- Email: [email protected]
The late stage following continuous amphetamine administration to rats is correlated with altered dopamine but not serotonin metabolism
Life Sciences, Vol. 31, pp. 771-777 Printed in the U.S.A.
Pergamon Press
THE LATE STAGE FOLLOWING CONTINUOUS AMPHETAMINE ADMINISTRATION TO RATS IS CORRELATED WITH ALTERED DOPAMINE BUT NOT SEROTONIN METABOLISM Gaylord Ellison and Raj Ratan Department of Psychology University of California, Los Angeles (Received in final form June 10, 1982) Summary In previous experiments rats pretreated with slow-release d-amphetamine (d-Amp) pellets for 4 1/2 days, given a 12-hr drug-free period, and then injected with d-Amp have been found to show a behavioral syndrome which has similarities to that induced by acute injections of the hallucinogens LSD and mescaline. The present results indicate that rats administered this same drug regimen have large decreases in Dopamine (DA), dihydroxyphenyl acetic acid (Dopac), and homovanillic acid (HVA) in caudate nucleus, smaller decreases in DA with no changes in Dopac and HVA levels in nucleus accumbens, but no alterations in 5-hydroxytryptamine (5HT) and 5-hydroxyindole acetic acid (5HIAA) levels in caudate, accumbens, brainstem and hippocampus. Increased 5HIAA levels are found in rats sacrificed with pellets intact following 3 days of continuous d-Amp administration, while sleep deprived and in motor stereotypies. The late and hallucinatory stage following continuous d-amp is correlated more closely with alterations in dopamine than of 5HT. Amphetamine psychosis, which is a close drug model of paranoid schizophrenia (1), develops most reliably in the "speed runs" of amphetamine addicts, during which the drug is self-administered every few hours for a number of days (2). Such prolonged bouts of constant drug intoxication can be mimicked in animals through the use of slow-release d-Amp pellets (3). Rats (4) and monkeys (5) implanted with such pellets exhibit stages of behavior alterations. An initial period of motor stereotypies is followed by a late phase, at 4 to 5 days after pellet implantation, during which several behaviors elicited by injections of the hallucinogens LSD and mescaline appear (sudden orienting responses, limb-flicks, and wet dog shakes)~ these are accompanied by bouts of highly excited grooming. These "late-stage" behaviors can also be elicited in a more highly concentrated fashion if rats are pretreated for 4 1/2 days with a slow-release d-Amp pellet, the pellet is removed and a 12 hr rest period given, and then a low dose of d-Amp is injected (the "Reactivation paradigm"1 cf ref
6). In the following experiment we sought to determine
0024-3205/82/080771-07503.00/0 Copyright (c) 1982 Pergamon Press Ltd.
the
772
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
neurochemical correlates of this behavloral syndrome, since it may provide an an~-ml model of the alterations present in schizophrenia. Because prior research on the biochemical correlates of chronic amphetamine administration has suggested that alterations in dopamine (7) or serotonin (8) might be present, we compared the effects of this drug regimen on these two monoamines. Methods Procedure Male Simonsen albino rats (N-64) weighing 400-500 g. were adapted to individual cages and a reversed light-dark cycle for three weeks and then divided into groups of 8 animals each equated for body weight. All animals were then implanted either with slow-release d-Amp pellets (constructed and implanted as described in re±. 3) or with control pellets containing only the vehicle polyethylene glycol (PEG). In the first experiment ('Reactivation study'), these pellets were removed 4 1/2 days after implantation and 16 hrs thereafter each animal was injected either with saline, or with 1.5 mg/kg or 4.0 mg/kg d-Amp sulfate (6 groups of n=8 each). Thirty minutes later the animals were decapitated, their brains were rapidly removed, frozen in dry ice and sectioned into Imm slabs and then dissected. When the results of this experiment indicated minimal alterations in 5HT metabolism in the pellet animals, a subsequent experiment was conducted in which two groups of N=8 each ('3-day animals') were implanted with either d-Amp or control pellets and the animals were sacrificed 72 hrs. later with pellets still implanted. Dissection The following dissection guide was used (with coordinates from Konig and Klippel (9) and mean tissue wt. in mg. ± SEM): Accumbens (A9.8 to A8.4 mm; 27.8f.6). While still frozen, the optic nerves were removed from the ventral surface and a bilateral wedge was knife cut using the dorsal lateral edge of the lateral olfactory tract and the anterior commissure as guides. This contained the nucleus accumbens and the mesolimbic region ventral to it. Caudate (A8.8 to A7.2 mm; 19.9±.5) The anterior cut was along the ventral edge of the corpus callosum and the lateral edges of the septum. Brainstem (A3.0 to P3.0 nun; 29.5±.7). The hippocampus, occipital cortex and the superior and inferior colliculi were removed as was dura. Hippocampus (A4.4 to A1.6! 34.0±1.1) The hippocampus was dissected free and all surrounding white matter was removed using tweezers. High Performance Liquid Chromatography (HPLC) Assay Accumbens, caudate and hippocampus samples were homogenized in 1 cc of ice cold IN formic acid/acetone 15:85 (v/v)1 brain stem was homogenized in 5 cc of same and a 1 cc aliquot taken. Dihydroxybenzylamine (DHBA) and N-methyl-5-HT (NMSHT), used as internal standards for the catecholes and indoles respectively, were added to the homogenizing medium before dispensing. For the accumbens and caudate sample 200 ng/cc DHBA and 30 ng/cc of NM5HT were used; for the hippocampus and brain stem, 20 ng/cc DHBA and i00 ng/cc NM5HT.
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
773
Homogenized samples were centrifuged at -5 C for i0 minutes at 30,000xg. The supernatants were removed, washed with .8cc of 8zl heptane/chloroform (v/v} and .5 cc aliquots of the acqueous layer removed and evaporated to dryness in a v a c u u m centrifuge. The residues were resuspended in 200 ul of the HPLC mobile phase, vortexed for 1 minute, and stored in dry ice. For analysis the s a ~ l e s were thawed, centrifuged for 2 minutes at 15,000xg, and a 50 ul sample injected. Samples were analyzed for DA, DOPAC, HVA, 5HT, and 5HIAA using a Hewlett Packard I084A Liquid Chromatograph and a glassy carbon electrochemical detector (Bioanalytical Systems, Inc.) with the working voltage set at 0.7 volts. Column packing was 10 M LIChrosorb RPIS. The mobile phase consisted of 10.5 g sodium dihydrogen phosphate dihydrate, 40 mg sodium octyl sulfate, 40 mg Na2 EDTA monohydrate, and 130 ml of MEOH in a total volume of 1 L. at ph 3.4. The flow rate was 1.7 ml/min. Separate daily standards were run for accumbens and caudate and for hippocampus and brain stem. Peak areas were integrated and converted to concentrations. Statistical analysis All data were analyzed using a factorial analysis of variancel significant effects were then subjected to Dunnett's tests. Results
In drug naive rats (Fig 1,"Control"), d-Amp injections at 4mg/kg produced decreases in dopamine levels in caudate but not accumbens, and in both regions there were dose-dependent decreases produced by acute d-amp injections in the dopamine metabolites
30
CAUDATE 20
~ "~i
~0 0 15 10 ACCUMBENS 5 0 AMPHETAMINE DOSE MG/KG
0 415 ~'0
O 1"54"0
0 ~"5 4"0
0415410
CONTROL POSTPELLET CONTROL POSTPELLET DOPAMINE
DOPAC
01154. 0
0 1.5 4.0
CONTROL POSTPELLET HVA
i. Average (±s.e.m.} concentration of DA, DOPAC, and HVA (in ng/mg) in caudate nucleus (on top) and in nucleus accumbens (bottom). "Controls" were pretreated with vehicle pellets, while "Postpellet" rats were pretreated with d-amphetamine pelletsl both groups were then injected with either saline or d-amphetamine (1.5 or 4.0 mg/kg}. FIG.
774
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
DOPAC and HVA (both p~.01). There were also significant increases produced by d-Amp injections in 5HT in caudate and accumbens but not in hippocampus and brainstem (p~.05), and no effects on 5HIAA (Fig. 2.) Fig 1 further shows that DA levels in the caudate nucleus were greatly affected by the d-Amp pellet treatment. Compared to controls, all pellet-pretreated animals showed large decreases in DA levels in caudate (to 32% of control levels in saline-injected animals, p~.001); levels of the DA metabolites DOPAC and HVA were also decreased in the caudate of pellet-pretreated animals (p~ .05). Given this overall background of depletion of DA and its metabolites in the caudate, the overall effects of further injections of d-Amp were not dissimilar from the effects of d-Amp injections on DA metabolism in drug-naive controls. That is, in both drug-naive and in rats which had been pretreated with the d-Amp pellet, d-Amp injections of 1.5 and 4.0 mg/kg produced dose-dependent decreases in DOPAC and HVA.
CAUDATE
0
HIPPOCAMPUS ~
4
I
i~1
o
BRAIN STEM 0
o
2
~
AMPHETAMINEDOSE 0 1.,540 0 1.54.0 v 0 1.54.0 0 '1.54.0 CONTROLPOSTI~ELLET CONTROLPOSTPELLET 5HT 5HIAA
Levels of 5HT and 5 HIAA in four brain regions. groups are as described in Fig. I. FIG 2.
The
In contrast to the large depletions produced by d - ~ pellet pretreatment in caudate, in the nucleus accumbens there were much smaller effects on DA levels (to 68% of control levels in sallne-injected animals, p-.05). DOPAC and HVA levels in the accumbens of the pellet-pretreated animals were not decreased, and in several cases were actually somewhat higher than controls° Again, the dose-dependent decreases in DA metabolites produced by d-Amp injections were similar in drug-naive and pellet-pretreated
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
775
rats. In drug naive Controls, d-Amp injections increased 5HT levels in all brain regions, but this was significant only for the largest dose in Caudate. 5-HIAA levels were similarly increased slightly in all regions. In both caudate and accumbens, there were no significant alterations in 5HT or 5HIAA levels in animals pretreated with the d-Amp pellet compared to those in drug-naive controls (Fig 2); this was also true in the hippocampus and brainstem. In the second experiment, when animals were sacrificed 3 days after pellet implantation (with the pellet intact and while still in incessant motor stereotypies), there were again large depletions of DA in the caudate of pellet animals (6.5 vs 17.9 ng/mg, p<.01) and not in DA in accumbens (7.03 vs 7.57 ng/mg, n.s.). The levels of DOPAC and HVA in these animals were also similar to those in the previous study, with significant decreases in both DA metabolites in caudate of d-Amp pellet animals but not in accumbens. Although 5HT levels in caudate was not significantly less than in controls, in these animals there was a significant depletion of 5HT in the accumbens (p<.05). Furthermore, the d-Amp pellet-treated animals now showed significant increases in 5HIAA in caudate, accumbens, hippocampus, and brain stem (p<.05 in each case). Discussion In the drug-naive controls the larger d-Amp injection led to decreased DA levels in caudate but not in nucleus accumbens. But in these controls, both 1.5 and 4.0 mg/kg d-Amp induced dose-dependent decreases in both DOPAC and HVA in both caudate and accumbens. These two effects of acute d-Amp relect increased DA release accompanied by MAO inhibition (I0). It is noteworthy that the decreased DA levels produced by acute d-Amp injections only occurred in caudate whereas the decreased DA metabolites were present in both DA-rich structures. Other animals were pretreated with the d-Amp pellets for 4 1/2 days, given a 12 hour drug-free period, and then sacrificed following either saline or d-Amp injections. All animals pretreated with the d-Amp pellet evidenced large decreases in the levels of DA, DOPAC, and HVA in caudate° This finding replicates other studies indicating that d-Amp can induce large and persisting alterations in caudate DA when the drug is administered chronically to higher mammals (11), or in rats when methamphetamine is administered in very high doses (12) or if the action of the drug d-Amp is prolonged using pharmacological agents (13). These depletions are especially persistant, and are accompanied by microscopic evidence of terminal destruction (14), when d-Amp administration is continuous. But compared to the effects on caudate, there were smaller depletions of DA in nucleus accumbens of d-Amp pellet-pretreated animals, and there were no decreases in DOPAC and HVA levels. These results indicate that DA in the nucleus accumbens is partially spared from the depleting effects of prolonged d-Amp administration. It is unclear why DA terminals in one brain region should be drastically affected by the chronic d-Amp,
776
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
whereas those of a closely adjacent region are not. The present results also indicate that the overall effects of d-Amp injections on DA metabolism in controls and in animals pretreated with a chronic d-Amp pellet are generally similar (i.e., d-Amp induced dose-dependent decreases in the levels of DA, DOPAC, and HVA). A major difference, then, between drug naive and pellet-pretreated animals is in the principal locus of control of amphetamine's effects on DA metabolism. In rats pre-exposed to the chronic amphetamine pellet, caudate DA is highly depleted, and a d-Amp injection has relatively small effects on DA metabolites compared to its effects in drug-naive animals. However, the effects of d-Amp on DA metabolism in nucleus accumbens are quite similar in pellet-pretreated and drug-naive animals. This finding is of significance, for d-Amp injections to animals preteated with the d-Amp pellet induce a novel behavioral syndrome which is more similar to that produced by injections of hallucinogens such as LSD and mescaline to that induced by d-Amp injections to a drug-naive animal ( 6 ) . The finding that one of the major neurochemical correlates of this altered, hallucinogen-like response to d-Amp is a decreased ability of d-Amp to alter caudate DA metabolism compared to its near-normal effects on DA metabolism in nucleus accumbens has been recently validated using other methods. Eison et.al. (15) recently reported that continuous d-Amp administration resulted in progressively increased retention of d-Amp and its metabolites by mesolimbic but not nigrostriatal brain regions, and this was accompanied by locally enhanced levels of glucose utilization which were predominantly localized in nucleus accumbens. The implication of these studies is that following several days of continuous d-Amp administration, the predominant site of action of d-Amp shifts from caudate dopaminergic systems to those of the nucleus accumbens and other olfactory forebrain structures. There were minimal alterations in serotonin metabolism in the five day "reactivation" animals even though in previous experiments we have demonstrated that such drug-treated animals show a behavioral syndrome like that induced by hallucinogens (6). The findings from this study, therefore, do not support the suggestion by Ellinwood (16) and by Trulson and Jacobs (8) that altered serotonin metabolism plays an important role in amphetamine psychosis. Although the cats in the latter study, when administered frequent and very large injections of d-Amp, did have decreased 5HT and 5HIAA levels in brain, the amount of depletions of caudate DA were much greater. The role of 5HT in the actions of hallucinogens is not clear, for similar behaviors have been reported to be induced both by 5HT agonists and 5HT depletions. While methamphetamine can produce long-term alterations in 5HT (17) this drug has a pharmacological profile very different from that of d-Amp and can induce psychoses immediately after injection (18). The present results using d-Amp appear to indicate that, at least in rats, altered 5HT levels are not necessary for the appearance of hallucinogen-like behaviors in the d-Amp model of amphetamine psychosis. Rather, the principal alterations in 5HT observed in the present experiment were in the animals sacrificed three days after pellet implantation, while still in constant motor stereotypy and in a state of sleep deprivation. Thus, these results appear to dissociate the effects
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
777
of chronic d-amp on 5HT metabolism from the hallucinogen-like behaviors induced by chronic administration of the drug. One implication of these results is that an altered locus of control of amphetamine's effects on dopaminerglc systems (from strlatal and towards mesolimbic systems) correlates better with the heightened ability of d-Amp to elicit a hallucinogen-like behavior than with its altered effects on serotonin metabolism. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. i0. ii. 12. 13. 14. 15. 16. 17. 18.
D.S. BELL, J. Psychiat. Res. 3, 701-707 (1965). J.C. KRAMER, V. GISCHMAN, and D. LITTLEFIELD, JAMA, 201, No.5, 89-93 (1967). H.S. HUBERMAN, M.S. EISON, K.S. BRYON, and G. ELLISON, Eur. J. Pharmacol, 45, 237, 97-100 (1972). G. ELLISON, M. EISON, H. HUBERMAN, and F. DANIEL, Science, 201, 276-278 (1978). G. ELLISON, E.B. NIELSEN, and M. LYON, J. Psychiat. Res. 16, 13-22 (1981). E.B. NIELSEN, T. LEE, and G. ELLISON, Psychopharm. 68, 197-200 (1980). S.W. SNYDER, S. BANERJEE, H. YAMAMURA, and D. GREENBERG, Science, 184, 1243-1253 (1979). M.E. TRULSON, and B.L. JACOBS, Science, 205, 1295-1297 (1979). J.F.R. KONIG, and R.A. KLIPPEL, The Rat Brainl A stereotaxic atlas, pp. 7-156, Krieger Publishig Co., Inc., New York (1963). H.H. MILLER, P. SHORE, and D. Clarke, Biochem. Pharm., 29, 1347-1354 (1980). M.S. LEVINE, D.D. HULL, E. GARCIA-RILL, L. ERNOFF, N.A. BUCHWALD, and A. HELLER, Brain Res. 194, 263-268 (1980). G.C. WAGNER, G.A. RICAURTE, C.E. HOHANSON, C.R. SCHUSTER, and L.S. SEIDEN, Neurology, 30, 547-550 (1980). R.W. FULLER, and S. HEMRICK-LUECKE, Science, 209, 305-306 (1980). E. NWANZE and G. JONSSON, Neuroscience Letters, 26, 163-168 (1981). M. EISON, A. EISON, and G. ELLISON, Exp. Brain Res.,43, 281-288 (1981). E.B. ELLINWOOD, M.KILBEY JR., In I. Hanin and E. Usdin (Eds), Animal Models of Psychiat. and Neurol. (1977). A. HOTCHKISS and J. GIBB, J. Pharmcol. Exp. Ther., 214, 257-262 (1980). D.A. BELL, ARCH. GEN. PSYCHIAT., 29, 35-40 (1973).
Pergamon Press
THE LATE STAGE FOLLOWING CONTINUOUS AMPHETAMINE ADMINISTRATION TO RATS IS CORRELATED WITH ALTERED DOPAMINE BUT NOT SEROTONIN METABOLISM Gaylord Ellison and Raj Ratan Department of Psychology University of California, Los Angeles (Received in final form June 10, 1982) Summary In previous experiments rats pretreated with slow-release d-amphetamine (d-Amp) pellets for 4 1/2 days, given a 12-hr drug-free period, and then injected with d-Amp have been found to show a behavioral syndrome which has similarities to that induced by acute injections of the hallucinogens LSD and mescaline. The present results indicate that rats administered this same drug regimen have large decreases in Dopamine (DA), dihydroxyphenyl acetic acid (Dopac), and homovanillic acid (HVA) in caudate nucleus, smaller decreases in DA with no changes in Dopac and HVA levels in nucleus accumbens, but no alterations in 5-hydroxytryptamine (5HT) and 5-hydroxyindole acetic acid (5HIAA) levels in caudate, accumbens, brainstem and hippocampus. Increased 5HIAA levels are found in rats sacrificed with pellets intact following 3 days of continuous d-Amp administration, while sleep deprived and in motor stereotypies. The late and hallucinatory stage following continuous d-amp is correlated more closely with alterations in dopamine than of 5HT. Amphetamine psychosis, which is a close drug model of paranoid schizophrenia (1), develops most reliably in the "speed runs" of amphetamine addicts, during which the drug is self-administered every few hours for a number of days (2). Such prolonged bouts of constant drug intoxication can be mimicked in animals through the use of slow-release d-Amp pellets (3). Rats (4) and monkeys (5) implanted with such pellets exhibit stages of behavior alterations. An initial period of motor stereotypies is followed by a late phase, at 4 to 5 days after pellet implantation, during which several behaviors elicited by injections of the hallucinogens LSD and mescaline appear (sudden orienting responses, limb-flicks, and wet dog shakes)~ these are accompanied by bouts of highly excited grooming. These "late-stage" behaviors can also be elicited in a more highly concentrated fashion if rats are pretreated for 4 1/2 days with a slow-release d-Amp pellet, the pellet is removed and a 12 hr rest period given, and then a low dose of d-Amp is injected (the "Reactivation paradigm"1 cf ref
6). In the following experiment we sought to determine
0024-3205/82/080771-07503.00/0 Copyright (c) 1982 Pergamon Press Ltd.
the
772
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
neurochemical correlates of this behavloral syndrome, since it may provide an an~-ml model of the alterations present in schizophrenia. Because prior research on the biochemical correlates of chronic amphetamine administration has suggested that alterations in dopamine (7) or serotonin (8) might be present, we compared the effects of this drug regimen on these two monoamines. Methods Procedure Male Simonsen albino rats (N-64) weighing 400-500 g. were adapted to individual cages and a reversed light-dark cycle for three weeks and then divided into groups of 8 animals each equated for body weight. All animals were then implanted either with slow-release d-Amp pellets (constructed and implanted as described in re±. 3) or with control pellets containing only the vehicle polyethylene glycol (PEG). In the first experiment ('Reactivation study'), these pellets were removed 4 1/2 days after implantation and 16 hrs thereafter each animal was injected either with saline, or with 1.5 mg/kg or 4.0 mg/kg d-Amp sulfate (6 groups of n=8 each). Thirty minutes later the animals were decapitated, their brains were rapidly removed, frozen in dry ice and sectioned into Imm slabs and then dissected. When the results of this experiment indicated minimal alterations in 5HT metabolism in the pellet animals, a subsequent experiment was conducted in which two groups of N=8 each ('3-day animals') were implanted with either d-Amp or control pellets and the animals were sacrificed 72 hrs. later with pellets still implanted. Dissection The following dissection guide was used (with coordinates from Konig and Klippel (9) and mean tissue wt. in mg. ± SEM): Accumbens (A9.8 to A8.4 mm; 27.8f.6). While still frozen, the optic nerves were removed from the ventral surface and a bilateral wedge was knife cut using the dorsal lateral edge of the lateral olfactory tract and the anterior commissure as guides. This contained the nucleus accumbens and the mesolimbic region ventral to it. Caudate (A8.8 to A7.2 mm; 19.9±.5) The anterior cut was along the ventral edge of the corpus callosum and the lateral edges of the septum. Brainstem (A3.0 to P3.0 nun; 29.5±.7). The hippocampus, occipital cortex and the superior and inferior colliculi were removed as was dura. Hippocampus (A4.4 to A1.6! 34.0±1.1) The hippocampus was dissected free and all surrounding white matter was removed using tweezers. High Performance Liquid Chromatography (HPLC) Assay Accumbens, caudate and hippocampus samples were homogenized in 1 cc of ice cold IN formic acid/acetone 15:85 (v/v)1 brain stem was homogenized in 5 cc of same and a 1 cc aliquot taken. Dihydroxybenzylamine (DHBA) and N-methyl-5-HT (NMSHT), used as internal standards for the catecholes and indoles respectively, were added to the homogenizing medium before dispensing. For the accumbens and caudate sample 200 ng/cc DHBA and 30 ng/cc of NM5HT were used; for the hippocampus and brain stem, 20 ng/cc DHBA and i00 ng/cc NM5HT.
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
773
Homogenized samples were centrifuged at -5 C for i0 minutes at 30,000xg. The supernatants were removed, washed with .8cc of 8zl heptane/chloroform (v/v} and .5 cc aliquots of the acqueous layer removed and evaporated to dryness in a v a c u u m centrifuge. The residues were resuspended in 200 ul of the HPLC mobile phase, vortexed for 1 minute, and stored in dry ice. For analysis the s a ~ l e s were thawed, centrifuged for 2 minutes at 15,000xg, and a 50 ul sample injected. Samples were analyzed for DA, DOPAC, HVA, 5HT, and 5HIAA using a Hewlett Packard I084A Liquid Chromatograph and a glassy carbon electrochemical detector (Bioanalytical Systems, Inc.) with the working voltage set at 0.7 volts. Column packing was 10 M LIChrosorb RPIS. The mobile phase consisted of 10.5 g sodium dihydrogen phosphate dihydrate, 40 mg sodium octyl sulfate, 40 mg Na2 EDTA monohydrate, and 130 ml of MEOH in a total volume of 1 L. at ph 3.4. The flow rate was 1.7 ml/min. Separate daily standards were run for accumbens and caudate and for hippocampus and brain stem. Peak areas were integrated and converted to concentrations. Statistical analysis All data were analyzed using a factorial analysis of variancel significant effects were then subjected to Dunnett's tests. Results
In drug naive rats (Fig 1,"Control"), d-Amp injections at 4mg/kg produced decreases in dopamine levels in caudate but not accumbens, and in both regions there were dose-dependent decreases produced by acute d-amp injections in the dopamine metabolites
30
CAUDATE 20
~ "~i
~0 0 15 10 ACCUMBENS 5 0 AMPHETAMINE DOSE MG/KG
0 415 ~'0
O 1"54"0
0 ~"5 4"0
0415410
CONTROL POSTPELLET CONTROL POSTPELLET DOPAMINE
DOPAC
01154. 0
0 1.5 4.0
CONTROL POSTPELLET HVA
i. Average (±s.e.m.} concentration of DA, DOPAC, and HVA (in ng/mg) in caudate nucleus (on top) and in nucleus accumbens (bottom). "Controls" were pretreated with vehicle pellets, while "Postpellet" rats were pretreated with d-amphetamine pelletsl both groups were then injected with either saline or d-amphetamine (1.5 or 4.0 mg/kg}. FIG.
774
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
DOPAC and HVA (both p~.01). There were also significant increases produced by d-Amp injections in 5HT in caudate and accumbens but not in hippocampus and brainstem (p~.05), and no effects on 5HIAA (Fig. 2.) Fig 1 further shows that DA levels in the caudate nucleus were greatly affected by the d-Amp pellet treatment. Compared to controls, all pellet-pretreated animals showed large decreases in DA levels in caudate (to 32% of control levels in saline-injected animals, p~.001); levels of the DA metabolites DOPAC and HVA were also decreased in the caudate of pellet-pretreated animals (p~ .05). Given this overall background of depletion of DA and its metabolites in the caudate, the overall effects of further injections of d-Amp were not dissimilar from the effects of d-Amp injections on DA metabolism in drug-naive controls. That is, in both drug-naive and in rats which had been pretreated with the d-Amp pellet, d-Amp injections of 1.5 and 4.0 mg/kg produced dose-dependent decreases in DOPAC and HVA.
CAUDATE
0
HIPPOCAMPUS ~
4
I
i~1
o
BRAIN STEM 0
o
2
~
AMPHETAMINEDOSE 0 1.,540 0 1.54.0 v 0 1.54.0 0 '1.54.0 CONTROLPOSTI~ELLET CONTROLPOSTPELLET 5HT 5HIAA
Levels of 5HT and 5 HIAA in four brain regions. groups are as described in Fig. I. FIG 2.
The
In contrast to the large depletions produced by d - ~ pellet pretreatment in caudate, in the nucleus accumbens there were much smaller effects on DA levels (to 68% of control levels in sallne-injected animals, p-.05). DOPAC and HVA levels in the accumbens of the pellet-pretreated animals were not decreased, and in several cases were actually somewhat higher than controls° Again, the dose-dependent decreases in DA metabolites produced by d-Amp injections were similar in drug-naive and pellet-pretreated
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
775
rats. In drug naive Controls, d-Amp injections increased 5HT levels in all brain regions, but this was significant only for the largest dose in Caudate. 5-HIAA levels were similarly increased slightly in all regions. In both caudate and accumbens, there were no significant alterations in 5HT or 5HIAA levels in animals pretreated with the d-Amp pellet compared to those in drug-naive controls (Fig 2); this was also true in the hippocampus and brainstem. In the second experiment, when animals were sacrificed 3 days after pellet implantation (with the pellet intact and while still in incessant motor stereotypies), there were again large depletions of DA in the caudate of pellet animals (6.5 vs 17.9 ng/mg, p<.01) and not in DA in accumbens (7.03 vs 7.57 ng/mg, n.s.). The levels of DOPAC and HVA in these animals were also similar to those in the previous study, with significant decreases in both DA metabolites in caudate of d-Amp pellet animals but not in accumbens. Although 5HT levels in caudate was not significantly less than in controls, in these animals there was a significant depletion of 5HT in the accumbens (p<.05). Furthermore, the d-Amp pellet-treated animals now showed significant increases in 5HIAA in caudate, accumbens, hippocampus, and brain stem (p<.05 in each case). Discussion In the drug-naive controls the larger d-Amp injection led to decreased DA levels in caudate but not in nucleus accumbens. But in these controls, both 1.5 and 4.0 mg/kg d-Amp induced dose-dependent decreases in both DOPAC and HVA in both caudate and accumbens. These two effects of acute d-Amp relect increased DA release accompanied by MAO inhibition (I0). It is noteworthy that the decreased DA levels produced by acute d-Amp injections only occurred in caudate whereas the decreased DA metabolites were present in both DA-rich structures. Other animals were pretreated with the d-Amp pellets for 4 1/2 days, given a 12 hour drug-free period, and then sacrificed following either saline or d-Amp injections. All animals pretreated with the d-Amp pellet evidenced large decreases in the levels of DA, DOPAC, and HVA in caudate° This finding replicates other studies indicating that d-Amp can induce large and persisting alterations in caudate DA when the drug is administered chronically to higher mammals (11), or in rats when methamphetamine is administered in very high doses (12) or if the action of the drug d-Amp is prolonged using pharmacological agents (13). These depletions are especially persistant, and are accompanied by microscopic evidence of terminal destruction (14), when d-Amp administration is continuous. But compared to the effects on caudate, there were smaller depletions of DA in nucleus accumbens of d-Amp pellet-pretreated animals, and there were no decreases in DOPAC and HVA levels. These results indicate that DA in the nucleus accumbens is partially spared from the depleting effects of prolonged d-Amp administration. It is unclear why DA terminals in one brain region should be drastically affected by the chronic d-Amp,
776
Amphetamine Psychosis in Rats
Vol. 31, No. 8, 1982
whereas those of a closely adjacent region are not. The present results also indicate that the overall effects of d-Amp injections on DA metabolism in controls and in animals pretreated with a chronic d-Amp pellet are generally similar (i.e., d-Amp induced dose-dependent decreases in the levels of DA, DOPAC, and HVA). A major difference, then, between drug naive and pellet-pretreated animals is in the principal locus of control of amphetamine's effects on DA metabolism. In rats pre-exposed to the chronic amphetamine pellet, caudate DA is highly depleted, and a d-Amp injection has relatively small effects on DA metabolites compared to its effects in drug-naive animals. However, the effects of d-Amp on DA metabolism in nucleus accumbens are quite similar in pellet-pretreated and drug-naive animals. This finding is of significance, for d-Amp injections to animals preteated with the d-Amp pellet induce a novel behavioral syndrome which is more similar to that produced by injections of hallucinogens such as LSD and mescaline to that induced by d-Amp injections to a drug-naive animal ( 6 ) . The finding that one of the major neurochemical correlates of this altered, hallucinogen-like response to d-Amp is a decreased ability of d-Amp to alter caudate DA metabolism compared to its near-normal effects on DA metabolism in nucleus accumbens has been recently validated using other methods. Eison et.al. (15) recently reported that continuous d-Amp administration resulted in progressively increased retention of d-Amp and its metabolites by mesolimbic but not nigrostriatal brain regions, and this was accompanied by locally enhanced levels of glucose utilization which were predominantly localized in nucleus accumbens. The implication of these studies is that following several days of continuous d-Amp administration, the predominant site of action of d-Amp shifts from caudate dopaminergic systems to those of the nucleus accumbens and other olfactory forebrain structures. There were minimal alterations in serotonin metabolism in the five day "reactivation" animals even though in previous experiments we have demonstrated that such drug-treated animals show a behavioral syndrome like that induced by hallucinogens (6). The findings from this study, therefore, do not support the suggestion by Ellinwood (16) and by Trulson and Jacobs (8) that altered serotonin metabolism plays an important role in amphetamine psychosis. Although the cats in the latter study, when administered frequent and very large injections of d-Amp, did have decreased 5HT and 5HIAA levels in brain, the amount of depletions of caudate DA were much greater. The role of 5HT in the actions of hallucinogens is not clear, for similar behaviors have been reported to be induced both by 5HT agonists and 5HT depletions. While methamphetamine can produce long-term alterations in 5HT (17) this drug has a pharmacological profile very different from that of d-Amp and can induce psychoses immediately after injection (18). The present results using d-Amp appear to indicate that, at least in rats, altered 5HT levels are not necessary for the appearance of hallucinogen-like behaviors in the d-Amp model of amphetamine psychosis. Rather, the principal alterations in 5HT observed in the present experiment were in the animals sacrificed three days after pellet implantation, while still in constant motor stereotypy and in a state of sleep deprivation. Thus, these results appear to dissociate the effects
Vol. 31, No. 8, 1982
Amphetamine Psychosis in Rats
777
of chronic d-amp on 5HT metabolism from the hallucinogen-like behaviors induced by chronic administration of the drug. One implication of these results is that an altered locus of control of amphetamine's effects on dopaminerglc systems (from strlatal and towards mesolimbic systems) correlates better with the heightened ability of d-Amp to elicit a hallucinogen-like behavior than with its altered effects on serotonin metabolism. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. i0. ii. 12. 13. 14. 15. 16. 17. 18.
D.S. BELL, J. Psychiat. Res. 3, 701-707 (1965). J.C. KRAMER, V. GISCHMAN, and D. LITTLEFIELD, JAMA, 201, No.5, 89-93 (1967). H.S. HUBERMAN, M.S. EISON, K.S. BRYON, and G. ELLISON, Eur. J. Pharmacol, 45, 237, 97-100 (1972). G. ELLISON, M. EISON, H. HUBERMAN, and F. DANIEL, Science, 201, 276-278 (1978). G. ELLISON, E.B. NIELSEN, and M. LYON, J. Psychiat. Res. 16, 13-22 (1981). E.B. NIELSEN, T. LEE, and G. ELLISON, Psychopharm. 68, 197-200 (1980). S.W. SNYDER, S. BANERJEE, H. YAMAMURA, and D. GREENBERG, Science, 184, 1243-1253 (1979). M.E. TRULSON, and B.L. JACOBS, Science, 205, 1295-1297 (1979). J.F.R. KONIG, and R.A. KLIPPEL, The Rat Brainl A stereotaxic atlas, pp. 7-156, Krieger Publishig Co., Inc., New York (1963). H.H. MILLER, P. SHORE, and D. Clarke, Biochem. Pharm., 29, 1347-1354 (1980). M.S. LEVINE, D.D. HULL, E. GARCIA-RILL, L. ERNOFF, N.A. BUCHWALD, and A. HELLER, Brain Res. 194, 263-268 (1980). G.C. WAGNER, G.A. RICAURTE, C.E. HOHANSON, C.R. SCHUSTER, and L.S. SEIDEN, Neurology, 30, 547-550 (1980). R.W. FULLER, and S. HEMRICK-LUECKE, Science, 209, 305-306 (1980). E. NWANZE and G. JONSSON, Neuroscience Letters, 26, 163-168 (1981). M. EISON, A. EISON, and G. ELLISON, Exp. Brain Res.,43, 281-288 (1981). E.B. ELLINWOOD, M.KILBEY JR., In I. Hanin and E. Usdin (Eds), Animal Models of Psychiat. and Neurol. (1977). A. HOTCHKISS and J. GIBB, J. Pharmcol. Exp. Ther., 214, 257-262 (1980). D.A. BELL, ARCH. GEN. PSYCHIAT., 29, 35-40 (1973).