Department of Pharmacology, Emory University School of Medicine,
Atlanta, Georgia
Opioids modulate brain dopaminergic function in various experimental
paradigms. This study used the rotational model of behavior in rats
with unilateral 6-hydroxydopamine-induced lesions of the nigrostriatal
pathway to investigate this interaction. Doses of two presynaptically
acting dopaminergic drugs, amphetamine and cocaine, were coadministered
with several doses of the mu opioid agonist, morphine.
Morphine, at 3.0 mg/kg, potentiated rotational behavior induced by each
dose of the stimulants. To determine the receptor specificity of the
actions of morphine, the mu opioid agonists buprenorphine,
fentanyl, levorphanol, meperidine, and methadone, and dextrorphan, the
non-opioid isomer of levorphanol, were administered alone and with 1.0 mg/kg amphetamine. Each of these drugs, as well as morphine, produced
circling behavior on its own. All of the mu opioid agonists
and dextrorphan increased amphetamine-induced turning; the
coadministration of dextrorphan, levorphanol, meperidine, methadone and
morphine with amphetamine produced turning greater than predicted by
simple additivity. To determine whether an opioid receptor was involved
in these interactions, the opioid antagonist, naltrexone, was
administered before the amphetamine/mu opioid receptor
agonist combination. Naltrexone blocked the potentiating effects of
morphine, but not those of the other drugs. Moreover, naltrexone alone
dose-dependently increased amphetamine-induced rotational behavior.
These studies show that some mu opioid receptor agonists can
potentiate stimulant-induced rotational behavior and that blockade of
opioid receptors can also produce a potentiation. The role of
mu opioid receptors in these effects remains unclear.
 |
Introduction |
Opioids
can modulate brain dopamine systems (Wood, 1983
). Of the three types of
opioid receptors, mu and delta receptor
activation increases dopamine release in the mesolimbic and
nigrostriatal tracts. Conversely, the stimulation of kappa
opioid receptors decreases dopamine levels in these brain regions (Di
Chiara and Imperato, 1988a,b). All three opioid receptors have been
localized within these dopamine systems (Goodman et al.,
1988; Mansour et al., 1995; Sharif and Hughes, 1989; Wood
and Iyengar, 1988).
Some effects exhibited by mu opioid agonists reflect
increases in dopaminergic activity. In mice, morphine, heroin,
levorphanol and meperidine increased locomotor activity (Longoni
et al., 1987; Oliverio and Castellano, 1974; Rethy et
al., 1971; Shippenberg et al., 1993), which is mediated
by dopaminergic neurons in the substantia nigra (Iwatsubo and Clouet,
1977) and in the ventral tegmental area (Matthews and German, 1984).
In vivo microdialysis indicated that morphine, methadone and
fentanyl increased extracellular dopamine levels in the nucleus
accumbens (Di Chiara and Imperato, 1988b; Kalivas and Stewart, 1991).
Behavioral sensitization to repeated administration of opioids was
attributed to the indirect stimulation of dopamine cell bodies in the
ventral tegmental area and the substantia nigra (Kalivas and Stewart,
1991). These results indicate that the stimulation of opioid receptors
influences processes thought to be mediated by brain dopamine systems.
Opioids also can modulate the effects of drugs that act via
brain dopamine systems. The effects of cocaine and amphetamine are
believed to be mediated largely by dopamine (Kalivas and Stewart, 1991). Behavioral interactions between mu opioid agonists
and psychomotor stimulants have been studied with various experimental paradigms. However, the results have not always been consistent among
these different techniques. In drug discrimination studies, heroin
substituted for cocaine in some rhesus monkeys (Mello et al., 1995), and morphine potentiated the discriminative stimulus effects of cocaine in squirrel monkeys (Spealman and Bergman, 1994).
Chronic treatment with the partial mu opioid receptor
agonist buprenorphine reduced cocaine self-administration in rats
(Carroll and Lac, 1992) and rhesus monkeys (Mello et al.,
1992, 1993, 1995). Doses of buprenorphine and cocaine, which were
ineffective by themselves, induced conditioned-place preference in rats
when administered together (Brown et al., 1991). In studies
of analgesia, cocaine potentiated the effects of morphine and the
mu opioid receptor-selective ligand DAMGO in mice (Sierra
et al., 1992) and rats (Kauppila et al., 1992),
and the effects of morphine and nalbuphine in rhesus monkeys (Gatch
et al., 1995). The coadministration of stimulants and
morphine-like compounds, commonly known as "speedballing," is
frequently seen in drug abusers (Kosten et al., 1986; Kreek, 1987). Whether the coadministration of the drugs enhances their euphoric effects or attenuates their dysphoric effects is not known. In
summary, mu opioid receptor agonists influence the effects of cocaine, but the direction of the change in the behavioral response
to cocaine is difficult to predict.
The effects of opioid antagonists on the behavioral effects of cocaine
and amphetamine have also been studied. In many cases, the antagonists
did not alter the effects of stimulants. When the antagonists had an
effect, it was usually to reduce the effects of the stimulant drugs.
For example, naloxone and naltrexone, two nonspecific opioid receptor
antagonists, attenuated amphetamine-induced locomotor activity in a
variety of animal species (Adams et al., 1981; Andrews and
Holtzman, 1978; Dettmar et al., 1978; Hitzemann et
al., 1982; Holtzman, 1974; Schad et al., 1995; Winslow
and Miczek, 1988), and blocked cocaine- and amphetamine-induced
conditioned place preference in rats (Gerrits et al., 1995;
Sala et al., 1995; Trujillo et al., 1991). These
two opioid receptor antagonists also attenuated cocaine
self-administration (Corrigall and Coen, 1991; Ramsey and van Ree,
1991). Sensitization to repeated cocaine injections was attenuated by
naltrexone (Sala et al., 1995). As the opioid antagonists
presumably were blocking effects of endogenous opioids, the results of
the experiments cited above suggest that endogenous opioids modulate
the effects of stimulants, probably via the dopaminergic
system.
To study further the relationship between opioids and dopaminergic
neurons in the brain, we used the rotational model of behavior in the
rat. This is a method of studying the nigrostriatal dopaminergic system
in which rats are given a unilateral lesion of the nigrostriatal tract
with 6-OHDA, thus creating a postsynaptic dopamine receptor supersensitivity (Ungerstedt, 1971). These animals circle toward the
lesion in response to presynaptically acting dopaminergic agonists
(i.e., amphetamine) (Lynch and Carey, 1989; Ungerstedt and
Arbuthnott, 1970; Zetterstrom et al., 1986a). This paradigm has been used infrequently to investigate interactions between the
opioid and dopaminergic systems of the rat brain, and results have not
been consistent across studies. Morphine produced rotational behavior
in one study (Cowan et al., 1975; Pert, 1978), but
amphetamine was used to determine the validity of the 6-OHDA-induced
lesion in this study. Apomorphine-induced turning has been a more
accurate indicator of the magnitude of the 6-OHDA-induced lesion
(Hudson et al., 1993). In other studies, injections of
morphine produced significant turning only in those rats that had been
made tolerant to the depressant effects of the drug by repeated
exposure to morphine (Kimmel et al., 1995; Pert 1978). In
one study, amphetamine-induced turning was reduced by morphine
(Blundell et al., 1976), whereas in another, it was reduced
by naloxone (Dettmar et al., 1978). Based on the effects of
mu opioid agonists on brain dopamine and on the
dopamine-mediated behaviors cited above, we hypothesized that these
drugs would potentiate the effects of psychomotor stimulants upon
rotational behavior, and would do so by activating the mu opioid receptor.
In this study, we first examined the effects of morphine on
amphetamine- and cocaine-induced rotational behavior. To do this, we
administered various doses of morphine immediately before doses of
amphetamine or cocaine. Because combinations of the highest doses of
morphine and cocaine were lethal in some rats, subsequent experiments
were carried out using only amphetamine. To determine the receptor
specificity of the effects of morphine on turning induced by
amphetamine, we tested the mu opioid receptor agonists fentanyl, levorphanol, meperidine and methadone. We also examined the
effects of dextrorphan, the optical isomer of levorphanol, to
investigate the stereoselectivity of this interaction. We measured rotational behavior for a 4-hr period to capture the full time course
of drug action.
Biochemical and behavioral effects of opioids that involve mesolimbic
and nigrostriatal dopamine systems can be blocked by the general opioid
antagonists naloxone and naltrexone (Di Chiara and Imperato, 1988b;
Spanagel et al., 1990). Thus, to determine whether the
effects of these mu opioid agonists were mediated by an
opioid receptor, we administered naltrexone before amphetamine alone or
the combination of amphetamine with the mu opioid agonists.
| |
Methods |
Subjects.
Male Sprague-Dawley rats (Sasco, Inc., Omaha, NE)
weighing 300 to 350 g at the time of surgery were used. All rats
were group housed in polycarbonate cages and maintained in a
temperature-controlled colony room with a 12 hr light:12 hr dark
lighting cycle, beginning with lights on at 7:00 A.M. Food
(Purina Rodent Chow, Purina Mills, St. Louis, MO) and water were
available ad libitum.
Stereotaxic surgery.
Rats were given unilateral lesions of
the right nigrostriatal pathway by a single injection of 6-OHDA. They
were first anesthetized with 3.3 mg/kg i.p. Equithesin and then placed
into a stereotaxic frame. Stereotaxic coordinates relative to bregma
were AP = 
4.8, ML = 2.2, DV = 8.0 (Paxinos and
Watson, 1986). A 25-µl Hamilton syringe was used to inject 8 µg/4
µl of 6-OHDA in a solution of 0.02% ascorbic acid and 0.9% saline
into the right substantia nigra at a rate of 1.0 µl/min for 4 min.
Upon completion, the injection needle was kept in place for an
additional minute to minimize backflow of the solution. No
desmethylimipramine pretreatment was used before the surgeries, so
norepinephrine nerve terminals may have been lesioned. However,
norepinephrine innervation of the striatum is relatively low, so this
should not have affected the experimental results.
Rotational behavior.
Rotational activity was measured in
stainless steel rotometer stations (MED Associates, Inc., East
Fairfield, VT). Each of eight stations consisted of a round stainless
steel bowl (40.6 cm diameter and 25.4 cm high) in a transparent
Plexiglas cover. A spring tether connected to a direction-sensitive
rotation sensor mounted above the bowl was attached to the rat by means
of a Velcro belt. Rotational activity was recorded by the Roto-Rat
Version 1.2 computer program (MED Associates). Measurements were taken of full (360°) turns in both the clockwise and counterclockwise direction. During experiments, counts were taken in 15-min intervals for 4 hr, resulting in 16 time points per animal for each session. All
sessions were conducted during the light phase of the lighting cycle.
Rats were allowed to recover from surgery for at least 14 days, then
they received 0.3 mg/kg R()-apomorphine s.c. twice weekly
for 2 weeks. Animals exhibiting at least 50 full contralateral turns/10
min for 1 hr were used for further behavioral testing. The amount of
turning observed in response to apomorphine is directly correlated
with the extent of the nigral lesion (Hudson et al., 1993).
The rats used in this study exhibited a mean (± S.E.M.) of 471 ± 56 turns/hr in tests with 0.3 mg/kg apomorphine.
Drug administration and behavioral testing.
On test days,
animals were weighed and placed into the test chambers and allowed to
habituate for approximately 5 min before drug injection. Rats
(n = 16) then received a 5-min pretreatment of doses of
saline or morphine (0.03-10 mg/kg) administered in a random sequence.
Five minutes later, half of the rats received saline or amphetamine
(0.1-1.0 mg/kg) and the other half received saline or cocaine (3.0-30
mg/kg), with the dose of each drug and saline administered in a random
order. Measurements of rotational behavior began 5 min after the second
injection. Each animal was tested twice a week, with a 3- to 4-day
interval between sessions, so that each animal received every possible
combination of opioid and either amphetamine or cocaine. Morphine,
amphetamine and saline were administered s.c., and cocaine was
administered i.p.
Based on results obtained in the experiments above, a 1.0 mg/kg dose of
amphetamine was selected for testing in combination with other
mu opioid receptor agonists and dextrorphan. Buprenorphine (0.01-1.0 mg/kg), dextrorphan (1.0-10 mg/kg), fentanyl (0.01-0.056 mg/kg), levorphanol (0.1-1.0 mg/kg), meperidine (3.0-30 mg/kg) and
methadone (0.3-3.0 mg/kg) were injected s.c., with doses of each drug
given in a random sequence. Amphetamine was injected 5 min later.
Testing was conducted in the same manner as described earlier.
To determine whether the interactions observed between the
mu opioid agonists and amphetamine could be blocked by
naltrexone, one group of animals was given 0.1 mg/kg naltrexone
followed by 3.0 mg/kg morphine and 1.0 mg/kg amphetamine. The dose of
each of the other agonists that produced the greatest amount of turning in combination with amphetamine was then tested with 1.0 mg/kg naltrexone and 1.0 mg/kg amphetamine, injected in the following order:
naltrexone, mu opioid agonist and amphetamine. Because this
dose of naltrexone was ineffective in combination with nearly all of
the agonists, 10 mg/kg naltrexone was tested in combination with 1.0 mg/kg methadone and 1.0 mg/kg amphetamine.
Statistical analysis.
Total rotational count data in the
initial experiments with morphine, amphetamine and cocaine were
analyzed using a two-way ANOVA with repeated measures on both factors
(morphine dose and stimulant dose). A Tukey's protected t
test was performed for multiple pairwise comparisons. Data obtained in
the experiments with additional mu opioid agonists and
dextrorphan were analyzed by a two-way ANOVA with repeated measures on
both the agonist dose and the amphetamine dose. Tukey's protected
t tests were performed on these data following the
statistical analysis. Two-hour rotation totals were subjected to a
one-way repeated measures ANOVA. The Friedman's ANOVA was used to
compare the overall theoretical additive values of the mu
opioid in combination with amphetamine to the observed behavior. The
Wilcoxon matched-pairs test was then performed on each pair of
theoretical and observed values for each of the mu agonist
doses. In the antagonism experiments, data were analyzed with a
repeated measures ANOVA with one factor, followed by a Tukey's
protected t test. All time-course data were analyzed by a
three-way ANOVA with repeated measures on all three factors (morphine
dose, stimulant dose, time period). Results were considered to be
statistically significant if the P value was 
.05.
Drugs.
Buprenorphine hydrochloride (National Institute on
Drug Abuse, Rockville, MD), dextrorphan tartrate and levorphanol
tartrate (Roche Laboratories, Nutley, NJ) were dissolved in distilled
water. d-Amphetamine sulfate, and naltrexone hydrochloride
(Sigma Chemical Co., St. Louis, MO), cocaine hydrochloride (National
Institute on Drug Abuse), fentanyl citrate (McNeil Laboratories, Fort
Washington, PA), methadone hydrochloride (Mallinkrodt, St. Louis, MO),
meperidine hydrochloride and morphine sulfate (Penick Corp., Newark,
NJ) were dissolved in 0.9% saline. R()-Apomorphine
hydrochloride (Research Biochemicals, Inc., Natick, MA) and 6-OHDA
hydrobromide (Sigma) were dissolved in a solution of 0.02% ascorbic
acid in 0.9% saline. All drugs except for 6-OHDA and 30 mg/kg of
meperidine were administered in a volume of 1.0 ml/kg b.wt., with all
doses expressed as the free base. The highest dose of meperidine (30 mg/kg) was given in 3.0 ml/kg b.wt. to avoid skin lesions, which occurred with concentrations of meperidine higher than 10 mg/ml.
| |
Results |
Effects of morphine on stimulant-induced turning behavior.
Amphetamine alone produced dose-dependent ipsilateral turning (fig.
1). The lowest doses of the drug (0.1 and
0.3 mg/kg) had only a slight effect on circling behavior, whereas 1.0 mg/kg produced approximately 250 full ipsilateral turns in a 4-hr
observation period, significantly greater than saline (P < .01).
In these animals, morphine alone produced significant ipsilateral
turning as an overall drug effect; however, no single dose elicited
turning greater than turning after saline + saline (fig. 1B).
View larger version (32K):
[in this window]
[in a new window]
|
Fig. 1.
Morphine (0.03-10 mg/kg) potentiated ipsilateral
turning behavior induced by amphetamine (0.1-1.0 mg/kg) in rats with
unilateral lesions of the nigrostriatal tract. Points on the
time-course curves (A) represent the mean number of turns per 15 min.
Only the saline curves and the 3.0 mg/kg morphine curves are shown for
simplicity. A three-way ANOVA of these data revealed significant F values for all factors and interactions: amphetamine:
F(3,21) = 30.78, P < .0001; morphine:
F(6,42) = 8.02, P <.0001; time: F(15,105) = 27.24, P < .0001; amphetamine × morphine:
F(18,126) = 3.95, P < .0001; amphetamine × time:
F(45,315) = 13.98, P < .0001; morphine × time:
F(90,630) = 5.13, P < .0001; amphetamine × morphine × time: F(270,1890) = 1.92, P < .0001. Dose-response data (B) show total amount of turning for the entire 4-hr
session (mean + S.E.). A two-way ANOVA of these data revealed a
significant effect by drug as well as a significant interaction between
the two drugs (amphetamine: F(3,28) = 12.87, P < .0001; morphine: F(6,168) = 16.88, P < .0001; and
amphetamine × morphine interaction: F(18,168) = 3.29, P < .0001). Asterisks indicate significant differences from
saline control plus that same dose of amphetamine (points above sal),
**P < .01, *P < .05. Daggers indicate significant differences of amphetamine or morphine alone from saline + saline, ++P < .01, +P < .05.
|
|
Morphine modified amphetamine-induced circling in a dose-dependent
manner that was more pronounced as the dose of amphetamine increased.
When 1.0 mg/kg amphetamine was administered with 3.0 mg/kg morphine,
the amount of full ipsilateral turning during 4 hr increased from 250 to nearly 1000 turns (fig. 1B). The time course of effects is
illustrated for the 3.0 mg/kg dose of morphine in figure 1A. The
combination of 3.0 mg/kg morphine and 1.0 mg/kg amphetamine shifted the
peak circling effect from 30 min after injection to 90 min after
injection, and increased the maximum turning in a 15-min interval from
45 turns to 110 turns. In this experiment and those that follow, no
contralateral turning was observed.
Cocaine alone also produced ipsilateral turning in a dose-dependent
manner (fig. 2). At 30 mg/kg, the highest
dose studied, 400 turns were recorded during the 4-hr observation
period, slightly more than the effects produced by 1.0 mg/kg
amphetamine. This dose of cocaine was the only one examined that
produced turning significantly greater than saline alone (P < .05). Morphine produced significant ipsilateral turning, and 3.0 mg/kg
produced turning significantly greater than saline did (P < .01)
(fig. 2B), in contrast to the results described in figure 1. The effect
of this dose of morphine had a plateau of approximately 45 turns/15 min for 150 min, which tapered off during the final 90 min (fig. 2A). The
effects of cocaine were dose-dependently affected by morphine. The
combination of 3.0 mg/kg morphine and 30 mg/kg cocaine was the most
effective of the doses examined in producing ipsilateral circling. The
3.0 mg/kg morphine dose did not shift the overall peak effect of 30 mg/kg cocaine from the first 15-min period, but it increased the
maximum turning observed during this 15-min period from 75 to 125 turns
(fig. 2A). Turning was also prolonged, lasting the entire 240 min,
rather than ceasing after 150 min, as it did after 30 mg/kg of cocaine
alone. This drug combination produced 1400 turns/4 hr, greater than a
3-fold increase in the amount of turning seen with this dose of cocaine
alone (fig. 2B). The combination of a higher dose of cocaine (56 mg/kg)
and 3.0 mg/kg morphine was lethal in some animals (data not shown), so we discontinued testing this pair of drug doses.
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 2.
Morphine (0.1-10 mg/kg) potentiated ipsilateral
turning behavior induced by cocaine (3.0-30 mg/kg). A three-way ANOVA
of the time-course values (A) showed that all three variables as well as all two- and three-factor interactions were significant: cocaine: F(3,21) = 19.39, P < .0001; morphine:
F(5,35) = 21.00, P < .0001; time: F(15,105) = 79.64, P < .0001; cocaine × morphine:
F(15,105) = 2.75, P < .0013; cocaine × time:
F(45,315) = 16.57, P < .0001; morphine × time:
F(75,525) = 5.33, P < .0001; cocaine × morphine × time: F(225,1575) = 2.77, P < .0001. Total full ipsilateral turning during the 4-hr experimental period (B)
was influenced by cocaine (F(3,28) = 16.55, P < .0001)
and morphine (F(5,191) = 34.85, P < .0001) as well as
their interaction (F(15,191) = 2.29, P = .0062). Other
details are as in figure 1.
|
|
Effects of mu opioid receptor agonists alone and on
amphetamine-induced turning.
When doses of the mu
agonists were tested alone, each one induced significant ipsilateral
turning across all doses during the 4-hr experimental session (fig.
3). Buprenorphine at 0.1 and 1.0 mg/kg
produced turning greater than that occurring after saline (P < .01) and the most of any of the mu opioid receptor agonists examined, with a peak effect of nearly 500 turns/4 hr. Levorphanol at
0.3 and 1.0 mg/kg also produced significant turning (P < .01), as
did the 30 mg/kg dose of meperidine (P < .05). These two
mu opioid agonists produced a maximum effect of
approximately 250 turns/4 hr. No single dose of fentanyl, methadone and
morphine produced turning significantly greater than saline + saline,
although there was a significant overall effect. Unlike the other drugs tested, fentanyl produced severe catalepsy and rigidity. Dextrorphan produced less turning than did its optical isomer, levorphanol, even at
a dose that was 10 to 30 times higher than doses of levorphanol that
produced turning. The nonselective opioid receptor antagonist, naltrexone, did not produce rotational behavior at doses of 0.1 to 10 mg/kg.
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 3.
Turning was induced by mu opioid
agonists and dextrorphan, but not naltrexone. Asterisks represent drug
doses that produced turning significantly greater than the vehicle
control, **P < .01, *P < .05.
|
|
Several doses of each of the mu opioid agonists were
administered along with 1.0 mg/kg amphetamine. The time-course data
were analyzed to determine whether there was a significant interaction between the mu opioid agonist of interest and amphetamine.
The interactions between time and amphetamine and methadone (fig. 4A) and morphine (fig.
5A) were significant, as determined by a
three-factor ANOVA (mu opioid agonist dose × amphetamine dose × time). However, in the cases of buprenorphine
(fig. 6A), fentanyl (fig.
7A), meperidine (fig.
8A) and levorphanol (fig.
9A) and its stereoisomer dextrorphan
(fig. 10A), this interaction term was
not significant. Nevertheless, with the exception of fentanyl, these
drugs increased the duration of action of amphetamine from approximately 180 min to more than 240 min. The data presented in
figure 5 for morphine are derived from those in figure 1, and they are
shown here to facilitate comparison with the other agonists studied.
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 4.
Methadone (0.3-3.0 mg/kg) potentiated
amphetamine-induced ipsilateral turning, with a significant effect of
its own. Part A shows the time-course effects of the saline control and
one test dose of methadone (1.0 mg/kg). (The others were omitted for clarity.) Each data point represents the mean full ipsilateral counts
during that 15-min time block (n = 8). The P value
shown represents the amphetamine × methadone × time
interaction determined by a three-factor ANOVA, with all doses of
methadone tested. Part B shows the same data as in A, but collapsed
into 2-hr time blocks in the first two panels, and then total
ipsilateral turns during the entire 4-hr observation time in the third
panel. In the first two panels of B, asterisks denote a significant
difference between that dose of methadone and the saline control
(within the saline group and the amphetamine group), **P < .01, *P < .05. In the third panel, asterisks denote a significant
difference between amphetamine alone and the amphetamine + methadone
combination, **P < .01, *P < .05. The P values shown here
were determined by a two-factor ANOVA (methadone dose × amphetamine dose).
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 5.
Morphine (1.0-10 mg/kg) enhanced
amphetamine-induced turning and had a considerable effect alone. (The
data in these graphs are redrawn from the data in figure 1 for
comparison with the other drugs used in this study.) Graphs are as
described in figure 4.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 6.
Buprenorphine (0.01-1.0 mg/kg) potentiated turning
induced by amphetamine although it had a significant effect alone.
Graphs are as described in figure 4.
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 7.
Fentanyl (0.01-0.056 mg/kg) potentiated
amphetamine-induced circling only slightly, and had a small effect of
its own. Graphs are as described in figure 4.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 8.
Meperidine (3.0-30 mg/kg) augmented
amphetamine-induced turning behavior and induced turning by itself.
Graphs are as described in figure 4.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 9.
Levorphanol (0.1-1.0 mg/kg) increased ipsilateral
turning observed with amphetamine, and produced a significant amount of turning on its own. Graphs are as described in figure 4.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 10.
Dextrorphan (1.0-10 mg/kg), a stereoisomer of
levorphanol, potentiated amphetamine-induced ipsilateral turning and
had an effect on circling behavior alone. Graphs are as described in figure 4.
|
|
We examined the effects of the mu opioid agonists alone and
combined with amphetamine on full ipsilateral turning during 2-hr time
blocks as well as during the full 4-hr observation. These data are the
same as those in the time-course graphs, but they are collapsed into
2-hr and 4-hr blocks, respectively, for further analysis. When
amphetamine was preceded by methadone (fig. 4B), morphine (fig. 5B),
buprenorphine (fig. 6B) and dextrorphan (fig. 10B), turning was
augmented significantly during both 2-hr periods. However, the effects
of fentanyl (fig. 7B) and meperidine (fig. 8B) occurred only during the
final 2 hr. Levorphanol did not significantly increase
amphetamine-induced turning in either 2-hr time block, although there
was a trend in that direction (fig. 9B).
A two-factor ANOVA (mu opioid agonist × amphetamine)
indicated that there was a significant main effect of amphetamine
alone, the mu opioid agonists alone and dextrorphan alone
upon turning over the full 4 hr. However, the interaction term with
amphetamine was significant only for methadone (fig. 4B) and morphine
(fig. 5B), which indicated that amphetamine changed the shape of the dose-response curves of these two mu opioid agonists. To
examine whether these drugs significantly potentiated the effects of
amphetamine, other statistical methods must be used.
Synergism of mu opioid agonists with amphetamine.
To determine whether the observed potentiation of
amphetamine-induced turning by the mu opioid agonists and
dextrorphan was a simple additive effect or an effect that was greater
than additive (i.e., synergistic), we summed the total turns
produced by each mu opioid agonist and dextrorphan alone
with the total turns produced by amphetamine alone for each individual
animal. These sums were averaged to produce a mean and standard error
for each drug combination (table 1).
Because the variance of the means in the table was not homogeneous, we
used nonparametric statistical methods to analyze these data. These
values were compared with the total of full ipsilateral turns observed
when the two drugs were actually coadministered.
View this table:
[in this window]
[in a new window]
|
TABLE 1
Synergism of mu opioid agonists and amphetamine
Theoretical number of full ipsilateral turns expected when the
mu opioid agonist or dextrorphan was administered with 1.0 mg/kg amphetamine vs. observed turning with the combination.
All values represent the mean ± S.E. P values in the last column
denote the statistical difference between the calculated and observed values at all doses of one mu opioid agonist plus 1.0 mg/kg
amphetamine as determined by Freidman two-way ANOVA by rank analyses.
|
|
A Friedman's ANOVA (calculated vs. observed rotations × dose) on these data indicated there was an overall significant
difference between the predicted and observed amounts of turning
induced by the drugs tested in combination with 1.0 mg/kg amphetamine, with the exception of buprenorphine and fentanyl (table 1). A follow-up
analysis of individual doses with a Wilcoxon matched-pairs test
revealed that the middle doses of dextrorphan (3.0 mg/kg), levorphanol
(0.3 mg/kg), meperidine (10 mg/kg), methadone (1.0 mg/kg) and morphine
(3.0 mg/kg), as well as the higher dose of dextrorphan (10 mg/kg) in
combination with amphetamine, produced turning greater than predicted,
P .05 (table 1).
Antagonism of mu opioid agonist effects.
Naltrexone (1.0 mg/kg) blocked the effects of 3.0 mg/kg morphine
upon amphetamine-induced rotational behavior (fig.
11). Post hoc tests revealed
that there was no statistical difference between turning induced by
amphetamine alone and by naltrexone combined with morphine and
amphetamine. Further testing revealed that the effects of 3.0 mg/kg
morphine on 1.0 mg/kg amphetamine were blocked by a lower dose of
naltrexone, 0.1 mg/kg. However, 1.0 mg/kg naltrexone did not
statistically alter the effects of the other mu opioid agonists or dextrorphan on amphetamine-induced turning. We tested a
higher dose of naltrexone, 10 mg/kg, with the combination of 1.0 mg/kg
methadone and 1.0 mg/kg amphetamine, but it did not significantly alter
rotational behavior induced by this drug combination (data not shown).
View larger version (43K):
[in this window]
[in a new window]
|
Fig. 11.
Naltrexone did not block potentiation of the
effects of amphetamine by dextrorphan or mu opioid agonists,
with the exception of 3.0 mg/kg morphine. The asterisk denotes a
significant difference between the mu agonist + amphetamine + naltrexone group and the mu agonist + 1.0 amphetamine
group, *P < .05.
|
|
Inspection of the data in figure 11 suggested that the combination of
1.0 mg/kg naltrexone and 1.0 mg/kg amphetamine often resulted in
greater turning than occurred after amphetamine alone. This apparent
interaction was not statistically significant within each individual
drug series. However, when data were pooled from all 24 subjects, 1.0 mg/kg naltrexone significantly potentiated turning induced by 1.0 mg/kg
amphetamine (P <.05) (fig. 12). The highest dose of naltrexone (10 mg/kg) also potentiated
amphetamine-induced circling (P <.01), but the lowest (0.1 mg/kg) did
not (fig. 12). None of these doses of naltrexone alone produced turning
behavior (fig. 3).
View larger version (26K):
[in this window]
[in a new window]
|
Fig. 12.
Naltrexone (0.1-1.0 mg/kg) did not induce turning
behavior by itself, but it potentiated amphetamine-induced circling in
a dose-dependent fashion, with the two higher doses producing
significant effects. Numbers above each pair of bars represent the
number of animals in that particular group. The 1.0 mg/kg naltrexone group data are pooled from the data shown in figure 11. Asterisks represent a significant difference from the corresponding 1.0 mg/kg
amphetamine alone group, **P < .01, *P < .05.
|
|
| |
Discussion |
The initial experiments in this study showed that morphine
potentiated rotational behavior induced by both amphetamine and cocaine
by 3- to 4-fold. Both of these stimulant drugs are indirect dopamine
agonists with different mechanisms of action. Amphetamine preferentially releases newly synthesized dopamine from presynaptic cells (Arbuthnott et al., 1991; Kuczenski and Segal, 1989;
Zetterstrom et al., 1986b). Cocaine acts by blocking the
presynaptic dopamine transporter, thus blocking the reuptake of
synaptic dopamine. Some differences have been found in the modulation
of locomotor effects of these two drugs by the nonselective opioid
receptor antagonist naloxone (Jones and Holtzman, 1994), which suggests that modulation of the effects of amphetamine by endogenous opioids also differs from that of cocaine. However, our data indicate that the
effects of both amphetamine and cocaine are enhanced by mu
opioid receptor agonists, although the exact mechanism(s) by which that
occurs is not clear. The results correspond with those of several other
studies in which there was an apparent synergy between cocaine and
morphine-like opioids (see the introduction).
To determine whether the effects of morphine on stimulant-induced
rotational behavior were specific to this drug or general to agonists
at the mu opioid receptor, we examined the effects of
several other mu opioid agonists. Each of these drugs
produced ipsilateral turning when administered alone, as reflected in
the significant main effect of dose. Mu opioid receptors are
localized on dopaminergic nerve terminals in the striatum (Mansour
et al., 1995), and increase the release of dopamine from
these cells upon stimulation (Kalivas and Stewart, 1991). Presumably,
these mu opioid agonists produced ipsilateral turning by
increasing synaptic levels of dopamine in an indirect fashion. The
order of potency of these drugs in producing turning paralleled their
order of potency in other assays of drug interactions with the
mu opioid receptor (Holtzman and Locke, 1988). Morphine
alone produced relatively little turning in the series of experiments
with amphetamine, in agreement with previous results (Kimmel et
al., 1995). However, morphine had a larger effect in the series of
experiments with cocaine. Each animal in that group was exposed to
cocaine 24 times during the randomized sequence of tests with cocaine
alone, morphine alone and morphine-cocaine combinations. It is possible
that this exposure to cocaine sensitized the animals to the effects of
morphine more than did comparable exposure to amphetamine, thus
accounting for the different outcomes.
Methadone had a significant overall effect on rotational behavior,
although no single dose produced significant turning, and increased
amphetamine-induced rotational behavior. A dose of 1.0 mg/kg produced
effects greater than predicted by simple additivity. In the
conditioned place preference paradigm, methadone enhanced the
reinforcing properties of cocaine in rats (Bilsky et al., 1992). Similarly, in humans, cocaine use was greater in
methadone-treated opioid addicts than in buprenorphine- or
naltrexone-treated patients (Kosten et al., 1989). The
discriminative stimulus effects of cocaine in squirrel monkeys were
also enhanced by methadone (Spealman and Bergman, 1992, 1994). This
mu opioid agonist increases synaptic dopamine levels in the
nucleus accumbens (Di Chiara and Imperato, 1988b). The data in the
present study concur with the findings that methadone enhances
dopaminergic functions.
Buprenorphine is a partial agonist at the mu opioid receptor
and an antagonist at the kappa receptor (Cowan et
al., 1977; Leander, 1987; Negus and Dykstra, 1988; Negus et
al., 1990). Evidence for interactions of buprenorphine with drugs
that act via the dopaminergic system has been conflicting.
For example, buprenorphine attenuated cocaine self-administration in
humans (Foltin and Fischman, 1994), rhesus monkeys (Mello et
al., 1992, 1993) and rats (Brown et al., 1991; Carroll
and Lac, 1992; Dykstra et al., 1992), and attenuated cocaine
conditioned place preference in rats in two studies (Kosten et
al., 1991; Suzuki et al., 1992), but augmented cocaine-induced conditioned place preference in another study (Brown
et al., 1991). Buprenorphine alone increased locomotor activity in mice, but did not potentiate or attenuate locomotor activity induced by cocaine (Jackson et al., 1993). Both
cocaine and buprenorphine increased dopamine release in the nucleus
accumbens when administered separately and together (Brown et
al., 1991). In the present experiments, buprenorphine produced
rotational behavior alone and increased amphetamine-induced circling.
These results lend support to the studies which indicate that
buprenorphine increases dopamine release, producing behaviors that are
associated with this neurochemical event. This augmented release of
dopamine is possibly a result of buprenorphine's actions on both the
mu and kappa opioid receptors. Based on
microdialysis data, stimulation of the mu opioid receptor
would result in increased synaptic dopamine levels, whereas inhibiting
the kappa opioid receptor would remove its inhibitory
effects upon dopamine release (Di Chiara and Imperato, 1988b). Perhaps
it is because of its dual actions at mu and kappa opioid receptors that buprenorphine produced the most turning of any of
the opioids that we tested.
Fentanyl did not increase the effects of amphetamine in the present
study, and in the rat drug discrimination paradigm, it did not
substitute for cocaine or alter the effects of cocaine when given as a
pretreatment (Broadbent et al., 1995). However, in monkeys,
fentanyl potentiated the discriminative effects of cocaine, as
evidenced by a leftward shift of the cocaine stimulus curve (Spealman
and Bergman, 1992, 1994). In vivo microdialysis studies
showed that this opioid agonist increased dopamine release in the
nucleus accumbens (Di Chiara and Imperato, 1988b).
In addition to examining the effects of mu opioid agonists
upon rotational behavior, we tested dextrorphan, the optical isomer of
levorphanol. Dextrorphan has little affinity for the mu
opioid receptors and lacks significant analgesic and other opioid
effects (Jaffe and Martin, 1985), and alone produced fewer rotations
than did levorphanol alone. Surprisingly, in combination with
amphetamine, dextrorphan produced slightly more turning behavior than
did levorphanol. The effect of dextrorphan on stimulant-induced turning
may be because of its interaction at the PCP recognition site of the NMDA glutamate receptor (Murray and Leid, 1984; Sun et al.,
1986).
Based on the model of simple arithmetic additivity, dextrorphan,
levorphanol, meperidine, methadone and morphine potentiated the effects
of amphetamine on circling. Although this "effect-addition" model
is not without limitations (Woolverton, 1987), it is a reasonable way
to approach the question of drug synergy in the absence of more
specific statistical methods. The fact that the mu opioid agonists often increased the effects of amphetamine at times when the
opioid itself had little or no effect (e.g., during the
second half of the 4-hr session in the case of most of the drugs, figs. 4B, 5B, 6B, 7B, 8B) suggests a true synergy.
Naltrexone, a nonspecific opioid receptor antagonist, blocked the
interaction between morphine and amphetamine, but not that between the
other mu opioid agonists and amphetamine. The reason for
this is not clear. However, the outcomes might have been confounded by
an interaction between naltrexone and amphetamine. When 0.1 mg/kg
naltrexone was administered with amphetamine, there were no significant
effects on the stimulant-induced turning behavior. However, higher
doses potentiated amphetamine-induced circling. This observation
conflicts with findings that the opioid antagonist naloxone can
decrease amphetamine-induced locomotor activity (Hooks et
al., 1992; Jones and Holtzman, 1992, 1994) and dopamine release in
rats (Schad et al., 1995). Kappa opioid receptor
activation decreases dopamine release (Di Chiara and Imperato, 1988b).
Perhaps by blocking kappa opioid receptors, naltrexone
augments the release of dopamine. Thus, naltrexone might have blocked
the potentiating effect of the mu opioid agonist while
increasing the dopamine response to amphetamine, which resulted in no
net change in turning. However, even if this speculation was to prove
correct, it would leave unresolved the reason that naltrexone did block
the amphetamine-potentiating effect of morphine.
In summary, morphine increases rotational behavior induced by
amphetamine and cocaine. However, the role of opioid receptors in this
interaction is unresolved. Results with other mu opioid agonists were consistent among drugs in that they all increased amphetamine-induced circling, but this interaction was not greater than
additive in all cases. In addition, each of the mu
opioid agonists studied induced turning by themselves, although not
always at any one dose. The opioid antagonist naltrexone blocked the effects of morphine but not those of the other mu opioids.
Although we cannot draw conclusions about the role of opioid receptors in the effects of mu opioid agonists on turning induced by
psychomotor stimulant drugs, it is, nevertheless, clear that these
opioids can enhance a dopamine-mediated behavioral effect of
amphetamine and cocaine.
The authors extend their appreciation for the generous
contribution of dextrorphan tartrate and levorphanol tartrate by Roche Laboratories, of fentanyl citrate by McNeil Laboratories, and of
meperidine hydrochloride and morphine hydrochloride by Penick Corp. We
also thank Dr. Ronald J. Tallarida for helpful discussions regarding
statistical analyses of the drug interaction data.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
-endorphin-induced alterations in dopamine release in the nucleus accumbens.
Eur. J. Pharmacol.
190: 177-184, 1990[Medline].
and 
agonists in squirrel monkeys discriminating low doses of cocaine.
Behav. Pharmacol.
5: 21-31, 1994.[Medline]
