 49, 50

51 39, 51

1 Membrane conductances 1 Membrane conductances 1 Membrane conductances 1 Glutamate-induced excitation

52-55 55 55-57 56 56 56

1 Membrane conductances 1 Membrane conductances 1 Membrane conductances; indirect via Sub P 1 Membrane conductances Blocked in Ca*+-free medium No change or I membrane conductances 1 Membrane conductances G-protein- IIP3- 1Ca2+ 1CI- conductance

59 58, 59 60-62

-

63 63 65 65

I , 68

NOTE:DPI = depolarization inactivation; ? = unclear with available data; 1 = increase; 1 = decrease. Since various NT fragments were not tested in some studies, it is not always clear whether the observed effect was mediated by NT receptors (see INTRODUCTION). (I

SHI & BUNNEY: ACTIONS OF NEUROTENSIN

131

EFFECTS OF NEURCITENSIN ON MIDBRAIN DOPAMINE NEURONS An interaction of NT with midbrain dopamine (DA) neurons has been suggested by two major pieces of evidence. First, central NT systems are anatomically in close proximity to midbrain DA systems. In particular, high concentrations of endogenous NT have been detected by radioimmunoassay within the midbrain substantia nigra and the ventral tegmental area. Within these two regions, NT has been localized, immunohistochemically, to both neuronal perikarya and axon terminals.2.3 Most NTpositive cells have been shown also to contain tyrosine hydroxylase (TH), a marker of DA neurons in these area^.^.^ NT axon terminals form a dense plexus throughout the ventral tegmentum, surrounding DA neurons in both the substantia nigra pars compacts and the ventral tegmental area. Using an electronmicroscopic double-antigen localization technique, NT-containing varicosities identified in the substantia nigra and the ventral tegmental area have been found to be in direct apposition to THimmunopositive perikarya and dendrites.5 In accordance, receptor-binding studies indicate that NT receptors are specifically located on DA neurons in these areas-9 and almost all DA cells contain NT receptors. 10 The second piece of evidence is that centrally administered NT produces many effects that resemble those of DA antagonists. I I One of the possible mechanisms by which centrally administered NT produces its DA antagonist-like effects is that NT directly or indirectly regulates the activity of DA cells located in the midbrain. We tested this hypothesis by recording the activity of midbrain DA neurons while NT was injected intracerebroventrically (i.c.v.). 12 We found that i.c.v. NT had no significant effect on either the basal firing rate or the firing pattern of DA cells recorded, suggesting that the DA antagonist-like effects of i.c.v. NT were not due to reduced DA release secondary to an inhibition of DA neurons. Biochemical studies, in fact, have shown that centrally administered NT increases DA synthesis and release in DA terminal areas.13 Since no increase in firing rate was observed, it is possible that i.c.v. NT exerts its stimulant effect on DA neurons by a mechanism that is independent of the firing rate. For example, i.c.v. NT may preferentially act on DA terminals without affecting the firing rate of the cell bodies located in the midbrain. Despite its ineffectiveness in altering the spontaneous activity of DA neurons, in a subpopulation of DA cells located in the ventral tegmental area, i.c.v. injection of NT markedly attenuated the inhibition induced by quinpirole, the selective D2/D3 DA agonist.12 A similar interaction of NT and DA may occur at postsynaptic sites and may account for the DA-agonist-like effects of NT. Indeed, i.c.v. NT has been shown to reduce the affinity of D2 receptors for its agonisti4and to block the behavioral effects induced by direct injection of DA or DA agonists into DA projection areas.15J6However, since NT was administered i.c.v. and quinpirole i.v., it was not clear whether the attenuation of quinpirole-induced inhibition by NT was due to a direct interaction between these two compounds on DA cells or an indirect one mediated through neurons that project to midbrain DA cells. In order to determine how NT and DA interact locally at DA cell body levels, we microiontophoretically applied NT and DA onto individual DA cells while their activity was being monitored. It was found that in almost all the cells tested, NT significantly attenuated DA-induced inhibition. This modulatory effect of NT could also be demonstrated when DA was replaced with the specific D2/& agonist, quinpirole. I7,I8

132

ANNALS NEW YORK ACADEMY OF SCIENCES

Because both NT receptors and Dz/D3 receptors in this area are selectively localized to DA cells, this result suggests that NT and DA interact directly on DA cells. In addition to its modulatory effect, microiontophoresis of NT also caused a significant increase in the basal firing rate of a few DA cells (see below). It is possible, therefore, that NT may simply attenuate DA-induced inhibition by nonspecific excitation. This possibility was tested in several ways, and the results suggest that this is unlikely.l* First, the effect of DA was measured while NT was ejected at low currents. Although under these conditions the basal firing rate of the cell was not altered by NT, the inhibition induced by DA was still attenuated. Second, the effect of NT was compared to that of glutamate and CCK (a neuropeptide also found in the DA cell area). Although they all had an excitatory effect on DA cells, only NT attenuated DA-induced inhibition. Glutamate did not alter the ability of DA to inhibit these cells; and CCK, in some cells, potentiated the effect of DA.'aI9 NMDA, a glutamate agonist, was also found to be unable to attenuate DA-induced inhibitionz0 (Shi and Bunney, unpublished data). Finally, if NT is acting through a nonspecific mechanism to attenuate the inhibition of DA, then the inhibition induced by other agents such as GABA should also be attenuated. Our results suggest that the effect of NT is selective, NT attenuated DA-induced inhibition without an effect on GABA-induced inhibition. In addition, NT was found to have no effect on glutamate-induced excitation.18 It has been hypothesized that NT may produce some of its neuroleptic-like effects by binding with DA, thereby preventing it from interacting with DA receptors on the cell surface.21*22 This hypothesis was supported by recent studies using voltammetry and ultraviolet/visible spectroscopy,ZI in which NT as well as NT(1-11) and NT(8-13) were found to form a complex with DA, whereas neuromedin N, a neurotensin analogue, did not. We used these same peptides to test whether NT attenuated DA-induced inhibition of DA neurons by forming an inactive complex with DA.23 We found that only the peptides that are capable of binding with the NT receptors mimicked the effect of native NT, whereas those peptides able to form a complex with DA but inactive in binding with NT receptors were inactive in attenuating the inhibition induced by DA. Thus, neuromedin N, which does not form a complex with DA but competes with NT for the same receptor, mimicked the effect of NT. NT(1-ll), which has a higher affinity for DA than NT(8-13)2' but is incapable of binding with NT receptors, did not mimic the effect of NT. NT(8-13), which can bind with NT receptors and form a complex with DA as well, produced effects similar to those of NT. These results suggest that the effects of NT on DA cells are likely to be mediated by an activation of NT receptors, rather than forming a complex with DA. DA receptors on DA neurons have the characteristics of DZreceptor. These receptors are believed to be negatively coupled to adenylate cyclase. Our studies, however, indicate that DA-induced inhibition of DA cells is not mediated solely by the inhibition of adenylate cyclase. Thus, 8-bromo-cAMP, a membrane-permeable analogue of cAMP that acts beyond the step of adenylate cyclase, attenuates but does not totally block the effect of DA.24 The similarity between the effects of 8-bromocAMP and those of NT has led us to investigate whether the modulatory effect of NT is mediated by an increase in intracellular cAMP.24.25 First, the effect of NT on DA inhibition was measured in the presence and absence of IBMX, a potent inhibitor of phosphodiesterase, the enzyme that normally degrades CAMP. If NT acts through CAMP systems, IBMX, which presents intracellularly produced CAMP from being metabolized, should potentiate the effect of NT. Indeed, coapplication of IBMX at concentrations at which IBMX did not produce a significant effect by itself further

SHI & BUNNEY ACTIONS OF NEUROTENSIN

133

potentiated the ability of NT to diminish the inhibition induced by DA. Second, the effect of NT was determined while the activity of adenylate cyclase was inhibited by SQ22536.26 Consistent with the hypothesis, inhibition of adenylate cyclase by SQ22536 markedly reduced the effect of NT. Finally, we used H8 to inhibit protein kinase A.27It was found that the modulatory effect of NT on DA cells could be effectively reversed by pretreatment with H8.The latter result is consistent not only with the postulate that intracellular CAMP plays an important role in mediating the effect of NT, but also suggests that an increase in protein phosphorylation mediated by protein kinase A is involved in the action of NT. Indeed, activation of adenylate cyclase and protein kinase A has been shown to phosphorylate D2 receptors and decrease their affinity for agonists, while leaving antagonist binding unchanged.28 Similar changes in the binding properties of D2 receptors by NT have also been r e p ~ r t e d . ~ ~ . ~ ~ A direct causal link between DZreceptor phosphorylation and activation of NT receptors, however, remains to be established. Inhibition of CAMP- and protein kinase A-mediated cellular processes by H8 has been reported in many studies. Studies in a broken cell preparation, however, indicated that H8 inhibits not only protein kinase A but also several other protein kinases, one of which is protein kinase C. To determine if protein kinase C is involved in NT’s action, H7 was tested. H7 is an analogue of H8 and is more potent than H8 in inhibiting protein kinase C. If the effect of NT is mediated by protein kinase C and H8 blocks the effect of NT by inhibiting this enzyme, H7 should be more effective than H8 in reversing the effect of NT. Contrary to this prediction, H7 was found to be much less effective than H8 in antagonizing the modulatory effect of NT.25 These results suggest that the reversal of the NT-induced effect by H8 is not due to its effect on protein kinase C. The inability of phorbal-12,13-dibutyrate,an activator of protein kinase C, to mimic the effect of NT in our preparation further supports the conclusion that protein kinase C is unlikely to be involved in the modulatory effect of NT on DA ~ e l l s . 2 ~ As discussed above, microiontophoresis of NT also caused an increase in the firing rate of a few DA cells in vivo. 17.18-31 This excitatory effect of NT was, however, more reliably observed in brain slice preparations.20.23.3z.33Our in vitro experiments indicated that both the modulatory and the excitatory effects of NT could be observed in the same cell.23 The latter occurred at higher concentrations (210 nM) than those needed to produce a modulatory effect. Because activation of DA autoreceptor inhibits DA neurons and NT attenuates this inhibition, it is possible that the excitatory effect of NT may be secondary to its modulatory effect on the inhibition induced by endogenously released DA, that is, due to a disinhibitory effect. This hypothesis is supported by the findings that perfusion of sulpiride also induced an increase in the firing rate of DA cells23 and caused an inward current.34According to the hypothesis, however, agents that are more potent in antagonizing DA-induced inhibition should produce more excitation, and the magnitude of the excitation induced by the proposed “disinhibitory effect” of NT should depend on the degree to which DA receptors are being stimulated. Thus, the more DA receptors are activated, the larger the excitation induced by NT. In contrast to these predictions, however, we found that the excitation induced by sulpiride was always smaller than that of NT, even though sulpiride was always more effective than NT in antagonizing the inhibition induced by DA application.25 Furthermore, the excitation induced by NT was unchanged, if not potentiated, during the blockade of DA receptors by ~ u l p i r i d eThese . ~ ~ results suggested that the increase in the spontaneous activity of DA cells induced by NT is not due to an

134

ANNALS NEW YORK ACADEMY OF SCIENCES

interaction between NT and the endogenously released DA, such as forming a complex with DA or competing with DA for the same receptors. Intracellular recording indicated that the excitatory effect of NT was associated with a depolarization and, in some cells, an increase in input resistance.32 At more hyperpolarized levels (close to the equilibrium potential of K+ in these cells), the effect of NT was abolished, implying that a decrease in K+ conductances might be involved. The persistence of the depolarizing response to NT in a CaZ+-free and highMg2+solution suggests that this effect was a direct action of NT on DA neurons. As has been reported for neurons in the hypothalamus, high concentrations of NT sometimes caused a sudden cessation of cell firing.20+23,33Whether this is due to depolarization inactivation remains to be determined.

EFFECTS OF NEUROTENSIN IN OTHER BRAIN AREAS Cortex The presence of NT-containing fibers and NT receptors throughout cortex suggests a widespread action. Of particular interest is the finding that many of the NT fibers in the prefrontal cortex are also stained for tyrosine hydroxylase, a marker for DAcontaining terrninals.4 Recent studies suggest that the mixed DA/NT projections extend rostrocaudally in layer VI of the whole cerebral cortex and in layers I1 and 111 of the entorhinal cortex.3s Audinat and colleagues studied the effect of NT in the medial prefrontal cortex using a rat brain slice preparation.36 In their study, 88% of the pyramidal neurons were found to respond to 0.1-10 pM NT showing a depolarization accompanied by an increase in input resistance. Under voltage-clamp, NT caused an inward current that reversed at -88 f 9.7 mV, suggesting a decrease in K+ conductance. However, the depolarization induced by NT was transient despite the continuous presence of NT, and the response of NT took more than 20 minutes to recover, suggesting the development of receptor desensitization. Since it persisted during the blockade of synaptic transmission, this depolarizing effect of NT is likely to be mediated by a direct action of NT on the recorded cells. In addition to the direct excitatory effect of NT, the authors also observed an increase in both inhibitory and excitatory PSPs in the recorded neurons during NT administration, indicating that NT may excite both inhibitory and excitatory neurons in the same slice. Acetyl-NT(8-13) mimicked the effect of native NT, whereas NT(1-8) did not, suggesting that the observed effects of NT might be mediated by an activation of NT receptors. Interestingly, these effects of NT are very similar to those of DA application on the same population of cells reported previously by the same laboratory.37 Because superfusion of fluphenazine or sulpiride, two DA antagonists, did not block the effects of NT, it is unlikely that NT produced its effects by releasing DA from mesocortical DA terminals. It is intriguing that NT, which has been shown to produce many biochemical and behavioral effects resembling those of neuroleptics," induced effects similar to those of DA in this major mesocortical DA target area. The effect of NT on neurons in the sensory-motor cortical area of the rat was examined by Phillis and Kirkpatrick using extracellular recording techniques combined with microiontophoresis.38 Again, NT (1 mh4, with ejecting currents up to 200 nA) caused an increase in cell firing. This effect of NT was observed in both the identified

SHI & BUNNEY ACTIONS OF NEUROTENSIN

135

corticospinal neurons as well as other neurons. However, the effect was characterized as weak and delayed and was only observed in about 25% of the cells tested. Both the onset and the offset time for the effect of NT were slow (in excess of 60 sec). Whether this excitation was direct or mediated indirectly by an action of NT on nearby neurons or terminals was not determined. When ejected in conjunction with glutamate, NT was found to have either no effect or a small potentiation of the excitation induced by glutamate.

Nucleus Accumbens The nucleus accumbens is another major target for midbrain dopamine neurons. Studies have suggested that NT produces many of its neuroleptic-like effects by interacting with DA in this area. 15.16 Using microiontophoretic techniques, McCarthy et ~ 1examined . ~ the ~ effect of NT on nucleus accumbens neurons in rats. Studying only cells that were sensitive to DA, they found that half were responsive to NT application (-7.6 mM, 50 nA). In most of the responding cells (13/15), NT mimicked the effect of DA and potently inhibited cell activity, only two cells were excited by NT. These results seemingly contradict those of biochemical and behavioral studies where NT has often been reported to block or counter DA action.40 However, since the techniques used in the study did not ascertain whether NT acted directly, it is possible that NT produced its inhibition by exciting the inhibitory neurons impinging on the recorded cell. In fact, most of the neurons in the nucleus accumbens are GABAergic. NT receptors have also been suggested to exist on mesolimbic DA terminals and potentiate DA so NT may cause DA release and thereby inhibit the cell indirectly. A similar paradigm has been observed in the cerebellum where the direct excitatory effect of NT was masked by an indirect inhibitory effect due to release of NE from locus ceruleus terminals. As noted in several other brain areas (see below), it is also possible that NT, at the dose used, may have overexcited the cell, resulting in depolarization inactivation.

Bed Nucleus of the Stria Terminalis Sawada ef al.42 studied the action of NT in the bed nucleus of the stria terminalis in rats using a brain slice preparation. They found that NT very potently increased the activity of the neurons recorded extracellularly. At low concentrations NT (0.35-3.5 nM) excited all the cells tested, while at higher doses (35-350 nM) NT increased the firing rate of about two-thirds of the cells. Cells that were not affected by NT were either silent or discharged in bursts. NT excited the neurons as effectively when synaptic transmission was blocked by perfusing the slice with medium containing low [Ca2+]and high [Mg*+], suggesting a direct effect of.NT on these cells.

Hypothalamus Using organotypic culture techniques, Baldino and Wolfson43 first reported an excitatory effect of NT on neurons in the preoptic-anterior hypothalamus of the rat. In 71%of the cells recorded, an increase in the spontaneous discharge rate was observed

136

ANNALS NEW YORK ACADEMY OF SCIENCES

during the application of NT. In agreement with receptor binding studies, only the COOH-terminal hexapeptide NT(8-13) and the N-acetylated NT(8-13) mimicked the effect of native NT. The NHZ-terminal fragment, NT(1-8) was without effect at concentrations as high as 1 mM.5 These results suggest that the observed effect was mediated by NT receptors. The excitatory effect of NT was apparently a direct effect of NT on the neurons recorded since it persisted in a Ca2+-free medium in which synaptic activity was completely blocked. The effect of NT was dose related and extremely potent; it could be observed at concentrations as low as SO pM. At high concentrations (21 mM), NT consistently abolished spontaneous activity. This decrease in activity was always preceded by a diminution in amplitude and an increase in duration of the action potential and could be routinely reversed by coadministration of the inhibitory neurotransmitter, GABA. Although only indirect, these results suggest that this inhibition was due to a development of depolarization block. In addition to its excitatory effect, NT produced a decrease in firing rate in 16% of the cells tested in the same cultures. In many of these cases, the reductions in spontaneous activity induced by NT were abolished or converted to excitation during synaptic blockade, suggesting that these actions are mediated presynaptically. Interestingly, in the same area, a larger proportion of neurons were inhibited when they were recorded in viva45 Thus, with local iontophoretic application of NT (0.52 mM; 10-40 nA), 51.6% of the cells were inhibited and 19.5% were excited. It is not known whether the inhibitory effect seen in vivo represents the action of NT at high concentrations in vitro, that is, depolarization inactivation, or an indirect effect. In addition to excitation and inhibition, application of NT was found to modulate glutamate-induced e ~ c i t a t i o nThus, . ~ ~ NT at low concentrations enhanced glutamateinduced excitation without influencing basal activity. In at least one of the cells shown in the paper, NT not only enhanced the excitation induced by glutamate but also decreased the background activity. This latter effect of NT actually improved the signalto-noise ratio more than 10 times, even though the increase in firing rate induced by glutamate was increased by only SO% in the presence of NT. Similar excitatory responses were reported in the cells of the arcuate nucleus.& About half of the cells were excited by NT at high concentrations (100 nM-1 m). In almost all cases, a postexcitatory inhibition was observed, even before washout of NT from the bath. At 1 @ theI brisk , and very transient excitations with their associated action potential changes were very similar to responses observed in preopticanterior hypothalamus neurons after application of NT doses greater than 1 mM.43 The postexcitatory inhibition may also be the result of depolarization block. At 100 nM, however, no alterations in action potential shape were noted before the cells became silent. At low concentration (1 nM), NT evoked simple excitatory and inhibitory responses without any of the large fluctuations in activity seen with higher peptide concentrations. Experiments conducted in low-Ca2+ and high-MgZ+ medium indicated that the inhibitory effects seen with low Concentrations of NT were indirect, as were about half of the excitatory actions.

Locus Ceruleus Guyenet and Aghajanian4’ reported that iontophoretic NT had no effect on the locus ceruleus neurons. However, a potent inhibitor effect of NT was observed by

SHI & BUNNEY: ACTIONS OF NEUROTENSIN

137

Young et al.48 In the latter study, microiontophoresis of NT produced a rapid and reversible inhibition of firing in about half of the cells tested. This effect of NT was not due to current artifacts, as similar currents used to eject either NaCl or HCl (the carrier solution) had no effect on cell firing. The inhibition induced by NT also seemed not to be due to depolarization inactivation as had been seen in the hypothalamus, since monitoring of the action potential size revealed no change. Young et al. suggested that at least two factors could account for the differencesbetween the two studies. First, the iontophoresis barrels used by Guyenet and Aghajanian were recessed 15 to 25 pm from the tip of the recording barrel, possibly allowing degradation of the peptide before it could reach the receptor. Second, Guyenet and Aghajanian tested only five cells. These cells may belong to the unresponsive half of the cells recorded by Young et al. In contrast to the inhibitory effect of NT observed in the cell body area, NT in the cerebellum has been reported to excite NE afferent terminals from the locus ceruleus, resulting in an increase in NE release (see below).

Nucleus Ractus Solitarius The possible involvement of NT in the control of breathing has been suggested by the finding that both NT-positive cell bodies and terminals, as well as NT receptors, have been localized in the nucleus tractus solitarius, which is thought to be one of the central nuclei involved in the control respiration. Morin-Surun et a1.4 studied the actions of NT in the nucleus tractus solitarius and adjacent areas. Application of NT by either pressure ejection or microiontophoresis was found to increase the discharge frequency of all cells studied. These cells included both the respiratory and nonrespiratory cells. In some cases, the discharge duration of the inspiratory-related neurons also was increased. The effect of NT was dose dependent. At the highest concentration used (1 mM), pressure injection of NT sometimes caused an increase in firing followed by a cessation of cell activity. A decrease of action potential amplitude associated with this change suggests that the cessation of cell firing may have been due to a depolarization inactivation. However, a local anesthesia effect was not ruled out.

Periaqueductal Gray One of the effects induced by centrally administered NT is analgesia. This effect of NT seems to be mediated by an opiate-independent mechanism since it cannot be blocked by naloxone. The high concentrations of NT and its receptors found in the periaqueductal gray suggest that this area might be one of the sites of action for NTinduced analgesia. Behbehani and Pert49tested this hypothesis. They found that local pressure ejection of NT excited most (64%) of the neurons in the periaqueductal gray area. Consistent with their hypothesis, local injection of NT in the periaqueductal gray area also induced a dose-dependent antinociceptiveeffect. Since lesioning of the nucleus raphe magnus significantly attenuated this effect of NT, they further proposed that NT may, by activating the periaqueductal gray neurons, indirectly activate the nucleus raphe magnus cells, which then inhibit the nociceptive dorsal horn neurons.

138

ANNALS NEW YORK ACADEMY OF SCIENCES

Supporting this hypothesis, they found that 62% of the cells in the nucleus raphe magnus were excited by the injection of NT into the periaqueductal area, about 25% were inhibited, and the rest were not affected. The direct excitatory effect of NT on the periaqueductal area cells was later confirmed in a brain slice preparation50 in which it persisted after blockade of synaptic transmission by replacement of Ca2+ with Co2+ in the perfusate. Intracellular recording indicated that the excitation induced by NT was associated with a gradual membrane depolarization. However, only a slight decrease in the membrane resistance was observed. The ionic mechanisms underlying these changes were not determined.

Cerebellum NT was initially reported to have only a weak inhibitory effect on cerebellar Purkinje neurons when it was applied by microiontophore~is.~~ This result was confirmed by Marwaha et al. 51 However, the latter study suggested that the weak effect was due to technical problems with iontophoresis, as NT produced consistent depressant effects when it was applied by pressure ejection from the same electrode. The same application occasionally exerted a biphasic response with an initial excitation followed by a potent depression. However, all cells were excited by NT after blockade of NE transmission either by treating the animals with haloperidol, a drug with a1 receptorblocking properties, or by destroying the NE neurons in the locus ceruleus using 6-OHDA. These results suggest that the inhibition was indirectly mediated by an action of NT on NE terminals. The fact that NT-induced excitation persisted during the application of Mg2+, which should reduce not only the release of NE but of other transmitters as well, further suggests a direct excitatory effect of NT on Purkinje cells.

Spinal Cord The excitatory effect of NT on spinal motoneurons was first reported by NicolP in the frog. NT was shown to cause a depolarization in three out of five cells with a threshold dose of approximately 1 mM. This excitatory effect of NT was later confirmed in toads and In rats, the effect of NT was found to be much more potent than in amphibians (about 106 times). In all three studies, however, the excitatory effect of NT was greatly reduced or entirely blocked by the addition of TTX to the perfusion medium, suggesting that the observed effect of NT on motoneurons may be an indirect one. One possible mechanism might be that NT inhibits the inhibitory interneurons that make synaptic contact with these motoneurons. Studies by Stanzion and Zieglgansberger suggest that this assumption is unlikely. In their studies, recordings were made not only from motoneurons, but also interneurons and neurons in the dorsal horn. In all three types of spinal neurons, only the excitatory effect was observed. These studies suggest that NT may excite motoneurons either directly via a TTXsensitive mechanism or indirectly by increasing the activity of the excitatory neurons. A NT-induced increase in the activity of dorsal horn neurons and sympathetic preganglionic neurons has also been reported.sk57 In addition to its excitatory effect, NT has been reported to modulate glutamateinduced excitation of the sensory neurons in the dorsal horn. Thus, application of NT suppressed glutamate-induced excitation without an effect on basal firing rate. Similar modulatory effects on the excitation induced by noxious stimulation of the

SHI & BUNNEY ACTIONS OF NEUROTENSIN

139

cutaneous receptive field have been observed. This modulatory effect of NT may partially account for its analgesic effect after central administration.56

EFFECTS OF NEUROTENSIN IN THE PERIPHERY AND OTHER CELLS Retina The actions of NT on amphibian retinal ganglion cells were first examined by Dick and Millers8 using microiontophoretic techniques. NT excited all the cells examined, while coadministration of GABA antagonized the NT-induced excitatory effect. Intracellular recording revealed a membrane depolarization upon application of NT.5yThis depolarization was accompanied by an increase in membrane conductance. Because the depolarization and conductance increases also occurred during cobalt blockade of synaptic transmission, the effect of NT was likely a direct one rather than transsynaptic. Since the equilibrium potential for C1- is at or below the resting membrane potential in these cells, and Ca2+channels were blocked by cobalt, the conductance increase associated with NT was probably due to an increase in sodium permeability. Similar excitatory effects of NT on amacrine cells were also 0bserved.5~

Znferior Mesenteric Ganglion In about 60% of the neurons studied by Stapelfeldt and Szurszewski,a-62 NT produced a depolarization. On the basis of different time courses, the depolarization induced by NT can be classified into two types, a transient type and a steady-state type. Both types of response were associated with a decrease in membrane conductance. The steady-state type of depolarization was unaltered during the blockade of synaptic transmission by lowering Ca2+and increasing Mg2+ concentrations in the medium, suggesting a direct postsynaptic action of NT. In contrast, the transient response of NT was entirely abolished under these conditions. Pretreatment with capsaicin, a substance-P-depleting agent, also attenuated this transient depolarization. Additional evidence suggests that NT may enhance the release of substance P from primary afferent nerve terminals and cause this transient depolarization. First, using radioimmunoassay, a substance-P-like material was released upon the stimulation of colonic afferents, and the addition of NT further increased the release. Second, stimulation of the same nerve produced a noncholinergic slow EPSP that was potentiated by application of NT. This potentiation of NT was blocked by pretreatment with capsaicin. Finally, the transient response was largely attenuated if the cell was desensitized to exogenously administered substance P. The authors further suggested that the central preganglionic nerve might be the source of NT, since stimulation of this nerve mimicked the effects of NT and these effects were abolished after the neuron became desensitized to NT.

Myenteric Ganglia The action of NT on single myenteric neurons was studied by Williams et al.63 using both extra- and intracellular recording techniques. NT at 100 to 300 nM increased

140

ANNALS NEW YORK ACADEMY OF SCIENCES

the firing rate of about 50% of the neurons tested. The excitation caused by NT persisted in solutions that contained EGTA (1 mh4) and no Ca2+. The proportion of cells that were excited by NT was the same in normal and in CA2+-free medium. These findings indicated that the excitation caused by NT was a direct action on the neurons and not caused by release of excitatory transmitters or inhibition of an inhibitory neuron. Intracellular recording revealed a depolarization and an increase in membrane resistance during the application of NT. A hyperpolarization response to NT was also observed in some neurons. This response was not related to the concentration of NT applied and was much reduced during perfusion of Ca2+-freeand high-Mgz+ solutions. In some cases, low concentrations ( 4 0 0 nM)caused depolarization, whereas higher concentrations(2100nM)hyperpolarized the cells. The hyperpolarization disappeared but the depolarization persisted in Caz+-free medium. Whether the hyperpolarization was induced by an indirect action of NT or by an activation of a Caz+-dependent K+ current is still unknown.

Neuroblastoma Cells Neuroblastoma cell lines have been used extensively as model nerve cells in both biochemical and electrophysiological investigations. Noradrenergic neuroblastoma cells of clone N1E 115 have the capacity to generate action potentials and possess several neurotransmitter receptors, including high-affinity NT receptors. In a preliminary experiment, NT showed no detectable effect on NlE 115 ce1ls.u On the other hand, NT (1-100 pM)produced a significant change in membrane potentials in NG 108-15 cells, initially hyperpolarizing and then depolarizing the cell.& Such changes have also been reported for bradykininM and PGD2.6’ The clear decrease in membrane resistance seen during hyperpolarization by NT suggested the involvement of an increase in ion flux. By measuring the responses at different concentrations of [K+],, the hyperpolarization induced by bradykinine has been shown to be caused by an increase in K+ conductance. The hyperpolarization induced by NT therefore may be mediated by the same mechanism. As changes in membrane conductance during sustained depolarization by NT were not significant, it is unknown which ions contribute to the depolarization. However, replacement of NA+ as well as Ca2+ in the recording medium depressed the depolarizing response.

Oocyte Expression System The electrophysiologicalactions of NT and the underlying intracellular mechanism were studied by Hirono et al.68 using &nopus laevis oocytes injected with poly (A)+ messenger ribonucleic acid (mRNA) isolated from rat brain. Under voltage-clamp conditions (-60 mV), application of NT (EDm = 92 nM)to mRNA-injected oocytes produced transient and oscillatory inward currents that were accompanied by an increase in membrane conductance. In the same cells, a similar response was also observed during acetylcholine application. The action of acetylcholine on these cells has been shown to be mediated through an activation of muscarinic subtype receptors, which leads to the formation of inositol 1,4,5,-trisphosphate@and mobilization of intracellular Ca2+,m resulting in increased permeability to C1 - . n P The similarities between NT- and acetylcholine-induced electrical responses of these cells suggest that

SHI & BUNNEY: ACTIONS OF NEUROTENSIN

141

NT may produce its effect by the same intracellular mechanism. In support of this suggestion, the response of NT and acetylcholine were both found to be inhibited by pertussis toxin and EGTA, mimicked by intracellularly injected inositol 1,4,5,-trisphosphate,and suppressed when cell response to inositol 1,4,5-trisphosphate was desensitized by a large dose of inositol 1,4,5-trisphosphate. Both responses were mediated by an increase in membrane permeability to C1-. Guanine nucleotidebinding regulatory protein (Gi or Go), may be the common target of NT and acetylcholine receptors. The pertussis toxin-sensitive G proteins may activate phospholipase C and cause the formation of inositol 1,4,5-trisphosphate.

SUMMARY AND CONCLUSIONS Three effects of NT were observed on midbrain DA cells. The modulatory effect of NT, that is, the attenuation of DA-induced inhibition, has been most extensively examined. Studies indicate that this effect of NT was not simply due to a nonspecific excitation. NT selectively attenuated DA-induced inhibition without affecting either GABA-induced inhibition or glutamate-induced excitation of the same cells, and the attenuation of DA-induced inhibition could be observed at the doses at which the basal activity of DA cells was not changed by NT. The attenuation of DA-induced inhibition by NT is also unlikely to result from the formation of a DA-NT complex, since neuromedin N, which competes with NT for the same receptor but does not bind to DA, mimicked the effects, and neurotensin(1-11), which forms a complex with DA but is inactive in competing for NT receptors, did not. The similarities between the effects of NT and those of 8-bromo-CAMP and forskolin suggest that intracellular CAMP and protein kinase A may be involved. This suggestion was supported by the findings that IBMX (an inhibitor of phosphodiesterases) potentiated the effect of NT; and SQ22536 (an inhibitor of adenylate cyclase) and H8 (an inhibitor of protein kinase A) antagonized it. Phorbal-12,13-dibutyrate (an activator of protein kinase C) did not mimic the effect of neurotensin, and H7 (an inhibitor of protein kinase C) did not reduce the effect, suggesting that protein kinase C is unlikely to be involved in the modulatory effect of neurotensin. Experiments in vitro indicated that the excitatory effect of NT on DA cells occurred at higher concentrations (>lo nM) than those needed for producing the modulatory effect. Its persistence during DA receptor blockade by sulpiride suggests that this effect was not entirely mediated by an attenuation of the inhibition induced by endogenously released DA. At even higher concentrations (>lo0 nM), a sudden cessation of cell activity preceded by an increase in firing rate was observed. Whether this effect of NT was due to depolarization inactivation or a toxic effect of the peptide at high concentrations remains to be determined. In most other areas studied, the excitatory effect of NT was most commonly observed. In many areas, this excitatory effect was apparently a direct postsynaptic effect of NT. However, different mechanisms may be involved (see TABLE1). For example, in some areas NT acted through a decrease in membrane conductance, while in others no change or an increase in the membrane conductance was observed. The physiological significance of these changes should be different since a decrease in membrane conductance makes cells electrically more compact, whereas an increase in membrane conductance will produce the opposite effect. The inhibitory effect of NT was less frequently observed. In some studies, this

142

ANNALS NEW YORK ACADEMY OF SCIENCES

effect seemed to be indirect since it disappeared when synaptic transmission was blocked. However, as solutions used to block synaptic transmission often block Ca2+channels, it is conceivable that in some of those areas, NT produced its inhibition through a Ca*+-mediated process (e.g., CaZ+-dependent K+ channels). As mentioned earlier, a decrease in the activity of a cell can also be brought about by depolarization inactivation, or a nonspecific local anesthetic effect. If an extracellular electrode is used, these particular types of inhibition are difficult to distinguish from the inhibition induced by hyperpolarization. Thus, the nature of the inhibitory effects of NT seen in some areas needs to be further examined. Modulatory effects of NT were also seen in two other areas. In both cases, NT modulates glutamate-induced excitation. In spinal dorsal horn neurons, NT attenuated, while in the hypothalamus NT potentiated glutamate-induced excitation. Whether these and other effects of NT are due to the activation of different types of NT receptors or of the same receptor coupled to different effector systems in different neurons remains to be investigated. REFERENCES I . TANAKA, K., M. MAW& S.NAKANISHI. 1990.Structure and functional expression of the cloned rat neurotensin receptor. Neuron 4: 847-854. T.,B. J. EVERITT,E. THEODORSSON-NORHEIM & M. GOLDSTEIN. 1984.Occurrence 2. HOKFELT, of neurotensin-like immunoreactivity in subpopulations of hypothalamic, mesencephalic and medullary catecholamine neurons. J. Comp. Neurol. 222: 543-560. 3. JENNES,L.,W. E. STUMPF& P. W. KALIVAS.1982. Neurotensin: Topographical distribution in rat brain by immunohistochemistry. J. Comp. Neurol. 210: 211-224. 4. STUDLER, J. M.,P. KITABGI, G. TRAMU,D. HERVE,J. GLOWINSKI & J. P. TASSIN.1988.Extensive colocalization of neurotensin with dopamine in rat meso-cortico-frontal dopaminergic neurons. Neuropeptides 11: 95-100. 5. WOULFE,J. & A. BEAUDET. 1989.Immunocytochemical evidence for direct connections between neurotensin-containing axons and dopaminergic neurons in the rat ventral midbrain tegmentum. Brain Res. 479: 402-406. J. M. & M. J. KUHAR.1981.Neurotensin receptors are located on dopamine-containing 6. PALACIOS, neurons in rat midbrain. Nature 294: 584-589. 7. Quirion, R., C. C. Chiueh, H. E. Everist & A. PERT.1985.Comparative localization of neurotensin receptors on nigrostriatal and mesolimbic dopaminergic terminals. Brain Res. 27: 385-389. R., P. GAUDREAU, S. ST. PIERRE,F. b o u x & C. B. PERT. 1982. Autoradiographic 8. QUIRION, distribution of 3H-neurotensin receptors in rat brain: visualization by tritium sensitive film. Peptides 3: 757-763. 9. WATERS,C. M., S. P. HUNT,P. JENNER& C. D.MARSDEN.1987. Localization of neurotensin receptors in the forebrain of the common marmoset and the effects of treatment with the neurotoxin I-methyl-4-phenyl-l,2,3,6-tetrahydropyridine. Brain Res. 412: 244-253. 10. SZIGETHY, E. & A. BEAUDET.1989. Correspondence between high-affinity 12SI-neurotensin binding sites and dopaminergic neurons in the rat substantia nigra and ventral tegmental area: A combined radioautographic and immunohistochemical light microscopic study. J. Comp. Neurol. 279: 128-137. 1 1. NEMEROFF, C. B. 1986.The interaction of neurotensin with dopaminergic pathways in the central nervous system: Basic neurobiology and implications for the pathogenesis and treatment of schizophrenia. Psychoneuroendocrinology 1 1 15-37. 12. SHI, W. X. & B. S. BUNNEY.1990. Neurotensin attenuates dopamine D2 agonist quinpiroleinduced inhibition of midbrain dopamine neurons. Neuropharmacology 29: 1095-1097. 13. NEMEROFF, C. B., D. LUTTINGER, D. E. HERNANDEZ, R. B. MAILMAN, G.A. MASON,S. D. DAVIS,E.WIDERLOV, G. D. FRYE,C. D. KILTS, K. BEAUMONT, G.R. BREESE& A. J. PRANCE,

SHI & BUNNEY: ACTIONS OF NEUROTENSIN

14. 15.

16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35.

36.

143

JR. 1983. Interactions of neurotensin with brain dopamine systems: Biochemical and behavioral studies. J. Pharmacol. Exp. Ther. 225: 337-345. VONEULER,G.. B. MEISTER, T. HOKFELT & K. FUXE.1989. Intraventricular injection of neurotensin reduces the binding of dopamine-D-2 receptors in the rat forebrain. Acta Physiol. Scand. 137 309-310. JOLICOEUR, F. B., G. DEMICHELE & A. BARBEAU. 1983. Neurotensin affects hyperactivity but not sterotypy induced by pre- and post-synaptic dopaminergic stimulation. Neurosci. Biobehav. Rev. 7 385-390. KALIVAS, P. W., C. B. NEMEROFF & A. J. PRANCE, JR. 1984. Neurotensin microinjection into the nucleus accumbens antagonizes dopamine-induced increase in locomotion and rearing. Neuroscience 11: 919-936. SHI,W. X . & B. S. BUNNEY. 1987. Neurotensin selectively attenuates DA inhibition of midbrain dopaminergic neurons. Soc. Neurosci. Abstr. 13: 934. SHI, W. X. & B. S. BUNNEY.1991. Neurotensin modulates autoreceptor mediated dopamine effects on midbrain dopamine cell activity. Brain Res. 543: 315-321. CHIODO,L. A,, A. S. FREEMAN & B. S. BUNNEY. 1987. Electrophysiological studies on the specificity of the cholecystokinin antagonist proglumide. Brain Res. 410: 205-211. & A. DRESSE.1989. Electrophysiological effects of neurotensin on SEUTIN,V., L. MASSOTTE dopaminergic neurons of the ventral tegmental area of the rat in vitro. Neuropharmacology 28: 949-954. ADACHI.D. K., P. W. KALIVAS& J. D. SCHENK.1990. Neurotensin binding to dopamine. J. Neurochem. 54: 1321-1328. SMYTHIES, J. R. 1981. An hypothesis of the molecular structure of the dopamine receptor. Med. Hypoth. 7 1449-1456. SHI,W. X. & B. S. BUNNEY. 1991. Effects of neurotensin on midbrain dopamine neurons: Are they mediated by formation of a neurotensin-dopamine complex? Synapse 9(2): 79-94. SHI,W. X. & B. S. BUNNEY. 1988. 8-bromo-CAMP mimics the effects of neurotensin in midbrain dopamine neurons: A study in brain slice. SOC.Neurosci. Abstr. 14: 78. SHI, W. X. & B. S. BUNNEY.1992. Roles of intracellular CAMP and protein kinase A in the actions of dopamine and neurotensinon midbrain dopamine neurons. J. Neurosci. 12: 2433-2438. & M. J. ANTONACCIO. 1979. HARRIS, D. N., M. M. ASAAD,M. B. PHILIPS,H. J. GOLDENBERG Inhibition of adenylate cyclase in human blood platelets by 9-substituted adenine derivatives. J. Cyclic Nucleotide Res. 5: 125-134. HIDAKA, H., M. INAGAKI, S. KAWAMOTO & Y. SASAKI. 1984. Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide dependent protein kinase and protein kinase C. Biochemistry 23: 5036-5041. 2 . & S. FUCHS.1991. Phosphorylation by cyclic AMP-dependent protein kinase modELAZAR, ulates agonist binding to the D2 dopamine receptor. J. Neurochem. 56: 75-80. AGNATI, L. F., K. FUXE,F. BENFENATI & N. BATTISTINI. 1983. Neurotensin in v i m markedly reduces the affinity in subcortical limbic 'H-N-propylnorapomorphine binding sites. Acta Physiol. Scand. 119: 459-461. VONEULER,G. & K. FUXE.1987. Neurotensin reduces the affinity of D-2 dopamine receptors in rat striatal membranes. Acta Physiol. Scand. 131: 625-626. 1981. Neurotensin selectively activatesdopaminergic neurons ANDRADE, R. & G. K. AGHAJANIAN. in the substantia nigra. SOC.Neurosci. Abstr. 7 537. R. D. 1985 Neurotensin depolarizes substantia nigra dopamine neurons. Brain Res. PINNOCK, 338: 151-154. F ~ Z Z AM. , F., E. KUNG,S. BISCHOFF & H-R. OLPE.1988. The neurotensin analog xenopsin excites nigral dopamine neurons. Eur. J. Pharmacol. 145: 341-343. LACEY,M. G., N. B. MERCURI & R. A. NORTH.1987. Dopamine acts on D? receptors to increase potassium conductance in neurons of the rat substantia nigra zona compacta. I. Physiol. 392: 387-416. & C. VERNEY. 1991. Further indication that distinct doFEBVRET, A,, B. BERGER,P. GASPAR paminergic subsets project to the rat cerebral cortex: Lack of colocalization with neurotensin in the superficial dopaminergic fields of the anterior cingulate, motor, retrosplenial and visual cortices. Brain Res. 547 37-52. AUDINAT, E., J. M. HERMEL & F. CREPEL.1989. Neurotensin-induced excitation of neurons

144

ANNALS NEW YORK ACADEMY OF SCIENCES

of the rat’s frontal cortex studied intracellularly in vim. Exp. Brain Res. 78: 358-368. 31. PENIT-SORIA, I., E. AUDINAT & F.CREPEL.1987. Excitation of rat prefrontal cortical neurons by dopamine: An in v i m electrophysiological study. Brain Res. 425: 263-274. J. W. & J. R. KIRKPATRICK. 1980. The actions of motilin, luteinizing hormone releasing 38. PHILLIS, hormone, cholecystokinin, somatostatin, vasoactive intestinal peptide, and other peptides on rat cerebral cortical neurons. Can. J. Physiol. Pharmacol. 58: 612-623. P. S., R. J. WALKER, H. YAJIMA,K. KITAGAWA & G. N. WOODRUFF. 1979. The 39. MCCARTHY, action of neurotensin on neurons in the nucleus accumbens and cerebellum of the rat. Gen. Pharmacol. 10: 331-333. M. P., D. MARLOT,J. P. KESSLER& M. DENAVIT-SAUBIE. 1986. The excitation 40. MORIN-SURUN, by neurotensin of nucleus tractus solitarius neurons induces apneustic breathing. Brain Res. 384: 106-113. & J. C. BLANCHARD. 1988. Neurotensin effects on evoked P. DUBEDAT 41. HETIER,E., A. BOIREAU, release of dopamine release from rat striatum, nucleus accurnbens and prefrontal cortex in rat. Naunyn-Schmiedeberg’s Arch. Pharmacol. 337 13-17. S., S. TAKADA & C. YAMAMOTO. 1980. Electrical activity recorded from thin sections 42. SAWADA, of the bed nucleus of the stria terminalis, and the effects of neurotensin. Brain Res. 188:578-581. JR., F. & B. WOLFSON. 1985. Postsynaptic actions of neurotensin on preoptic-anterior 43. BALDINO, hypothalamic neurons in v i m . Brain Res. 325: 161-170. JR., F., L. G. DAVIS& B. WOLFSON.1985. Structure-activity studies with carboxy44. BALDINO, and amino-terminal fragments of neurotensin on hypothalamic neurons in v i m . Brain Res. 342: 266-212. 45. HORI,H., M. YAMASAKI, T. KIYOHARA & M. SHIBATA.1986. Responses of preoptic thermosensitive neurons to poikilothermia-inducing peptides-bombesin and neurotensin. Pflugers Arch. 407: 558-560. 46. HERBISON, & N. E. SIRETT.1986. Neurotensin excites neurons in the arcuate A,, J. I. HUBBARD nucleus of the rat hypothalamus in v i m . Brain Res. 364: 391-395. 41. GUYENET, P. G. & G. K. AGHAJANIAN. 1977. Excitation of neurons in the nucleus locus coeruleus by substance P and related peptides. Brain Res. 136: 178-184. 48. YOUNG111, W. S., G. R. UHL& M.J. KUHAR.1978. Iontophoresis of neurotensin in the area of the locus coeruleus. Brain Res. 150: 431-435. M. M. & A. PERT.1984. A mechanism for the analgesic effect of neurotensin as 49. BEHBEHANI, revealed by behavioral and electrophysiological techniques. Brain Res. 324: 35-42. M. M., M. T. SHIPLEY 50. BEHBEHANI, & J. H. MCLEAN.1987. Effect of neurotensin on neurons in the periaqueductal gray: An in vim study. J. Neurosci. 7: 2035-2040. 51. MARWAHA, 1980. Electrophysiological actions of neurotensin J., B. HOFFER& R. FREEDMAN. in rat cerebellum. Regul. Peptides 1: 115-125. 52. NICOLL,R. A. 1978. The action of thyrotropin-releasing hormone, substance P and related peptides on frog spinal rnotoneurons. J. Pharmacol. Exp. Ther. 207: 817-824. 53. PHILLIS,J. W. & J. R. KIRKPATRICK.1979. Actions of various gastrointestinal peptides on the isolated amphibian spinal cord. Can. J. Physiol. Pharrnacol. 57: 887-899. 54. SUZUE,T., N. YANAIHARA & M. OTSUKA.1981. Actions of vasopressin, gastrin releasing peptide and other peptides on neurons of newborn rat spinal cord in v i m . Neurosci. Lett. 26: 137-142. 55. STANZIONE, P. & W. ZIEGLGANSBERGER. 1983. Action of neurotensin on spinal cord neurons in the rat. Brain Res. 268: 111-118. 56. HENRY,J. L. 1982. Electrophysiological studies on the neuroactive properties of neurotensin. Ann. N.Y. Acad. Sci. 400:216-226. 57. MILETIC,V. & M. RANDIC.1979. Neurotensin excites cat spinal neurons located in laminae 1-111. Brain Res. 169: 600-604. 58. DICK,E. & R. F. MILLER.1981. Peptides influence retinal ganglion cells. Neurosci. Lett. 26: 131-135. 59. ZALUTSKY, R. A. & R. F. MILLER.1986. Neurotensin actions in the retina: Mechanisms and variability. Brain Res. 371: 360-363. W. H. & J. H. SZURSZEWSKI. 1989. Neurotensin facilitates release of substance 60. STAPELFELDT, P in the guinea-pig inferior mesenteric ganglion. J. Physiol. 411: 325-345. 61. STAPELFELDT, W. H. & J. H. SZURSZEWSKI. 1989. Central neurotensin nerves modulate colocolonic reflex activity in the guinea-pig inferior mesenteric ganglion. J. Physiol. 411:347-365.

SHI & BUNNEY ACTIONS OF NEUROTENSIN

145

W. H. & J. H. SZURSZEWSKI. 1989. The electrophysiological effects of neurotensin 62. STAPELFELDT, on neurones of guinea-pig prevertebral sympathetic ganglia. J. Physiol. 411 301-323. J. T., Y. KATAYAMA & R. A. NORTH.1979. The action of neurotensin on single my63. WILLIAMS, enteric neurones. Eur. J. Pharmacol. 59: 181-186. 64. POUSTIS,C., J. MAZELLA, P. KITABGI& J.-P. VINCENT.1984. High-affinity neurotensin binding sites in differentiated neuroblastoma NIE115 cells. J. Neurochem. 42: 1094-1100. Y., H. HIGASHIDA & N. MIKI.1984. A single class of neurotensin receptors with 65. NAKAGAWA, high affinity in neuroblastoma x glioma NG108-15 hybrid cells that mediate facilitation of synaptic transmission. J. Neurosci. 4: 1653-1661. G. & B. HAMPRECHT. 1982. Bradykinin induces hyperpolarizations in rat glioma cells 66. REISER, and in neurobiastoma x glioma hybrid cells. Brain Res. 239: 191-199. H., Y. NAKAGAWA & N. MIKI.1984. Facilitation of synaptic transmission by pros67. HIGASHIDA, taglandin D2 at synapses between NG108-15 hybrid and muscle cells. Brain Res. 295: 113-119. 1987. Neurotensin and acetylcholine evoke common re68. HIRONO,C., I. ITO & H. SUGIYAMA. sponses in frog oocytes injected with rat brain messenger ribonucleic acid. J. Physiol. 382: 523-535. E. NADLER & M. LUPU. 1985. Inositol 1,4,5,-trisphosphate mimics mus69. ORON,Y.,N. DASCAL, carinic response in Xenopus oocyte. Nature 3l3: 141-143. N., 8. GILLCI & Y.LASS.1985. Role of calcium mobilization in mediation of acetylcholine70. DASCAL, evoked chloride currents in Xenopus laevis oocytes. J. Physiol. 366: 299-313. N., E. M. LANDAU& Y.LASS.1984. Xenopus oocyte resting potential, muscarine re71. DASCAL, sponses and the role of calcium and guanosine 3: 5’-cyclic monophosphate. J. Physiol. 352: 551-574. 1982. Cholinergic and catecholaminergic receptors 72. KUSANO,K.,R. MILEDI& J. STINNAKRE. in the Xenopus oocyte membrane. J. Physiol. 328: 147-170.