Brain Research Bulletin, Vol. 24, pp. 565-568. Q Pergamon Press plc, 19% Printed in the U.S.A. 0361-9230/W $3.00 + .OD

Discontinuation of Sustained Sucrose-Feeding Aggravates

Morphine Withdrawal

GEOFFREY M. SCHOENBAUM, R. J. MARTINI AND D. S. ROANE

Department of Foods and Nutrition, Dawson Hall, The University of Georgia, Athens, GA 30602

Received 24 July 1989

SCHOENBAUM, G. M., R. J. MARTIN AND D. S. ROANE. Discontinuation of sustained sucrose-feeding aggravates morphine withdrawal. BRAIN RES BULL 24(4) 565-568, 1990.-Sprague-Dawley rats were allowed ad lib access to a 20% sucrose solution in addition to their normal diet to investigate the relationship between the prolonged consumption of a high carbohydrate diet and opioid function as evidenced by opioid dependence and withdrawal. Morphine dependence, assayed by tailflick, was induced, followed by naloxone-pr~ipi~~ widows, gauged by weight loss. Sucrose-fed animals developed lowered pain thresholds prior to dependence induction relative to those of control animals, but failed to exhibit any differences from controls in the development of dependence. Weight loss during withdrawal was increased by the discontinuation of sustained sucrose-feeding. In addition, the induction of dependence first decreased, then increased the animals’ preference for sucrose. It is concluded that changes in opioid function caused by sustained sucrose-feeding are insufficient to affect the development of tolerance to morphine analgesia, but do aggravate the symptoms of precipitated withdrawal when access to sucrose is denied prior to the injection of naloxone.

Sucrose Opioid dependence Opioid withdrawal Nociception Naloxone Mo~hine

THIS present study explores the realm of opioids and food intake. Previous research has evaluated both the effect of opioids on food intake and preference and the effect of food intake on the endogenous opioid system (EOS). We present data pertaining to both areas, but focused on the latter relationship involving carbohydrate-feeding and morphine withdrawal.

Previous research on the former relationship has established that opioid agonists promote food intake (1 l), while opioid antagonists suppress food intake in a stereospecific manner (13). Opioids have also been proven to alter food preference. For example, morphine exerts a dose-dependent effect on carbohy- drate preference (9,lO). When chronically administered ICV, morphine will eventually promote carbohydrate preference (1). Similarly our own earlier work has shown chronic morphine injection IP to have a biphasic effect upon sucrose ingestion (15). In response to the repeated application of morphine, sucrose intakes were observed to decrease significantly in the fist 24 hours then, by the seventh day, si~~c~~y surpass their preinjection levels, while chow consumption declined steadily.

Studies exploring the effects of carbohydrates on subsequent opioid function have confiied that food intake and opioids are strongly linked, particularly regarding the effects of simple sugars on opioid-mediated pain perception. Most of this work has focused on the acute effect of simple sugars and suggests the activation of endogenous opioids in response to their consumption. Dum et aI. (6) found an immediate release of beta-endorphin from the

‘Requests for reprints should be addressed to Dr. R. J. Martin,

hypothalamus following the consumption of a highly palatable food. Blass et al. (3) administered an oral infusion of sucrose to cause an immediate, naltrexone-reversible analgesia of short duration. Furthermore, both ex~~rnen~ly induced hyperglyce- mia (12,16) and exposure to a dextrose-saccharine cocktail for several hours (7) were reported to produce tolerance to mor- phine analgesia, again suggesting a period of increased activity in the EOS.

In exploring the effect of refined sugars on pain perception, our work has focused on the long-term effects of ad lib sucrose- feeding on pain perception and morphine analgesia. We have found that long-term consumption of a high sucrose diet causes an apparent opioid-mediated decrease in pain thresholds and an increase in morphine potency (13). Since morphine potency increased with sucrose-feeding, we considered it a natural exten- sion to investigate the effect of a high-sucrose diet on the development of tolerance to morphine analgesia and the severity of subsequent precipitated w~~draw~. In this pursuit, we found no alteration of the development of morphine dependence or the severity of withdrawal due to sucrose-feeding when the animal’s access to sucrose is not restricted during withdrawal (15). Given two preliminary reports indicating that simple sugars in the diet had a protective effect against withdrawal symptoms (5,17), the question arose as to whether the continued consumption of sucrose during widows had prevented the more severe weight loss we had anticipated in the sucrose-fed animals. Reinforcing this

565

SCHOENBAUM, MARTIN AND ROANE

proposal was the large sucrose intake observed during naloxone- precipitated withdrawal. In normal animals opioid antagonists, such as naloxone, are documented to decrease consumption of sweet fluids (8). Our present experiment addresses this question investigating the development of morphine dependence and the severity of withdrawal in animals maintained on a high sucrose diet with emphasis on the role of sucrose feeding during with- drawal precipitated by opioid receptor blockade. To correct our earlier experiment’s shortcoming, sucrose-feeding was terminated in a portion of the animals during withdrawal to better evaluate any alterations in the withdrawal’s severity.

METHOD

Forty-five male Sprague-Dawley rats weighing between 155- I85 g (42-45 days of age) were obtained from Harlan Sprague- Dawley andCo. LaboratoryAnimals(P.0. Box29176, Indianapolis, IN 46229). All animals were housed individually on a 12-hour light/dark cycle and allowed ad lib access to water and a semipurified diet modified from American Institute of Nutrition (AIN) standards (2) consisting of 20% casein, 0.35% DL-methio- nine, 65% corn starch, 5.15% fiber, 5% corn oil, 3.5% AIN minerals, and 1% AIN vitamins. The animals were given two days to acclimate to their diet and handling conditions. At 0745 hours, on day 0, the pain threshold of each animal was assessed by the tailflick method (4). In this method, the animal is held lightly while wrapped in a towel to expose only the tail. The rat’s tail is then placed in a trough beneath a radiant heat source. The common “on” switch activates both the heat source and a timer. A photocell beneath the animal’s tail senses removal of the tail and stops both timer and heat source, thereby determining the animal’s latency to tailflick. For this experiment, a predetermined cut-off latency of 10 seconds was chosen to avoid tissue damage.

Following the determination of pain thresholds, the forty-five rats were divided into two groups of fifteen and thirty so that the average weights and tailflick latencies were similar. The larger group of rats was given access to a 20% sucrose solution at 1030 hours on day 0. From this time forward, the animals in this group had ad lib access to both sucrose solution and water. The pain thresholds were measured by the tailflick method at 0745 hours on days 1, 2, 4, 5, and daily during the induction of morphine dependence. Body weights and food and sucrose consumption were measured at 0815 hours on days 1, 2, 4, 5, daily during the induction of morphine dependence, and at 1.5, 3, and 18 hours after withdrawal was precipitated. Water intakes were also mea- sured at 1.5, 3, and 18 hours during withdrawal.

AdministraGon of morphine was begun on the evening of day 5 at 1730 hours. In order to induce morphine dependence, the animals were given IP injections of 20 mgikg morphine (Roxanol, Roxane Laboratories, Inc., Columbus, OH} at approximately 0715 and 1730 hours each day. Pain thresholds were measured each morning by the tailflick method thirty minutes ~stin~ction. These values were used to assess the development of dependence. We chose to administer 40 mgfkg of morphine daily by IP injection based on the observations of Yaksh et al. (18) and those from our own earlier experiment (IS). In both cases, the mainte- nance of this dosing schedule over a period of 7-9 days produced a near 100% decline in the percent analgesia caused initially by the morning morphine injection. Why, in this experiment, only a 50% decline in percent analgesia was observed over the 8 days of tolerance induction is not clear; however, the degree of tolerance that developed was still sufficient to cause significant withdrawal symptoms, so we do not feel that our design has compromised the experimental results.

On day 13, wi~drawal was precipitated by two IP injections of

100 1

80

u) 60 i

ii 40

20

0 124s 6 7 8 9 10 11 12 13 14

DAY

FM;. 1. Total daily intakes in kilocalories @al) for both control and sucrose-fed animals for the 24 hours prior to the date of measurement. The black bars represent the kcal consumed in chow by the control group. The diagonally striped bars represent the kcal consumed in chow by the sucrose group, and the white bars represent the kcals consumed in sucrose by the sucrose group. ‘Ihe horizontally striped bar added during withdrawat represents the chow intake of the sucrose-fed animals denied sucrose during withdrawal. The groupings of days l-5, 6-13, and 14 correspond to the periods preceding and during tolerance induction, and during wi~aw~, respectively.

I mgikg naloxone at 1400 and 1730 hours. Prior to the first injection of naloxone the sucrose-fed rats were divided into two equal groups of similar mean weights and tailflicks. Sucrose was removed from one of these two groups during withdrawal. The severity of withdrawal was gauged by the measurement of weight loss over the 18-hour period following the first injection of naloxone .

RESULTS

The food intakes are iilus~ated in Fig. 1. The intakes are divided into three parts: pre~j~tion, those during dependence induction, and those during wi~~aw~. During the preinjection phase, the animals with access to sucrose tended to be hyperphagic relative to the controls, consuming a large portion of their kilocalories (kcal) in sucrose and becoming less hyperphagic over time.

During the induction of morphine dependence, the sucrose group continued to be slightly hyperphagic. As the morphine treatment progressed, both groups decreased their chow consump- tion while the animals with access to sucrose exhibited a biphasic change in sucrose preference, tending to decrease, then increase, the proportion of the diet consumed in sucrose. ANOVA and post hoc analysis by Dunnett’s procedure showed that the percentage of the total kilocalories consumed in the form of sucrose was increased si~i~c~tly over p~in~ction levels from day 8 until the final day of the morphine injections, day 13, F(8,255)=5.41, pco.01.

When withdrawal was precipitated, intakes in each group fell. The total caloric intake of the animals from the sucrose group allowed access to sucrose during withdrawal was similar to those of the controls. Those animals denied access to sucrose consumed significantly fewer kcal than did the controls (Student’s t-test, r=2.23, p<O.OS). In the group with access to sucrose, ANOVA showed the proportion of the total caloric intake consumed in sucrose did not change from either the previous day or the preinjection level, F(2,73) =0.58, N.S., while the actual amount decreased significantly (Student’s f-test, r=2.13, pr0.05).

TABLE 1

TAILFLICK LATRNCIES IN SECONDS ( +: STANDARD ERROR) ANJJ r VALUES FROM DAYS SHOWING SIGNIFWANT DIFFERENCES

I

I I Sucrose-Fed

a I5 li I I 1 7.15 * 0.28 6.45 rt 0.22 1.93

10 2 4

5 6 7 9

6.64 -c 0.23 6.01 z!z 0.21 1.89

6.96 r 0.17 6.42 t 0.17 2.03 9.58 2 0.24 10.0 c 0 2.58 9.53 rt 0.27 9.95 * 0.05 2.13 9.42 & 0.28 9.87 rt 0.11 1.83

CONTROL SUCROSE SUCROSE VITHDRAVU

FW. 2. Total fluid intake (ml) during the 18-hour period of withdrawal for controls (control), animals with continued access to sucrose (sucrose), and animals whose access to sucrose was discontinued at the onset of withdrawal (sucrose withdrawn). Diagonal striping represents water in- take, and white represents sucrose intake.

Fluid consumption during withdrawdl is represented in Fig. 2. As the graph shows, there were no differences in water intakes between the tbree groups despite the sucrose group’s consumption of a relatively large amount of the sucrose solution. ANOVA with post hoc analysis showed that the total fluid intake of this group was significantly larger than either the controls or the group from whom sucrose was withheld, F(2,41)=21.03, pcO.01.

The tailflick latencies of the two groups are shown in Fig. 3. Prior to the morphine injections, ANOVA with repeated measures on the tailflick data collected preceding tbe period of dependence induction shows the overall latency of the sucrose group to be signi~c~dy lower than the controls. ~~1~~ Student’s t-tests were performed to evaluate differences between the groups on individual days. The days on which significant differences were present are marked in Fig. 3, and the mean tailflick latencies from those days are given in Table 1 along with standard errors and r values from the Student’s f-tests. The tailflick latencies of the animals given sucrose were signi~c~dy lower than the controls 21 hours after the introduction of the 20% sucrose solution, and the latencies remained so until the day of the Fist morphine injection. During the induction of dependence, morphine was significantly more potent in the sucrose group initially; however, the interaction [diet x days, F(7,294) =0..521, N.S.] between the

FIG 3. TaiIfiick latencies (set) prior to and duriug the iuduction of morphine tolerance for control (white squares) and sucrose-fed (black squares) animals (*p<o.o5; **p<o.ol).

two groups over the eight day period was not significant. There- fore no change in dependence development is believed to have occurred.

The percent weight lost by the animals during the l&hour period of withdrawal is shown in Fig. 4. Each group contained one outlying value (>2 standard deviations of the mean), and data from these animals were not included in the statistical analysis. On average, the control group and the group with continued access to sucrose each experienced a 7.8% weight loss, while the animals denied sucrose lost 8.7% of their weight. ANOVA, F(2,40)= 1.64, with post hoc analysis by Dumtett’s procedure (test value = 0.896, p=O.O5) showed the percentage of weight lost by the sucrose-fed rats denied sucrose during witbdrawal to be signifi- cantly higher than that lost by the controls, pcO.05.

DISCUSSION

Repeated rno~~ne injections had a biphasic effect on the animals’ preference for sucrose. The proportion of the daily kilocalorie intake that was consumed as sucrose during tolerance induction was increased over its preinjection level after three days of morphine injections, following an initial decrease. As men- tioned earlier, the acute injection of morphine has been demon- strated to decrease carbohydrate preference in high doses (9) and increase carbohydate preference at a dose of 1 mg&g (lo), while the repeated application of a large dose of morphine ICV has also been shown to eventually facilitate carbohydrate preference (1). Gur data agree with the conclusion that carbohydrate, and specif- ically sucrose, preference is decreased by the acute injection of a large dose of morphine and increased by the chronic application, either centrally or peripherally, of the same dose of motphine.

Our ~~~rnen~ during n~oxo~-lipid withdrawal differed from the findings in our first experiment (1 S), perhaps due to the addition of a second naloxone injection in an effort to prolong and intensify withdrawal. In the present experlment, n~oxone-wiping withdrawal decreased the absolute quantity of sucrose consumed but failed to affect preference for sucrose in terms of percentages compared to both day 5 and 13. The decline in quantity consumed seen during withdrawal is not sutprising in light of reports that naloxone diminishes the c6nsumption of sweet solutions for a period of several hours (8). Unchanged sucrose preference suggests that, during naloxone-precipitated morphine withdrawal, naloxone’s attenuation of absolute sucrose intake may result from a decrease in overall food intake, rather than the specific i~bition of sucrose ~ns~p~on. A longer period of observation following the second naloxone injection might have shown a recovery of sucrose consumption. During withdrawal, sucrose consumption occurred in addition to the normal intake of water.

568 SCHOENBAU~~. MARTIN AND ROANE

9.5 -

8 3 9.0-

5 0 8.S- q L 2 8.0 -

Control Sucrose i%!l3iQt:” PIG. 4. Mean percent weight loss over the 18-hour period of naloxone- precipitated withdrawal from morphine dependence for controls (control), animals with continued access to sucrose (sucrose), and animals whose access to sucrose was discontinued at the onset of withdrawal (sucrose withdrawn).

As in our previous study (15), we again failed to find differences in the development of morphine dependence due to long-term sucrose consumption. Although significant differences were seen between the tailflick late&es of the two groups, no change in the development of dependence was evident; the differences found by Student’s f-test merely confirm the findings

1.

2.

3.

4.

5.

6.

7.

8

9

of earlier research (13) finding that sustained sucrose-feeding lowers pain thresholds and increases morphine potency. It is possible that the artificial limit imposed on the tailflick latencies (10 seconds) to prevent tissue damage caused our failure to see an interaction, as the sucrose group initially registered the maximum value more often that the controls.

Measurement of weight loss during precipitated withdrawal showed significant alterations resulting from the discontinuation of sustained sucrose consumption. Reports that both sucrose-feeding (5) and glucose administered IP (17) decrease the severity of precipitated withdrawal and indications from our own research (I 5) gave us reason to expect increased weight loss due to the removal of sucrose from the sucrose-fed animals. As expected. these animals did lose more weight over the is-hour period of withdrawal than the controls. Animals in both the control group and the group given access to sucrose lost similar amounts of weight. It is our observation that sustained sucrose-feeding during the induction of morphine dependence increases the severity of naloxone-precipitated withdrawal. and the continuation of sus- tained sucrose consumption during withdrawal from morphine dependence guards against this increased severity.

The more severe withdrawal seen when the availability of both morphine and sucrose were restricted would be expected if prolonged sucrose feeding somehow impinged upon the endoge- nous opioid system to produce a dependence syndrome similar to that induced by repeated mo~hine injection. Such dependence could result if a transient release of endogenous opioids, such as that observed by Dum cv al. (6f, occurred in response to the initiation of sucrose feeding each night and waned over time.

REFERENCES

Belluzzi, J. D.; Stein, L. Endorphin mediation of feeding. In: Hoebel, B. G.; Novin, D., eds. The neural basis of feeding and reward. Brunswick, ME: Haer Institute; 1982:479484. Bieri, I. G.; Stoewsand, G. S.; Briggs, G. M.; Phillips, R. W.; Woodard, J. C.; Knapka, J. J. Report of the American Institute of Nutrition Ad Hoc Committee on Standards of Nutritional Studies. J. Nutr. 107(7):13~1348; 1977. Blass, E.; Fitzgerald, E.; Kehoe, P. Interactions between sucrose, pain and isolation distress. Pharmacol. B&hem. Behav. Z(3): 483-489; 1987. D’Amour, F. E.; Smith, D. L. A method for determining loss of pain sensation. J. Pharmacol. Exe. Ther. 72:74-79; 1941. Delanty, S. N.; Marks-Kaufman, R.; Kanarek, R. B. Role of sweet-tasting solutions in morphine addiction and withdrawal. FASEB J. 2(5):A1385; 1988. Dum, J.; Gramsch, C. H.; Herz, A. Activation of hypothalamic beta-endorphin pools by reward induced by highly palatable food. Pharmacol. Biochem. Behav. 18(3):443-447; 1983. Klein, S. P.; Green, K. F. Tolerance to morphine analgesia from brief exposure to a palatable solution. Brain Res. Bull. 21(6):963-965; 1988. Levine, A. S.; Murray, S. S.; Kneip, J.; Grace, M.; Morley, J. E. Flavor enhances the antidip~genic effect of naloxone. Physiol. Behav. 28(1):23-Z; 1982. Marks-Kaufman, R, Increased fat consumption induced by morphine administration in rats. Pharmacol. Biochem. Behav. 16(6):949-955; 1982.

10.

11.

12.

13.

14.

1.5.

16.

17.

18.

Milano, W. C.; Wild, K. D.; Hui, Y.; Hubbell, C. L.; Reid, L. D. PCP, THC, ethanol, and morphine and consumption of palatable solutions. Pharmacol. Biochem. Behav. 31(4):893-897; 1988. Morley, J. E.; Levine, A. S.; Yim, G. K.; Lowy, M. T. Opioid modulation of appetite. Neurosci. Biobehav. Rev. 7(2):281-305: 1983. Raz, I.; Hasdai, D.; SeItzEr, 2.; Meimed. R. N. Effect of hypergfy- cemia on pain perception and on efficacy of morphine analgesia in rats. Diabetes 37:1253-1259; 1988. Roane, D. S.: Martin, R. J. Continuous sucrose feeding decreases pain threshold and increases morphine potency. Pharmacol. Biochem. Behav. 35~255-229; 1990. Sanger, D. J.; McCarthy, P. S.; Metcalf, G. The effects of opiate antagonists on food intake are stereospecific. Neuropharmacology 20:45-47; 1981. Schoenbaum, G. M.; Martin, R. J.; Roane, D. S. Relationships between sustained sucrose-feeding and opioid tolerance and with- drawal. Pharmacol. Biochem. Behav. 34:91 l-914; 1989. Simon. G. S.; Dewey, W. L. Narcotics and diabetes. 1. The effects of Streptozotocin-induced diabetes on the antinociceptive potency of morphine. J. Pharmacol. Exp. Ther. 218:318-323; 1981. Simpkins, J. W.; Millard, W. J.; Katovich, M. J. Glucose modifies the opiate withdrawal response. FASEB J. 2(5):Ai 197; 1988. Yaksh, T. I_..: Kohl, R. L.; Rudy, T. A. Induction of tolerance and withdrawal in rats receiving morphine in the spinai subarachnoid space. Eur. J. Pharmacol. 42:275-284; 1977.