Neuronal and behavioral correlates of intrastriatal infusions of amphetamine in freely moving rats

79

Brain Research, 627 (1993) 79-88 © 1993 Elsevier Science Publishers B.V. All rights reserved 0006-8993/93/$06.00

BRES 19367

Neuronal and behavioral correlates of intrastriatal infusions of amphetamine, in freely moving rats Zhongrui Wang, George V. Rebec

*

Program in Neural Science, Department of Psychology, Indiana University, Bloomington, IN 47405, USA (Accepted 8 June 1993)

Key words: Amphetamine; Freely moving rat; Haloperidol; Intrastriatal infusion; Single-unit recording; Striatum

When injected systemically in rats, amphetamine routinely activates striatal neurons that increase firing rate in close temporal association with movement but suppresses nonmotor-related neurons. To assess the role of striatal mechanisms in these opposing effects, D-amphetamine (20 t z g / # l ) was infused (10 p.l/h) directly into the striatum of awake, behaving rats and single-unit activity was recorded simultaneously at the infusion site. Intrastriatal amphetamine reliably activated motor-related, but suppressed nonmotor-related neuronal activity shortly after infusion onset. These changes in firing rate preceded overt behavioral changes, in most cases by several minutes. When they did emerge, behavioral responses were characterized mainly by focused sniffing and head bobbing. Interestingly, the strongest behavioral responses, as measured by onset latency and response magnitude, were likely to result from infusions into motor-related rather than nonmotor-related recording sites. Systemic injection of haloperidol (1.0 mg/kg) shortly after infusion offset suppressed both behavior and striatal neuronal activity. Control infusions of intrastriatal saline had no consistent effect on either striatal neuronal activity or behavior. Collectively, these results indicate that the divergence in firing rate between motor- and nonmotor-related striatal neurons reflects an intrinsic action of amphetamine in the striatum rather than a secondary effect of behavioral feedback. Moreover, the linkage of motor-related striatal areas with the strongest behavioral responses to amphetamine suggests important functional differences between motor- and nonmotor-related striatal neurons.

INTRODUCTION The behavioral effects of amphetamine, which in rats are manifest as a series of repetitive or stereotyped movements, often serve as a model of some symptoms of schizophrenic psychosis38'42'44. In fact, blockade of the behaviors induced by amphetamine and related psychomotor stimulants is a widely used screen in the development of new antipsychotic drugs 1'6°. Amphetamine is believed to exert its behavioral effects by enhancing dopamine transmission in specific areas of the forebrain, including the striatum. Depletion of striatal dopamine, for example, blocks amphetamineinduced behavioral activation in rats 9'11, whereas intrastriatal infusions of amphetamine elicit many of the behavioral patterns characteristic of systemic administration 31'52. Moreover, certain aspects of the amphetamine-induced behavioral response parallel the release of dopamine from axon terminals in the striatum 35,5~.

* Corresponding author. Fax: (1) (812) 855-4520.

Despite these advances in the psychopharmacology of amphetamine, relatively little information is available on the neuronal mechanisms by which the striatum mediates amphetamine-induced behavioral effects. Early attempts to record single-unit activity in the striatum of freely moving rats revealed that the neuronal response to this drug is characterized by both excitations and inhibitions t2'5°. Further analysis has shown that excitatory responses to amphetamine predominate in striatal neurons known to increase activity during spontaneous movement relative to periods of quiet rest, whereas neurons with firing rates unrelated to motor behavior are likely to respond with an inhibition 22'45'56. Because amphetamine increases movement, however, the drug-induced acceleration of striatal neuronal activity may be secondary to the change in behavior rather than reflect an intrinsic action of amphetamine on striatal neurons. This issue has been addressed with behavioral clamping techniques, which compare pre- and post-amphetamine firing rates during matched behavioral episodes 24'43. Although this analysis has shown a net amphetamine-induced increase in neuronal activity, it is impossible to rule out

80 some contribution of behavioral feedback effects, including those associated with simultaneous covert behavioral changes. Moreover, with systemic injections, it is difficult to assess the extent to which mechanisms intrinsic to the striatum can account for the opposing effects of amphetamine on motor- and nonmotor-related striatal neurons. To address these issues, we infused amphetamine directly into the striatum and simultaneously recorded both neuronal activity at the infusion site and open-field behavioral changes in freely moving rats. MATERIALS

AND METHODS

Animals A total of 50 male, Sprague-Dawley rats (350-500 g), bred in our animal colony, were used as subjects. They were housed individually under standard laboratory conditions with food and water available continuously. All animal-use protocols were approved by the Indiana University Institutional Animal Care and Use Committee in accordance with the guidelines established by the National Institutes of Health.

Surgery In preparation for surgery, rats were anesthetized with Equithesin (0.33 m l / k g , i.p.) and secured in a stereotaxic device. A 2-mm hole was drilled through the skull over the right striatum (approximately 0.7 m m anterior and 2.5 m m lateral to bregma, according to the coordinates of Paxinos and Watson 41) and covered with a thin layer of silicon film. A plastic hub, designed to mate with a micromanipulator for single-unit recordings and simultaneous intrastriatal infusions, was placed over the hole and cemented to the skull. A metal post, also anchored to the skull, was used to support a voltage follower and to serve as a ground electrode for electrophysiology. Some rats were implanted with a subcutaneous catheter for systemic drug injections. Following surgery, all animals were treated with 0.2 ml (25 mg, i.m.) ceftriaxone (Roche) to minimize infection. A recovery period of 6 - 1 0 days intervened before experimentation.

sensitive spike discriminator and audio monitor. For off-line analysis and subsequent synchronization with behavior, neuronal data were stored on an audio track of the videotape used to record behavioral activity. In some cases, it was possible to record more than one neuron simultaneously per animal. For these situations, BrainWave Systems software was used to isolate and monitor the activity of each neuron. Spontaneous neuronal discharges were recorded as the animal engaged in typical open-field behavior (e.g., quiet rest, head turning, forward locomotion, sniffing, etc.). Many neurons could be classified during this period as motor- or nonmotor-related depending on whether firing rate changed in close temporal association with movement as in previous work 22'55'56. Attempts also were made to manipulate neuronal activity by tactile stimulation of the animal's vibrissae, snout, limbs, and dorsum. Each rat then was allowed a 5 - 1 0 min period of quiet rest followed by activation of the infusion pump (flow rate of 1 0 / z l / h ) . Most rats (n = 35) received 2 0 / z g / j z l D-amphetamine sulfate, although 10 or 15 ~ g / / x l D-amphetamine was infused in some animals (n = 11). No concentration-dependent differences were observed in the striatum with respect to either the onset or magnitude of the amphetamine-induced neuronal and behavioral response during the recording session, and thus the a m p h e t a m i n e data were combined. Control animals (n = 4) received infusions of physiological saline (0.9%). In all cases, neuronal activity and behavior were monitored throughout the course of the infusion (20-30 min). In some animals (n = 9), 1.0 m g / k g haloperidol (McNeil) was administered (s.c.) 10 min after the a m p h e t a m i n e infusion, and neuronal and behavioral data were recorded for another 30 min.

Behavioral ratings Videotapes were analyzed to assess the effects of the amphetamine infusion on behavior. An independent observer rated individual components of the behavioral response according to standard procedures described previously 47. Behavioral responses, including locomotion, rearing, sniffing, grooming, and head movements, were rated for 1-min intervals according to duration (1 = discontinuous, 2 = continuous) and intensity (0 = not present, 1 = mild, 2 = moderate, 3 = intense). Ratings were made every 5 min beginning with the onset of the a m p h e t a m i n e infusion. The duration and intensity scores for each interval were multiplied to yield a single value from 0 to 6. Total scores were obtained by summing each individual score.

Data analysis Electrophysiology A n open-field arena (1.3 m2), housed inside a sound-attenuating chamber within view of a videotaping system, served as the recording location. Following an habituation period of at least 2 h, each animal was readied for single-unit recording. Two tungsten microelectrodes, flame-formed to a tip and coated with Epoxylite, were prepared to impedances of 1 - 2 M/2 measured at 1 kHz s. Both microelectrodes and a 23-gauge stainless steel infusion cannula were fitted into the micromanipulator, which then was attached to the hub previously mounted on the animal's head. A distance of less than 400 /~m separated the tips of the recording electrodes and the tip of the infusion cannula 46. The micromanipulator is equipped with a threaded assembly that allows the entire recording apparatus (microelectrodes and infusion cannula in tandem) to be raised and lowered through the brain. After positioning 4 m m below the cortical surface, the recording apparatus was advanced in increments of approximately 30 ~tm until single-unit activity was encountered. Recordings were obtained from the microelectrode that provided the best isolation (signal-to-noise ratio of at least 3:1). From the head-mounted voltage follower, neuronal activity was passed to a differential preamplifier (band pass: 0.3-10.0 kHz) via shielded 33-gauge wire in conjunction with an electronic swivel, which allowed the animal complete freedom of movement in the open-field arena. Polyethylene tubing connected the infusion cannula with a syringe controlled by a minipump. Neuronal activity was displayed by conventional m e a n s and counted on-line with a computer system connected to an amplitude-

Neuronal data were used to quantify drug-induced changes in single-unit activity relative to baseline firing rates. These data were analyzed with analysis of variance ( A N O V A ) and Fisher's LSD multiple comparisons. Those cases in which neuronal discharges decreased in amplitude during infusion were excluded from subsequent data analysis.

Histology W h e n the recording session was completed, each rat received sodium pentobarbital (200 m g / k g , i.p.) and a 3-mA current was passed through the microelectrode to mark the recording site. Following a transcardial perfusion with formosaline, brains were removed, frozen, sectioned, and stained with cresyl violet. Electrode placements were identified by microscopic examination.

RESULTS

Single-unit recordings in the striatum were obtained from 47 awake, behaving rats. In 5 of these animals, it was possible to isolate 2 cells simultaneously, resulting in a total sample of 52 striatal neurons. In all cases, spike activity during quiet rest was relatively slow (less than 20 spikes/s) and characterized by initially positive

81 TABLE I

Baseline activity of striatal neurons during quiet rest Mean firing rate ( + S.E.M.) was calculated for each neuronal type as the animals rested quietly prior to the intrastriatal infusion. Motorrelated neurons discharged significantly slower than nonmotor-related neurons ( P < 0.05). n = number of neurons in each case.

I I ..... I-"

Fig. 1. Computer-generated record of the firing pattern of a striatal motor-related neuron responding to spontaneous sniffing behavior. Neuronal discharges (middle line) and the corresponding rate-meter plot (top line) are shown for individual sniffs, which occur at each 'S' (bottom line). The depicted time period is approximately 30 s.

biphasic or triphasic waveforms with amplitudes of between 400 and 1800 /xV. More than half of our neuronal sample (n = 28) was classified as motor-related in that firing rate increased above the resting baseline rate in close temporal association with movement. Consistent with previous reports 24'56, these cells typically changed firing rate in response to general motor activity (e.g., rearing, locomotion, sniffing), though some cells responded only to specific movements such as head turning. An example of the firing pattern of a motor-related neuron is shown in Fig. 1. Note in this case the frank increase in firing rate immediately prior to sniffing behavior. A much smaller group of striatal neurons (n = 8) failed to change activity during overt behavior, and thus were classified as nonmotor-related. The remaining neurons in our sample (n = 16) could not be classified either because the animal failed to engage in sufficient spontaneous movement during the baseline period or because firing rate was inconsistent, alternately showing excitation, inhibition, or no change during multiple behavioral episodes. The resting baseline firing rate of each category of cells is listed in Table I. An overall A N O V A revealed significant category differences in firing rate (F2,48 = 5.39, P < 0.01) with a subsequent pairwise comparison indicating that the resting baseline rate of motor-related neurons was significantly slower than that of nonmotor-related cells ( P < 0.05). It was often the case that during the search for spontaneously active neurons, these three cell classifications were encountered in separate clusters or groups. As the electrode was lowered through the striatum, for example, several motor-related (or non-

Neuronal type

n

Spikes / s ( + S.E.M.)

Motor-related Nonmotor-related Unclassified

28 8 16

2.5 ( + 0.51) 18.2 ( + 8.3) 4.8 ( + 1.7)

motor-related or unclassified) neurons could be isolated in succession. Moreover, in all 5 cases in which 2 neurons were isolated simultaneously, both ceils received the same classification.

Effects of intrastriatal infusions on behavior In all cases, the infusion pump was activated during quiet rest. For amphetamine infusions (n = 43 rats), focused stereotyped behaviors, such as head bobbing and sniffing, were predominant, but as shown in Table II some animals also engaged in episodes of grooming as well as locomotion and rearing. These behavioral patterns began between 5 and 15 min after infusion onset. As behavior increased in intensity, all infusions were terminated between 20 and 30 min, and the behavioral response continued undiminished for at least another 30 min. In contrast, saline infusions (n = 4 rats) produced no overt behavioral activation either during or after the 30-min infusion period.

Effects of intrastriatal infusions on striatal neurons During amphetamine infusions, the activity of individual neurons (n = 46) began to change in a direction consistent with that reported following systemic amphetamine injections 24'56. Thus, motor-related neurons (n = 24) routinely showed a progressive, monotonic increase in firing rate (23 of 24 cases) that peaked at a mean of 823 ( + 359)% of resting baseline, whereas all nonmotor-related cells (n = 8) were inhibited to a mean

TABLE II

Percentage of animals showing specific behavioral responses during intrastriatal infusions of amphetamine (AMPH) Percentage calculations are based on 43 animals.

Type of behavior

Percentage of AMPH-infused rats

Sniffing Head bobbing Grooming Rearing Locomotion

93 86 67 53 30

82 30

100 '

'

'

,

F

e,

4,

,

I 20

I 25

o ' /

Y E-* .<

•

10

o

'b e

o ~

20

O ~3

=

r~ z .<

Q3 m r=,l

E ~3

0.1

10

•

0

0//

0

0.01

0 0.001 0.001

I 0.01

I 0.1

I 1

I 10

I 5

I 10

r 15

O0

TN

30

(min)

Fig. 3. Scanerplots of the onset times (min) of the neuronal response (TN) and stereotyped behavior (TB) during intrastriatal amphetamine infusions for motor-related (e), nonmotor-related (©) and unclassified neurons (open triangles). The regression equation is TB = 2.65 + 1.07 TN with correlation coefficient r = 0.89. The onset of the neuronal response precedes a change in behavior in virtually all cases as indicated by the number of points above the diagonal.

LOG BASELINE RATE

(SPreES/S) Fig. 2. Scatterplots of the log baseline firing rate during quiet rest vs. the log firing rate during the peak neuronal response to intrastriatal amphetamine infusions. Points above or below the diagonal indicate drug-induced increases or decreases, respectively, in neuronal activity. Note that virtually all motor-related neurons (*) are excited by amphetamine, whereas nonmotor-related neurons (©) are inhibited and unclassified neurons (open triangles) show both types of responses.

Correlation of behavioral and neuronal effects during amphetamine infusions

of 33 ( + 6)%. Neurons that remained unclassified (n = 14) responded with either excitations (n = 11) or inhibitions (n = 3), and the magnitude of these responses was consistently less than that recorded for either motor- or nonmotor-related neurons. Fig. 2 illustrates the peak response of each neuron during the amphetamine infusion with respect to the resting baseline rate. Saline infusions with either motor-related (n = 2) or unclassified (n = 2) neurons failed to produce any consistent change in neuronal activity.

Following infusion onset, amphetamine-induced changes in neuronal activity of at least 20% from the resting baseline rate occurred with a mean latency of 7.85 _+ 1.1 min, whereas the mean latency of behavioral activation was 11.04 _+ 1.3 min. This difference is highly significant (t = 5.04, P < 0.001). A scatterplot of behavior and firing rate onset times for individual animals is shown in Fig. 3. Note that for neurons in every category a change in firing rate occurs prior to behavioral

BEFORE A M P H E T A M I N E I N F U S I O N NEURONAL

ACTIVIT,

I !

TIME

I/

III m I

I

-6

(rain)

-2

~!!J

BEHAVIOR MARK

id

id

O. 1 min

DURING AMPHETAMINE INFUSION NEURONAL ACTIVITY I TIME

(mln)

BEHAVIOR MARK

II_

I

+4

+13

,,I ,, ,,I

.1

J!!

Fig. 4. Representive example of the firing pattern of a striatal motor-related neuron at selected times before ( - m i n ) and during ( + m i n ) intrastriatal amphetamine infusion. 'S', 'H', 'R', and 'L' represent sniffing, head bobbing, rearing, and locomotion, respectively. Note that the marked increase in neuronal activity during the amphetamine infusion occurs well in advance of behavioral activation.

83

hi, h ~

/

~

[

•

1

mediuml

TABLE III Behavioral scores for rats receiving infusions of amphetamine into striatal areas identified by neuronal type

Scores are presented at the mean total score (+ S.E.M.) during the amphetamine infusion. Infusions into motor-related areas elicited a significantly higher behavioral score than infusions into areas with other neuronal types (P < 0.05).

1.7

0.7

-0.3

Fig. 5. Schematic representation of the distribution of histologically

verified infusion and recording sites in striatum. The approximate locations of identified motor-related, nonmotor-related, and unclassified neurons are indicated by filled, open, and half-filled symbols, respectively. The size of the symbols shown in the legend reflects the classification of the amphetamine-induced behavioral response in each case as high, medium, or low based on the magnitude of the total behavioral score. Note the predominance of high behavioral scores with infusions into motor-related areas. The S depicts the location of recording sites challenged with saline infusions; in none of these cases was a neuronal or behavioral response evident. Illustrations are after Paxinos and Watson41; numbers indicate distance (mm) from bregma.

activation. An example of the firing pattern of a motor-related neuron before and during an amphetamine infusion is shown in Fig. 4. In this case, pre-infusion movements, such as sniffing, elicit a reliable increase in firing rate above the relatively slow baseline of quiet rest. During the amphetamine infusion, however, a change in neuronal activity is evident within 4 min, but approximately 13 rain elapse before behavioral activation. Histological analysis revealed a widespread distribution of infusion sites throughout anterior striatum, especially medial and central areas. For each site, we calculated an overall ranking based on the total behavioral score. This procedure allowed us to identify the relative sensitivity of each infusion site for amphetamine-induced behavioral effects. Fig. 5 illustrates the location of each infusion site as well as its position in the upper (most sensitive), middle, or lower (least sensitive) third of the ranking. Note the heterogeneous distribution of the highest and lowest ranking infusion sites throughout the striatum. Note also that the most behaviorally sensitive infusion sites coincided most often with recordings from motor-related neurons. Thus, 16 of 24 infusions into motor-related areas earned the highest ranking, but this was the case for only 5 of 12 unclassified recording sites and none of the 8 nonmotor-related recordings. Consistent with this finding, an overall analysis of the total behavioral scores for mo-

Neuronal type

Mean ( +_S.E.M.) total behavioral score

Motor-related Nonmotor-related Unclassified

9.26 ( ± 0.74) 5.33 ( ± 0.95) 6.87 ( +_0.53)

tor-related, nonmotor-related, and unclassified neuronal infusion sites revealed a significant difference (F2•42 = 5.56, P < 0.01), as shown in Table III. Subsequent pairwise comparisons indicated a significantly higher score for motor- than nonmotor-related infusion sites ( P < 0.05).

Effect of haloperidol Nine animals were injected (s.c.) with 1.0 m g / k g haloperidol shortly after termination of the intrastriatal amphetamine infusion. All 9 neurons recorded from these animals (6 motor-related, 2 nonmotor-related, and 1 unclassified) were inhibited by haloperidol (overall mean of 13.4 + 6.0% of the pre-haloperidol rate) regardless of the direction of the response of these cells to amphetamine. Thus, as shown for individual neurons in Fig. 6, haloperidol inhibited striatal activity following either an amphetamine-induced excitation or inhibition. In all cases, haloperidol also suppressed the behavioral response to intrastriatal amphetamine infusions. Within 15 min after haloperidol, the total behavior score was reduced to 18.6 + 3.6% of that during the peak behavioral response to amphetamine. In 3 neurons recorded from 3 separate animals, an injection (s.c.) of physiological saline failed tO alter either the

30.0 O UJ O9

i,

tU O9

1

=i A

H MINUTES

IO.O

A

14 MINUTES

uv.0

Fig. 6. Rate-meter records of representative examples of the response of motor-related (left) and nonmotor-related (right) neurons to a local infusion of amphetamine (A at arrow) for 30 rain followed by s.c. injection of 1.0 mg/kg haloperidol (H at arrow) Note that haloperidol suppressed neuronal activity in both cases, regardless of the direction of the response to the amphetamine infusion.

84 firing rate or behavior elicited by intrastriatal amphetamine infusions.

Extrastriatal infusions In 3 additional animals, histological analysis revealed that single-unit recordings were made in infusion sites outside the striatum. In 2 of these sites, located in cortical areas overlying the striatum, amphetamine infusions failed to produce any overt behavioral activation, though both neurons showed biphasic changes: an initial increase in firing rate that began within 10 min after infusion onset followed within 5 min by a suppression of neuronal activity until infusion offset. A third site, located ventral to the globus pallidus in the nucleus preopticus magnocellularis, also failed to elicit behavioral activation and produced a similar biphasic neuronal response. In all 3 extrastriatal recordings, baseline neuronal activity ranged between 0.5 and 6.3 spikes/s during quiet rest and failed to show consistent motor-related changes in firing rate. DISCUSSION Intrastriatal infusions of amphetamine reliably excite motor-related but inhibit nonmotor-related neurons in the striatum well before the onset of overt behavioral changes. These results suggest that the bidirectional effect of amphetamine on striatal neurons in freely moving animals involves an action of this drug on mechanisms intrinsic to the striatum rather than a secondary response to behavioral activation. We also found that sensitivity to amphetamine, as measured by the magnitude and onset of the behavioral change, was not distributed uniformly across the striatum. In fact, the strongest behavioral responses to amphetamine emerged from infusions into motor-related areas rather than into nonmotor-related areas. Moreover, consistent with data on systemic amphetamine injections 24'5~, haioperidol blocked the behavioral effects of intrastriatal amphetamine and reliably inhibited striatal neurons regardless of the direction of the amphetamine-induced neuronal response.

Amphetamine-induced excitations and inhibitions of striatal neurons Our classification of striatal neurons as motor- or nonmotor-related is based on the relationship between firing rate and spontaneous behavioral activity, as in previous work 22"24"56. Consistent with these reports, motor-related neurons, which comprised the clear majority of striatal recording sites, typically increased activity in association with either whole-body movements or movements of specific body parts. Such neuronal

responses, which have been reported in numerous investigations of striatal activity in behaving animals, have helped to support a role for the striatum in motor function 57. The nature of this role, however, is extremely complex since the firing pattern of striatal neurons appears to reflect the confluence of motor events with cognitive 29' motivational 4, and sensory 3'28'39 processes. Indeed, the striatum receives input from virtually all areas of cerebral cortex as well as several brainstem structures (for reviews see refs. 19, 20, 40). Thus, although striatal neurons are undoubtedly involved in movement, motor-related changes in firing rate are likely to reflect the integration of many different types of information into behavioral output. Our results indicate that motor-related neurons discharge at a slower rate than nonmotor-related cells during periods of quiet rest, suggesting fundamental differences between these neuronal populations. In behaving monkeys, slow-discharging, motor-related cells have been identified as striatal output neurons, whereas faster firing cells less closely linked to movement may represent striatal interneurons 33. Projection and interneuron populations in rat striatum also have been differentiated on the basis of firing rate 6'59. We also found, consistent with previous reports of rat striatal activity ~°'~2'z2"-~7'58, that motor-related neurons clearly outnumbered other neuronal categories in our sample. In light of anatomical evidence that striatal projection neurons outnumber interneurons by a large margin 25, one could argue that our sample of motor-related units is comprised of striatal output neurons. It is important to stress, however, that many striatal projection neurons are silent during resting behavior 33'5° and thus may not be included in our sample of spontaneously active cells. To the extent that such silent cells respond in a different direction to amphetamine than more active cells, as some evidence suggests 5°, striatal output neurons may not comprise a homogeneous population. In addition, we found many striatal neurons that could not be easily classified as motor- or nonmotor-related. This third group, which discharged at an intermediate baseline rate, may include both motorand nonmotor-related neurons that could not be adequately identified in animals with a low level of spontaneous behavior. Indeed, this group of neurons responded to amphetamine infusions with either excitations or inhibitions in roughly the same proportion as the number of motor- to nonmotor-related neurons. Alternatively, this third group may represent a unique category of neurons that signal more complex aspects of motor functioning such as arousal, anticipation, or attention. Hikosaka and Sakamoto 27, for example, found such cells in the striatum of monkeys trained to

85 perform a series of complex behavioral tasks. Indeed, many striatal neurons showing movement-related activity in trained animals respond in a cue- or context-dependent fashion (see refs. 13, 58). In striatal recordings from spontaneously behaving rats, therefore, it may not be unusual to find a relatively large population of striatal neurons that appear to discharge inconsistently to movement. In any case, these cells, like our group of readily identifiable motor-related neurons, respond to intrastriatal amphetamine administration primarily with an increase in firing rate. The predominance of excitatory striatal responses to systemic amphetamine injections has been noted repeatedly for recordings obtained from awake, behaving rats, but not from animals prepared for stereotaxic recording (for review see ref. 43). One explanation for this difference involves the influence of behavioral feedback on striatal neurons. In the freely moving preparation, for example, the amphetamine-induced excitation of striatal activity occurs most frequently in neurons excited during movement 22, and thus the increase in behavior induced by amphetamine, rather than an intrinsic action of this drug on striatal neurons, may explain the increase in striatal firing rate following amphetamine administration. Our results,, however, argue against this interpretation. We found that, like systemic injections, intrastriatal infusions of amphetamine activated motor-related neurons, but the infusions increased firing rate prior to the onset of behavioral activation, and in most cases the neuronal change preceded the behavioral change by several minutes. These results make it extremely unlikely that behavioral feedback is responsible for amphetamine-induced excitations of striatal neuronal activity in awake, behaving animals. Moreover, motor-related neurons responded with an excitation, even though intrastriatal amphetamine did not elicit the same behavioral pattern in all rats (see Table II). Our results also suggest that the inhibitory effects of amphetamine in the striatum, which occur relatively infrequently but follow the same time course as amphetamine-induced neuronal excitations, are similarly independent of behavioral feedback. The opposing effects of amphetamine in the striatum of the freely moving and stereotaxic preparations may reflect a complex interaction between amphetamine-induced dopamine release and the level of afferent input, both of which are likely to be higher in awake, behaving rats than in anesthetized or immobilized animals 2l'24. The level of cortical afferent activity is especially important because it conveys excitatory motor-related information to the striatum 2'~4'36, most likely via glutamate release 26. Interestingly, iontophor-

esis of dopamine, which by itself may exert excitatory or inhibitory effects on striatal neurons 5'48, has been shown to enhance the activating effects of glutamate 7. Thus, we have proposed that an amphetamine-induced release of dopamine may selectively increase the activity of striatal neurons receiving substantial cortical input 24. To the extent that such input is directed to motor-related neurons, the amphetamine-induced activation of these cells should be sensitive to cortical manipulations. Consistent with this model, cerebrocortical lesions significantly attenuate the amphetamineinduced excitation of motor-related striatal neurons 55, but fail to alter the inhibitory action of this drug on nonmotor-related neurons in the striatum 23. Amphetamine, therefore, appears to enhance information flow through the striatum by facilitating the activity of striatal neurons receiving excitatory motor-related information, but suppressing striatal neurons receiving little such input. Our results with haloperidol confirm the ability of this drug to reverse both the activation of striatal motor-related neurons and the corresponding change in behavior induced by amphetamine 24'45'56. The reversal of these effects is consistent with haloperidol's role as a dopamine receptor antagonist. Indeed, recent evidence indicates that either SCH-23390 or eticlopride, selective DI and D 2 dopamine receptor antagonists respectively, also blocks the acceleration of striatal neuronal activity and the behavioral activation induced by amphetamine 49. These data, however, cannot rule out a role for other neurochemical systems. Haloperidol, for example, also has a high affinity for striatal sigma receptors, and BMY-14802, a sigma ligand with relatively little affinity for dopamine receptors 53, mimics haloperidol in blocking the excitatory effects of amphetamine on both striatal neuronal activity and behavior 56. Our results also confirm the failure of haloperidol to block the amphetamine-induced inhibition of nonmotor-related striatal neurons 24. The inhibition cannot be explained as a local anesthetic effect of the amphetamine infusion since neuronal discharges maintained a stable amplitude and waveform. Moreover, the inhibitions were mainly confined to nonmotor-related neurons, which also are suppressed by systemic amphetamine injections 22'45 and, indeed, haloperidol is largely ineffective in reversing these inhibitions as well 24. Rather, amphetamine-induced inhibitions of striatal activity may involve non-dopaminergic mechanisms, including an increased release of GABA from intrastriatal axon collaterals 3°. Because striatai activity is regulated, at least in part, by GABAergic lateral inhibitory networks 2°, an amphetamine-induced inhibi-

86 tion may reflect the activation of adjacent motor-related neurons. Further research is required to test this hypothesis and to assess the extent to which motorand nonmotor-related neurons differ in their intrinsic and extrinsic afferent connections.

Striatal involvement in the behavioral effects of amphetamine Behaviorally, our intrastriatal amphetamine infusions most commonly elicited focused sniffing and head bobbing with relatively little locomotion. These results are consistent with evidence that the behavioral effects of amphetamine are not distributed uniformly across the striatum 3~. In fact, amphetamine-induced locomotion appears to involve ventral striatum (nucleus accumbens), whereas focused stereotypy reflects an action of this drug on more dorsal areas 52. Accordingly, histological analysis revealed that virtually all of our striatal infusion sites were located either dorsal to the nucleus accumbens or along its most dorsal extent. Amphetamine infusions into areas outside the striatum failed to activate behavior, as did intrastriatal saline. Saline infusions also failed to alter striatal neuronal activity, arguing against a nonspecific infusion effect on the discharge rate of individual neurons. Not all striatal infusion sites were equally sensitive to amphetamine as measured by the magnitude of the behavioral changes that we recorded. In central and medial regions of anterior striatum, where most of our amphetamine infusions occurred, areas of high sensitivity were adjacent to areas of low sensitivity, suggesting a complex heterogeneity. Most areas of high sensitivity, however, were associated with motor-related neurons. Because similarly classified neurons were often encountered together in the striatum, it seems likely that rather than occur randomly, such neurons occupy distinct striatal areas, which may reflect differences in neurochemical organization and afferent input. Noteworthy in this regard is the patch-matrix organization of the striatum L5'~8, but without additional information about our recording sites, it is difficult to speculate on the mechanisms underlying the relationship between motor-related neurons and striatal sensitivity to amphetamine. Interestingly, however, preliminary data on this issue indicate that motor-related striatal neurons are largely confined to matrix 54, which receives considerable input from sensorimotor cortex 16 and which differs from the patch (striosome) compartment in the number of dopamine uptake sites ~7, dopamine receptor density 37, and dopamine release 32'34. Taken together, these results suggest that the differential sensitivity of motor- and nonmotor-related striatal areas to amphetamine is likely to reflect neurochemical and neu-

roanatomical differences in the patch-matrix compartments of the striatum.

Conclusions Amphetamine acts directly in the striatum of freely moving rats to increase the discharge rate of motor-related neurons and to inhibit nonmotor-related neuronal activity. Although virtually all intrastriatal infusions of amphetamine elicited focused sniffing and head bobbing, the magnitude of these behavioral responses was likely to be greater during infusions into motor-related than nonmotor-related areas. Subsequent injections of haloperidol blocked amphetamineinduced behavioral activation, but reversed only the excitatory neuronal responses to amphetamine. Collectively, these results suggest important functional distinctions between motor- and nonmotor-related striatal neurons that may mediate the behavioral effects of this and related psychomotor stimulants. Acknowledgements. This research was supported by USPHS Grant DA 02451. We also acknowledge the technical assistance of Paul Langley and the secretarial assistance of Faye Caylor.

REFERENCES 1 Ahlenius, S., Pharmacological evaluation of new antipsychotic drugs. In P. Willner (Ed.), Behavioural Models in Psychopharmacology, Cambridge University Press, Cambridge, 1991, pp. 311330. 2 Aldridge, J.W. and Gilman S., The temporal structure of spike trains in the primate basal ganglia: afferent regulation of bursting demonstrated with precentral cerebral cortical ablation, Brain Res., 543 (1991) 123-138. 3 Aldridge, J.W., Jaeger, D. and Gilman, S.A., Comparison of single unit activity in primate caudate nucleus and putamen in a sensory cued motor task. In G. Bernardi, M.B. Carpenter, G. Di Chiara, M. Morelli and P. Stanzione (Eds.), The Basal Ganglia III, Advances in BehaL'ioral Biology, 1Iol. 39, Plenum, New York, 1991, pp. 303-312. 4 Apicella, P., Scarnati, E. and Schultz, W., Tonically discharging neurons of monkey striatum respond to preparatory and rewarding stimuli, Exp. Brain Res., 84 (1991) 672-675. 5 Arbuthnott, G.W., Brown, J.R., MacLeod, N.K., Mitchell, R. and Wright, A.K., The action of dopamine on synaptic transmission through the striatum. In M. Sandier, C. Feuerstein and B. Scatton (Eds.), Neurotransmitter Interactions in the Basal Ganglia, Raven, New York, 1987, pp. 71-81. 6 Chang. H.T. and Wilson, C.J., Anatomical analysis of electrophysiologically characterized neurons in the rat strio-pallidal system. In A. Bj6rklund, T. H6kfelt, F.G. Wouterlood and A.N. van den Pol (Eds.), Handbook of Chemical Neuroanatomy, Vol. 8, Analysis of Neuronal Microcircuits and Synaptic Interactions, Elsevier, Amsterdam, 1990, pp. 351-402. 7 Chiodo, L.A. and Berger, T.W., Interactions between dopamine and amino acid-induced excitation and inhibition in the striatum, Brain Res., 375 (1986) 198-203. 8 Ciancone, M.T. and Rebec, G.V., A simple device for the reliable production of varnish-insulated high impedance tungsten microelectrodes, J. Neurosci. Methods, 27 (1989) 77-79. 9 Creese, I. and Iverson, S.D., The pharmacological and anatomical substrates of the amphetamine response in the rat, Brain Res., 83 (1975) 419-436.

87 10 Dolbakyan, E., Hernandez-Mesa, N. and Bures, J., Skilled forelimb movements and unit activity in motor cortex and caudate nucleus in rats, Neuroscience, 2 (1977) 73-80. 11 Fink, J.S. and Smith, G.P., Relationships between selective denervation of dopamine terminal fields in the anterior forebrain and behavioral responses to amphetamine and apomorphine, Brain Res., 201 (1980) 107-127. 12 Gardiner, T.W., Iverson, D.A. and Rebec, G.V., Heterogeneous responses of neostriatal neurons to amphetamine in freely moving rats, Brain Res., 463 (1988) 268-274. 13 Gardiner, T.W. and Kitai, S.T., Single-unit activity in the globus pailidus and neostriatum of the rat during performance of a trained head movement, Exp. Brain Res., 88 (1992) 517-530. 14 Gardiner, T.W. and Nelson, R.J., Striatal neuronal activity during the initiation and execution of hand movements made in response to visual and vibratory cues, Exp. Brain Res., 92 (1992) 15-26. 15 Gerfen, C.R., The neostriatal mosaic: the reiterated processing unit. In M. Sandier, C. Feuerstein and B. Scatton (Eds.), Neurotransmitter Interactions in the Basal Ganglia, Raven, New York, 1987, pp. 19-29. 16 Gerfen, C.R., The neostriatal mosaic - multiple levels of compartmental organization in the basal ganglia, Annu. Rev. Neurosci., 15 (1992) 285-320. 17 Graybiel, A.M. and Moratalla, R., Dopamine uptake sites in the striatum are distributed differentially in striosome and matrix compartments, Proc. Natl. Acad. Sci. USA, 86 (1989) 9020-9024. 18 Graybiel, A.M. and Ragsdale, C.W., Jr., Biochemical anatomy of the striatum. In P.C. Emson (Ed.), Chemical Neuroanatomy, Raven, New York, 1983, pp. 427-504. 19 Grofova, I., Extrinsic connections of the neostriatum. In I. Divac and R,G.E. Obert (Eds.), The Neostriatum, Pergamon, Oxford, 1979, pp. 37-51. 20 Groves, P.M., A theory of the functional organization of the neostriatum and the neostriatal control of voluntary movement, Brain Res. Rev., 5 (1983) 109-132. 21 Hamilton, M.E., Mele, A. and Pert, A., Striatal extracellular dopamine in conscious vs anesthetized rats - effects of chloral hydrate anesthetic on responses to drugs of different classes, Brain Res., 597 (1992) 1-7. 22 Haracz, J.L., Tschanz, J.T., Greenberg, J. and Rebec, G.V., Amphetamine-induced excitations predominate in single neostriatal neurons showing motor-related activity, Brain Res., 489 (1989) 365-368. 23 Haracz, J.L., Tschanz, J.T., Griffith, K.G,, Wang, Z. and Rebec, G.V., Behavioral clamping analysis of amphetamine-induced effects on striatal single-unit activity in freely moving rats: role of cortical afferents, Soc. Neurosci. Abstr., 17 (1991) 1247. 24 Haracz, J.L., Tschanz, J.T., Wang, Z., White, I. and Rebec, G.V., Striatal single-unit activity responses to amphetamine and neuroleptics in freely moving rats, Neurosci. Biobehav. Rev., 17 (1993) 1-12. 25 Heimer, L., Alheid, G.F., and Zaborszky, L., Basal ganglia. In G. Paxinos (Ed.), The Rat Nervous System, Vol. 1, Forebrain and Midbrain, Academic Press, Orlando, FL, 1985, pp. 37-118. 26 Herrling, P.L., Pharmacology of the corticocaudate excitatory postsynaptic potential in the cat: evidence for its mediation by quisqualate- or kainate-receptors, Neuroscience, 14 (1985) 417426. 27 Hikosaka, O. and Sakamoto, M., Neural activities in the monkey basal ganglia related to attention, memory and anticipation, Brain Dev., 8 (1986) 454-462. 28 Hikosaka, O., Sakamoto, M. and Usui, S., Functional properties of monkey caudate neurons II. Visual and auditory responses, J. Neurophysiol., 61 (1989) 799-813. 29 Hikosaka, O., Sakamoto, M. and Usui, S., Functional properties of monkey caudate neurons III. Activities related to expectation of target and reward, J. Neurophysiol., 61 (1989) 814-832. 30 Katayama, Y., Miyazaki, S. and Tsubokawa, T., Electrophysiological evidence favoring intracaudate axon collaterals of GABAergic caudate output neurons in the cat, Brain Res., 216 (1981) 180-186.

31 Kelley, A.E., Gautheir, A.M. and Lang, C.G., Amphetamine microinjections into distinct striatal subregions cause dissociable effects on motor and ingestive behavior, Behac. Brain Res., 35 (1989) 27-39. 32 Kemel, M.-L., Desban, M., Glowinski, J. and Gauchy, C., Distinct presynaptic control of dopamine release in striosomal and matrix areas of the cat caudate nucleus, Proc. Natl. Acad. Sci. USA, 86 (1989) 9006-9010. 33 Kimura, M., Behaviorally contingent property of movement-related activity of the primate putamen, J. Neurophysiol., 63 (1990) 1277-1296. 34 Krebs, M.-O., Trovero, F., Desban, M., Gauchy, C., Glowinski, J. and Kemel, M.-L. Distinct presynaptic regulation of dopamine release through NMDA receptors in striosome- and matrix-enriched areas of the rat striatum, J. Neurosci., 11 (1991) 1256-1262. 35 Kuczenski, R. and Segal, D., Concomitant characterization of behavioral and striatal neurotransmitter response to amphetamine using in vivo microdialysis, J. Neurosci., 9 (1989) 2051-2065. 36 Levine, M.S., Schneider, J.S., Lloyd, R.L., Hull, C.D. and Buchwald, N.A., Aging reduces somatosensory responsiveness of caudate neurons in the awake cat, Brain Res., 405 (1987) 389-394. 37 Loopuijt, L.D., Sebens, J.B. and Korf, J., A mosaic-like distribution of dopamine receptors in rat neostriatum and its relationship to striosomes, Brain Res., 405 (1987) 405-408. 38 Lyon, M., Animal models with parallels to schizophrenia, Animal Models in Psychiatry, I, 18 (1991) 25-65. 39 Manetto, C. and Lidsky, T.I., Caudate neuronal activity in cats during head turning: selectivity for sensory triggered movements, Brain Res. Bull., 16 (1986) 425-428. 40 Mehler. W.R. and Nauta, W.J.H., Connections of the basal ganglia and of the cerebellum, Confin. Neurol., 36 (1974) 205-222. 41 Paxinos, G. and Watson, C., The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1982, 91 pp. 42 Randrup, A. and Munkvad, I., Biochemical, anatomical, and physiological investigations of stereotyped behavior induced by amphetamine. In E. Costa and S. Garattini (Eds.), Amphetamines and Related Compounds, Raven, New York, 1970, pp. 695-713. 43 Rebec, G.V., Neuropharmacology of amphetamine: changes in the activity of striatal neurons during behavior. In R.R. Watson (Ed.), Biochemistry and Physiology of Substance Abuse, Vol. IV, CRC Press, Boca Raton, FL, 1992, pp. 149-174. 44 Rebec, G.V. and Bashore, T.R., Critical issues in assessing the behavioral effects of amphetamine, Neurosci. Biobehav. Rev., 8 (1984) 153-159. 45 Rebec, G.V., Haracz, J.L., Tschanz, J.T., Wang, Z. and White, I., Responses of motor- and nonmotor-related neostriatal neurons to amphetamine and neuroleptic drugs. In G. Bernardi, M.B. Carpenter, G. Di Chiara, M. Morelli and P. Stanzione (Eds.), The Basal Ganglia III, Advances in Behavioral Biology, Vol. 39, Plenum, New York, 1991, pp. 463-470. 46 Rebec, G.V., Langley, P.E., Pierce, R.C., Wang, Z. and Heidenreich, B.A., A simple micromanipulator for multiple uses in freely moving rats: electrophysiology, voltammetry, and simultaneous intracerebral infusions, J. Neurosci. Methods, 47 (1993) 53-59. 47 Rebec, G.V., Peirson, E.E., McPherson, F.A. and Brugge, K., Differential sensitization to amphetamine following long-term treatment with haloperidol or clozapine, Psychopharmacology, 77 (1982) 360-366. 48 Rolls, E.T., Thorpe, S.J., Boytim, M., Szabo, I. and Perrett, D.I., Responses of striatal neurons in the behaving monkey. 3. Effects of iontophoretically applied dopamine on normal responsiveness, Neuroscience, 12 (1984) 1201-1212. 49 Rosa-Kenig, A., Puotz, J.K. and Rebec, G.V., The involvement of D1 and D2 dopamine receptors in amphetamine-induced changes in striatal unit activity in behaving rats, Brain Res., 619 (1993) 347-351. 50 Ryan, L.J., Young, S.J., Segal, D.S. and Groves, P.M., Antidromically identified striatonigral projection neurons in the chronically implanted behaving rat: Relations of cell firing to amphetamineinduced behaviors, Behav. Neurosci., 103 (1989) 3-14. 51 Sharp, T., Zetterstrom, T., Ljungberg, T. and Ungerstedt, U., A

88

52

53 54

55

56

direct comparison of amphetamine-induced behaviours and regional brain dopamine release in the rat using intracerebral dialysis, Brain Res., 401 (1987) 322-330. Staton, D.M. and Solomon, P.R., Microinjections of d-amphetamine into the nucleus accumbens and caudate-putamen differentially affect stereotypy and locomotion in the rat, Physiol. Psychol., 12 (1984) 159-162. Taylor, D.P. and Dekleva, J., Potential antipsychotic BMY-14802 selectively binds to sigma sites, Drug Dev. Res., 11 (1987) 65-70. Trytek, E.S., White, I.M., Schroeder, D.M. and Rebec, G.V., Characterization of neuronal activity in the patch-matrix compartments of the striatum in freely moving rats, Soc. Neurosci. Abstr., in press. Tschanz, J.T., Haracz, J.L., Griffith, K.E. and Rebec, G.V., Bilateral cortical ablations attenuate amphetamine-induced excitations of neostriatal motor-related neurons in freely moving rats, Neurosci. Lett., 134 (1991) 127-130. Wang, Z., Haracz, J.L. and Rebec, G.V., BMY-14802, a sigma

57

58

59

60

ligand and potential antipsychotic drug, reverses amphetamineinduced changes in neostriatal single-unit activity in freely moving rats, Synapse, 12 (1992) 312-321. West, M.O., Carelli, R.M., Pomerantz, M., Cohen, S.M., Gardner, J.P., Chapin, J.K. and Woodward, D.J., A region in the dorsolateral striatum of the rat exhibiting single-unit correlations with specific locomotor limb movements, J. NeurophysioL, 64 (1990) 1233-1246. White, I.M. and Rebec, G.V., Responses of rat striatal neurons during performance of a lever-release version of the conditioned avoidance response task, Brain Res., 616 (1993) 71-82. Wilson C.J. and Groves P.M., Spontaneous firing patterns of identified spiny neurons in the rat neostriatum, Brain Res., 220 (1981) 67-80. Worms, P. and Lloyd, K.G., Predictability and specificity of behavioral screening tests for neuroleptics, Pharmac. Ther., 5 (1979) 445-450.