Introduction

The hypothalamic melanocortins (MCs) have received increasing experimental attention during the last several years. Importantly, this neuropeptide system is now thought to be critical for the regulation of food intake and energy balance. First, evidence that supports this suggestion is the finding that central administration of MC ligands produces changes in food intake.^1,2,3 Second, expression of MC peptides and their precursor proteins is regulated by energy balance.^4,5,6,7 Third, disruption of MC signaling leads to obesity.⁸ Finally, MC neurons appear to mediate the anorexic effects of the adipocyte hormone, leptin.⁹ Collectively, these data suggest that the hypothalamic MC system is an important regulator of food intake and body weight. For more complete reviews of the MC system and food intake, see Cone,¹⁰ Cone,¹¹ and Woods and Seeley.¹²

These data also suggest that pharmacotherapies based on the MC system might provide useful tools in the treatment of obesity. However, at least one potential difficulty in utilizing MC agonists as anorexic agents in humans is that the mixed MC agonist melanotan-II (MTII) reportedly produces a mild conditioned taste aversion (CTA). The general interpretation of the existence of a CTA is that the agent that caused it to develop (eg an MC agonist in this case) produces aversive effects like those of visceral illness. In a typical CTA paradigm, subjects are presented with a novel taste to sample and this is followed by the administration of the agent in question. Drugs that produce vomiting in humans and other emetic species, when used in this paradigm, generally cause a long-lasting avoidance of the novel taste to develop in both emetic and nonemetic species (eg the rat) (see Bures et al¹³ for a complete review of this literature). The relevant point is that since some MC agonists have been reported to support CTAs,¹ if an agonist for the same receptor were to be developed as a pharmacotherapy for the treatment of obesity in humans, it would be important to know whether or not any potential reduction in food intake was due to the production of visceral illness. This is especially important because visceral illness is an important predictor of drug-taking compliance (eg Hoebe et al¹⁴).

Several recent reports have found that acute administration of the nonselective MC agonist, MTII, elicits reports of nausea and discomfort in humans.^15,16,17 MTII has been assessed as a potential treatment for erectile dysfunction (ED). In those studies, MTII was found to significantly improve erectile function in human males. However, men in those studies also reported experiencing nausea and visceral discomfort, as well as reduced appetite. These data would seem to suggest that peripheral administration of MTII might not be an optimal pharmaceutical intervention for ED or obesity. However, those studies employed single- or dual-administration paradigms. It remains possible that a different route of administration or a longer-term administration of MTII might result in lessened aversive consequences.

The purpose of the present experiments was to assess the aversive consequences of the MC3/4 agonist, MTII, by varying the CTA paradigm and routes of drug administration. As a result of the potential therapeutic applications of MC agonists, we reasoned that some paradigms might minimize any aversive properties. That is, the development of a CTA may be specific to a central route of administration, and the first set of experiments was designed to assess this possibility. Alternatively, aversive consequences, including CTA development, may be specific to receiving an acute bolus of MTII or to the novelty of the compound. The second set of experiments was designed to assess these possibilities.

Experiment 1

The purpose of Experiment 1 was to assess the ability of MTII to support development of a CTA using single-injection paradigms with three routes of administration. To assess whether the delivery method of MTII administration is an important factor in the development of CTA, in three separate subexperiments MTII was administered subcutanouesly (s.c.), intraperitoneally (i.p.), or intravenously (i.v.).

Subjects and materials

All animal procedures were approved by the Institutional Animal Care and Use Committee of the University of Cincinnati. Male Long-Evans rats (Harlan, Indianapolis, IN, USA) were individually housed in clear plastic cages and maintained on a 12:12-h light/dark cycle. Pelleted chow (Harlan Teklad, Indianapolis, IN, USA) and water were provided ad libitum (except where noted). The flavor stimulus was a 0.10% (v/v) concentration of saccharin. Unless otherwise noted, sample size was eight subjects per treatment condition.

Experiment 1a (i.p.)

Rats were first habituated to 1-h daily access to water. During this hour, two bottles, each containing unflavored water, were placed on each home cage. After 12 days, all rats received two bottles containing 0.1% saccharin instead of water. Immediately following this 1-h exposure, rats received one of several doses of MTII. In Experiment 1a (i.p.), doses of MTII were 0.0, 0.3, 1.0, 3.0, and 10.0 mg/kg. MTII was dissolved in 50 mM sodium acetate/saline (pH=5.0) and administered at 1 ml/kg. On the following day, rats again received 1-h access to two bottles with unflavored tap water. On the subsequent day, a two-bottle choice test was administered in which all rats were allowed 1-h access to tap water and the saccharin solution, with the relative position of the two solutions counterbalanced across subjects.

Experiment 1b (i.v.)

Experiment 1b was identical to Experiment 1a except that MTII was administered i.v. The doses of MTII were 0.0, 0.03, 0.1, 0.3, and 1.0 mg/kg. MTII was dissolved in 50 mM sodium acetate/saline (pH=5.0) and administered at 1 ml/kg via a tail vein.

Experiment 1c (s.c.)

Experiment 1c was unlike the first two experiments in that rats were not water deprived at the time of initial saccharin exposure. However, all other details are identical, including time between training and test day. In Experiment 1c (s.c.), doses of MTII were 0.0, 3.0, 10.0, and 30.0 mg/kg. MTII was dissolved in 50 mM sodium acetate/saline (pH=5.0) and administered at 1 ml/kg.

Results

For each test, preference ratios were calculated by dividing the amount of saccharin that was consumed on test day by the total amount of liquid consumed (ie saccharin plus water intake). Preference ratios were then subjected to one-way analysis of variance (ANOVA) using drug as the independent variable. Tukey's HSD post hoc tests were then used to compare each drug dose to vehicle.

Experiment 1a (i.p.)

MTII appeared to cause a CTA in a dose-dependent fashion. Rats that received MTII had reduced preference ratios relative to saline (Figure 1). Rats that received the two highest doses of MTII had significantly lower preference ratios than vehicle-treated rats (P<0.05).

Figure 1.

Mean preference ratio (saccharin intake/total fluid intake) after i.p. administration of MTII. ^*P<0.05.

Full figure and legend (41K)

Experiment 1b (i.v.)

Overall, preference ratios in Experiment 1b appeared increased relative to Experiment 1a. However, MTII still appeared to cause the formation of a CTA, in a dose-dependent fashion. As seen in Figure 2, rats that received the highest i.v. doses of MTII (1 mg/kg) had preference ratios significantly lower than those of vehicle-treated controls (P's<0.05).

Figure 2.

Mean preference ratio (saccharin intake/total fluid intake) after i.v. administration of MTII. ^*P<0.05.

Full figure and legend (42K)

Experiment 1c (s.c.)

Subcutaneous administration of MTII dose-dependently reduced preference ratios relative to rats receiving vehicle. As depicted in Figure 3, all doses of MTII (3, 10, and 30 mg/kg) resulted in decreased saccharin intake relative to vehicle. Statistical analysis confirmed that 10 and 30 mg/kg MTII-treated rats had lower preference ratios than vehicle-treated control rats (P's<0.05).

Figure 3.

Mean preference ratio (saccharin intake/total fluid intake) after s.c. administration of MTII. ^*P<0.05.

Full figure and legend (37K)

Experiment 2a

The purpose of Experiment 2 was to assess the possible aversive quality of MTII when it is administered continuously over a 7-day period via osmotic minipump. That is, if the aversive effects of MTII are mainly short-lived and tend to subside with more chronic exposure, rats might be expected to extinguish any newly formed CTA across an entire chronic exposure period.

Methods

Phase I

Animals were comparable to those used in Experiment 1. During phase 1 of the experiment, flavor 1 (0.25% grape or cherry Kool-Aid in 0.1% saccharin, counterbalanced across rats) was available ad libitum for 7 days in place of water. Following this period of adaptation to flavor 1, unflavored tap water was available for 48 h.

Surgical procedure

Osmotic minipumps that delivered 10 l/h/day for 7 days were then implanted into the intraperitoneal cavity of all of the rats. For this procedure, the rats were briefly anesthetized with oxygen and 3% isoflurane. Incisions were made through the skin and muscle lateral to the ventral midline. One group of rats received minipumps containing MTII to deliver a dose of 1 mg/kg/day, one group received minipumps containing lithium chloride (LiCl) to deliver a dose of 1 mmol/kg/day, and a third group received minipumps containing NaCl (0.9% w/v). The incision was closed with suture and surgical staples.

Phase II

Immediately after surgery, water was removed from the home cage and replaced with flavor 2 (the alternate flavor for each rat relative to what it received in phase 1, that is, sweetened grape or cherry Kool-Aid, counterbalanced). After 7 days, all rats next received 48 h ad libitum access to unflavored tap water. Finally, all rats were given a 24-h flavor test during which flavors 1 and 2 were available without access to unflavored water. Intake of both flavors was recorded after 24 h. Body weight, food intake, and fluid intake were recorded daily throughout all phases of the experiment, and pelleted lab chow was available ad libitum throughout the experiment.

Results

Phase 1

Although no treatments were administered during phase 1, it was important to document that the groups had equivalent intakes and body weight. As expected, no differences among subsequent groups were found in these parameters. The statistical validity of these conclusions was assessed via two-way ANOVAs on (1) food intake, (2) Kool-Aid intake, and (3) body weight change. No reliable main effects or interactions were found (all P's>0.10).

Phase 2

At 24 h after surgery, all rats had a decline in food intake and body weight, and then gradually regained body weight and increased their food intake until they had reached or surpassed presurgical levels. Figure 4 represents 7-day cumulative food intake (panel a) and fluid intake (panel b). During phase 2, rats with MTII minipumps consumed less food and lost more body weight than the other groups (P's<0.05). However, both LiCl- and MTII-treated groups had small reductions in Kool-Aid intake, relative to vehicle-treated controls. No other significant differences were found between groups.

Figure 4.

Mean cumulative food and fluid intakes during phase II of Experiment 2a: (a) 7-day food intake and (b) 7-day fluid intake. ^*P<0.05.

Full figure and legend (64K)

Two-bottle choice test

To the degree that rats developed an aversion to the flavor consumed while the minipumps were implanted, intake of that flavor would be reduced on the two-bottle test. Figure 5 represents the 24-h preference ratio (flavor 1 divided by total fluid intake). As seen in the figure, rats that had flavor 2 paired with NaCl infusion had no flavor preference. However, rats that had flavor 2 paired with LiCl infusion had a significant CTA, consuming almost their entire 24-h liquid intake as flavor 1. Of particular interest were the rats that received MTII paired with flavor 2. These rats also had a preference for flavor 1 implying that MTII supported the development of a CTA. The statistical validity of these conclusions was assessed with a one-way ANOVA using drug (NaCl, LiCl, or MTII) as the independent variable. That analysis yielded a reliable main effect (F(2,17)=5.22, P<0.05. Subsequent Tukey's post hoc tests reveled that both LiCl- and MTII-treated rats had significantly lower preference ratios than vehicle-treated rats (P's<0.05).

Figure 5.

Mean preference ratio during the 24-h test after Experiment 2a. ^*P<0.05.

Full figure and legend (29K)

Experiment 2b

Experiment 2b was run concurrently with Experiment 2a. The purpose of this experiment was to determine if MTII or lithium, when administered as an acute bolus, would cause an aversion to a flavor that would continue to be present for several days following the treatment; that is, would an acutely formed CTA extinguish when the flavor remained present and no further drug (MTII or lithium) were administered?

Two additional groups of rats were included in the previous experiment. Until the time of surgery, they were treated identically to those rats, having tap water replaced with Kool-Aid (counterbalanced flavors) for 7 days. They then received 2 days access to tap water. However, these groups did not receive minipumps. Instead, tap water was replaced with the second Kool-Aid flavor. After 1 h, these rats received a single i.p. injection of either LiCl (0.15 M at 2% body weight) or/and s.c. injection of MTII (1 mg/kg in a volume of 2 ml/kg). As with the groups in Experiment 2a, flavor 2 remained the only source of liquid for the subsequent 7 days.

Figure 6 depicts the 24-h intake data for rats in the acute drug-administration groups. We did not expect rats that had acute LiCl paired with 7 days of flavor 2 to show a preference. The left bar of Figure 6 supports the hypothesis that 7 days access to flavor 2 in the absence of LiCl was sufficient to extinguish any CTA developed on the first day. However, the results from rats that received an acute administration of MTII are more interesting. The right bar of Figure 6 represents choice-test data for rats that received a single administration of MTII on the first day of flavor 2 access. Unlike LiCl, this dose of MTII was effective at supporting the development of a CTA that was resistant to extinction, even after several days of flavor 2 access, in the absence of MTII. The statistical validity of these conclusions was assessed with two t-tests. MTII-treated rats had a significant lower preference ratio than LiCl-treated rats (t(7)=3.18, P<0.05). Further, MTII-treated rats had a significantly lower preference ratio than the 0.50% expected by chance (t(7)=4.33, P<0.05). These data support the conclusion that MTII caused a CTA that is resistant to even several days of continuous extinction treatment (ie presence of the second Kool-Aid flavor in the absence of the effects of MTII).

Figure 6.

Mean preference ratio during the 24-h test after Experiment 2b. ^*P<0.05.

Full figure and legend (30K)

Experiment 3

A growing body of research demonstrates that the CNS MC system mediates the hypophagic effects of the adipocyte hormone, leptin.^4,5,18,19 Thus, the MC system appears to be a key mediator of an important adiposity signal. As it seems unlikely that illness would be a primary mechanism for the target of an adiposity hormone, we sought to assess whether leptin-induced release of the endogenous MC agonist, a-MSH, would also support a CTA. To overcome the confound that exogenously administered leptin might interact with endogenous leptin, we used ob/ob mice that lack biologically active leptin. In this experiment, a large bolus of leptin was paired with a novel flavor in both wild-type and ob/ob mice. We predicted that if leptin's effects on hypothalamic POMC cells (ie MC-releasing neurons) was also aversive, a high dose of leptin would also cause CTA development.

Methods

Subjects and materials

Mice (12 ob/ob and 12 wild type) were obtained from Jackson Labs (Bar Harbor, MA, USA). They were individually housed in air flow-controlled units and maintained on regular Purina Lab Chow 5001.

Procedure

All mice were trained to a 23-h water deprivation schedule until 1-h intakes reached asymptote. There were 4 training days. On the training days, the mice were given two water bottles filled with a novel flavor (20% sucrose flavored with either grape or cherry Kool-Aid). Following the 1-h access, the mice received either i.p. 10 g/kg leptin or i.p. 0.15 M LiCl (2% body weight). After a recovery day (access to water), mice were exposed to a second novel flavor, followed by i.p. saline (2% body weight). The mice were allowed a second exposure to each flavor followed by its respective injection with recovery days between exposures. Flavor pairings were counterbalanced across subjects. On the test day, all mice received access to both flavors for 24 h. The results were expressed as a preference ratio of the drug-paired flavor to total fluid intake.

Results

Figure 7 depicts data from Experiment 3. As seen in the figure, both ob/ob and wild-type mice were able to learn a CTA to a flavor paired with i.p. injection of LiCl. Consistent with previous reports, wild-type mice showed no evidence of CTA formation when the flavor was paired with leptin. Additionally, ob/ob mice also showed no evidence of CTA formation due to leptin. The statistical validity of these conclusions was assessed by two-way ANOVA using genotype (ob/ob vs wt) and paired-drug (LiCl vs leptin) as factors. This analysis yielded only a significant main effect of drug (F(1,19)=3.22, P<0.05). Subsequent Tukey's post hoc tests revealed that in both genotypes, the preference ratio was significantly lower in the LiCl-paired groups than in the leptin-paired groups (P's<0.05). No differences were found between ob/ob mice and controls.

Figure 7.

Mean preference ratio during the 24-h test of Experiment 3. Left-hand bars represent data for ob/ob mice and right-hand bars represent data for wild-type controls. ^*P<0.05.

Full figure and legend (40K)

Discussion

Interest in the hypothalamic MC system has increased dramatically during the last several years.¹⁰ There is little disagreement that the MC system is important for the control of food intake and body weight. Indeed, most current discussions hinge only on how the MC system might exert this control and whether they mediate the effects of adiposity signals such as leptin.²⁰ One reason for this current scientific interest is the possibility that a small molecule might be developed that could cross the blood–brain barrier and act as an MC3/4 agonist. Such a molecule might prove useful as well as lucrative as a pharmacotherapy for the treatment of obesity. One difficulty with targeting the MC receptors as a weight-loss treatment is the finding that MC agonists also elicit aversive responses, including the development of a CTA.

Thiele et al¹ reported that intracerebroventricular infusion of the nonselective MC agonist, MTII, supported the development of a CTA using an intraoral paradigm for the delivery of a novel taste stimulus. The present experiments were designed (1) to assess the ability of MTII to support CTA development using different routes of administration and (2) to assess the aversive consequences of chronic, 7-day administration of MTII. Results from Experiment 1 demonstrated that MTII, administered via i.p. and s.c. at doses capable of reducing food intake and body weight, was capable of supporting a CTA. Results from i.v. dosing were more equivocal, with only the highest doses causing a mild CTA. However, the relative efficacy between the doses was not compared directly in these experiments. Collectively, these data suggest that acute doses of the nonselective MC3/4 agonist are accompanied by aversive consequences that may impede development of pharmacotherapies for obesity based on MC agonism. This is consistent with several previous reports of MTII administration in humans.^15,16,17,21

We also investigated the ability of MTII, delivered chronically, to support the development of a CTA to a flavor that was copresent for the duration of MTII administration. In a modification of the protocol used by Chavez et al,²² rats received MTII (delivered via osmotic minipump) while water was replaced with Kool-Aid on the home cage. We had hypothesized that the aversive effects of MTII might be transient and, given the continuous 7-day administration, animals would come to tolerate its effects while maintaining the effect on body weight. During the subsequent 24-h test, however, rats that had the second Kool-Aid flavor paired with MTII showed a tremendous preference for the first flavor. These data indicate that MTII, even when administered chronically, is capable of supporting CTA development. Finally, test data from rats that received acute drug administration and chronic Kool-Aid exposure would appear to speak to the strength of the aversive consequences of MTII. Rats that received an i.p. bolus of LiCl and 7-day Kool-Aid access showed no preference whatsoever. This finding is consistent with our hypothesis that long-term access to Kool-Aid would extinguish any CTA that might have developed on the first day. However, rats that received an acute injection of MTII and 7-day access to Kool-Aid showed no evidence of CTA extinction. In fact, the magnitude of the CTA between chronic and acute MTII appears identical, even though the acute MTII was followed by 5–6 days recovery and opportunity for extinction. These data suggest that 1 mg/kg MTII can provide a powerful aversive stimulus.

The present results have important implications for the development of pharmacotherapies for obesity that target MC receptors. While the nature of the stimuli that caused CTA formation here has not been directly assessed, it is clear that MTII treatment was accompanied by stimuli capable of producing a powerful CTA. MTII has been administered to human males in the study of ED. While the variable of most interest in those studies was erectile function, the researchers also measured other psychological variables such as appetite and nausea. In those reports, MTII administration in humans has been described to cause frequent nausea and discomfort as well as decreased appetite.^15,16,17 However, men in these studies received only a single injection of larger doses of MTII than that used here. In a smaller study, repeated injections of 0.025 mg/kg MTII were also associated with mild nausea.²¹ These results and those found in the present experiments call into question the utility of MTII as a treatment for obesity. Specifically, the occurrence of visceral illness and nausea during long-term and continuous administration of low doses of MTII in humans should be carefully evaluated.

References

1. Thiele T, van DG, Yagaloff K, Fisher S, Schwartz M, Burn P, Seeley R. Central infusion of melanocortin agonist MTII in rats: assessment of c-Fos expression and taste aversion. Am J Physiol 1998; 274: R248–R254. | PubMed | ISI | ChemPort |
2. Tsujii S, Bray GA. Acetylation alters the feeding response to MSH and beta-endorphin. Brain Res Bull 1989; 23: 165–169. | Article | PubMed | ISI | ChemPort |
3. Fan W, Boston B, Kesterson R, Hruby V, Cone R. Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 1997; 385: 165–168. | Article | PubMed | ISI | ChemPort |
4. Schwartz MW, Seeley RJ, Weigle DS, Burn P, Campfield LA, Baskin DG. Leptin increases hypothalamic proopiomelanocoritin (POMC) mRNA expression in the rostral arcuate nucleus. Diabetes 1997; 46: 2119–2123. | Article | PubMed | ISI | ChemPort |
5. Mizuno T, Kleopoulos S, Bergen H, Roberts J, Priest C, Mobbs C. Hypothalamic pro-opiomelanocortin mRNA is reduced by fasting and in ob/ob and db/db mice, but is stimulated by leptin. Diabetes 1998; 47: 294–297. | Article | PubMed | ISI | ChemPort |
6. Bergendahl M, Wiemann JN, Clifton DK, Huhtaniemi I, Steiner RA. Short-term starvation decreases POMC mRNA but does not alter GnRH mRNA in the brain of adult male rats. Neuroendocrinology 1992; 56: 913–920. | PubMed | ChemPort |
7. Thornton JE, Cheung CC, Clifton DK, Steiner RA. Regulation of hypothalamic proopiomelanocortin mRNA by leptin in ob/ob mice. Endocrinology 1997; 138: 5063–5067. | Article | PubMed | ISI | ChemPort |
8. Huszar D, Lynch C, Fairchild-Huntress V, Dunmore J, Fang Q, Berkemeier L, Gu W, Kesterson R, Boston B, Cone R, Smith F, Campfield L, Burn P, Lee F. Targeted disruption of the melanocortin-4 receptor results in obesity in mice. Cell 1997; 88: 131–141. | Article | PubMed | ISI | ChemPort |
9. Seeley R, Yagaloff K, Fisher S, Burn P, Thiele T, van DG, Baskin D, Schwartz M. Melanocortin receptors in leptin effects. Nature 1997; 390: 349. | Article | PubMed | ISI | ChemPort |
10. Cone RD (ed) The melanocortin receptors.Totowa, NJ: Humana Press; 2000.
11. Cone RD. The central melanocortin system and energy homeostasis. Trends Endocrinol Metab 1999; 10: 211–216. | Article | PubMed | ISI | ChemPort |
12. Woods SC, Seeley RJ. Adiposity signals and the control of energy homeostasis. Nutrition 2000; 16: 894–902. | Article | PubMed | ISI | ChemPort |
13. Bures J, Bermudez-Rattoni F, and Yamamoto T. Conditioned taste aversion: Memory of a special kind. New York: Oxford University Press; 1998.
14. Hoebe C, de Munter J, Thijs C. Adverse effects and compliance with mefloquine or proguanil antimalarial chemoprophylaxis. Eur J Clin Pharmacol 1997; 52: 269–275.
15. Wessells H, Levine N, Hadley ME, Dorr R, Hruby V. Melanocortin receptor agonists, penile erection, and sexual motivation: human studies with Melanotan II. Int J Impot Res 2000; 12: S74–S79. | Article | PubMed |
16. Wessells H, Gralnek D, Dorr R, Hruby VJ, Hadley ME, Levine N. Effect of an alpha-melanocyte stimulating hormone analog on penile erection and sexual desire in men with organic erectile dysfunction. Urology 2000; 56: 641–646. | Article | PubMed | ISI | ChemPort |
17. Wessells H, Fuciarelli K, Hansen J, Hadley ME, Hruby VJ, Dorr R, Levine N. Synthetic melanotropic peptide initiates erections in men with psychogenic erectile dysfunction: double-blind, placebo controlled crossover study. J Urol 1998; 160: 389–393. | Article | PubMed | ISI | ChemPort |
18. Cowley MA, Smart JL, Rubinstein M, Cerdan MG, Diano S, Horvath TL, Cone RD, Low MJ. Leptin activates anorexigenic POMC neurons through a neural network in the arcuate nucleus. Nature 2001; 411: 480–484. | Article | PubMed | ISI | ChemPort |
19. Mizuno TM, Makimura H, Silverstein J, Roberts JL, Lopingco T, Mobbs CV. Fasting regulates hypothalamic neuropeptide Y, agouti-related peptide, and proopiomelanocortin in diabetic mice independent of changes in leptin or insulin. Endocrinology 1999; 140: 4551–4557. | Article | PubMed | ISI | ChemPort |
20. Benoit SC, Schwartz MW, Baskin DG, Woods SC, Seeley RJ. CNS melanocortin system involvement in the regulation of food intake and body weight. Hormones Behav 2000; 37: 299–308.
21. Dorr RT, Lines R, Levine N, Brooks C, Xiang L, Hruby VJ, Hadley ME. Evaluation of melanotan-II, a superpotent cyclic melanotropic peptide in a pilot phase-I clinical study. Life Sci 1996; 58: 1777–1784. | Article | ISI | ChemPort |
22. Chavez M, Seeley RJ, Woods SC. A comparison between the effects of intraventricular insulin and intraperitoneal LiCl on three measures sensitive to emetic agents. Behav Neurosci 1995; 109: 547–550. | Article | PubMed | ISI | ChemPort |