Facilitory effect of Δ9-tetrahydrocannabinol on hypothalamically induced feeding

  • Published on
    10-Jul-2016

  • View
    212

  • Download
    0

Embed Size (px)

Transcript

<ul><li><p>Psychopharmacology (1991 ) 103:17~176 003331589100006S Psychopharmacology </p><p> Springer-Verlag 1991 </p><p>9 </p><p>Facilitory effect of A -tetrahydrocannablnol induced feeding </p><p>on hypothalamically </p><p>Weronika Trojniar* and Roy A. Wise </p><p>Center for Studies in Behavioral Neurobiology, Department of Psychology, Concordia University, Montreal, Quebec, Canada </p><p>Received May 23, 1990 / Final version August 20, 1990 </p><p>Abstract. Six male Lewis rats were tested for the effect of Ag-tetrahydrocannabinol (A 9 THC) on feeding evoked by electrical stimulation of the lateral hypothalamus. Treatment with A9-THC (0.4 mg/kg IP) decreased fre- quency threshold for feeding by 20.5% (_+4.3), causing a leftward shift in the function relating stimulation fre- quency to the latency to begin eating 45-mg food pellets upon stimulation onset; there was no change in the asymptotic performance that was approached with suf- ficiently high stimulation frequencies. Naloxone (1 and 2 mg/kg) reduced the facilitory effect of A9-THC, but did so at doses that can inhibit feeding in the no-drug con- dition. These data are consistent with evidence implicat- ing endogenous opioids in feeding, and suggest (but do not confirm) that the facilitation of feeding by A9-THC may be mediated by endogenous opioids. The facilitation of stimulation-induced feeding by doses of A9-THC that have been found to facilitate brain stimulation reward is consistent with evidence suggesting common elements in the brain mechanisms of these two behavioral effects of medial forebrain bundle stimulation. </p><p>Key words: A9-tetrahydrocannabinol- A9-THC Feed- ing -Naloxone - Opioids </p><p>The main psychoactive component of marijuana and hashish is A9-tetrahydrocannabinol (Ag-THC). While this substance is of interest because of its potential for abuse, the mechanism of action of A9-THC is still un- clear (Martin 1986). It is only recently that specific bind- ing sites for A 9 THC have been identified in the central </p><p>* Present address." Department of Animal Physiology, University of Gdansk, Gdansk, Poland Offprint requests to: R.A. Wise, Department of Psychology, 1455 de Maisonneuve Boulevard West, Montreal, Quebec, Canada H3G 1 M8 </p><p>nervous system (Devane et al. 1988; Herkenham et al. 1988), and little is known about the functional circuitry within which these binding sites are localized. It is of theoretical interest to know whether Ag-THC shares some properties and perhaps has an overlapping mech- anism of action - with other, better studied habit-form- ing drugs (Wise and Bozarth 1987). </p><p>While it has been difficult to demonstrate direct re- warding actions of A9THC in lower animals, Gardner et al. (1988a, b, 1989) have recently shown that Ag-THC can facilitate medial forebrain bundle brain stimulation reward. Most drugs of abuse have similar effects (Kor- netsky et al. 1979; Wise 1980; Wise and Bozarth 1987). Amphetamine (Broekkamp et al. 1975; Colle and Wise 1988b) and morphine (Broekkamp et al. 1976; Rompr6 and Wise 1989) have such effects when centrally injected at sites and doses that are known to be rewarding in their own right (Phillips and LePiane 1980; Bozarth and Wise 1981; van der Kooy et al. 1982; Carr and White 1983; Hoebet et al. 1983). Thus it has been suggested that the reward-facilitating effects of drugs of abuse reflect summation of the rewarding actions of the drugs with the rewarding actions of the brain stimulation, and that a common mechanism mediates the rewarding effects of the drugs and the stimulation (Stein and Wise 1973; Kornetsky et al. 1979; Wise 1980, 1988; Wise and Bo- zarth 1987). In this context, the findings of Gardner et al. (1988a, b, 1989) offer an animal model of habit-form- ing actions of A9-THC. Because the effect of A9-THC in this model was reversed by naloxone at doses that did not affect self-stimulation in their own right, Gardner et al. (1988a, b, 1989) have suggested the possible involvement of an opioid mechanism in the euphorigenic action of A9-THC. </p><p>Lateral hypothalamic brain stimulation not only in- duces rewarding effects; it also induces feeding in sated animals (Wise 1974). A common medial forebrain bundle substrate is thought to be involved in self-stimulation and stimulation-induced feeding (Glickman and Schiff 1967; Hoebel 1969; Wise and Bozarth 1987; Wise 1988). Several habit-forming and reward-facilitating drugs </p></li><li><p>173 </p><p>stimulate feeding, including morphine, barbiturates, ben- zodiazepines (Wise and Bozarth 1987; Wise 1988), and even, under some conditions, amphetamine (Holtzman 1974; Dobrzanski 1976; Blundell and Latham 1980). Most of these drugs have been shown to facilitate stimu- lat ion- induced as well as deprivat ion- induced or spon- taneous feeding (Soper and Wise 1971; Jenck et al. 1986b, 1987b; Colle and Wise 1988a). </p><p>The effects of A9-THC on st imulat ion- induced feed- ing are particularly interesting, because if A9-THC facili- tates the brain mechanisms of natural reinforcement processes the mechanisms of food reinforcement are am- ong the most likely to be facilitated. St imulation of appe- tite is a wel l -known effect of mar i juana and hashish; appetite st imulating effects of mar i juana smoke or A9-THC injection in humans have been described in numerous anecdotal reports (Siler et al. 1933; Al lentuck and Bowman 1942; Tart 1970; Haines and Green 1970; Halikas et al. 1971), laboratory studies (Hollister 1971; Greenberg et al. 1976; Folt in et al. 1986, 1988), and clinical investigations (Noyes et al. 1976; Regulson et al. 1976; Gross et al. 1983). The present study was designed to determine if Ag-THC would facilitate st imulation- induced feeding, and, if so, whether the effect would be attenuated by doses of naloxone that attenuate this agent's reward-facil itating effects (Gardner et al. 1989). </p><p>Materials and methods </p><p>Animals and surgery. Male Lewis rats weighing approximately 400 g at the time of surgery were used. They were housed in individual cages with free access to food and water under a normal 12 h light-12 h dark illumination cycle. Ten rats were implanted, under pentobarbital anesthesia (60 mg/kg, IP), with bilateral, monopolar, stainless steel electrodes (254 gm diameter) insulated with Formvar, except for the square-cut tip. Pellegrino et al. (1979) coordinates were: 0.8 mm posterior to bregma, 1.7 mm lateral to the midline, and 8.3-8.8 mm ventral to the dural surface. The exact depth of each placement was determined on the basis of the effects of brain stimulation during surgery. Trains of cathodal square-wave con- stant current pulses (0.1 ms pulses at 100-300 gA and 60-100 Hz) were delivered as the electrodes were lowered through the brain by 0.2-0.5 mm steps, starting from 7.0 mm below dura. Animals were observed for sniffing responses (nose and vibrissae movements ac- companied by an increase in respiration rate). Each electrode was fixed at the locus of the most vigorous stimulation-induced sniffing. The electrodes were anchored to four stainless steel skull screws with dental acrylic; a stainless steel wire wrapped around two of the screws served as the anode for electrical stimulation. </p><p>Behavioral tests. After a 1-week recovery period, the rats were screened for stimulation-induced feeding. The testing was carried out in a 250 x 350 mm box with 45-rag food pellets covering the floor. The rats were taken from their home cages, where they had free access to food, and were allowed to explore the test box for 30 min before testing to allow for habituation and complete satiation. Trains of square-wave, constant current, 0.1 ms duration cathodal pulses were conducted from the stimulator to the electrode by way of a mercury commutator and flexible wire leads. Pulse duration, pulse frequency and stimulation intensity were monitored by oscil- loscope. Screening was carried out using a fixed stimulation fre- quency of 50 Hz; current intensity was raised incrementally in 20-s trials until forward search, sniffing, and eventually eating were observed. Stimulation through each electrode was tested in a separate block of trials; the one through which stimulation induced </p><p>more reliable eating was chosen for further experiment. For each rat a stimulation intensity was determined which would, at a stimu- lation frequency of 50 Hz, induce feeding with a mean latency of 10 s; the range of such frequencies was 80-400 gA. Once deter- mined, this stimulation intensity was used for all subsequent tests. Animals that did not show signs of stimulation-induced eating within 10 days of daily screening were discarded. </p><p>Six rats that ate in response to stimulation were next tested in a latency paradigm, where frequency of stimulation was varied from trial to trial. Latencies to eat were measured for each 30-s trial; stimulation was maintained for 30 s or until 5 s after the animal began to eat. Rest time of 20 s was given between trials. Four blocks of trials were given each day; stimulation frequency was progres- sively increased in the first and third blocks and decreased in the second and fourth. The between-trial increments in stimulation frequency were 5% of each previous value. The range of tested frequencies was from 18 to 50 Hz in control conditions and was adjusted as required under drug conditions. A total of 18 stimula- tion frequencies was tested per block; each block of trials took about 10 rain to complete. The four tests were averaged to obtain a daily mean latency at each stimulation frequency. Once stimula- tion-induced feeding stabilized, the animals were tested, on separate days, under drugs or drug vehicles. </p><p>Feeding threshold was defined as the stimulation frequency at which an animal began to eat with a latency of 20 s. Threshoid was derived from each rat's latency-frequency function by a method of linear interpolation. </p><p>Drugs. All drugs were administered by IP injection. The A 9 THC (DHHS, N1DA, Research Triangle Institute) was dissolved by the method of Fenimore and Loy (1971). First, A9THC was dissoived in a 20% solution of polyvinylpyrrolidone (PVP) (Sigma) prepared with ethanol. This solution was evaporated under nitrogen and the concentrated A9-THC-PVP complex was then suspended in 0.9% NaCI at a A 9 THC concentration of 1 mg/ml. A vehicle solution of 20% PVP in 0.9% NaC1 was tested on separate days (injection volumes and other parameters identical to the A9-THC injections). Vehicle and Ag-THC were each administered 15 rain before behav- ioral testing. </p><p>The animals were first tested with A9-THC at a dose of 1.5 mg/kg, the dose that Gardner et al. (1988a, b) found to signifi- cantly increase the sensitivity of their animals to rewarding brain stimulation. This dose caused a syndrome of hyper-reflexia (hyper- reactivity to environmental sounds) and catalepsy in our animals, and we subsequently tested lower doses. A dose of 0.4 mg/kg was found to be effective, both in the animals tested for stimulation- induced feeding and in other rats (of both Lewis and Long-Evans strains) tested for self-stimulation; this dose was used for subse- quent testing. </p><p>Each rat was tested once or twice under A9-THC and 1-4 times under saline and PVP. Injection days were separated by 2 days of control testing (no injections). The effect of A9-THC on stimulation induced feeding was challenged with 1 and 2 mg/kg naloxone (NAL) administered 10 min before A9-THC injection. On separate test days the naloxone effect on stimulation-induced feeding was also evaluated under vehicle injections. </p><p>Histology. After completion of behavioral testing, the animals were intracardially perfused with saline followed by 10% formalin solu- tion. The brains were removed from the skulls and stored in 10% formalin solution. The brains were frozen and sectioned at 40 lam to determine the electrode placements. </p><p>Results </p><p>Stimulat ion- induced feeding was facilitated by Ag-THC; the drug caused a leftward shift of the funct ion relating feeding latency to st imulat ion frequency, decreasing the amount of st imulation required to produce responding </p></li><li><p>174 </p><p>20" </p><p>15" </p><p>&gt;- o Z 10 UJ I'-" ,&lt; _J </p><p>5 </p><p>0 5 ' 2'0 ' 2'5 ' 3'0 3'5 4'0 4'5 5'0 </p><p>STIMULATION FREQUENCY (Hz) </p><p>Fig. 1. Effect of A9THC on latency to eat in response to stimulation of the lateral hypothalamus as a function of stimulation frequency (mean-~ SE; n = 6). Maximum time allowed for a response was 30 s. Data from trials in which eating did not occur in 20 s are not shown. </p><p>o NACL; * no inject;--~-- PVP;---- THC </p><p>20 </p><p>15' </p><p>&gt;.- o lO. </p><p>W </p><p>&lt; J </p><p>5 </p><p>0 5 ' 2'0 ' 2'5 ' 3~0 3'5 4'0 4'5 5'0 </p><p>STIMULATION FREQUENCY (Hz) </p><p>Fig. 2. Naloxone [1 mg/kg (n= 4) and 2 mg/kg (n = 6)] challenge of the effects of AgTHC on hypothalamically induced feeding. </p><p>n NAL 2 mg/kg; THC+NAL 2; - zx-- NAL 1 mg/kg; THC+NAL 1; THC </p><p>IJJ Z </p><p>LU </p><p>CO :E O r r I.J._ LU </p><p>Z &lt; </p><p>O </p><p>20- </p><p>10 </p><p>0 -V </p><p>-10 </p><p>-20 - </p><p>-30 j </p><p>NACL PdP ~o N2o NL2 ,H&amp;AL1 TH&amp;A~ TREATMENT </p><p>Fig. 3. Percentage changes (mean=l: SE) of frequency threshold for hypothalamically-induced feeding under THC and naloxone treat- ment. A change was calculated as a per cent of noninjection control. Frequency thresholds were derived from each rat's latency- frequency function by a method of graphical interpolation </p><p>with a given latency, but not altering the minimum (asymptotic) latency seen with the highest stimulation levels (Fig. 1). Analysis of variance revealed that both the effect of stimulation (F17,ss= 136.6; P</p></li><li><p>175 </p><p>feeding, but - while the effects on baseline feeding were not statistically reliable - it did so to approximately the same extent in the vehicle conditions and in the A9-THC conditions. Since statistically reliable effects on baseline stimulation-induced feeding have been reported by others with similar or lower doses of naloxone, it would not be prudent to ignore the statistically insignificant effects on baseline responding seen in the present study. It is established in the literature that naloxone exerts a small but repeatable effect on stimulation-induced eating at doses as low as 0.2 (Carr and Simon 1983) or 0.5 mg/kg (Jenck et al. 1986a). Thus it should not be quickly concluded from our data that the effects of A9-THC were mediated by an endogenous opioid mech- anism. Rather, the conservative interpretation would be that stimulation-induced feeding is, itself, partially de- pendent on an endogenous opioid mechanism and that naloxone eliminates the contribution of that mechanism to performance under both baseline and A9-THC con- ditions. </p><p>This appears to be quite different from the case of self-stimulation. In general, stimulation-induced eating seems considerably more sensitive to naloxone challenge than is self-stimulation; naloxone does not have clear-cut effects on lateral hypothalamic self-stimulation until doses of 5 mg/kg or higher (Wauquier et al. 1974; van der Kooy et al. 1977; Lorens and Sainati 1978; Esposito et al. 1980; Perry et al. 1981 ; Seeger et al. 1981 ; West and Wise 1988; Gardner et al. 1989), while it begins to have marked effects on stimulatio...</p></li></ul>