Effect of dose on cocaine self-administration behavior and dopamine levels in the nucleus accumbens

Brain Research. 539 (1991) 94-102

94

Elsevier BRES 16246

Effect of dose on cocaine self-administration behavior and dopamine levels in the nucleus accumbens Hugh O. Pettit and Joseph B. Justice Jr. Emory University, Department of Chemistry, Atlanta, GA 30322 (U.S.A.) (Accepted 14 August 1990)

Key words: Microdialysis; Reinforcement; Reward; Cocaine; Dopamine; Self-administration

The reinforcing properties of cocaine are thought to be primarily mediated by the release of dopamine (DA) in the nucleus accumbens (N ACC). The extraceUular concentration of DA in the N ACC was monitored with in vivo microdialysis procedures during ongoing cocaine self-administration to achieve a more detailed understanding of how DA mediates the reinforcing effects of cocaine. A dose-dependent decrease in lever pressing behavior occurred as the dose of cocaine was increased. The mean number of lever presses (in 20 min intervals) for 0.25, 0.50 and 0.75 mg/infusion doses was 5.6 + 0.7, 3.3 + 0.3 and 2.4 + 0.3, respectively. However, a simple inverse relationship did not occur between lever pressing behavior and the total amount of cocaine injected. Lever pressing behavior significantly increased cocaine intake as the dose of cocaine was increased. The total amount of cocaine intake that occurred during the 3 h self-administration period of the 0.25, 0.50 and 0.75 mg/infusion doses was 12.0 + 1.8 mg, 14.6 + 0.37 mg and 16.6 + 1.2 rag. Correspondingly, the extracellular concentration of DA in the N ACC was increased and maintained at significantly higher levels as the dose of cocaine was increased. The average concentration of DA that occurred during the self-administration of 0.25, 0.50 and 0.75 mg/infusion doses of cocaine was 269 + 26%, 381 + 21% and 464 + 49% of the basal DA concentration. As dose is increased, a corresponding increase occurs in both cocaine intake and in the extracellular concentration of DA in the N ACC. These results indicate that the level of reinforcement that occurs during cocaine self-administration is dependent on dose. Although within a single self-administration session, DA levels are stable, between self-administration sessions, the level does change with dose. Across doses, animals do not titrate cocaine infusions to maintain a specific level of DA in the N ACC. Apparently, just maintaining a specific extracellular level of DA in the N ACC does not maximize the reinforcing effects of cocaine. INTRODUCTION Drugs of abuse have specific molecular structures that in either direct or indirect ways act on the neuronal mechanisms that can produce reinforcing effects. Cocaine is one such drug of abuse that has a physiological effect that prevents the reuptake of the neurotransmitter dopamine ( D A ) into presynaptic terminals28'33'34. Molecules of D A are typically removed from the extracellular fluid into presynaptic terminals by an active uptake process 2'19'38. Cocaine is thought to block the reuptake of D A by interfering with the binding of an extracellular s o d i u m - D A complex to reuptake sites 14'28. By inhibiting uptake, cocaine increases the extracellular concentration of D A and thus the actions of D A on receptors. The reinforcing effects of cocaine have been suggested to be primarily mediated by dopaminergic effects in the nucleus accumbens (N A C E ) 15'40. Cocaine self-administration is attenuated by the administration of D A antagonists5, and by specific 6-hydroxydopamine (6O H D A ) lesions of D A terminals in the N A C C 20'29'30. During intravenous psychomotor stimulant self-administration, the response rate varies inversely with dose 26.

Animals will decrease the rate of responding for cocaine administration as dose is increased z5"27. A n i m a l s have been proposed to compensate response rate for drug dose in order to produce a constant drug level in both the blood and brain 27'42'43. We have previously reported that the titrated administration schedule observed during cocaine self-administration can maintain the concentration of D A in the N A C C at an increased and steady level 23. The selfselection of a particular schedule of drug administration implies that drug-induced reinforcement may be maximal under self-administration conditions, and during cocaine self-administration the extracellular concentration of D A in the N A C C is m a i n t a i n e d at approximately 350% of basal levels23. If the release of D A in the N A C C 15 40 primarily mediates cocaine-induced reinforcement ' , then a 350% increase in the concentration of D A in the N A C C may represent a level at which D A - i n d u c e d reinforcement is maximal. In the present experiment we have examined the extracellular concentration of D A in the N A C C during the self-administration of several doses of cocaine. If animals regulate lever pressing behavior in order to

Correspondence: J.B. Justice Jr., Emory University, Department of Chemistry, 1515 Pierce Drive, Atlanta, GA 30322, U.S.A. 0006-8993/91/$03.50 © 1991 Elsevier Science Publishers B.V. (Biomedical Division)

95 maintain a fixed concentration of DA in the N ACC, then this level should not change as the dose of cocaine is altered in a self-administration paradigm. If, however, the degree of reinforcement that is produced by cocaine is not regulated by the effects of a specific concentration of DA in the N ACC, then different levels of DA should be observed during the self-administration of different doses of cocaine. MATERIALS AND METHODS

Subjects Male Wistar rats (SASCO, n = 6) weighing 275-325 g at the time of surgery were used as subjects. Animals were initially group housed in a animal facility on a 12 h light-dark cycle (lights off from 19.00 h to 07.00 h). Subjects were given continuous access to food and water, except during a one week period in which food (Purina Lab Chow) was limited to approximately 15 g/rat/day. During the one week food deprivation period animals were trained to lever press for food reinforcement. Subsequently surgical jugular catheter and microdialysis probe guide cannula implantation procedures were implemented (see below). Thereafter, animals were individually housed in self-administration cages for the remainder of the study. Lighting was controlled on a reversed 12 h light-dark cycle (lights off from 09.00 h to 09.00 h). Food and water were provided on an ad libitum basis.

Apparatus The high pressure liquid chromatography (HPLC) system with electrochemical detection and details on the construction of microdialysis probes have been described elsewhere 3'23"24. Self-administration apparatus consisted of Razei infusions pumps (Model A), an Esterline Angus cumulative recorder, and Coulbourn electronic logic modules.

Procedure Surgical procedures. Animals were anesthetized with 50 mg/kg intraperitoneai injections of sodium pentobarbital (Nembutal). Intravenous catheters were implanted into the right external jugular vein according to previously published methods 2°'23'36. The above procedures were modified, however so that the free end of the jugular catheter passed subcutaneously to a mount (Plastics One, 313-001) that was secured to the dorsal surface of the skull. A screw on fluid lead (Plastics One, C313CS) connected the mount to a liquid swivel assembly. This modification serves to reduce trauma, circumvents bacterial infections, and removes a structural weak point that is inherent in a subcutaneous mesh harness design. After catheter implantation, stainless steel microdialysis probe guide cannulas (Plastics One, C313GP, 20 gauge) were implanted 2.0 mm above the ventral surface of the N ACC (AP +3.0 from bregma, L + 1.7, DV -6.1 from skull at the point of penetration, incisor bar +5.0). Intramuscular penicillin (60,000 units) was administered immediately following surgery. All subjects were allowed a 3 day recovery period. Self-administration procedure. In order to aid acquisition of self-administration behavior, animals were initially trained to lever press for food pellets on a continuous reinforcement schedule (FR-1). Following surgical procedures and the 3 day recovery period, animals were allowed to lever press for cocaine administration on a FR-1 schedule of reinforcement (3 h per day, 7 days a week). All trials began 2-3 h after house lights had been turned off. Two free priming injections signaled the beginning of each 3 h self-administration trial. Each infusion of cocaine was delivered in a 0.1 ml volume over a 4-s period. Following the initiation of each infusion, a cue light was illuminated and levers were deactivated for a 20-s period to prevent overdose from continuous infusion. Cocaine (0.25, 0.5 or 0.75 mg/infusion (mg/inf); see experimental section for

details) was dissolved in a sterile 0.9% saline solution. These doses correspond to approximately 0.75, 1.5 and 2.25 mg/kg/inf (assuming a body weight of 333 g). Doses were not corrected for body weight because a daily change in solution can deliver both lethal amounts of air into the bloodstream and cocaine infusions that are not contingent on lever pressing behavior. The dose range from 0.25 to 0.75 mg/inj is on the ascending limb of dose-response functions25'32. Microdialysis procedure. The active area of the dialysis probe was 2 mm in length and 300/~m in diameter. The inflow line of each microdialysis probe was spliced to one channel of a dual channel liquid swivel (BAS; the remaining channel was used for cocaine delivery). The microdialysis outflow line passed from the probe tip to a small sample collection vial (0.2 ml) mounted directly below the swivel, outside of the self-administration cage. In this manner samples of the extracellular fluid of the N ACC could be collected during cocaine self-administration without a disruption of behavior. The dialysis fluid flow rate delivered approximately 2.1 #l in a 10 min collection period (i.e. 210 nl/min). The samples were then injected into a 0.5/zl sample loop of a smallbore HPLC-EC system for analysis. Artificial CSF was composed of 0.13 M sodium chloride, 2.68 mM potassium chloride, 1.77 mM calcium chloride, 0.98 mM magnesium chloride, 0.25 mM ascorbic acid, and 10 mM glucose. Other microdialysis and chromatographic procedures have been reported elsewhere 23'24. Experimental procedure. The design of the present selfadministration study included a two-factor within-subject analysis of the effects of cocaine dose (factor 1: 0.25, 0.5 and 0.75 mg/inf) over time (factor 2 : 1 0 min measures over a 3 h period) on the concentration of DA in the extracellular fluid of the N ACC (dependent variable). All animals were initially allowed to lever press for a 0.5 mg/inf dose of cocaine until titrated responding was observed. This initial procedure enhanced the acquisition of subsequent self-administration behavior across doses by minimizing both overdosing with high doses, and extinction with low doses. Each animal was then allowed to self-administer one of 3 doses of cocaine (0.25, 0.5 and 0.75 mg/inf). The order in which each animal self-administered each dose of cocaine was randomly determined. Once stable self-administration behavior was observed, microdialysis procedures were used to monitor the level of DA in the N ACC during cocaine self-administration. Stable self-administration behavior was defined as 3 consecutive days/trials in which the number of lever presses for any individual trial did not deviate by more than 10% of the mean. The following day animals were allowed to self-administer the next dose of cocaine, and the above procedures were continued for each remaining dose. On completion of the above dose regimen, a subset of animals (n = 3) was placed on an extinction schedule in which saline was substituted for cocaine. After approximately 20 extinction trials, DA levels in the N ACC were analyzed during ~aline selfadministration conditions. Data analysis. A within-subjects design was used to assess the effects of cocaine dose on self-administration behavior and the DA concentration in the N ACC. Two-way analyses of variance (ANOVAs) with repeated measures on both factors were used to determine the effects of cocaine dose on the dependent variables of (1) self-administration lever press rates, (2) the DA concentration in the N ACC, (3) cocaine intake that occurred during selfadministration on the microdialysis test day, and (4) cocaine intake on the 4 days that preceded the microdialysis test day. The factors with repeated measures were cocaine dose and time. The effect of multiple insertions of a microdialysis probe into the N ACC on the basal concentration of DA in the N ACC was assessed with a one-way ANOVA with repeated measures over insertion days. Newman-Keuls aposteriori comparisons were used to define significant results of the ANOVAs. RESULTS D i f f e r e n t levels o f D A in t h e N A C C w e r e p r o d u c e d

96

~z

7

r~

5

Z

3

z

"..

z

~~

.........

40

flO

• ......... ~ . . . . . .

1 -40

," i 0

120

160

200

TIME (min) Fig. 1. The effects of cocaine dose on lever pressing behavior in a self-administration paradigm (n = 6). Increasing cocaine dose (0.25, 0.50 and 0.75 mg/inf) decreased lever pressing rates for cocaine administration in a dose-dependent manner. Open circles represent the lever press rates that were observed during extinction conditions in a subset of animals (n = 3). The first time point includes the two priming lever presses for each dose. Vertical bars illustrate the standard error of the mean (S.E.M.).

during the self-administration of different doses of cocaine. Also, cocaine intake was enhanced as dose of cocaine was increased. Although animals decreased their response rate as the dose of cocaine was increased, a simple inverse relationship did not occur between response rate and cocaine dose. These results reveal that during cocaine self-administration a positive relationship occurs between cocaine dose and both (1) cocaine intake and (2) the extracellular concentration of D A in the N ACC. The effect of cocaine dose on self-administration behavior is depicted in Fig. 1. As cocaine dose was increased (from 0.25 to 0.75 mg/inf) animals decreased the rate of responding for cocaine administration (F = 17.1, df = 2/10, P < 0.001). The mean amount of lever presses that occurred in 20 min intervals for 0.25, 0.50 and 0.75 mg/inf doses was 5.57 + 0.73, 3.28 + 0.28 and 2.44 + 0.26 lever presses, respectively. Newman-Keuls post-hoe comparisons indicated that the lever pressing rate for the 0.25 mg/inf dose was significantly greater than rates produced by cocaine doses of 0.50 and 0.75 mg/inf (P < 0.01). Interinfusion intervals were 4.21 + 0.62 min, 6.19 + 0.16 min and 8.35 + 0.66 min for 0.25, 0.5 and 0.75 mg/inf doses, respectively. A n inverse relationship occurred between lever pressing behavior and dose. The effect of cocaine dose in a self-administration paradigm on the extracellular concentration of D A in the N A C C is illustrated in Fig. 2. As cocaine dose was increased the D A concentration in the N A C C increased in a significantly different manner over time (dose by time interaction, F = 2.4, df = 34/170, P < 0.0001). The average concentration of D A that occurred during the

self-administration of 0.25, 0.50 and 0.75 mg/inf doses of cocaine was 269 + 26%, 381 + 21% and 464 + 49% of the basal D A concentration, respectively. The average basal D A concentration in the N A C C was 19.0 _+ 5.0 nM. As saline extinction data were obtained in a subset of the animals, these data were not included in the withinsubject analyses cited above. The data are, however, included in Figs. 1 and 2 for purposes of clarity. Statistical analyses of lever press rates and D A levels observed in the animals that received 0.0, 0.25 0.50 and 0.75 mg/inf doses of cocaine revealed findings that were essentially the same as those findings cited previously (main effect of dose on lever pressing: F = 9.9, df = 3/6, P < 0.01; interaction of dose over time on D A levels: F = 3.2, df = 51/102, P < 0.00001). Exceptions were that during extinction the lever press rates and D A levels were significantly lower than either response rates at the 0.25 mg/inf dose (P < 0.01), or D A levels observed during self-administration of the 0.75 mg/inf dose of cocaine (P < 0.05). A main effect of dose on D A levels was observed (F = 7.2, df = 3/6, P < 0.03). The lever press rates that occurred for the 0.25 mg/inf dose were significantly higher than the rates that occurred for 0.50 and 0.75 doses of cocaine (P < 0.05). The average number of lever presses and D A concentration that occurred in 20 min intervals during the extinction period was 1.18 + 0.47 lever presses, and 121.6% + 13.4% of the basal D A concentration. These control data indicate that either saline infusions or the motoric effects of lever pressing behavior (3.5 + 1.4/60 min) do not substantially increase the D A concentration in the N ACC. The amounts of cocaine intake that occurred over time

f~ Z

7oo 600

o~ .~

500

K~ s

~

0.75

ST 0.50

400 o

~TT~

300

T T ~-~J~i~i-~

O.Z5

200

10o 0

o.oo '

-40

'

'

'

0

.

.

.

.

.

40

.

.

.

.

.

.

80

.

.

.

.

.

.

120

.

.

.

160

200

TIME (min) Fig. 2. The extracellular concentration of DA in the N ACC during the self-administration of 0.25, 0.50 and 0.75 rag/infusion doses of cocaine (n = 6). The DA concentration in the N ACC was increased and maintained at higher levels as the dose of cocaine was increased in the self-administration paradigm. Open circles represent the DA concentrations that were observed during extinction conditions in a subset of animals (n = 3). The arrow depicts the beginning of the self-administration (S/A) period. Vertical bars represent the S.E.M.

97 ~; o

3.0 v

600 50O

2.5

2o

o.75 I

u Z o

4O0

"

300 2O0

m 0.0

r= 0.9998

100

o ,

0

'

'

20

40

60

80

TIME

0

100 120 140 160 180 200

(rain)

Fig. 3. The amount of cocaine intake during the self-administration of 0.25, 0.50 and 0.75 mg/inf doses of cocaine. A significant increase in cocaine intake occurred as the dose of cocaine was increased. Lever pressing rates that occurred during high dose self-administration conditions resulted in higher amounts of cocaine intake (asterisk indicates a significant difference from the 0.25 mg/inf dose). Note that intake was increased throughout the entire 3 h self-administration trial, and not just during the initial 'loading dose' period. Intake resulting from the two priming injections is included in the data for each dose. Vertical bars represent S.E.M.

during the self-administration of each dose of cocaine are shown in Fig. 3. A significant difference in cocaine intake occurred during the self administration of 0.25, 0.50 and 0.75 mg/inf doses of cocaine (F = 5.2, df = 2/10, P < 0.03). Newman-Keuls analyses indicated that significantly more cocaine intake occurred during the selfadministration of the 0.75 mg/inf dose as compared to the 0.25 mg/inf dose of cocaine (P < 0.05). Cocaine intake was increased throughout the entire 3 h period of self-administration. Hence, increased intake was not restricted to just the initial 'loading dose' period (see Fig. 3). The total amount of cocaine intake that occurred during the 3 h cocaine self-administration period of 0.25, 0.50 and the 0.75 mg/inf doses was 12.0 + 1.8 mg, 14.6 + 0.37 mg and 16.6 + 1.2 mg, respectively. A high correlation occurred between cocaine dose and the total amount of cocaine intake during the 3 h period of self-administration (r = 0.997). When the dose of cocaine was doubled (from 0.25 to 0.50 mg/inf) a 22% increase in cocaine intake occurred. A 38% increase in cocaine intake occurred when the dose of cocaine was tripled (from 0.25 to 0.75 mg/inf). The 22% increase in cocaine intake was accompanied by a 42% increase in the D A concentration, and the 38% increase in cocaine intake was accompanied by a 73% increase in the D A concentration. An analysis of the % increase of basal D A concentration that was divided by the amount of cocaine intake that occurred at the same time period (in 20 min intervals for the entire 3 h self-administration session) revealed no significant differences across doses (F = 0.43, df = 2/10, n.s.). These

i

0

2

,

i

,

4

i

6

,

i

8

,

i

10

,

i

12

,

i

i

14

i

16

i

i

18

TOTAL COCAINE INTAKE ( m g / 3 h o u r s ) 20

18

!

t~ 14

12

N

B

~ 0 ~

4

0E-~

0

fi--

8

0.25

0.50

0.75

SELF-ADMINISTRATION DOSE ( m g / i n f u s i o n ) Fig. 4. The correlation between cocaine intake and the level of DA in the N ACC (top panel), and the effect of cocaine dose in a self-administration paradigm on normalized lever pressing behavior (bottom panel). Cocaine intake accounted for the majority of the increase in the DA concentration during the self-administration of different doses of cocaine (0.25, 0.50 and 0.75 mg/inf). The bottom panel indicates how cocaine intake was increased as dose was increased. Lever press rates that occurred for each dose were standardized for the amount of cocaine delivered per infusion (amount of lever presses per mg cocaine). Normalized responding significantly increased as the dose of cocaine was increased. The asterisk represents a significant difference from 0.25 mg/inf lever pressing rates. Vertical bars illustrate the S.E.M.

results indicate that the differences in cocaine intake accounted for the majority of the differences observed in D A levels between doses. The top panel of Fig. 4 illustrates the high correlation (r = 0.9998) between the amount of cocaine intake and the level of extracellular DA in the N ACC. In order for cocaine intake to be increased as the dose of cocaine was increased there could not have been a simple inverse relationship between the lever press rate and dose (i.e. if 10 lever presses occur for a 0.25 dose this does not indicate that 5 lever presses will occur for a 0.5 mg/inf dose). The bottom panel of Fig. 4 indicates the normalized lever pressing behavior that was observed across doses. The total number of lever presses for every dose was normalized for the amount of cocaine administered (the total number of lever presses per mg of cocaine). Normalized lever pressing behavior signifi-

98 ~-~

2O

,~ o'3

18 16

i 0,75 i 0.50

r

14

M < Z Z <

lO

8 6 4 2 o

0.25

0.50

Correlation=

0.47

0,98

0.86

Slope=

0.09

0.37

0.22

i

-4

i

-3

~

i

SUCCESSIVE

i

i

-2

S/A

0.75

0

TRIALS

Fig. 5. The intake of cocaine during the 4 days prior to and the day of microdialysis. Regression lines indicate the stability of the acquisition to cocaine self-administration at the doses of 0.25, 0.5 and 0.75 mg/inf. Extending the acquisition time period would not appear to have resulted in a single common level of cocaine intake during the self-administration the different doses of cocaine. Intake values observed on the day of microdialysis (Day 0) are not included in regression calculations. Vertical bars depict the S.E.M.

cantly increased as the dose of cocaine was increased (F = 6.4, df -- 2/10, P < 0.02). Significantly more normalized lever pressing behavior occurred at the 0.75 as compared to the 0.25 mg/inf dose (P < 0.05). Essentially, these results indicate that a simple inverse relationship did not occur between lever pressing behavior and dose. Cocaine intake was enhanced by lever pressing behavior as the dose of cocaine was increased. After the dose of cocaine has been changed, differences in intake could be observed if animals are tested at a time point in which self-administration behavior is not stable. The mean amount of cocaine intake that occurred for each dose on the 4 days that preceded the microdialysis measurements is illustrated in Fig. 5. Statistical analysis of these data revealed that significantly different levels of cocaine intake occurred between doses (F = 13.6, df = 2/10, P < 0.02; each dose differed from the remaining doses, P < 0.05). However, intake did not change significantly during the 4 days that preceded microdialysis measures (F -- 2.1, df = 3/15, n.s.), or in a significantly different manner between doses over time (F = 0.33, df = 6/30, n.s.). Examination of the positive slopes of the first order regression lines through the data points in Fig. 5 indicates that levels of cocaine intake may not have been completely stable during the self-administration of the two high doses of cocaine (0.5 and 0.75 mg/inf). However, the level of intake for the low 0.25 mg/inf dose of cocaine was very stable prior to the microdialysis measures. The percent change in intake per day for the 0.25, 0.5 and the 0.75 mg/inf doses was 0.8%, 2.5% and 1.3%, respectively. The small rates of change, especially in the low dose relative to the others, suggest that a common level of cocaine intake would not have

been produced by extending the acquisition period (see Fig. 5). Apparently, the changes produced by extending the acquisition period would have increased, rather than decreased, the differences in cocaine intake between doses. More than 4 days are not included in the regression calculations because in two out of the 18 cases microdialysis was done after 4 days of acquisition. The mean amount of time that occurred between dialysis and a change in dose (the acquisition period) was 7.9 +_ 0.9 days/3 h trials. In each individual animal microdialysis probes were inserted into the N ACC 3 times (one microdialysis test day for each dose). Multiple insertions of a dialysis probe into the N ACC can produce damage and gliosis that could limit the detection of neurochemical responses at later time points. To test for these effects a one-way A N O V A with repeated measures was used to determine if the basal D A concentration changed over the 3 dialysis days. Results indicated that no significant change occurred in the basal D A concentration (or in detecting the basal concentration) as a result of inserting a microdialysis probe into the N ACC 3 times (F -- 0.55, df = 2/10, n.s,). Although the random order in which doses were given did control for damage effects, the above analysis revealed that a significant amount of error was not produced by multiple insertions of a dialysis probe into the N ACC. DISCUSSION

We have previously reported that during cocaine self-administration (0.5 mg/inf) D A levels in the N ACC are increased and maintained at approximately 350% of the basal D A concentration 23. In the current study the extracellular concentration of D A was also observed to be maintained at a similar steady level during selfadministration of the same dose of cocaine (381% for 0.5 mg/inf). However, present results also indicate that different levels of extracellular D A are maintained during the self-administration of different doses of cocaine. The present findings indicate that an increase in the extracellular level of D A occurs during cocaine selfadministration while other investigators have reported that during cocaine self-administration (after 10 days of experience) the level of D A in the N ACC does not increase above saline control levels 13. In the present study substantially increased levels of D A were observed during cocaine self-administration in animals that had experienced 33.3 + 4.8 days/3 h trials of cocaine selfadministration. The exact reason why the results of Hurd et al. 13 differ from those reported here (and elsewhere 23) are unknown. However the findings may result from differences in the probe location within the brain. The

99 dialysis site of Hurd et al. 13 was in a site of DA and cholecystokinin coexistence in a posterior area of the N ACC. The present dialysis site was in a more anterior region of the N ACC. The results indicate that some mechanism of tolerance may exist in the posterior N ACC. In the anterior N ACC, on the other hand, DA levels remain substantially above control levels in animals that self-administer cocaine on a daily basis. Animals have been suggested to compensate lever pressing behavior for dose in order to maintain a specific drug concentration in the blood and brain 27'42'43. If a specific concentration of the drug is maintained across doses, then equivalent neurochemical effects would also be expected to occur across doses. The present study reveals that different levels of DA in the N ACC occur during the self-administration of different doses of cocaine. Different neurochemical effects would not be expected to be produced by equivalent cocaine concentrations. Extracellular concentrations of cocaine and dopamine have been found to be linearly related in the striatum 12"1s. When dose is changed animals do not titrate lever pressing behavior in order to maintain a specific concentration of DA, and apparently cocaine, in the N ACC. Differences in cocaine intake would be expected to produce a difference in the concentration of cocaine in the blood and brain. However no differences in amphetamine levels in the blood have been reported to occur when dose is changed in a self-administration paradigm. Yokel and Pickens 43 state, in the first experiment of a study, that increases in amphetamine dose lead to increases in intake. However in the second experiment of the study, whole body amphetamine levels were not observed to change as dose was increased. Unfortunately, statistical tests that compared intake between doses were not provided. In the present study dose was changed by altering the concentration of cocaine in a 0.1 ml infusion volume (delivered in 4 s). Yokel and Pickens 43 altered dose by changing the infusion duration and volume. Infusions were delivered in durations of 7-32 s and in volumes of 0.12-0.53 ml. The selfadministration response rate has been shown to decrease with increases in the infusion duration 1. A possibility exists that this type of effect could have retarded intake at higher doses and this may have been why no difference was seen in blood levels as dose was changed. Different amounts of cocaine intake would be expected to produce different concentrations of cocaine in the brain, and thus the observed different levels of DA in the N ACC. Alterations in the intake of cocaine and in the intake of other drugs of abuse have been reported in past research. As dose is increased, increases in intake have been reported to occur during the self-administration of

cocaine 1,27,39,amphetamine42,43, methylphenidate39, morphine 35'37, methadone 37, nicotine 4, ethanol 41, pentobarbital 1°, and chlordiazepoxide 6. The intake of most drugs of abuse is enhanced as dose is increased in a self-administration paradigm. The present study is not the first to report that cocaine intake is increased as dose is increased 1'27'39. These findings indicate that between self-administration trials, a simple inverse relationship does not exist between lever pressing behavior and dose. Animals will decrease their lever pressing rate as dose is increased, but not to rates that are sufficient enough to maintain an equivalent amount of drug intake. In a specific close range (perhaps only on the ascending limb of the dose response curve) animals alter lever pressing behavior, and consequently increase drug intake. The reason why animals increase drug intake is not known. However, the 3 following hypotheses are possible explanations for the observed effect of dose on cocaine intake: (1) the number of accidental lever presses is increased by the administration of high doses of cocaine; (2) as dose is increased the magnitude of cocaine-induced reinforcement is decreased; and (3) as dose is increased the reinforcement magnitude of cocaine is increased. Each of the above hypotheses are discussed below. The effects of large doses of cocaine could enhance the frequency of accidental lever presses. Both locomotor activity and stereotypy would be expected to be enhanced as dose is increased and, by these means, the number of inadvertent lever presses may be augmented as a result of the enhanced motoric activity. However this explanation appears unlikely because the response interval becomes larger as dose is increased. If increased levels of locomotor activity or stereotypy had produced lever presses, then these effects would be expected to occur at the time points when the in vivo drug concentration is the highest (i.e. within a short time period after an infusion). In this situation interresponse intervals would decrease rather than increase. Furthermore, accidental lever presses would not be expected to occur at regular time points. The interresponse intervals that occurred during the self-administration of the highest dose used in the present study were quite regular over time (8.35 + 0.66 min). Thus the above hypothesis cannot adequately explain the increased amount of cocaine intake observed in the present study. Another hypothesis could explain increased intake by suggesting that cocaine-induced reinforcement is decreased as the dose of cocaine is increased. Animals may compensate for a decrease in reinforcement magnitude by increasing intake. The administration of low doses of DA antagonists can increase rates of psychomotor stimulant self-administration 5,44. A partial blockade of DA

100 receptors by low doses of an antagonist may be compensated for by increasing the amount of DA that can compete with the antagonist for receptor binding 44'45. In a self-administration paradigm increases in lever pressing behavior can be interpreted as a sign of a decrease in the reinforcing efficacy of a drug. However, some research does indicate that, within a specific dose range, cocaineinduced reinforcement does not decrease as the dose of cocaine is increased. Present findings reveal that normalized lever pressing behavior is augmented during highdose self-administration, and high doses of cocaine have been shown to increase the breaking point for cocaine administration on a progressive ratio schedule 11'32. Thus, the above hypothesis does not adequately explain why cocaine intake was increased in the present study. Increased cocaine intake can also be explained if an increase in dose increases the reinforcement magnitude of cocaine (in a specific dose range). Animals may increase cocaine intake with access to higher doses, because the reinforcing effects produced by a 'large' dose are greater than those produced by smaller doses of cocaine. Pharmacokinetic effects may alter cocaineinduced reinforcement. This hypothesis is supported by findings that when dose is increased both normalized lever pressing behavior increases and higher breaking points occur in a progressive ratio schedule for cocaine self-administration 11,32. According to the above hypothesis cocaine intake was enhanced because the self-administration of high doses of cocaine produces levels of reinforcement that are greater than levels produced by low doses of cocaine. An increase in normalized lever pressing behavior occurred because of an increase in the magnitude of reinforcement that was administered after the performance of the operant response. Under these conditions the level of DA in the N ACC was also observed to increase as the self-administration dose of cocaine was increased. A corresponding increase in the level of D A in the N ACC occurred as the level of cocaine-induced reinforcement was increased. Although these findings do not provide evidence of causality, the correlation supports the hypothesis that the reinforcing properties of cocaine are primarily mediated by DA release in the N ACC. The intravenous administration of cocaine would be expected to induce an increase in DA levels throughout the brain, and not just in the N ACC. The extracellular concentration of D A would he expected to be increased to different levels (during the self-administration of different doses of cocaine) in brain areas such as the medial prefrontal cortex and the caudate nucleus. However, dopaminergic effects in these other brain areas do not primarily mediate the reinforcing effects of cocaine. Specific 6 - O H D A lesions of the medial prefrontal cortex,

and apparently the caudate, do not affect the intravenous self-administration of cocaine 17'31. Evidence that the reinforcing effects of cocaine are primarily mediated by D A release in the N ACC has been provided by findings that cocaine self-administration can be attenuated by specific 6 - O H D A lesions of D A terminals in the N ACC 2°'29'3°. Therefore, although intravenous cocaine can affect the concentration of D A in many areas of the brain, the reinforcing effects of cocaine are primarily mediated by changes that occur in the D A concentration in the N ACC. The present results have implications for exactly how D A in the N ACC produces reinforcing effects. The results indicate that there is not a single fixed concentration of D A in the N ACC that maximizes reinforcing effects. Lever pressing behavior does result in maintaining the concentration of D A in the N ACC at an increased and steady level, however the same level is not maintained during the self-administration of different doses of cocaine. To achieve a more detailed understanding of how D A in the N ACC mediates cocaine-induced reinforcement it may be crucial to understand why (1) different levels of D A in the N ACC occur during the self-administration of different doses of cocaine, and (2) why, during low dose self-administration, animals do not simply increase lever pressing rates in order to increase the concentration of D A in the N ACC to levels achieved during high dose self-administration. Cocaine-induced reinforcement may, in part, be mediated by neurochemicals other than DA, and this hypothesis might be suggested as an explanation of present results. Psychomotor stimulant reinforcement can be affected by interactions of the effects of D A with the effects produced by serotonin 16, benzodiazepine/ G A B A 9'21, and peptides 7's'22. Additionally, high doses of cocaine may increase the effects of active cocaine metabolites, or may have greater effects on low affinity DA, serotonin or norepinephrine receptors. Although the effects of non-dopaminergic substrates are contributing factors in mediating the reinforcing effects of cocaine, these types of effects do not explain why cocaine intake is increased during high dose self-administration, or why animals that self-administer low doses of cocaine do not simply increase lever pressing rates. If cocaine intake during low dose self-administration was increased to levels seen during high dose self-administration, then the increased intake would presumably increase the reinforcing effects induced by all of the above factors. However, animals self-regulate self-administration behavior, and consequently produce a fixed level of cocaine intake and a steady level of DA in the N ACC that differs according to the dose of cocaine that is self-administered. The present results indicate that at high doses animals

101 increase cocaine intake. O n e difference between cocaine doses m a y be the rate of change in D A concentration. Following a single infusion the neurochemical effects of a high (as c o m p a r e d to low) dose of cocaine occur in a much greater magnitude over a similar p e r i o d of time 23. Infusions of equivalent doses of cocaine delivered in s h o r t e r infusion intervals do p r o d u c e greater rates of responding and increased intake during cocaine selfadministration 1. D u r i n g low dose self-administration, animals could increase lever pressing behavior in o r d e r to increase cocaine intake and D A levels in the N A C C . H o w e v e r , the resulting cocaine and D A levels (that would be equivalent to those that occur during high dose self-administration) would not occur in the same p e r i o d of time. The rates of change would be different. Thus, rates of change in the D A (or cocaine) concentration may play a role in mediating the reinforcing effects of cocaine. The neurochemical events that mediate the reinforcing effects of cocaine primarily involve the actions of D A in

REFERENCES 1 Balster, R.L. and Schuster, C.R., Fixed-interval schedule of cocaine reinforcement: effect of dose and infusion duration, J. Exp. Anal. Behav., 20 (1973) 119-129. 2 Bogdanski, D.E and Brodie, B.B., The effects of inorganic ions on the storage and uptake of 3H-norepinephrine by rat heart slices, J. Pharmacol. Exp. Ther., 165 (1969) 181-189. 3 Church, W.H. and Justice, J.B., Jr., Rapid sampling and determination of extracellular dopamine in vivo, Anal. Chem., 59 (1987) 712-716. 4 Deneau. G.A. and Inoki, R., Nicotine self-administration in monkeys, Ann. N.Y. Acad. Sci., 142(1967)277-279. 5 Ettenberg, A., Pettit, H.O., Bloom, EE. and Koob, G.F., Heroin and cocaine intravenous self-administration in rats: mediation by separate neural systems, Psychopharmacology, 78 (1982) 204-209. 6 Findley, J.D., Robinson, W.W. and Peregrino, L., Addiction to secobarbital and chlordiazepoxide in the rhesus monkey by means of a self-infusion preference procedure, Psychopharmacologia, 26 (1972) 93-114. 7 Glimcher, P.W., Margolin, D.M., Giovino, A.A. and Hoebel, B.G., Neurotensin: a new 'reward peptide', Brain Research, 291 (1984) 119-124. 8 Glimcher, P.W., Giovino, A.A. and Hoebel, B.G., Neurotensin self-injection in the ventral tegmentai area, Brain Research, 403 (1987) 147-150. 9 Goeders, N.E., McNulty, M.A., Mirkis, S. and McAllister, K.H., Chlordiazepoxide alters intravenous cocaine self-administration in rats, Pharmacol. Biochem. Behav., 33 (1989) 859-866. 10 Goldberg, S.R., Hoffmeister, E, Schlichting, V. and Wuttke, W., A comparison of pentobarbital and cocaine self-administration in rhesus monkeys: effects of dose and fixed ratio parameters, J. Pharmacol. Exp. Ther., 179 (1971)277-283. 11 Griffiths, R.R., Bradford, L.D. and Brady, J.V., Progressive ratio and fixed ratio schedules of cocaine-maintained responding in baboons, Psychopharmacology, 65 (1979) 125-136. 12 Hurd, Y.L., Kehr, J. and Ungerstedt, U., In vivo microdialysis as a technique to monitor drug transport: correlation of extracellular cocaine levels and dopamine overflow in the rat brain, J. Neurochem. 51 (1988) 1314-1316. 13 Hurd, Y.L., Weiss, E, Koob, G.E, Anderson, N.E. and

the N A C C . It has b e e n generally assumed in the past that cocaine-induced r e i n f o r c e m e n t is regulated by the level of D A in the N A C C . R e c e n t technological advances in the use of in vivo microdialysis p r o c e d u r e s have m a d e it possible to m o n i t o r the concentration of D A in the N A C C that occurs during cocaine selfadministration. These d a t a indicate that during cocaine self-administration an increased level of D A is maintained in the N A C C . T h e reinforcing p r o p e r t i e s of cocaine may well be primarily m e d i a t e d by the release of D A in the N A C C . H o w e v e r , b o t h the intake of cocaine and the level of D A in the N A C C change according the dose of cocaine that is self-administered. These results raise new questions a b o u t the p h a r m a c o k i n e t i c m a n n e r in which D A in the N A C C m e d i a t e s the reinforcing effects of cocaine. Acknowledgements. This research was supported by NIDA Grant DA-05827 and by a Merck Sharp & Dohme postdoctoral grant.

Ungerstedt, U., Cocaine reinforcement and extracellular dopamine overflow in rat nucleus accumbens: an in vivo microdialysis study, Brain Research, 498 (1989) 199-203. 14 Kennedy, L.T. and Hanbauer, I., Sodium-sensitive cocaine binding to rat striatal membrane: possible relationship to dopamine uptake sites, J. Neurochem., 41 (1983) 172-178. 15 Koob, G.E, Vaccarino, F.J., Almaric, M. and Swerdlow, N.R., Neural substrates for cocaine and opiate reinforcement. In S. Fischer, A. Raskin and E.H. Uhlenhuth (Eds,), Cocaine: Clinical and Biobehavioral Aspects, Oxford University Press, New York, Oxford, 1987, pp. 80-108. 16 Lyness, W.H., Friedle, N.M. and Moore, K.E,, Increased self-administration of D-amphetamine after destruction of 5hydroxytryptaminergic neurons, Pharmacol. Biochem. Behav., 12 (1980) 937-941. 17 Martin-Iverson, M.T., Szostak, C. and Fibiger, H.C., 6-Hydroxydopamine lesions of the medial prefrontal cortex fail to influence intravenous self-administration of cocaine, Psychopharmacology, 88 (1986) 310-314. 18 Nicolaysen, L.C., Pan, H. and Justice, J.B., Jr., Extracellular cocaine and dopamine concentrations are linearly related in the rat stfiatum, Brain Research, 456 (1988) 317-323. 19 Paton, D.M., Mechanism of efflux of noradrenaline from adrenergic nerves in rabbit atria, Br. J. Pharmacol., 49 (1973) 614-627. 20 Pettit, H.O., Ettenberg, A., Bloom, EE. and Koob, G.E, Destruction of dopamine in the nucleus accumbens selectively attenuates cocaine but not heroin self-administration in rats, Psychopharmacology, 84 (1984) 167-173. 21 Pettit, H.O., Batsell, R.W. and Mueller, K., Triazolam attenuates amphetamine but not morphine conditioned place preferences, Psychopharmacology, 98 (1989) 483-486. 22 Pettit, H.O. and Mueller, K., Infusions of cholecystokinin octapeptide into the ventral tegmental area potentiate amphetamine conditioned place preferences, Psychopharmacology, 99 (1989) 423-426. 23 Pettit, H.O. and Justice, J.B., Jr., Dopamine in the nucleus accumbens during cocaine self-administration as studied by in vivo microdialysis, Pharmacol. Biochem. Behav., 34 (1989) 899-904. 24 Pettit, H.O., Pan, H., Parsons, L.H. and Justice, J.B., Jr., Extracellular concentrations of cocaine and dopamine are enhanced during chronic cocaine administration, J. Neurochem.,

102 55 (1990) 798-804. 25 Pickens, R. and Thompson, T., Cocaine-reinforced behavior in rats: effects of reinforcement magnitude and fixed-ratio size, J. Pharmacol. Exp. Ther., 161 (1968) 122-129. 26 Pickens, R. and Harris, W.C., Self-administration of o-amphetamine by rats, Psychopharmacology, 12 (1968) 158-163. 27 Pickens, R., Meisch, R.A. and Thompson, T., Drug selfadministration: an analysis of the reinforcing effects of drugs, In S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 12, Plenum Press, New York, 1978, pp, 1-37. 28 Ritz, M.C., Lamb, R.J., Goldberg, S.R. and Kuhar, M.J., Cocaine receptors on dopamine transporters are related to self-administration of cocaine, Science, 237 (1987) 1219-1223. 29 Roberts, D.C.S., Corcoran, M.E. and Fibiger, H.C., On the role of ascending catecholaminergic systems in intravenous self-administration of cocaine, Pharmacol. Biochem. Behav., 6 (1977) 615-620. 30 Roberts, D.C.S., Koob, G.E, Klonoff, P. and Fibiger, H.C., Extinction and recovery of cocaine self-administration following 6-hydroxydopamine lesions of the nucleus accumbens, Pharmacol. Biochem. Behav., 12 (1980) 781-787. 31 Roberts, D.C.S. and Koob, G.E, Disruption of cocaine selfadministration following 6-hydroxydopamine lesions of the ventral tegmental area in rats, Pharmacol. Biochem. Behav., 17 (1982) 901-904. 32 Roberts, D.C.S., Loh, E.A. and Vickers, G., Self-administration of cocaine on a progressive ratio schedule in rats: doseresponse relationship and effect of haloperidol pretreatment, Psychopharmacology, 97 (1989) 535-538. 33 Ross, S.B. and Renyi, A.L., Inhibition of the uptake of dopamine and ~4C-5-hydroxytryptamine in mouse striatal slices, Acta Pharmacol. Toxicol., 36 (1975) 56-66. 34 Sershen, H., Reith, M.E.A. and Lajtha, A., Comparison of the properties of central and peripheral binding sites for cocaine, Neuropharmacology, 21 (1982) 469-474.

35 Weeks, J.R. and Collins, R.J. Factors affecting voluntary morphine intake in self-maintained addicted rats, Psychopharmacologia, 6 (1964) 267-279. 36 Weeks, J.R., Long-term intravenous infusion. In R.D. Myers (Ed.), Methods in Psychobiology, Vol. 2, Academic Press. London, 1972, pp. 155-168. 37 Werner, T.E., Smith, S.G. and Davis, W.M., A dose-response comparison between methadone and morphine self-administration, Psychopharmacologia, 47 (1976) 209-211. 38 White, T.D., Models for neuronal noradrenaline uptake. In D.M. Paton (Ed.), The Mechanism of Neuronal and Extraneuronal Transport of Catecholamines, Raven Press, New York, 1976, pp. 175-193. 39 Wilson, M.C., Hitomi, M. and Schuster, C.R., Psychomotor stimulant self-administration as a function of dosage per injection in the rhesus monkey, Psychopharmacology, 22 (1971) 271-281. 40 Wise, R.A. and Bozarth, M.A., A psychomotor stimulant theory of addiction, Psychol. Rev., 94 (1987) 469-492. 41 Woods, J.H., Ikomi, E and Winger, G., The reinforcing property of ethanol. In M.K. Roach, W.M. Mclsaac and P.J. Creaven (Eds.), Biological Aspects of Alcohol, University of Texas Press, Austin, 1971, pp. 371-388. 42 Yokel, R.A. and Pickens, R., Self-administration of optical isomers of amphetamine and methylamphetamine by rats, J. Pharmacol. Exp. Ther., 187 (1973) 27-33. 43 Yokel, R.A. and Pickens, R., Drug level of D- and Lamphetamine during intravenous self-administration, Psychopharmacology, 34 (1974) 255-264. 44 Yokel, R.A. and Wise, R.A., Increased lever pressing for amphetamine after pimozide in rats: implications for a dopamine theory of reward, Science, 187 (1975) 547-549. 45 Yokel, R.A. and Wise, R.A., Attenuation of intravenous amphetamine reinforcement by central dopamine blockade in rats, Psychopharmacology, 48 (1976) 311-318.