Intrathecal administration of NTS1 agonists reverses nociceptive behaviors in a rat model of neuropathic pain.

Chronic neuropathic pain arising from peripheral nerve damage is a severe clinical issue where there is a major unmet medical need. We previously demonstrated that both neurotensin (NT) receptor subtypes 1 (NTS1) and 2 (NTS2) are involved in mediating the naloxone-insensitive antinociceptive effects of neurotensin in different analgesic tests including hotplate, tail-flick, and tonic pain. However, the role of these receptors in neuropathic pain management has been poorly investigated. In the present study, we therefore examined whether intrathecal delivery of NTS1 agonists was effective in reducing neuropathic pain symptoms in rats. Neuropathy was induced by sciatic nerve constriction (CCI model), and the development of mechanical allodynia and thermal hyperalgesia on the ipsi- and contralateral hind paws was examined 3, 7, 14, 21, and 28 days post-surgery. CCI-operated rats exhibited significant increases in thermal and mechanical hypersensitivities over a 28-day testing period. Spinal injection of NT to CCI rats alleviated the behavioral responses to radiant heat and mechanical stimuli, with a maximal reversal of 91% of allodynia at 6 µg/kg. Intrathecal administration of the NTS1-selective agonist, PD149163 (30-90 µg/kg) also produced potent anti-allodynic and anti-hyperalgesic effects in nerve-injured rats. Likewise, heat hyperalgesia and tactile allodynia produced by CCI of the sciatic nerve were fully reversed by the NTS1 agonist, NT69L (5-25 µg/kg). Altogether, these results support the idea that the NTS1 receptor subtype is involved in pain modulation, and the potential use of NTS1 agonists for the treatment of painful neuropathies.

NT69L, a novel analgesic, shows synergy with morphine.

BACKGROUND: Neurotensin (NT) is a neuropeptide with antinociceptive effects that are mediated through NT receptors, of which there are three known subtypes (NTS1, NTS2, and NTS3). Morphine is a mu-opioid receptor agonist commonly used for pain treatment but is associated with side effects that can be serious. We hypothesize that selective NT receptor agonists may represent a novel class of analgesics and their use in conjunction with morphine will have synergistic properties which may reduce the dose of morphine administered and its side effects. METHODS: The antinociceptive activity of an NT agonist (NT69L) and morphine was studied in rats using the hot plate test to determine if there is synergism between the two drugs in reducing pain. The NTS2 receptor antagonist, levocabastine, was used to determine the receptor subtype involved in the analgesic effect of NT69L and morphine. RESULTS: The administration of both NT69L and morphine resulted in a dose-dependent analgesic effect. The isobolographic analysis demonstrated that the combination of sub-analgesic doses of NT69L and morphine was synergistic in the hot plate test. Pretreatment with the NTS2 receptor antagonist, levocabastine attenuated the antinociceptive effect of NT69L and the combined effect of NT69L and morphine in the hot plate test. CONCLUSION: The results support the hypothesis that the synergistic combination of NT69L and morphine would improve the pharmacological treatment of pain while minimizing specific adverse effects of each of the drugs at a higher dose. NTS2 is important for the antinociceptive effect of NT69L and morphine.

Antinociceptive, hypothermic, hypotensive, and reinforcing effects of a novel neurotensin receptor agonist, NT69L, in rhesus monkeys.

Neurotensin (NT) is a tridecapeptide found in the nervous system, as well as elsewhere in the body. It has anatomic and functional relationships to dopaminergic neurons in brain. NT has been implicated in the actions of antipsychotic drugs and psychostimulants, and animal studies suggest that neurotensin directly injected into brain has reinforcing effects. Previously, we showed that one of our brain-penetrating analogs of neurotensin, NT69L (N-methyl-L-Arg, L-Lys, L-Pro, L-neo-Trp, L-tert-Leu, L-Leu), has many pharmacological effects in rats including antinociception, hypothermia, and blockade of the hyperactivity caused by psychostimulants (cocaine, D-amphetamine, and nicotine). Since these studies in rats suggest that this compound may have clinical use in humans, we were interested to know what effects NT69L had in primates. NT69L caused a potent antinociceptive effect against capsaicin (0.1 mg)-induced allodynia in 46 degrees C water in rhesus monkeys, inducing 40% of the maximal possible effect at an intravenous dosage of 0.03 mg/kg; its hypotensive effects precluded evaluation of higher dosages. Core temperature measured by rectal probe was modestly reduced at 0.01 and 0.03 mg/kg. In an intravenous self-administration procedure, NT69L was without reinforcing effects at any dose, including those that caused other pharmacological effects, and did not alter cocaine-maintained behavior when administered as a pretreatment.

Neurotensin agonists block the prepulse inhibition deficits produced by a 5-HT2A and an alpha1 agonist.

RATIONALE: Neurotensin (NT) agonists have been proposed as potential antipsychotics based exclusively upon their ability to inhibit dopamine-2 (D2) receptor transmission. Several other pharmacological mechanisms have been implicated in enhancing the antipsychotic profile produced by D2 inhibition alone. These include inhibition of 5-HT2A and alpha1-adrenoceptors. Recently, we reported that systemic administration of the neurotensin agonist PD149163 blocks deficits in prepulse inhibition (PPI) of the startle reflex produced by the 5-HT2A receptor agonist DOI. This suggested that NT agonists could inhibit 5-HT2A modulation of neurotransmission. OBJECTIVE: To determine if other peripherally administered NT agonists shared this effect, we examined the effects of NT69L, another NT agonist, on DOI-induced PPI deficits. In addition, to determine if NT agonists also inhibit alpha1-adrenoceptor neurotransmission, we examined the effects of PD149163 and NT69L on PPI deficits induced by the alpha1-adrenoceptor agonist, cirazoline. METHODS: In the NT69L/DOI study, rats received subcutaneous (SC) injections of NT69L (0, 0.1, 1, or 2 mg/kg) followed 30 min later by SC saline or DOI (0.5 mg/kg). In the NT agonist/cirazoline studies, animals received SC injections of either PD149163 (0, 0.01, 0.1, or 1 mg/kg) or NT69L (0, 0.01, 0.1, or 1 mg/kg) followed 30 min later by SC saline or cirazoline (0.7 mg/kg). Animals were tested in startle chambers 20 min later. RESULTS: In all three experiments the PPI disruption produced by DOI and cirazoline was blocked by the NT agonists. CONCLUSIONS: These findings provide strong evidence that NT agonists inhibit 5-HT2A and alpha1-adrenoceptor modulation of neurotransmission, pharmacological effects that, in conjunction with their known inhibition of dopamine transmission, strengthen the antipsychotic potential of NT agonists.

The effects of systemic NT69L, a neurotensin agonist, on baseline and drug-disrupted prepulse inhibition.

Centrally administered neurotensin (NT) produces behavioral and biochemical effects that are very similar to the effects of antipsychotic drugs. Therefore, there is much interest in the potential use of NT agonists as antipsychotic drugs. We have previously reported that PD149163, a NT(8-13) analogue, produced effects on prepulse inhibition (PPI) of startle after systemic administration that were suggestive of an atypical antipsychotic-like drug profile. To determine if these effects are shared by other peripherally administered NT agonists, we tested the effects of NT69L, a recently developed NT agonist that penetrates the CNS, on drug-induced PPI deficits. In the first experiment, rats received subcutaneous (s.c.) injections of NT69L (vehicle, 0.08, 0.25, and 1.0mg/kg) followed 30min later by subcutaneous saline or D-amphetamine (2.0mg/kg). In the second experiment, NT69L injections were followed by saline or the non-competitive NMDA antagonist dizocilpine (0.1mg/kg). Both D-amphetamine and dizocilpine significantly decreased PPI as expected. In the first experiment, NT69L significantly increased PPI levels at baseline and after D-amphetamine. In the second experiment, NT69L attenuated PPI deficits produced by dizocilpine, without increasing baseline PPI. In addition, NT69L had no effect on startle magnitude. The effects of NT69L in these studies were similar in some ways to the effects of PD149163 and were also consistent with the preclinical effects of atypical antipsychotic drugs. These data provide further support for the notion that NT agonists may have use as novel antipsychotic drugs. Furthermore, the ability of NT69L and PD149163 to attenuate dizocilpine-disrupted PPI, an antipsychotic drug effect not mediated by dopamine, suggests that NT agonists may produce some of their antipsychotic-like effects by modulating neurotransmitter systems other than dopamine, such as serotonin, noradrenaline or glutamate.

Neurotensin analog NT69L induces rapid and prolonged hypothermia after hypoxic ischemia.

OBJECTIVE: To determine whether the neurotensin analog NT69L, administered systemically, could induce mild brain hypothermia after asphyxial cardiac arrest (ACA) in rats. METHODS: The study design was experimental, blinded, randomized, and approved by the animal use committee. All rats had continuous monitoring of brain temperature and sustained 8 minutes of ACA, resuscitation, and either saline or NT69L intravenously after return of spontaneous circulation (ROSC). Rats surviving 14 days after ACA had a neurological deficit score (NDS) and a Morris Water Maze (MWM) test. RESULTS: Seven of eight rats in each group survived 14 days. Brain temperature was less than 35 degrees C 13.1 +/- 3 minutes (mean +/- standard deviation) after NT69L vs controls that remained 37.5 degrees C at the same ambient temperature (p < 0.05 ANOVA). The NT69L group remained below 35 degrees C for 300 +/- 100 minutes while the controls remained at 37.5 +/- 0.5 degrees C. The NDS in the NT69L rats was 3 +/- 3% vs controls 26 +/- 8% (p < 0.05, Kruskal-Wallis, 0% = normal, 100% = brain dead). The NT69L rats performed better on the MWM vs the controls (22 +/- 8 sec vs 45 +/- 26 sec, respectively, p < 0.05 ANOVA). CONCLUSIONS: NT69L induced rapid and prolonged mild brain hypothermia after ACA in this rat model and reduced neurological deficits.

Neurotensin analog selective for hypothermia over antinociception and exhibiting atypical neuroleptic-like properties.

Neurotensin (NT) is a tridecapeptide neurotransmitter in the central nervous system. It has been implicated in the therapeutic effects of neuroleptics. Central activity of NT can only be demonstrated by direct injection into the brain, since it is readily degraded by peptidases in the periphery. We have developed many NT(8-13) analogs that are resistant to peptidase degradation and can cross the blood-brain barrier (BBB). In this study, we report on one of these analogs, NT77L. NT77L induced hypothermia (ED(50)=6.5 mg/kg, i.p.) but induced analgesia only at the highest dose examined (20 mg/kg, i.p.). Like the atypical neuroleptic clozapine, NT77L blocked the climbing behavior in rats induced by the dopamine agonist apomorphine (600 microg/kg) with an ED(50) of 5.6 mg/kg (i.p.), without affecting the licking and the sniffing behaviors. By itself NT77L did not cause catalepsy, but it moderately reversed haloperidol-induced catalepsy with an ED(50) of 6.0 mg/kg (i.p.). Haloperidol alone did not lower body temperature, but it potentiated the body temperature lowering effect of NT77L. In studies using in vivo microdialysis NT77L showed similar effects on dopamine turnover to those of clozapine, and significantly different from those of haloperidol in the striatum. In the prefrontal cortex, NT77L significantly increased serotonergic transmission as evidenced by increased 5-hydroxyindole acetic acid:5-hydroxytryptamine (5-HIAA:5-HT) ratio. Thus, NT77L selectively caused hypothermia, over antinociception, while exhibiting atypical neuroleptic-like effects.

Antiparkinson-like effects of a novel neurotensin analog in unilaterally 6-hydroxydopamine lesioned rats.

Parkinson's disease is a neuropathological disorder involving the degeneration of dopamine neurons in the substantia nigra, with the resultant loss of their terminals in the striatum. This dopamine loss causes most of the motor disturbances associated with the disease. One animal model of Parkinson's disease involves destruction of the nigrostriatal pathway with a neurotoxin (6-hydroxydopamine) injected into this pathway. In unilaterally lesioned animals, injection of D-amphetamine causes rotation towards the lesioned side, while injection of apomorphine acting upon supersensitive postsynaptic dopamine receptors causes rotation away from the lesioned side. In this study, we tested the effects of acute and subchronic injection of a neurotensin analog (NT69L) on the rotational behavior induced by D-amphetamine (5 mg/kg) or apomorphine (600 microg/kg) in unilaterally 6-hydroxydopamine lesioned rats. Pretreatment of animals with intraperitoneal injections of NT69L (1 mg/kg) resulted in a significant reduction of apomorphine-induced contralateral rotation and D-amphetamine-induced ipsilateral rotation in these lesioned rats with an ED(50) of 40 and 80 microg/kg, respectively. After three daily injections of NT69L, its effects on this rotational behavior were unchanged, suggesting that no tolerance develops to this effect of NT69L.

Altering behavioral responses and dopamine transporter protein with antisense peptide nucleic acids.

The dopamine transporter (DAT) plays a role in locomotion and is an obligatory target for amphetamines. We designed and synthesized an antisense peptide nucleic acid (PNA) to rat DAT to examine the effect of this antisense molecule on locomotion and on responsiveness to amphetamines. Rats were injected intraperitoneally daily for 9 days with either saline, an antisense DAT PNA, a scrambled DAT PNA, or a mismatch DAT PNA. On days 7 and 9 after initial motility measurements were taken, the animals were challenged with 10 mg/kg of amphetamine and scored for motility. On day 7, there was no significant difference between the baseline levels of activity of any of the groups or their responses to amphetamine. On day 9, the antisense PNA-treated rats showed a statistically significant increase in their resting motility (P < 0.01). When these rats were challenged with amphetamine, motility of the saline-, scrambled PNA-, and mismatch PNA-treated animals showed increases of 31-, 36-, and 20-fold, respectively, while the antisense PNA-treated animals showed increases of only 3.4-fold (P < 0.01). ELISA results revealed a 32% decrease in striatal DAT in antisense PNA-treated rats compared with the saline, scrambled PNA, and mismatch PNA controls (P < 0.001). These results extend our previous findings that brain proteins can be knocked down in a specific manner by antisense molecules administered extracranially. Additionally, these results suggest some novel approaches for the treatment of diseases dependent upon the function of the dopamine transporter.

A novel neurotensin analog blocks cocaine- and D-amphetamine-induced hyperactivity.

Neurotensin is a tridecapeptide that exhibits selective anatomic and neurochemical interactions with dopaminergic systems. Since dopaminergic neurotransmission underlies many of the behavioral properties of psychostimulants, and since neurotensin has been implicated in modulating dopaminergic neurotransmitter systems, we tested the effect of our novel neurotensin analog, NT69L (N-methyl-Arg(8),L-Lys(9),L-neo-Trp(11),tert-Leu(12)]neurotensin-(8-13)), on hyperactivity caused by cocaine and D-amphetamine. Previously, we showed that NT69L reduces body temperature, blocks apomorphine-induced climbing, and haloperidol-induced catalepsy. In this study, NT69L blocked the hyperactivity induced by both cocaine and D-amphetamine administered at three different doses each, when this peptide was injected intraperitoneally. These results provide further evidence for the involvement of the neurotensin system in some of the behavioral properties of psychostimulants and suggest that NT69L may find clinical application in patients who abuse this class of compounds.

Intraperitoneal injection of antisense peptide nucleic acids targeted to the mu receptor decreases response to morphine and receptor protein levels in rat brain.

To determine the effectiveness of peptide nucleic acids (PNAs) in vivo, we designed and synthesized PNAs antisense to the mu receptor, the molecular target of morphine for inducing antinociception. Responsiveness of rats to morphine and the levels of mu receptor expression after treatment was measured. We delivered intraperitoneal injections of antisense PNAs targeted to the mu receptor (AS-MOR), mismatch PNAs (AS-MOR MM), antisense PNAs targeted to the neurotensin receptor subtype 1 (AS-NTR1), or saline and then challenged the rats with 5 mg/kg morphine (intraperitonally) or neurotensin directly into the periaqueductal gray region of the brain. To avoid tolerance, separate groups of animals were tested at 24, 48, and 72 h post-PNA treatment. Only animals treated with the AS-MOR showed a reduction in their antinociceptive response to morphine. The lack of effect of morphine on the AS-MOR rats was profound at 24 and 48 h, but animals tested at 72 h were similar to control groups. At 24 h the AS-MOR rats had a significant 55% decrease in the levels of mu receptor in their periaqueductal gray region, while AS-MOR MM rats showed no significant change. Lastly, the AS-MOR rats continued to show a normal antinociceptive response to neurotensin. This study, therefore, provides additional support for the use of PNAs to target proteins within brain by systemically administered PNAs.

Pharmacology of antidepressants.

Presently in the United States, 21 compounds have been approved by the Food and Drug Administration as antidepressants. Two additional drugs marketed outside the United States as antidepressants have been approved for obsessive-compulsive disorder. Nearly one half of all these compounds became available within the past 12 years, whereas the first antidepressant was available more than 40 years ago. After the clinical aspects of depression are introduced in this article, the pharmacology of the newer generation drugs is reviewed in relationship to the older compounds. The information in this review will help clinicians treat acute depression with pharmacological agents.

Binding of antipsychotic drugs to human brain receptors focus on newer generation compounds.

Using radioligand binding assays and post-mortem normal human brain tissue, we obtained equilibrium dissociation constants (K(d)s) for nine new antipsychotic drugs (iloperidone, melperone, olanzapine, ORG 5222, quetiapine, risperidone, sertindole, ziprasidone, and zotepine), one metabolite of a new drug (9-OH-risperidone), and three older antipsychotics (clozapine, haloperidol, and pimozide) at nine different receptors (alpha1-adrenergic, alpha2-adrenergic, dopamine D2, histamine H1, muscarinic, and serotonin 5-HT1A, 5-HT1D, 5-HT2A, and 5-HT2C receptors). Iloperidone was the most potent drug at the two adrenergic receptors. ORG 5222 was the most potent drug at dopamine D2 and 5-HT2c receptors, while ziprasidone was the most potent compound at three serotonergic receptors (5-HT1A, 5-HT1D, and 5-HT2A). At the remaining two receptors, olanzapine was the most potent drug at the histamine H1 receptor (Kd=0.087 nM); clozapine at the muscarinic receptor (Kd=9 nM). Certain therapeutic and adverse effects, as well as certain drug interactions can be predicted from a drug's potency for blocking a specific receptor. These data can provide guidelines for the clinician in the choice of antipsychotic drug.

Neurotensin: peptide for the next millennium.

Neurotensin is an endogenous tridecapeptide neurotransmitter (pGlu-Leu-Tyr-Glu-Asn-Lys-Pro-Arg-Arg-Pro-Try-Ile-Leu-OH) that was discovered by Carraway and Leeman in bovine hypothalami in the early 1970s. Since then this peptide has been the subject of a multitude of articles detailing discoveries related to its activity, receptors, localization, synthesis, and interactions with other systems. This review article does not intend to summarize again all the history of this fascinating peptide and its receptors, since this has been done quite well by others. The reader will be directed to these other reviews, where appropriate. Instead, this review attempts to provide a summary of current knowledge about neurotensin, why it is an important peptide to study, and where the field is heading. Special emphasis is placed on the behavioral studies, particularly with reference to agonists, antagonists, and antisense studies, as well as, the interaction of neurotensin with other neurotransmitters.

A single amino acid of the human and rat neurotensin receptors (subtype 1) determining the pharmacological profile of a species-selective neurotensin agonist.

The neurotensin (NT) receptor, subtype 1 (NTR1), is a 7-transmembrane-spanning receptor, forming 3 extracellular and 3 intracellular loops. Previously, we showed that the third outer loop (E3) is the binding site for NT and its analogs, several of which bind with higher affinity to rat NTR1 (rNTR1) than to human NTR1 (hNTR1). In particular, NT34 [3,1'-naphthyl-l-Ala(11)]NT(8-13) has greater than 60-fold higher affinity for rNTR1 (46 and 60 pM for transiently- and stably-transfected cells, respectively) than for hNTR1 (2.8 and 5.8 nM for transiently- and stably-transfected cells, respectively) isolated from transfected cell membranes. Previously, our molecular modeling studies of rNTR1 and hNTR1 showed that the binding pocket in the human receptor for NT34 is smaller in volume from the bulky residue Tyr(339) in the pocket center, as compared with the corresponding residue Phe(344) in the rat binding pocket. Therefore, with site-directed mutagenesis, we derived mutant forms of rNTR1(F344Y) and hNTR1(Y339F). Examination of the mutant receptors from membranal preparations of transfected cells in radioligand binding assays and with intact cells in functional assays (phosphatidyl-4,5-bisphosphate turnover) showed that the human-like rat receptor and the rat-like human receptor bound NT34 with a predicted reverse of binding compared with its binding to the wild-type receptors. These results strongly affirm our molecular modeling studies and demonstrate the importance of the study of even minor structural variations in proteins to determine the basis of significantly different drug responses, an area of focus for pharmacological research in the 21st century.

A novel neurotensin peptide analog given extracranially decreases food intake and weight in rodents.

Neurotensin decreases food intake in the rat when injected into the cerebral ventricles. We tested the effect of a novel neurotensin analog (NT69L), injected intra-peritoneally (i.p.), on weight gain and food intake in rats. Sprague-Dawley rats (270 g) were injected i. p. with either saline or NT69L at 0.001 or 0.010 mg/kg. In further experiments, larger rats at a more steady state on the growth curve (400 g) were injected with either saline or 0.010 or 1 mg/kg NT69L. Food intake, water consumption and body weight were recorded daily. Weight gain was significantly reduced in the smaller rats injected with 0.001 or 0.010 mg/kg, showing only a 8.5 and 9.0% increase in original weight, respectively, as compared to a 29% increase for the controls. The larger rats injected with 1 mg/kg, had a significant reduction in body weight with a 3.0% decrease in original body weight as compared to a 2.4% increase for the controls. Food intake was significantly reduced suggesting that the weight loss observed after injection of NT69L was attributable in part to a reduction in food intake. The genetically obese Zucker rats injected with NT69L (1 mg/kg) had a significant reduction in weight gain and food intake. NT69L significantly increased blood glucose and corticosterone levels and decreased TSH and T4 in Sprague-Dawley and Zucker rats, an effect that was only transitory. NT69L also caused a decrease in norepinephrine in both the hypothalamus and nucleus accumbens, and an increase in dopamine, 3,4-dihydroxyphenylacetic acid (DOPAC) and serotonin. In this study, NT69L exhibited a consistent and dramatic effect on body weight and food intake in Sprague-Dawley and obese Zucker rats, and enabled us to study the role that NT plays in weight control and the functional interactions of NT with brain amines, and metabolic and endocrinological parameters.

Effects of a novel neurotensin peptide analog given extracranially on CNS behaviors mediated by apomorphine and haloperidol.

Neurotensin (NT) is a neuropeptide neurotransmitter in the central nervous system. It has been implicated in the therapeutic and in the adverse effects of neuroleptics. Activity of NT in brain can only be shown by direct injection of the peptide into that organ. However, we have developed a novel analog of NT(8-13), NT69L, which is active upon intraperitoneal (i.p.) injection. Like atypical neuroleptics, NT69L blocked the climbing behavior in rats, but not the licking and sniffing behaviors of a high dose (600 microgram/kg) of the non-selective dopamine agonist apomorphine. Its blockade of climbing was very potent with an ED(50) (effective dose at 50% of maximum) of 16 microgram/kg. Both apomorphine and NT69L caused a long-lasting hypothermia, which was greater with the peptide but not synergistic in combination with apomorphine. The ED(50) of NT69L for hypothermia was 390 microgram/kg. NT69L (up to 5 mg/kg i.p.) did not produce catalepsy. However, when given before haloperidol, NT69L, but not clozapine, completely prevented catalepsy. When given after haloperidol, NT69L, but not clozapine, reversed haloperidol's cataleptic effects with an ED(50) of 260 microg/kg. There was no significant difference between the ED(50)s for hypothermia and anticataleptic effects of NT69L. However, the ED(50) for blocking the effects of apomorphine was significantly lower than the other two. These data suggest that NT69L may have neuroleptic properties in humans and may be useful in the treatment of extrapyramidal side effects caused by typical neuroleptics such as haloperidol.

Highly potent neurotensin analog that causes hypothermia and antinociception.

The tridecapeptide neurotensin has long been proposed as an endogenous neuroleptic. However, for neurotensin [or neurotensin(8-13) [NT(8-13)], the active fragment] to cause its effects, it must be administered centrally. Here, we report on an analog of NT(8-13), (N-methyl-Arg),Lys,Pro,L-neo-Trp,tert-Leu,Leu (NT69L), which contains a novel amino acid, L-neo5 degrees C (rectal), with a significant effect persisting for over 7 h. NT69L also caused a rapid (within 15 min) and persistent (for over 5 h) antinociceptive effect, as determined by the hot plate test. NT69L was overall the most potent and longest lasting neurotensin analog that has been reported. These studies provide the background for further testing of a stable, potent and long lasting neurotensin analog as a potential neuroleptic.