Effects of tifluadom on food consumption compared with chlordiazepoxide and kappa agonists in the rat

Neuropharmacology Vol. 24, No. 9, pp. 877-883, Printed in Great Britain. All rights reserved

0028-3908/85 $3.00 + 0.00 Copyright 0 1985 Pergamon Press Ltd

1985

EFFECTS OF TIFLUADOM ON FOOD CONSUMPTION COMPARED WITH CHLORDIAZEPOXIDE AND KAPPA AGONISTS IN THE RAT S. J. COOPER, W. R. MOORES, ANNE JACKSON and D. J. BARBER Department

of Psychology,

University

of Birmingham,

Birmingham

B15 2TT, England

(Accepfed 9 January 1985) Summary-Tifluadom (0.625-10.0 mg kg-‘) was administered to non-deprived male rats which had been accustomed to eating a highly palatable diet in a 30 min test period. This compound, an opioid benzodiazepine, produced a significant increase in consumption of food when administered by the subcutaneous route, but not after intraperitoneal injection. Both chlordiazepoxide (1.25-20.0 mg kg-‘) and the selective kappa opiate receptor agonist U-50,488 (0.3125-2.5 mg kg-‘) also produced significant hyperphagic effects in the same feeding situation. In contrast, the two kappa opiate receptor agonists, ethylketocyclazocine (0.1-3.0 mg kg-‘) and bremazocine (0.078-1.25 mg kg-‘) brought about a doserelated suppression of food intake. Hence, the effects of kappa opiate receptor agonists in the feeding situation described here were not uniform. Furthermore, tifluadom could be likened either to a benzodiazepine or to a selective kappa receptor agonist. The hyperphagia induced by tifluadom was antagonized by naloxone, suggesting that the effect was mediated by an action at opiate receptors. It was not antagonized however by Ro15-1788 (10.0 and 20.0mg kg-‘), a selective benzodiazepine receptor antagonist, ruling out possible mediation by benzodiazepine receptors. The benzodiazepine receptor antagonist, CGS 8216, exhibited intrinsic activity when administered alone, and significantly reduced food consumption in tifluadom-treated and control animals. Key words: bremazocine,

CGS 8216, chlordiazeDoxide,

ethylketocyclazocine,

food intake, hyperphagia,

naloxbne, rats, Ro15-17881 tifluadom; U-50,488. -

Tifluadom is an atypical benzodiazepine derivative, which displaces [3H]naloxone from its binding sites in homogenates of rat brain, and is effective as an analgesic in the hot-plate test in mice (Riimer, Buscher, Hill, Maurer, Petscher, Zeugner, Benson, Finner, Milkowski and Thies, 1982). Furthermore, it selectively displaces the binding of [3H]( -)bremazotine from homogenates of guinea-pig brain, and is effective in inhibiting the electrically-induced twitch response in isolated vasa deferentia of the rabbit. Tifluadom, it is thought, acts predominantly as an agonist at opiate kappa receptors (Riimer et al., 1982). In a recent study on the isomers of tifluadom, (-)tifluadom was shown to possess high selectivity for opiate sites labelled with [3H]naloxone, whereas (+)tifluadom was a much weaker opiate, which has about equal affinity for opiate and benzodiazepine sites (Kley, Scheidemantel, Bering and Muller, 1983). Tifluadom is an unusual compound and it is of considerable pharmacological interest to compare its effects with kappa opiate receptor agonists, on the one hand, and with typical effects of benzodiazepines on the other. Such comparisons assume particular importance in the study of feeding behaviour. Tifluadom has been shown to increase consumption of food in nondeprived male rats which had access to a standard animal diet and water (Morley, Levine, Grace, Kneip and Zeugner, 1983a). A hyperphagic effect was detected 2 hr after injection of 5.0 mg kg-’ tifluadom, when spontaneous food intake was increased by

slightly less than 2g. Benzodiazepines, in general, also produce hyperphagia in rats (Cooper, 1980; Cooper and Estall, 1985), and it was recently demonstrated that they produced a strong, immediate stimulation of food consumption in non-deprived rats fed a familiar, highly-palatable diet (Cooper and Gilbert, 1985; Cooper and Moores, 1985a, b). The first aim of the present study was therefore to determine whether tifluadom would also stimulate food consumption under the same experimental conditions. For the purpose of comparison with a typical benzodiazepine agonist, the effect of chlordiazepoxide was also assessed. To complete the set of comparisons, the effects on feeding of several other kappa receptor agonists were measured: the benzomorphan, ethylketocyclazocine; bremazocine (RBmer, Buscher, Hill, Maurer, Petscher, Welle, Bake1 and Akkerman, 1980); and U-50,488 (Von Voigtlander, Lahti and Ludens, 1983). In an attempt to characterize the effect of tifluadom on feeding responses further, it was also administered in combination with the opiate receptor antagonist, naloxone, or with the benzodiazepine receptor antagonists, Ro 15- 1788 and CGS 8216 (Boast, Bernard, Barbaz and Bergen, 1983; Czernick, Petrack, Kalinsky, Psychoyos, Cash, Tsai, Rinehart, Granat, Lovell, Brundish and Wade, 1982; Hunkeler, Miihler, Pieri, Pole, Bonetti, Cumin, Schaffner and Haefely, 1981). The data reveal a clear dichotomy separating tifluadom and U-50,488H, on the one hand, from ethylketocyclazocine and bremazocine, on the other. 877

878

S. J. COOPERer al. METHODS

The subjects were 172 adult male hooded rats (General strain) which were bred in the animal laboratory of the Psychology Department. They were housed individually in stainless-steel cages with continuous access to standard laboratory food pellets (modified diet 41B, Heygate & Sons, U.K.). They were maintained under a 12 hr light-12 hr dark cycle (lights on at 7a.m.) and the room temperature was kept constant at 21-22°C. The animals were weighed and handled regularly before drug testing. At the start of the studies, they weighed in the range 19&250 g. The animals were first familiarized with the highly palatable diet. Each day, animals were transferred to individual test cages, identical to the home cages, for 30min during the morning light period. About 354Og of freshly prepared diet placed in a clean Perspex petri dish was positioned inside each test cage. The diet was made up according to the following formula: 50ml Nestles brand sweetened condensed milk, 150 ml ground rat maintenance diet No. 1 (Special Diet Services Ltd, Essex, U.K.), and 200 ml distilled water. Within 10 min of blending, this food sets to a relatively firm consistency and can be scooped into small portions. The animals injected with ethylketocyclazocine were treated identically, except that they were trained to consume the palatable diet in individual opaque wooden boxes, the dimensions of which were closely similiar to those of the home cages. After animals have been adapted to this procedure over a period of l&14 days, the latency to begin consumption of the diet falls to an absolute minimum in all cases, the food is consumed rapidly, and the intake of food in a 30min period stabilizes at an asymptotic high level. During the palatable food consumption test, the water supply and the standard diet were not available. Consumption of the diet was measured by successive weighings of the food container using a sensitive electronic top-loading balance (Sartorius 1203 MP) and was recorded to an accuracy of 0.1 g. Care was taken to collect any food spillage and to make appropriate corrections to the weighings. In the first series of experiments, the effects of individual treatments with drugs were tested. Animals were allocated at random to equal groups according to the injection conditions. Racemic tifluadom hydrochloride was dissolved in isotonic saline and administered in doses of 0,0.625, 1.25, 2.5, 5.0 and 10.0 mg kg-‘. Since it has been reported that the effect of tifluadom on feeding was more potent when it was administered by subcutaneous route, compared with intraperitoneal route (Morley et al., 1983a), both routes of administration were investigated in separate groups of subjects (N = 5 per group). Chlordiazepoxide hydrochloride was dissolved in isotonic saline, and administered in doses of 0, 1.25, 2.5, 5.0, 10.0 and 20.0mg kg-’ by the

intraperitoneal route (N = 10 per group). The three remaining kappa agonists were administered by the subcutaneous route: bremazocine hydrochloride in doses of 0, 0.0781, 0.1563, 0.3125, 0.625 and 1.25 mg kg-’ (N = 10 per group), ethylketocyclazotine methanesulphonate in doses of 0, 0.1, 0.3, 1.0 and 3.0 mg kg-’ (N = 8 per group), and U-50,488 (trans-3,4-dichloro-N-methyl-N-[2-(1-pyrrolidinyl)cyclohexyll-benzeneacetamide) in doses of 0, 0.3125, 0.625, 1.25 and 2.5 mg kg-’ (N = 10 per group). In each case the vehicle was isotonic saline. All solutions of drug were injected in a volume of 1 ml kg-‘, 2@25 min before the feeding test and all doses are expressed as salts. In the drug-interaction studies, each rat received two injections prior to the 30 min feeding test. In the first, 56 animals were allocated at random to 7 equal groups. In three of the groups the animals were first injected with isotonic saline vehicle, three with 2.5 mg kg-’ tifluadom hydrochloride, and one with 5.0 mg kg-’ tifluadom hydrochloride. The second injection was vehicle, or naloxone hydrochloride in doses of either 0.1 or 0.3 mg kg-‘. Table 1 provides details of the injection schedule for the 7 groups. Naloxone was dissolved in isotonic saline, and was administered subcutaneously. In the second interaction study, 56 animals were chosen and allocated at random to 7 equal groups. Once again, in three of the groups the animals were first injected with isotonic saline vehicle, three with 2.5 mg kg-’ tifluadom hydrochloride and one with 5.0 mg kg-’ tifluadom hydrochloride. The second injection was vehicle, or CGS 8216 in doses of either 5.0 or lO.Omgkg-‘. Table 1 gives further details of the injection schedule for the 7 groups. The drug CGS 8216 (2-phenylpyrazolo[4,3-clquinolin-3(5H)-one) is a benzodiazepine receptor antagonist and was chosen because it completely reversed the hyperphagia induced by chlordiazepoxide in a separate study (Cooper and Moores, 1985a). Since it is insoluble in water, it was dispersed ultrasonically in distilled water to which Tween 80 (2 drops per 10 ml) had been added. It was administered intraperitoneally. In the third interaction study, 72 additional animals were trained and allocated at random to nine equal groups. In three of the groups the animals were first injected with isotonic saline vehicle, three with 2.5 mg kg-’ tifluadom hydrochloride, and three with 5.0 mg kg-’ tifluadom hydrochloride. The second injection was vehicle, or Ro15-1788 in doses of either 10.0 or 20.0 mg kg-‘. Table 1 gives further details of the injection schedule for the nine groups. The drug Ro15-1788 (ethyl-gfluoro-5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a] [ 1,4] benzodiazepine-3-carboxylate) is a benzodizepine receptor antagonist which has previously been shown to reverse the hyperphagia induced by clonazepam (Cooper and Gilbert, 1985). It was dispersed ultrasonically in distilled water to which Tween 80 (2 drops per 10 ml) had been added and was administered intraperitoneally. Injections were

879

Tifluadom and food intake Table 1. Design of drug-interaction N

First iniection

2

8

3

8 8 8 8 8

2.5 mg kg-’ Tifluadom 5.0 mg kg-’ Tifluadom Vehicle Vehicle 2.5 mg kg-’ Tifluadom 2.5 mg kg-’ Tifluadom

Grow

4

5 6 7

studies Second iniection

Route

Vehicle” Vehicle Vehicle 0.1 mg kg’ I Naloxone 0.3 mg kg-’ Naloxone 0.1 mg kg-’ Naloxone 0.3 mg kg-’ Naloxone

(se.)

Vehicle? Vehicle Vehicle

fi.p.f

Route

(ii) 7’1&x&x?t and CGS 8216

(iii)

I

8

2 3 4 5 6 7

8 8 8 8 8 8

T@adom

Vehicle* 2.5 mg kg-’ Tifluadom 5.0 mg kg-’ Tifluadom Vehicle Vehicle 2.5 mg kg-’ Tifluadom 2.5 mg kg-’ TiRuadom

5.0mgkg-’ lO.Omgkg5.0 mg kg-’ lO.Omgkg-

CGS 8216 CGS 8216 CGS 8216 CGS 9216

and Ro IS-1 788

I 2

8 8

3 4 5 6 I 8 9

8 8 8 8 8 8 8

Vehiclet Vehicle Vehicle 10.0 mg kg-’ Ro15-1788 10.0 mg kg-’ Rol5-1788 lO.OmgkgRol5-1788 20.0 mg kg-’ Ro15-1788 2O.Om~kg-’ Ro15-1788 20.0 mg kg-’ Ro15-1788

Vehicle* 2.5 mg kg-’ Tifluadom 5.0 mg kg-’ Tifluadom Vehicle 2.5 mg kg-’ Tifluadom 5.0 mg kg-’ Tifluadom Vehicle 2.5 mg kg-’ Tiflnadom 5.0 mg kg-’ Tifluadom

*Isotonic saline. tDistilled water + Tween 80. S.C.= Subcutaneous

given in a volume of 1 mg kg-‘, 20-25 min before the feeding test. The food intake data after single treatment with drugs were analysed using a one-way analysis of variance (ANOVA) for independent groups and comparisons between individual dose-treatments and the corresponding control group were made using Dunnett’s f-test (Winer, 1971). Orthogonal components were used in tests for linear, quadratic and cubic trends. In the drug-interaction experiments with naloxone and CGS 8216, data for six of the groups were analysed using a two-way ANOVA for independent groups. Factor 1 had two levels (0 and 2.5 mg kg-’ tifluadom), and factor 2 had three (for naloxone, 0, 0.1 and 0.3 mg kg-. ‘; for CGS 8216, 0, 5.0 and 10.0 mg kg--‘). Individual group comparisons with the control group were made using Dunnett’s t-test. The tifluadom-Ro1.5-1788 experiment was analysed using a two-way ANOVA for independent groups. Factor 1 had three levels (0, 2.5 and 50mg kg-’ ti~uadom) and factor 2 had three (0, 10.0 and 20.0 mg kg-’ Rol5-1788). It should be noted that the drug-interaction studies were designed to include replications of the effects of tifluadom (2.5 and 5.0 mg kg-‘), when administered alone. RESULTS T~$4adom

As Fig. 1 shows, tifluadom (0.625-lO.Omg kg-‘, s.c.) caused a significant increase in consumption of food (F5,24= 4.16, P < 0.01). A reliable stimulation of food intake was detected at 2.5 mg kg-’ and more.

(i.p.1

route; i.p. = intraperitoneal.

When tifiuadom was administered by the intraperitoneal route, consumption of food was significantly suppressed (Fs,24= 3.78, P < 0.05). These data are shown in Table 2. The intraperitoneal injections appeared to cause distress vocalization, however, and

l

Tifluadom

5-

V

0.070

0 15 0.313 0.625

1.25 2.5

5

IO

20

mg kg-’

Fig. I. Tifluadom, U-50,488 and chlordiazepoxide (CDP) each caused significant increases in the consumption of a highly palatable sweetened diet, in non-deprived male rats. The potencies appeared to be similar, with a significant hyperphagic effect being detected at 2.5 mg kg-’ in each case. In clear contrast, the kappa agonists bremazocine and ethylketocyclazocine (EKC) produced strong dose-related suppression of feeding. Bremazocine was more potent. Results are shown as mean (+SEM) food intake (g) in a 30min test. Levels of significance in comparisons with the corresponding vehicle (V) condition: *P i 0.05; **P < 0.01 (Dunnett’s r-test).

S. _I. COOPER et al.

880

Table 2. Effects of intraperitoneal administration of tifluadom hydrochloride on palatable food consumption (a) by non-deprived male rats (30 min test, N = 5 per group) Tifluadom (mg kg-‘) 0

0.625

1.25

2.5

5.0

10.0

17.5 f 2.1

17.4 i 1.4

19.3 i 1.9

17.5 i 2.6

13.1 f 1.6

10.0 f 2.4*

Results shown as mean *SEM. *I’ < 0.01 in comparison with control level of intake (Dunnett’s

I-test). therefore this route of administration further experiments.

was not used in

U-50,488 The specific kappa receptor agonist, U-50,488, also produced an increase in intake of food (F5,54= 3.36, P < 0.05). At 2.5 mg kg-‘, the intake was significantly greater than the baseline level (Fig. 1). Trend analysis of the feeding data after treatment with U-50,488H, showed significant linear (F,.54 = 7.79, P < 0.01) and quadratic (F,,,, = 5.46, P < 0.05) trends. Chlordiazepoxide The benzodiazepine receptor agonist stimulated consumption of food (F5,54= 8.25, P < 0.001). There were significant increases in the level of food intake over the range 5.0-20.0 mg kg-’ (Fig. 1). There was a highly significant linear trend in the data (F,,,, = 38.705, P < 0.001). Bremazocine and ethylketocyclazocine These two compounds produced clear dosedependent suppression of food intake (Fig. 1). Bremazocine was the more potent of the two. Feeding was practically abolished by bremazocine at

0.625 and 0.125 mg kg-‘. Feeding was severely reduced by ethylketocyclazocine at 3.0 mg kg-’ (Fig. 1). T@adom-naloxone Naloxone alone at 0.1 and 0.3 mg kg-’ had no effect on consumption of food (Fig. 2). At 0.1 mg kg-’ it had no effect on the hyperphagia caused by tifluadom, but at 0.3 mg kg-’ it completely antagonized the hyperphagic effect. In a two-way ANOVA carried out on the feeding data, tifluadom (2.5 mg kg-‘) had a significant hyperphagic effect (F,,d, = 7.50, P < 0.01) and there was a significant drug interaction (F2,42= 3.36, P < 0.05). The interaction occurred because of the antagonism by naloxone at 0.3 mg kg-’ of the hyperhagic effect of tifluadom. Tifluadom (5.0’mg kg-‘) produced a significant increase in consumption of food. TiJluadom-CGS

8216

At 2.5 and 5.0mg kg-‘, CGS 8216 significantly suppressed the level of food consumption, and abolished the hyperphagic effect of tifluadom (Fig. 3). In a two-way analysis of variance, tifluadom (2.5 mg kg-‘) had a significant hyperphagic effect P <0.05), CGS 8216 (2.5 and (FM, = 4.10,

25 20

20 3 d 2 15 c

5 15 d e .G z 10

:: 2

9

10

5 5 0 2.5

5

v

v

v

u

2.5

2.5 lTlfluadomj

0.1 0.3 mg kg-’

0.1

0.3 -1

0

-

1 V v

v

v

5

10

5

‘0 [CGS]

mg kg-’

Fig. 2. Hyperphagic effect of tifluadom (2.5 mg kg-‘) was unaffected by 0.1 tng kg-’ naloxone, but was fully reversed by 0.3 mg kg-’ naloxone. Naloxone, alone, had no effect on consumption of food. Histogram key: (0) control (vehicle-vehicle) group; (m) tifluadom-alone groups; (m) naloxone-alone groups; (m) drug-combination groups. Results are shown as mean (+SEM) food intake (g) in a 30min test. For levels of significance see Fig. 1 legend.

Fig. 3. Hyperphagia induced by tifluadom (S) and anorexia induced by CGS 8216 (m) in non-deprived male rats given access to a highly palatable diet. In combination (m), CGS 8216 completely prevented the hyperphagia induced by tifluadom. Results are shown as mean (+ SEM) food intake (g) in a 30min test, For levels of significance see Fig. 1 legend.

Tifluadom and food intake Table 3. Lack of effect of the benzodiazepine receptor antagonist Ro15-1788, on the hyperphagia produced by tifluadom is nondeprived male rats consuming a palatable diet (30 min test, N = 8 per group) Tifluadom RolS-1788 (mg kg-‘) 0 10.0 20.0

(mg kg-‘)

0

2.5

5.0

16.3 f 0.5 18.6 + 1.1 17.7 f 1.4

20.0 f 1.1 21.6 + 2.2 20.9 + I .6

21.1 f 1.3 20.2 f 1.5 20.3 f I .4

Results are shown as mean a significant tifluadom Rol5-1788.

*SEM. A two-way ANOVA yielded effect (P < 0.05). but no effect of

5.0 mg kgg’) produced a significant reduction in feeding (F2,42= 41.27, P < O.OOl), but the interaction was not significant (F2,42= 1.84, NS). Tifluadom did not counteract the marked effect of CGS 8216 to suppress consumption of food (Fig. 3). Tifluadom (5.0 mg kg-‘), as before, produced a reliable increase in consumption of food. Tjhadom-RolS-1788 The data for this experiment are shown in Table 3. Tifluadom produced a significant overall effect (F2,6, = 4.33, P < 0.05) resulting in increased consumption of food. Neither the main effect of Ro 15- 1788 nor the drug-interaction were significant (F ratio < 1.0 in both cases). Hence, Rol5-1788 did not affect the hyperphagic effect of tifluadom.

DlSCIJSSlON

The present study demonstrated that tifluadom (following subcutaneous but not intraperitoneal administration) stimulated the ingestion of a highly palatable diet in non-deprived animals. The hyperphagic effect was comparable with that produced by the benzodiazepine agonist, chlordiazepoxide (Fig. 1). The present results also point to a marked dissociation between different kappa receptor agonists in terms of their effects on intake of food. Like tifluadom, the selective kappa receptor agonist, U50,488, produced a significant elevation in the level of food consumption. In contrast, both bremazocine and ethylketocyclazocine suppressed eating in a doserelated manner (Fig. 1). These data can be compared with the results of a diuresis test for the activity of drugs at kappa opiate receptors in vivo. Bremazocine, proxorphan, ethylketocyclazocine, ketocyclazocine, tifluadom and U50-488 dose-dependently increased the output of urine in rats, which were initially in water balance (Leander, 1983a, b, c; Leander, 1984; Leander and Zimmerman, 1984; Slizgi and Ludens, 1982; Von Voigtlander et al., 1983). Doses of tifluadom and U-50,488 which stimulated consumption of food, and doses of bremazocine and ethylketocyclazocine which markedly suppressed intake of food, produced very similar elevations in the output of urine in the diuresis test (Leander, 1984). Hence, the diuretic effect pro-

881

duced by kappa receptor agonists and which appears to be due to the inhibition of release of vasopressin (Leander, 1984; Slizgi and Ludens, 1982) is dissociable from the hyperphagic effect described in the present report. The behavioural specificity of the suppression of feeding by bremazocine and ethylketocyclazocine can be assessed by examination of their effect on consumption of water. Leander (1984) has reported that bremazocine at 0.08 and 0.32mg kg-’ produced a marked suppression of water consumption in rehydrating rats. Similarly, ethylketocyclazocine (0.1-3.0 mg kg-‘) caused a dose-related inhibition of drinking in a 30 min test period (Turkish and Cooper, 1984). In the two largest doses, ethylketocyclazocine stopped drinking completely in the water-deprived animals. Hence, these two compounds produce, at least initially, a marked non-specific suppression of ingestional responses. In contrast, the present data rule out a similar action for either tifluadom or U-50,488. Endorphinergic mechanisms have been implicated in the control of appetite (Morley, Levine, Yim and Lowy, 1983b; Sanger, 1983), and recently attention has been particularly directed towards actions of drugs and peptides at kappa opiate receptors (Morley, Levine, Gosnell and Billington, 1984). Ethylketocyclazocine and ketocyclazocine are the kappa receptor agonists which have received most attention, although the results are not altogether consistent. Sanger and McCarthy (1981) reported that ethylketocyclazocine (0.1 and 0.3 mg kg-‘) produced a small increase (between 1 and 2g) in the intake of food over an initial 1 hr period in non-deprived male rats given access to standard laboratory food and water. Locke, Brown and Holtzman (1982) found that ethylketocyclazocine at 0.03 mg kg-‘, but not over the range 0. l-l .O mg kg-‘, produced an increase in consumption of palatable milk in non-deprived male rats. Ketocyclazocine also caused a stimulation of the consumption of milk at doses over the range 0.1-l .O mg kg-‘. Interestingly, both compounds produced only dose-related suppression of consumption of milk in squirrel monkeys (Locke et al., 1982). Increases in spontaneous feeding have also been reported following treatment with 10.0 mg kg-’ ethylketocyclazocine, but the effect was small and was reported to occur after 4 and 6 hr of the test period (Morley, Levine, Grace and Kniep, 1982). Other kappa receptor agonists have received comparatively little attention. Over a 6 hr test period, U-50,488 has been reported to produce small increases (not exceeding approx. 1 g) in spontaneous consumption of food in non-deprived rats (Morley and Levine, 1983). Similarly, over a 6 hr test period, tifluadom (5.0 and 10.0 mg kg-‘) has been reported to produce small increases in spontaneous consumption of food (Morley et al., 1983a). Bremazocine (1.&8.0mg kg-‘) produced increases in the consumption of food and water over 24 hr in non-

882

S. J. COOPERet al.

deprived female mice (Hartig and Opitz, 1983). With such a variety of procedures, species, measurement periods and doses, it is at present difficult to draw firm conclusions from such evidence, concerning the functions of kappa receptor mechanisms in the control of food intake. The present data, however, indicate that the kappa agonists, tifluadom and U50,488 produced appreciable increases in the consumption of a highly-palatable diet at a dose of 2.5 mg kg-’ and within 30 min of access to the food (Fig. 1). Hence a relatively short-latency and robust hyperphagic response could be elicited. In striking contrast, bremazocine and ethylketocyclazocine produced, within the same short time period, a marked suppression of food consumption. It would be premature to assume, therefore, that drugs which are currently recognized as kappa agonists are homogeneous with respect to their effects on appetite for food. Tifluadom is an interesting compound, which has a benzodiazepine structure and yet exhibits opiate activity (RGmer er al., 1982). At a dose of naloxone (0.3 mg kg-‘) which itself did not affect consumption of food, the hyperphagic effect of tifluadom was completely antagonized (Fig. 2). These data are consistent with the mediation of opiate receptors, in the hyperphagic effect. The benzodiazepine receptor antagonist CGS 8216 exhibited some intrinsic activity and significantly reduced consumption of food (Fig. 3). It may have acted, therefore, as an inverse agonist at benzodiazepine receptors which are involved in the control of feeding responses. The hyperphagic effect of tifluadom was completely blocked by CGS 8216, while there was no indication that tifluadom attenuated the anorectic effect of CGS 8216. In contrast to the effects of CGS 8216, the benzodiazepine receptor antagonist RolS-1788 did not exhibit intrinsic activity when administered alone (Table 3), in confirmation of earlier results (Cooper and Estall, 1985). Furthermore, it did not affect the hyperphagic effect of tifluadom, although it has been shown previously that Ro 1%1788 reversed the hyperphagic effect of the benzodiazepine receptor agonist, clonazepam (Cooper and Gilbert, 1985). Similarly, Rol5-1788 antagonized the hyperphagic effect of diazepam in rabbits (Mansbach, Stanley and Barrett, 1984). Previously, it has been shown that Rol5-1788 did not antagonize the antinociceptive effect of tifluadom in animals (Riimer et al., 1982). On the basis of the present data, therefore, it seems likely that tifluadom produced its hyperphagic effect through an action at opiate receptors, but not at benzodiazepine receptors. Further work is required to clarify the basis of the intrinsic activity of CGS 8216

and

suppressed

to determine

is the possibility effects

mediated

diazepine-receptors

(Cooper,

the

the hyperphagic

mechanism effect

of a functional by

by

of tifluadom. relationship

opiate-receptors in

the

1983), so that

control

further

which

it

There between

and

benzo-

food intake work on possible of

interactions between drugs active at the respective receptors may prove to be informative. Acknowledgements+ f )Tifluadom hydrochloride and bremazocine hydrochloride were provided by courtesy of Dr D. RGmer, Sandoz Ltd, Base], Switzerland; U-50,488 by The Upjohn Company, Kalamazoo, Michigan; ethylketocyclazocine methanesulphonate by Sterling-Winthrop Research, Surrey, U.K.; chlordiazepoxide hydrochloride by Roche Products Ltd, U.K.; CGS 8216 by CIBA-Geigy, Summit, New Jersey; Rol5-1788 by courtesy of Dr W. Haefely, Hoffman-LaRoche, Basel, Switzerland, naloxone hydrochloride by Du Pont de Nemours & Co., Glenolden. A. Jackson is supported by a postgraduate training award from the Medical Research Council of Great Britain. We thank Dr G. T. Shearman (Sandoz Ltd, Basel) for his thoughtful comments on an earlier draft. REFERENCES

Boast C. A., Bernard P. S., Barbaz B. S. and Bergen K. M. (1983) The neuropharmacology of various diazepam antagonists. Neuropharmacology 22: 151 l-l 521. Cooper S. J. (1980) Benzodiazepines as appetite-enhancing compounds. Appetite 1: 7-16. Cooper S. J. (1983) Minireview. Benzodiazepine-opiate antagonist interactions in relation to feeding and drinking. Life Sci. 32: 1043-1051. Cooper S. J. and Estall L. B. (1985) Behavioral pharmacology of food, water and salt intake in relation to drug actions at benzodiazepine receptors. Neurosci. Biobehav. Rev. 9: S-19. Cooper S. J. and Gilbert D. B. (1985) Clonazepam-induced hyperphagia in non-deprived rats: tests of pharmacological specificity with Ro5-4864, Ro5-3663, Ro15-1788 and CGS $896. Pharmac. Biochem. Behav. In press. Cooner S. J. and Moores W. R. (1985a) Chlordiazenoxideinhuced hyperphagia in non-fobd-deprived rats: elects of Rol5-1788, CGS 8216 and ZK 93 426. Eur. J. Pharmac. In press. Cooper S. J. and Moores W. R. (1985b) Benzodiazepineinduced hyperphagia in the nondeprived rat: Comparisons with CL 218,872, zopiclone, tracazolate and phenobarbital. Pharmac. Biochem. Behav. In press. Czernik A. J., Petrack B., Kalinsky H. J., Psychoyos S., Cash W. D., Tsai C., Rinehart R. K., Granat F. R., Love11 R. A., Brundish D. E. and Wade R. (1982) CGS 8216: receptor binding characteristics of a potent benzodiazepine antagonist. Life Sci. 30: 363-372. Hartig U. and Opitz K. (1983) The influence of the kappaagonist bremazocine on ingestive behaviour in mice and rats. Archs int. Pharmacodyn. Ther. 262: 4-12. Hunkeler W., Mijhler H., Pieri L., Pole P., Bonetti E. P., Cumin R., Schaffner R. and Haefely W. (1981) Selective antagonists of benzodiazepines. Nature 290: 51&516. Kley H., Scheidemantel U., Bering B. and Muller W. E. (1983) Reverse stereo-selectivity of opiate and benzodiazepine receptors for the opioid benzodiazepine tifluadom. Eur. J. Pharmac. 87: 503-504. Leander J. D. (1983a) A kappa opioid effect: increased urination in the rat. J. Pharmac. exp. Ther. 24: 89-94. Leander J. D. (1983b) Evidence that nalorphine, butorphano1 and oxilorphan are partial agonists at a K-opioid receptor. Eur. J. Pharmac. 86: 467-470. Leander J. D. (1983~) Further study of kappa opioids on increased urination. J. Pharmac. exp. Ther. 227: 35-41. Leander J. D. (1984) Kappa opioid agonists and antagonists: effects on drinking and urinary output. Appetite 5: 7-14. Leander J. D. and Zimmerman D. M. (1984) Antagonism of bremazocine-induced urination as a test for kappaopioid receptor antagonists within the phenyl-piperidine series. Drug Decl Res. 4: 421-427.

Tifluadom

and food intake

Locke K. W., Brown D. R. and Holtzman S. G. (1982) Effects of opiate antagonists and putative mu- and kappaagonists on milk intake in rat and squirrel monkey. Pharmac. Biochem. Behav. 17: 1275-1279. Mansbach R. S., Stanley J. A. and Barrett J. E. (1984) Ro 1S- 1788 and /I-CCE selectively eliminate diazepaminduced feeding in the rabbit. Pharmac. Biochem. Behav. 20: 763-766. Morley J. E. and Levine A. S. (1983) Involvement of dynorphin and the kappa opioid receptor in feeding. Peptides 4: 797-800. Morley J. E., Levine A. S., Grace M. and Kniep J. (1982) An investigation of the role of kappa opiate receptor agonists in the initiation of feeding. Life Sci. 31: 26 17-2626. Morley J. E., Levine A. S., Grace M., Kniep J. and Zeugner H. (1983a) The effect of the opioid-benzodiazepine, tifluadom, on ingestive behaviors. Eur. J. Pharmac. 93: 265-269. Morley J. E., Levine A. S., Yim G. K. and Lowy M. T. (1983b) Opioid modulation of appetite. Neurosci. Biobehav. Rev. 7: 281-305. Morley J. E., Levine A. S., Gosnell B. A. and Billington C. J. (1984) Which opioid receptor mechanism modulates feeding? Appetite 5: 61-68. Rijmer D., Biischer H., Hill R. C., Maurer R., Petscher T. J., Welle H. B. A., Bake1 H. C. C. K. and Akkerman

883

A. M. (1980) Bremazocine: a potent, long-lasting opiate kappa agonist. Life Sci. 27: 971-978. Romer D., Buscher H., Hill R. C., Maurer R., Petscher T. J., Zeugner H., Benson W., Finner E., Milkowski W. and Thies P. W. (1982) An opioid benzodiazepine. Nafure 298: 759-760. Sanger D. J. (1983) Opiates and ingestive behaviour. In: Theory in Psychopharmacology, Vol. 2 (Cooper S. J., Ed.), pp. 75-l 13. Academic Press: London. Sanger D. J. and McCarthy P. S. (1981) Increased food and water intake produced in rats by opiate receptor agonists. Psychopharmacology 74: 2 17-220. Slizgi G. R. and Ludens J. H. (1982) Studies on the nature and mechanism of the diuretic activity of the opioid analgesic ethylketocyclazocine. J. Pharmac. exp. Ther. 220: 585-591. Turkish S. and Cooper S. J. (1984) Effects of a kappa receptor agonist, ethylketocyclazocine, on water consumption in water-deprived and nondeprived rats in diurnal and nocturnal tests. Pharmac. Biochem. Behav. 21: 47-5 1. Von Voigtlander P. F., Lahti R. A. and Ludens J. H. (1983) U-50,488: a selective and structurally novel nonmu (kappa) opioid agonist. J. Pharmac. exp. Ther. 224: 7-12. Winer B. J. (1971) Statisrical Principles in Experimental Design, 2nd edn. McGraw-Hill, New York.