Biol. Pharm. Bull. 25(12) 1570ﾑ1576 (2002)

1570 Biol. Pharm. Bull. 25(12) 1570—1576 (2002) Vol. 25, No. 12

The Pharmacological Characterization of Attentional Processes Using a Two-lever Choice Reaction Time Task in Rats

a a a a a Kenichi MISHIMA, Megumi FUJII, Naoya AOO, Tetsuya YOSHIKAWA, Yoshihiko FUKUE, a a a b Yo k o H ONDA, Nobuaki EGASHIRA, Katsunori IWASAKI, Yukihiro SHOYAMA, and ,a Michihiro FUJIWARA* a Department of Neuropharmacology, Faculty of Pharmaceutical Sciences, Fukuoka University; 8–19–1 Nanakuma, Jonan-ku, Fukuoka 814–0180, Japan: and b Department of Medicinal Resources Regulation, Graduate School of Pharmaceutical Sciences, Kyushu University; 3–1–1 Maidashi, Higashi-ku, Fukuoka 812–8582, Japan. Received May 7, 2002; accepted August 26, 2002

Activating the noradrenergic and cholinergic systems is known to enhance attentional processes, while stimulating dopaminergic, serotonergic, and GABAergic systems suppresses them. The objective of the present study was to investigate the pharmacological characterization in the attentional processes of a two-lever choice reaction time (CRT) task using different centrally acting drugs. We designed seven parameters in this task: the correct response (CR) rate; error response rate; nonresponse (NR) rate; differential reinforcement of other behavior (DRO) responses; number of incorrect lever pressings during both the intertrial interval and DRO periods; the mean CRT of CR; and activity during 30 trials. The compounds produced different profiles at each dose. 1) Fa- cilitative and disruptive effects on attentional processes occurred with changes in CRT alone. Scopolamine (0.1 mg/kg) and prazosin (0.3—1 mg/kg) prolonged the CRT, whereas methamphetamine (0.3 mg/kg) shortened the CRT. 2) Attentional deficits occurred with abnormal behavior showing premature response or perseverative behavior. Scopolamine (0.2—1 mg/kg), methamphetamine (3 mg/kg), D9-tetrahydrocannabinol (10 mg/kg), and MK-801 (0.1—0.3 mg/kg) produced a marked increase in the number of total lever pressings. 3) Motor function deficits rather than attentional deficits occurred. 8-OH DPAT (1 mg/kg) and muscimol (1 mg/kg) produced a decrease in CR and an increase in NR with a marked decrease in activity and prolonged the CRT. Activating nora-

drenergic a1 receptors was found to enhance the attentional processes, while blocking muscarinic receptors, a1 receptors, and NMDA receptors, and stimulating cannabinoid receptors and the dopaminergic systems impaired the attentional processes in the two-lever CRT task. Key words attention; choice reaction time task; D 9-tetrahydrocannabinol (THC); scopolamine; methamphetamine; MK-801

Information processes through the learning and memory a variety of neurons and neurotransmitters have thus been system consist of three steps: attention (arousal, decision- implicated in attentional processes in the ﬁve-choice serial making, motivation, etc.); short-term memory (working mem- reaction time task, the pharmacological characterization in ory); and long-term memory (reference memory).1) Attention the attentional processes of a two lever CRT task is not un- is used in experimental and human clinical studies to mean derstood well. the ability to perceive or focus on certain stimuli, but to ig- The objective of the present study was to investigate the nore others in the environment. When people and animals are pharmacological characterization in the attentional processes confronted with a stimulus in the environment, they must of a two-lever CRT task using 10 different centrally acting usually decide whether they need to remember it and they drugs: scopolamine, a nonselective muscarinic receptor an- memorize it as a part of two memory systems (short-term tagonist; methamphetamine, a psychostimulant; D 9-tetrahy- memory and long-term memory) through attentional drocannabinol (THC), the major psychoactive component of processes if necessary. Thus most investigators have found marijuana; prazosin, a noradrenergic a 1 receptor antagonist; that this mechanism plays an important role in information 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH DPAT), a se- processing and memory processing. lective 5-HT1A receptor agonist; muscimol, a GABAA recep- A choice reaction time (CRT) task has been regarded as a tor agonist; dizocilpine (MK-801), an N-methyl-D-aspartate good measure in experimental and human clinical studies on (NMDA) receptor antagonist; 6-(1H-imidazol-1-yl)-7-nitro- cognitive performance capability (attention, arousal, deci- 2,3-(1H,4H)-quinoxalinedione hydrochloride (YM90K), an sion-making, motivation, etc.), especially visual attention.2—4) alpha-amino-3-hydroxy-5-methyl-4-isoxazoleproprionic acid The pioneering work of the group of Robbins in attentional (AMPA) receptor antagonist; N-nitro-L-arginine methyl ester processes using a ﬁve-choice serial reaction time task has (L-NAME), a nitric oxide (NO) synthase inhibitor; and nil- shown that cholinergic innervation of the forebrain is impor- vadipine, an L-type Ca2ϩ channel blocker. tant in attentional processes, noradrenergic innervation per- mits accurate performance during arousing situations, MATERIALS AND METHODS dopaminergic innervation of the nucleus accumbens is involved in behavioral activation, and forebrain serotonin (5- Animals Male Wistar rats weighing 200—250 g were HT) is involved in the control of impulsive behavior.5—10) We obtained from the Kyu-Do Co., Ltd. (Saga, Japan), and were also reported that ischemic-hypoxic insult to neonatal rats housed in groups of 4 to 5 per cage, in a room with a con- and methamphetamine produced severe impairment of atten- trolled temperature of 23Ϯ2 °C and a relative humidity of tion in a two-lever correct response (CR) task.11,12) Although 60Ϯ10% with the lights on from 07:00 to 19:00. The animals

* To whom correspondence should be addressed. e-mail: [email protected] © 2002 Pharmaceutical Society of Japan December 2002 1571 were placed on a restricted diet (10—12 g/d, CE-2, Crea, presses is known to increase during this period. Then the Japan). Their body weights were maintained at approxi- CRT period (maximum 8 s) was started, and a cue light above mately 80% of the free feeding level during the experimental one of the levers was illuminated. A single pressing of either period. Water was freely available in their home cages. All lever turned the cue light off, the house light was turned on, procedures regarding animal care and use were carried out and the ITI period of 20 s occurred. If the lever beneath the based on the regulations dictated by the Experimental Ani- cue light onset was pressed CR, a food pellet was delivered mal Care and Use Committee of the Center for Experimental in the tray located midway between the levers, the house light Animals in Fukuoka University. was turned on, and the ITI period occurred. The duration be- Apparatus The apparatus and CRT task procedure tween the cue light onset and lever pressing was defined as (Neuroscience Inc., Tokyo, Japan) were basically identical to the CRT. If the other lever was pressed (error response [ER]), that described previously.11,12) Two Skinner boxes (31ϫ24ϫ no food was received, the house light was illuminated, and 28 cm, Neuroscience Inc.) with two levers, which were inter- the ITI period began for 2 s. If rats did not press levers (non- faced with two SUB-CPU interfaces (Neuroscience Inc.) response [NR]), the cue light was illuminated for a maximum were used. They were controlled by personal computers (PC- of 8 s and the trial ended. The position of the cue light was 9801RS, NEC, Tokyo, Japan) for experimental events and randomly determined, except that there were no more than collected data. Two levers were mounted on one wall of each five successive trials with the light in the same position, and box at a height of 5 cm from the bottom grid. Located be- within a session the total number of trials with each cue light tween the two levers was a food tray, into which single 45-mg onset was 50—50. To examine how the change in response (Bio-serv, Frenchtown, NJ, U.S.A.) food pellets were deliv- rate induced by test compounds affected the CRT perfor- ered as reinforcers. The box was illuminated by a 3-W house mance, activity was measured every 1 min during 30 trials light located in the center above the frontal panel, which was using an automated activity box on each Skinner box. A daily equipped with two levers and a food tray. Above each lever, session consisted of 30 trials, and the total period for the three cue lights were aligned horizontally. An automated ac- training was approximately 2 months. tivity box (NS-AS01, Neuroscience Inc.) for measuring ac- Parameters Measured In each session, the following tivity was placed on each Skinner box in a dark, soundproof behavioral measurements were derived for each rat: 1) CR, compartment (70ϫ60ϫ60 cm) with a ventilator fan provid- ER, NR, and DRO response rates in 30 trials; 2) number of ing white noise. The counter for activity has a pyroelectric incorrect lever pressings during both the ITI and DRO peri- sensor that generates voltage-change by changing the amount ods; 3) mean CRT of CR in a 100-ms unit (duration between of infrared radiation from the sensor window. The infrared cue light and pressing correct lever); and 4) activity during radiation amount depends on: 1) the volume of the object 30 trials. under observation; 2) temperature of the object; 3) distance Drug Testing After the rats had been trained for 2 between the sensor and the object; and 4) direction and ve- months, the stable baseline of CRT performance was estab- locity of the object. The voltage change becomes larger when lished. If the rats achieved at least an 80% CR rate, fewer the object is large, has high temperature, and moves quickly. than 20 lever presses for the number of total lever presses We digitized this voltage change of specific thresholds using during both the ITI and DRO periods, and about 1.30 s for a personal computer and measured the activity of each ani- CRT in three successive sessions, vehicle was administered mal. The software was programmed by Neuroscience. for the first time. If the above criteria were met in that ses- Training The experiments were usually begun at 08:30 sion, the effects of drugs on the CRT performance were ex- and extended over a period of 8 h. At the end of the 2-week amined the following day. From the next drug treatment, the food restriction period, rats were placed inside the experi- drug testing schedule was vehicle on the first day and test mental compartment. For shaping, they were trained to press compounds on the second day, and was conducted once a levers for a food pellet with a continuous reinforcement week. Test compounds were injected only in rats which met schedule of a fixed ratio 1 (FR1). At this stage, both cue the same criteria as during the vehicle treatment phase. lights were illuminated and both levers were effective. For a Drugs Methamphetamine and prazosin were obtained period of 1—2 weeks, the animals were trained on the same from Dainippon Pharmaceutical Co. (Osaka, Japan) and schedule until they were able to receive 100 pellets within a Tokyokasei Co. (Tokyo, Japan), respectively. Scopolamine limited session of 60 min. and L-NAME were purchased from Sigma Co. (Poole, When rats achieved the above criterion, the CRT perfor- Dorset, U.K.). 8-OH DPAT, muscimol, and MK801 were pur- mance training was started. One trial consisted of 3 periods, chased from Research Biochemicals, Inc., Natick, MA, the differential reinforcement of another behavior (DRO) pe- U.S.A. YM90K and nilvadipine were obtained from Ya- riod, the CRT period, and an intertrial interval (ITI) period. A manouchi Pharmaceutical Co. (Tokyo, Japan) and Fujisawa trial began with the DRO period. Prior to cue light presenta- Pharmaceutical Co. (Osaka, Japan), respectively. Metham- tion, rats had to refrain from pressing either lever randomly phetamine, scopolamine, 8-OH DPAT, muscimol, MK801, for between 2 and 5 s. If rats pressed either lever during this and L-NAME were diluted with 0.9% saline. Prazosin, period (premature response), the DRO period was restarted. YM90K, and nilvadipine were suspended in 0.5% CMC and When rats repeated pressing levers during this period (DRO polyethylene glycol : ethanol : distilled water 1:1:2 (PED), response), the cue light was not illuminated and the trial respectively. THC was isolated from cannabis by Professor ended. This period was designed to prevent premature re- Y. Shoyama (Department of Pharmacognosy, Graduate sponses and to discriminate the action of psychostimulant School of Pharmaceutical Sciences, Kyushu University) and drugs. For example, when d-amphetamine, methampheta- emulsified in a 1% Tween 80 solution. All injections were mine, and cocaine are administered, the number of lever given in a volume of 1.0 ml/kg. All drugs except 8-OH DPAT 1572 Vol. 25, No. 12 and THC were injected intraperitoneally 30 min prior to testing, while 8-OH DPAT was injected 15 min and THC 60 min before testing. Data Analysis The results are expressed as meanϮ S.E.M. Significant differences were evaluated using one-way analysis of variance (ANOVA) followed by Dunnett’s test for parametric multiple comparisons. Differences were consid- ered to be significant with a probability of less than 0.05.

RESULTS

Acquisition Curves of CRT Performance in Intact Rats The acquisition curves of CRT performance are shown in Fig. 1. After meeting the criterion for shaping, rats were trained for CRT performance. On the first day, CR, ER, and NR rates were 24.6Ϯ1.9%, 20.9Ϯ2.4%, and 47.6Ϯ3.3%, respectively. The total number of lever presses during both periods was 107.0Ϯ7.7 and the mean CRT of CR was 2.5Ϯ0.2 on day 1. When rats had been trained for 1 month, CR was greater than 90% and the number of total lever presses was less than 40, and stable curves were observed from the first month. The mean CRT showed stable values from day 18 and was about 1.40 s from day 32. When rats had been trained for 1—2 months, stable values for all parameters were achieved in this task. The number of rats meeting the criteria for eval- uation of drugs described in “Methods” was 9 of 11 (81.8%) within about 40 trials. Effect of Scopolamine Scopolamine (Table 1) dose dependently produced a decrease in CR [F(4,31)ϭ7.217, pϽ0.001], an increase in ER [F(4,31)ϭ2.811, pϽ0.05], NR [F(4,31)ϭ3.601, pϽ0.05], CRT [F(4,31)ϭ6.132, pϽ0.001], and the total number of lever presses [F(4,31)ϭ4.888, pϽ0.01]. Scopolamine at dose of 0.1 mg/kg significantly prolonged only CRT (pϽ0.05). At a dose of 0.2 mg/kg, scopolamine significantly decreased CR (pϽ0.01) and increased CRT (pϽ0.01) and the total number of lever presses (pϽ0.01). At the higher dose of 1 mg/kg, it significantly de- Fig. 1. Acquisition Curves of Seven Parameters of a Two-Lever CRT Task creased CR (pϽ0.01), and increased NR (pϽ0.05), CRT in Intact Rats (pϽ0.01), and the total number of lever presses (pϽ0.05). No significant difference between vehicle and each dose of 4.905, pϽ0.01], and decreased activity [F(4,37)ϭ3.885, scopolamine was observed in the ER rate. pϽ0.01]. THC at a dose of 10 mg/kg significantly decreased Effect of Methamphetamine Methamphetamine (Table CR (pϽ0.05) and activity (pϽ0.01) and increased NR 1) dose dependently produced a decrease in CR [F(4,35)ϭ (pϽ0.05) and CRT (pϽ0.01). A significant difference in total 8.885, pϽ0.001], an increase in the DRO response [F(4,35)ϭ lever presses was also observed [F(4,37)ϭ3.326, pϽ0.05]. 9.732, pϽ0.001], total lever presses [F(4,35)ϭ13.932, THC at lower doses (2—6 mg/kg) tended to increase the total pϽ0.001], and activity [F(4,35)ϭ14.799, pϽ0.001]. A sig- number of lever presses and at a dose of 10 mg/kg signifi- nificant difference for CRT was also observed [F(4,35)ϭ cantly increased the total number of lever presses (pϽ0.01). 2.908, pϽ0.05]. Methamphetamine at a dose of 0.3 mg/kg Effect of Prazosin Prazosin (Table 1) dose dependently significantly shortened CRT (pϽ0.05). At a dose of 1 mg/kg, produced an increase in CRT [F(4,25)ϭ7.412, pϽ0.001]. methamphetamine showed increased activity (pϽ0.01). At Prazosin at doses of 0.3 and 1 mg/kg significantly increased the higher dose of 3 mg/kg, methamphetamine decreased CR CRT (pϽ0.01). (pϽ0.01) and markedly increased total lever presses (pϽ Effect of 8-OH DPAT 8-OH DPAT (Table 2) produced a 0.01) and activity (pϽ0.01). Moreover, methamphetamine at decrease in CR [F(2,20)ϭ4.104, pϽ0.05] and activity this dose increased the DRO response by 13.3Ϯ4.7% (pϽ [F(2,20)ϭ9.634, pϽ0.01], an increase in NR [F(2,20)ϭ 0.01), suggesting a premature response as described in the 3.766, pϽ0.05], and prolongation in CRT [F(2,19)ϭ5.876, Methods. Methamphetamine was thus found to produce a pϽ0.05]. 8-OH DPAT at a dose of 3 mg/kg significantly de- biphasic effects on CR performance: a facilitative effect at a creased CR (pϽ0.05) and activity (pϽ0.01) and markedly in- low dose and conversely a disruptive effect at a high dose. creased NR (pϽ0.05) and prolonged the CRT (pϽ0.01). Effect of THC THC (Table 1) dose dependently produced Effect of Muscimol Muscimol (Table 2) produced a de- a decrease in CR [F(4,37)ϭ2.552, pϽ0.05], an increase in crease in CR [F(3,18)ϭ1042.04, pϽ0.001] and activity NR [F(4,37)ϭ3.335, pϽ0.05], prolonged CRT [F(4,37)ϭ [F(3,18)ϭ44.698, pϽ0.001], an increase in NR [F(3,18)ϭ December 2002 1573

Table 1. Effect of Scopolamine, Prazosin, Methamphetamine, THC, and MK-801 on CRT Task in Rats

Agent No. of rats CR (%) ER (%) NR (%) CRT (s) Total lever pressings Activity

Scopolamine Vehicle 8 97.9Ϯ0.9 1.3Ϯ0.9 0.4Ϯ0.4 1.1Ϯ0.2 4.5Ϯ1.8 564.4Ϯ21.2 0.03 mg/kg 5 98.7Ϯ0.8 0.0Ϯ0.0 0.7Ϯ0.7 1.2Ϯ0.1 10.2Ϯ2.5 529.8Ϯ32.1 0.1 mg/kg 6 98.4Ϯ0.7 1.1Ϯ0.7 0.6Ϯ0.6 1.7Ϯ0.1* 8.0Ϯ1.7 528.3Ϯ26.3 0.2 mg/kg 9 88.5Ϯ1.8** 5.1Ϯ1.7 5.9Ϯ2.2 1.8Ϯ0.2** 21.2Ϯ1.9** 594.8Ϯ19.6 1 mg/kg 8 88.7Ϯ3.2** 4.2Ϯ1.5 6.7Ϯ2.2* 1.8Ϯ0.1** 18.8Ϯ5.9* 544.5Ϯ18.2 Prazosin Vehicle 5 96.0Ϯ0.6 2.6Ϯ0.7 0.7Ϯ0.7 1.2Ϯ0.1 7.2Ϯ2.2 528.8Ϯ28.0 0.03 mg/kg 6 96.1Ϯ3.9 0.6Ϯ0.6 4.4Ϯ3.8 1.3Ϯ0.1 3.5Ϯ1.3 497.5Ϯ25.7 0.1 mg/kg 6 98.3Ϯ1.1 0.6Ϯ0.6 1.1Ϯ0.7 1.6Ϯ0.1 3.5Ϯ0.6 515.3Ϯ77.3 0.3 mg/kg 6 96.1Ϯ1.0 0.6Ϯ0.6 2.8Ϯ1.0 1.8Ϯ0.1** 4.3Ϯ1.2 421.2Ϯ26.6 1 mg/kg 7 92.9Ϯ3.0 1.4Ϯ0.7 6.7Ϯ2.6 1.9Ϯ0.1** 2.4Ϯ1.2 376.7Ϯ30.1 Methamphetamine Vehicle 10 97.4Ϯ1.0 2.0Ϯ1.0 0.3Ϯ0.3 1.3Ϯ0.2 7.0Ϯ1.7 569.5Ϯ25.3 0.1 mg/kg 8 97.1Ϯ1.2 2.5Ϯ1.2 0.0Ϯ0.0 1.1Ϯ0.1 8.8Ϯ1.6 594.8Ϯ20.4 0.3 mg/kg 8 97.1Ϯ1.7 1.7Ϯ1.3 1.3Ϯ0.9 0.9Ϯ0.1* 8.3Ϯ2.1 644.6Ϯ15.1 1 mg/kg 7 96.7Ϯ1.0 1.4Ϯ1.0 0.5Ϯ0.5 1.1Ϯ0.1 25.1Ϯ13.8 751.3Ϯ25.3** 3 mg/kg 7 80.0Ϯ5.5** 4.3Ϯ1.0 2.4Ϯ1.0 1.3Ϯ0.1 171.4Ϯ45.0** 832.3Ϯ50.1** THC Vehicle 10 96.3Ϯ0.9 2.0Ϯ0.9 1.3Ϯ0.7 1.0Ϯ0.1 7.0Ϯ0.7 581.4Ϯ16.6 2 mg/kg 9 97.0Ϯ0.9 0.7Ϯ0.5 0.4Ϯ0.4 1.0Ϯ0.1 17.9Ϯ4.0 548.9Ϯ23.0 4 mg/kg 9 92.6Ϯ4.2 3.7Ϯ2.2 3.0Ϯ2.1 1.1Ϯ0.2 22.4Ϯ8.2 501.7Ϯ58.0 6 mg/kg 8 90.4Ϯ2.6 3.3Ϯ1.5 4.6Ϯ1.5 1.4Ϯ0.1 18.4Ϯ4.5 433.8Ϯ65.3 10 mg/kg 7 83.9Ϯ6.1* 3.3Ϯ1.7 11.1Ϯ5.3* 1.7Ϯ0.2** 32.8Ϯ3.4** 332.3Ϯ69.8** MK801 Vehicle 7 96.7Ϯ1.6 0.5Ϯ0.5 2.4Ϯ1.4 1.3Ϯ0.1 13.6Ϯ2.1 648.4Ϯ26.5 0.01 mg/kg 7 93.8Ϯ1.3 3.3Ϯ1.3 1.4Ϯ0.7 1.2Ϯ0.1 17.9Ϯ3.3 682.1Ϯ24.7 0.03 mg/kg 7 96.7Ϯ1.5 1.9Ϯ1.0 1.0Ϯ1.0 1.2Ϯ0.2 7.0Ϯ1.7 600.7Ϯ44.1 0.1 mg/kg 7 87.1Ϯ3.4 3.3Ϯ1.0 1.4Ϯ1.0 0.9Ϯ0.1 54.4Ϯ7.5* 755.3Ϯ33.0 0.3 mg/kg 7 78.1Ϯ7.1** 6.7Ϯ2.2** 7.1Ϯ4.0 1.5Ϯ0.3 66.1Ϯ19.3** 840.3Ϯ29.3**

All drugs except THC were injected i.p. 30 min prior to testing and THC was injected i.p. 60 min prior to testing. The values are expressed as meanϮS.E.M. * pϽ0.05, ** pϽ0.01 vs. vehicle (Dunnett’s test).

Table 2. Effects of 8-OH DPAT and Muscimol on CRT Task in Rats

No. of rats CR (%) ER (%) NR (%) CRT (s) Total lever pressings Activity

8-OH DPAT Vehicle 9 97.8Ϯ0.8 1.1Ϯ0.8 0.7Ϯ0.5 1.2Ϯ0.2 9.6Ϯ2.4 583.6Ϯ25.1 0.3 mg/kg 6 96.1Ϯ1.8 3.3Ϯ1.7 0.6Ϯ0.6 1.4Ϯ0.2 6.2Ϯ1.8 518.3Ϯ44.3 1 mg/kg 8 66.3Ϯ14.6* 2.1Ϯ1.1 31.3Ϯ15.2* 2.5Ϯ0.5** 8.3Ϯ3.2 320.8Ϯ62.0** Muscimol Vehicle 6 99.5Ϯ0.6 0.2Ϯ0.2 0.0Ϯ0.0 1.3Ϯ0.2 6.7Ϯ1.6 598.0Ϯ22.5 0.3 mg/kg 6 96.1Ϯ1.6 0.3Ϯ0.2 0.8Ϯ0.5 1.3Ϯ0.2 7.2Ϯ1.4 604.0Ϯ19.1 1 mg/kg 4 99.2Ϯ0.8 0.0Ϯ0.0 0.8Ϯ0.8 1.4Ϯ0.1 6.5Ϯ3.0 644.5Ϯ26.4 3 mg/kg 6 3.3Ϯ2.1** 2.2Ϯ1.4 93.9Ϯ3.0** 3.3Ϯ1.2** 7.5Ϯ4.4 175.5Ϯ50.9**

All drugs were injected i.p. 30 min prior to testing. The values are expressed as meanϮS.E.M. * pϽ0.05, ** pϽ0.01 vs. vehicle (Dunnett’s test).

783.824, pϽ0.001], and prolongation in CRT [F(3,15)ϭ 8.1Ϯ4.5%, respectively, suggesting a premature response as 4.933, pϽ0.05]. Muscimol at a dose of 3 mg/kg significantly described in the Methods, although ANOVA did not show a decreased CR (pϽ0.01) and activity (pϽ0.01), and markedly significant difference for the DRO response. increased NR (pϽ0.01) and prolonged the CRT (pϽ0.01). Effects of YM90K, L-NAME, and Nilvadipine YM90K Effect of MK-801 MK-801 (Table 1) dose dependently produced a decrease in activity [F(2,16)ϭ7.125, pϽ0.01, produced a decrease in CR [F(4,30)ϭ4.658, pϽ0.01], and an Table 3]. YM90K at a dose of 10 mg/kg significantly de- increase in ER [F(4,30)ϭ3.058, pϽ0.05], the total number of creased activity (pϽ0.05), but did not affect other parame- lever presses [F(4,30)ϭ7.917, pϽ0.001], and activity ters. L-NAME and nilvadipine did not affect CRT perfor- [F(4,30)ϭ8.514, pϽ0.001]. MK-801 at a dose of 0.3 mg/kg mance (Table 3). significantly decreased CR (pϽ0.01) and increased ER (pϽ0.01), the total number of lever presses (pϽ0.01), and DISCUSSION activity (pϽ0.01). A significant increase in the total number of lever presses was also observed at a dose of 0.1 mg/kg A CRT task has been regarded as a good measure in exper- (pϽ0.05). Moreover, MK-801 at higher doses of 0.1 and 0.3 imental and human clinical studies of cognitive performance mg/kg increased the DRO response by 8.1Ϯ2.9% and capability (attention, arousal, decision-making, motivation, 1574 Vol. 25, No. 12

Table 3. Effect of YM-90K, L-NAME, and Nilvadipine on CRT Task in Rats

No. of rats CR (%) ER (%) NR (%) CRT (s) Total lever pressings Activity

YM90K Vehicle 5 98.7Ϯ0.8 0.7Ϯ0.7 0.7Ϯ0.7 1.1Ϯ0.1 6.6Ϯ0.7 626.2Ϯ27.4 1 mg/kg 6 94.5Ϯ2.2 3.9Ϯ1.6 0.6Ϯ0.6 1.3Ϯ0.1 14.3Ϯ4.9 626.0Ϯ27.8 10 mg/kg 8 96.7Ϯ1.3 1.7Ϯ0.6 1.7Ϯ0.9 1.5Ϯ0.2 6.9Ϯ1.7 455.9Ϯ44.8* L-NAME Vehicle 3 100.0Ϯ0.0 0.0Ϯ0.0 0.0Ϯ0.0 1.3Ϯ0.1 5.6Ϯ2.6 581.8Ϯ20.0 100 mg/kg 3 99.3Ϯ0.7 0.0Ϯ0.0 0.0Ϯ0.0 1.1Ϯ0.1 15.2Ϯ3.2 596.6Ϯ20.8 Nilvadipine Vehicle 5 96.7Ϯ0.0 3.3Ϯ0.0 2.0Ϯ2.0 1.3Ϯ0.2 7.2Ϯ2.2 528.4Ϯ62.0 10 mg/kg 5 98.7Ϯ0.8 0.7Ϯ0.7 0.7Ϯ0.7 1.3Ϯ0.1 6.0Ϯ0.7 494.0Ϯ30.7

All drugs were injected i.p. 30 min prior to testing. The values are expressed as meanϮS.E.M. * pϽ0.05, vs. vehicle (Dunnett‘s test). etc.), especially visual attention.2—4) A CRT, one of the para- deficits. 8-OH DPAT (1 mg/kg) and muscimol (1 mg/kg) pro- meters in the CRT task, has been used as a useful psychomet- duced a decrease in CR and an increase in NR with a marked ric index of mental performance in human clinical studies.2,4) decrease in activity and prolonged the CRT of CR. Both Experimental evidence that a functional relation between compounds were found to impair motor function rather than CRT shortening and the regulation of selective attention ex- attention. ists has been accumulating.3,6,11) On the other hand, the CRT In the present study, scopolamine (0.1 mg/kg) and prazosin has also been valuable in evaluating the role of basal ganglia (0.3—1 mg/kg) prolonged the CRT, whereas methampheta- in motor programming and response output, both in human mine (0.3 mg/kg) shortened it. Behavioral studies with exper- diseases such as Parkinson’s and Huntington’s and in animal imental animals demonstrated that some neurotoxin lesions studies involving experimental disruption of the neostriatum of the basal forebrain-cortical cholinergic system induced and its connections.13—16) In the present study, to examine in profound deficits of attentional processes in a five-choice ser- detail the effects of several centrally acting drugs on atten- ial reaction time task, and scopolamine, a muscarinic recep- tional processes, we designed seven parameters of the two- tor antagonist, but not mecamylamine, a nicotinic receptor lever CRT task, the CR, ER, NR, and DRO response rates, antagonist, impaired attentional processes.9,17—20) Thus these the number of incorrect lever presses during both the ITI and facts suggest that the cholinergic system, especially mus- DRO periods, the mean CRT of CR, and activity during 30 carinic receptors, may be essential for attention and arousal. trials. NR is generally disregarded in the visual discrimina- The low dose (0.3 mg/kg) of methamphetamine significantly tion task, although omissions are used in some tasks, because shortened the CRT, in agreement with the findings of a previ- a CR is calculated not by dividing the number of CR trials by ous study.6,11,21) It is known that this facilitative effect of the number of total trials, but by dividing it by the total num- methamphetamine on attentional processes is due to en- ber of response trials.7,8,17,18) In the present study, we calcu- hanced noradrenergic transmission.22) Moreover, pharmaco- lated each response rate by dividing the number of each four logical studies using a 2 receptor antagonists and lesion stud- response trials by the number of total trials, that is, 30 trials. ies suggest a role for the dorsal ascending noradrenergic pro- Therefore the present method had advantages allowing the jection in the processes of attentional processes.5,23,24) There- influence of various drugs and experimental treatments on fore additional studies were carried out using prazosin, an a 1 the performance of a two-lever CRT task to be evaluated by receptor antagonist, to gain further insight into the possible measuring NR and DRO response with activity during 30 tri- role of noradrenergic a 1 receptors in attentional processes in als. the two-lever CRT task. The results demonstrated that pra- The effects of several centrally acting drugs on attentional zosin prolonged the CRT alone without affecting the other processes were investigated using a two-lever CRT task. The parameters. These results suggest that cholinergic muscarinic test compounds produced different profiles of action at each receptors and noradorenergic a 1 receptors may be important dose in terms of the seven parameters which were classified in the modulation of attentional processes and the CRT-short- into the following three types: 1) Facilitative and disruptive ening effect of methamphetamine may be related to a 1 recep- effects on attentional processes with changes in CRT alone. tors. Scopolamine (0.1 mg/kg) and prazosin (0.3—1 mg/kg) pro- Scopolamine (0.2—1 mg/kg), methamphetamine (3 mg/ longed the CRT, whereas methamphetamine (0.3 mg/kg) kg), THC (10 mg/kg), and MK-801 (0.1—0.3 mg/kg) were shortened the CRT. 2) Attentional deficits with abnormal be- found to increase markedly the total number of lever presses havior showing a premature response or perseverative behav- as measured by the premature response or perseverative be- ior. Scopolamine (0.2—1 mg/kg), methamphetamine (3 mg/ havior, which was characteristically observed with psychos- kg), THC (10 mg/kg), and MK-801 (0.1—0.3 mg/kg) pro- timulant drugs. Similar observations of methamphetamine duced a decrease in CR and an increase in ER, NR, DRO re- have been reported in several previous studies investigating sponse, and CRT and changed activity. These doses of com- the effect of amphetamine on choice behavior.6,7) Further- pounds were interestingly found to increase markedly the more, it was reported that this abnormal behavior was sup- total number of lever presses, by repeatedly pressing levers, pressed by the dopaminergic antagonist alpha-flupenthixol as measured by the premature response or perseverative be- and aniracetam, which interacts with dopaminergic neurons havior. 3) Motor function deficits rather than attentional as discussed previously.7,11) These results suggest that the December 2002 1575 dopaminergic system might be related to attentional process of these compounds at which pharmacological effects were deficits with abnormal behavior such as a change in the num- demonstrated were investigated.40—43) The results demon- ber of lever presses. Interestingly, THC at all doses used (2— strated that these compounds were inactive in attentional 10 mg/kg) also increased the total number of lever presses in processes, indicating that the MK-801-induced deficits of at- the CRT task, although it was not significantly different from tentional processes may be induced by the secondary activa- controls. Almost all cognitive studies on THC using animals tion of the dopaminergic system through NMDA receptors. has focused on memory, such as short-term memory and In the present study, the two-lever CRT task was classified working memory.25,26) However, there have been few studies into the following three types: 1) facilitative and disruptive on the effects of THC on attention in nonhumans, despite effects on attention with changes only in the CRT; 2) atten- many human studies.27—29) The present results raised the pos- tional deficits with abnormal behavior showing a premature sibility that the attentional deficits may account for the work- response or perseverative behavior; and 3) motor function ing memory deficits, which is a characteristic of THC-in- deficits rather than attentional deficits. As only one type of duced deficits of cognitive function. MK-801 significantly in- centrally acting drug for each receptor and enzyme was used creased the total number of lever presses in agreement with in the present study, a further experiment is required to eval- other studies.30—32) For example, in a differential reinforce- uate the effects of these drugs on attentional processes using ment of a low-rate 15-s (DRL-15 s), MK-801 (0.1—0.2 each agonist and antagonist. In conclusion, activation of no- mg/kg) produced markedly higher rates of lever pressing, es- radrenergic a 1 receptors was found to enhance the attentional pecially during a period of interresponse times of 5 s or less, processes, while blocking muscarinic receptors, a 1 receptors, suggesting that NMDA receptor antagonists disrupt timing and NMDA receptors, and stimulating cannabinoid receptors behavior, leading to a large proportion of premature re- and the dopaminergic systems impaired the attentional sponses.30) Moreover, MK-801 (0.1 mg/kg) increased the re- processes in the two-lever CRT task. sponse rate of schedule-controlled behavior in rats.33) It is likely that the stimulatory effects of MK-801 may be medi- Acknowledgments This study was supported in part by ated through the activation of the central dopaminergic sys- Grants-in-Aid for Scientific Research (No. 12771472 and tem.31,32) The present study also demonstrated that the behav- No. 14572171) from the Ministry of Education, Science and ioral change induced by MK-801 resembled the action of the Culture of Japan, and a Research Grant (13A-3) for Nervous methamphetamine, suggesting the activation of dopaminergic and Mental Disorders from the Ministry of Health and Wel- system through NMDA receptors. Thus the attentional fare. We are also grateful to Miss A. Saita, Mr. I. Hirakawa, processes with abnormal behavior showing a premature re- Miss M. Nakashima, Mr. T. Kaneko, Mr. K. Sakou, Miss H. sponse or perseverative behavior may be mediated by mus- Okada, Miss S. Ohta, Mr. K. Funaishi, and Mr. Y. Kawabe carinic receptors, the dopaminergic system, cannabinoid re- for their excellent technical assistance and proofreading the ceptors, and NMDA receptors. manuscript.

8-OH DPAT, a 5-HT1A receptor agonist, and muscimol, a GABAA receptor agonist, produced motor function deficits REFERENCES rather than attentional deficits in the two-lever CRT task. It was reported that 8-OH DPAT causes a severe 5-HT behav- 1) Atkinson R. C., Shiffrin R. M., Sci. Am., 225, 82—90 (1971). ioral syndrome such as flat body posture and that muscimol 2) Sims N. R., Bowen D. M., Smith C. C. T., Flack R. H. A., Davison A. also produces profound behavioral changes such as sedation, N., Snowden J. S., Neary D., Lancet, 16, 333—335 (1980). muscle relaxation, and ataxia.34,35) These behavioral changes 3) Grilly D. M., Gowans G. C., McCann D. S., Grogan T. W., Pharmacol. Biochem. Behav., 33, 733—739 (1989). cause motor function deficits making it difficult to assign the 4) Walker M. P., Ayre G. A., Cummings J. L., Wesnes K., McKeith I. G., effects observed to specific impairment of various task per- O’Brien J. T., Ballard C. G., Neurology, 54, 1616—1625 (2000). formances.36,37) In the present study, flat body posture was 5) Carli M., Robbins T. W., Evenden J. L., Everitt B. J., Behav. Brain observed at a dose of 1 mg/kg of 8-OH DPAT, and muscimol Res., 9, 361—380 (1983). 3 mg/kg produced sedation and ataxia. For this reason, both 6) Evenden J. L., Robbins T. W., Psychopharmacology, 85, 361—366 (1985). compounds markedly decreased activity during CRT perfor- 7) Col B. J., Robbins T. W., Psychopharmacology, 91, 458—46 (1987). mance, produced a decrease in CR with an increase in NR, 8) Muir J. L., Robbins T. W., Everitt B. J., Psychopharmacology, 107, and prolonged the CRT of CR. 8-OH DPAT and muscimol 541—550 (1992). thus affected the two-lever CRT performance alone. 9) Muir J. L., Everitt B. J., Robbins T. W., J. Neurosci., 14, 2313—2326 Long-term potentiation (LTP) has been widely used as a (1994). 10) Harrison A. A., Everitt B. J., Robbins T. W., Behav. Brain Res., 89, cellular model of learning and memory and there is consider- 135—149 (1997). able evidence of pre- and postsynaptic changes indicating 11) Himori N., Mishima K., Pharmacol. Biochem. Behav., 47, 219—225 that glutamate release is enhanced within minutes of stimula- (1994). tion, NMDA and AMPA receptors are upregulated, intracel- 12) Ikeda T., Mishima K., Yoshikawa T., Iwasaki K., Fujiwara M., Xia Yi. lular Ca2ϩ is evaluated through the Ca2ϩ channel, and Ca2ϩ- X., Ikenoue T., Behav. Brain Res., 118, 17—25 (2001). 13) Pullman S. L., Watts R. L., Juncos J. L., Chase T. N., Sanes J. N., Neu- dependent protein kinases, NO production, and the expres- rology, 38, 249—254 (1988). sion of the immediate-early genes c-fos and c-jun are acti- 14) Jahanshahi M., Brown R. G., Marsden C. D., J. Neurol. Neurosurg. vated.38,39) Therefore we chose YM90K, an AMPA receptor Psychiatry, 56, 1169—1177 (1993). antagonist, L-NAME, an NO synthase inhibitor, and nil- 15) Dobrossy M. D., Dunnett S. B., Behav. Brain Res., 87, 159—171 2ϩ (1997). vadipine, an L-type Ca channel blocker, to examine the ef- 16) Brasted P. J., Dobrossy M. D., Robbins T. W., Dunnett S. B., Brain Res. fects of intracellular signal processes from NMDA receptors Bull., 46, 487—493 (1998). on attentional processes in a two-lever CRT task. All doses 17) Andrews J. S., Grutzner M., Stephens D. N., Psychopharmacology, 1576 Vol. 25, No. 12

106, 523—530 (1992). rochem., 54, 1157—1162 (1990). 18) Ruotsalainen S., Miettinen R., MacDonald E., Koivisto E., Sirvio J., 32) Verma A., Kulkarni S. K., Psycopharmacology, 107, 431—436 Psychopharmacology, 148, 111—123 (2000). (1992). 19) Muir J. L., Page K. J., Sirinathsinghji D. J., Robbins T. W., Everitt B. J., 33) Sanger D. J., Jackson A., J. Pharmacol. Exp. Ther., 248, 1215—1221 Behav. Brain Res., 57, 123—131 (1993). (1988). 20) Gill T. M., Sarter M., Givens B., J. Neurosci., 20, 4745—4757 (2000). 34) Tricklebank M. D., Forler C., Fozard J. R., Eur. J. Pharmacol., 106, 21) Steckler T., Sahgal A., Psychopharmacology, 118, 226—229 (1995). 271—282 (1984). 22) Grilly D. M., Gowans G. C., Psychopharmacology, 96, 73—80 (1988). 35) Gundlach A. L., Beart P. M., Psycopharmacology, 74, 71—73 (1981). 23) Devauges V., Sara S. J., Behav. Brain Res., 39, 19—28 (1990). 36) Grech D. M., Balster R. L., Psychopharmacology, 129, 339—347 24) Jakala P., Sirvio J., Riekkinen J. R. P., Haapalinna A., Riekkinen P., (1997). Pharmacol. Biochem. Behav., 42, 903—907 (1992). 37) Nakamura K., Kurasawa M., Naunyn-Schmiedeberg Arch. Pharmacol., 25) Hampson R. E., Deadwyler S. A., J. Neurosci., 20, 8932—8942 361, 521—528 (2000). (2000). 38) Luscher C., Nicoll R. A., Malenka R. C., Muller D., Nat. Neurosci., 3, 26) Sullivan J. M., Learning Memory, 7, 132—139 (2000). 545—550 (2000). 27) Presburger G., Robinson J. K., Behav. Brain Res., 99, 27—34 (1999). 39) Rose S. P. R., Learning Memory, 7, 1—17 (2000). 28) Ehrenreich H., Rinn T., Kunert H. J., Moeller M. R., Poser W., 40) Takakura S., Susumu T., Satoh H., Mori J., Shiino A., Handa J., Jpn. J. Schilling L., Gigerenzer G., Hoehe M. R., Psychopharmacology, 142, Pharmacol., 56, 547—550 (1991). 295—301 (1999). 41) Takakura S., Sogabe K., Satoh H., Mori J., Fujiwara T., Totsuka Z., 29) Marks D. F., MacAvoy M. G., Psychopharmacology, 99, 397—401 Tokuma Y., Kohsaka M., Neurosci. Lett., 141, 199—202 (1992). (1989). 42) Bannerman D. M., Chapman P. F., Kelly P. A. T., Butcher S. P., Morris 30) Sanger D. J., Behav. Pharmacol., 3, 593—600 (1992). R. G. M., J. Neurosci., 14, 7404—7414 (1994). 33) Sanger D. J., Jackson A., J. Pharmacol. Exp. Ther., 248, 1215—1221 43) Shimizu-Sasamata M., Kawasaki-Yatsugi S., Okada M., Sakamoto S., (1988). Yatsugi S., Togami J., Hatanaka K., Ohmori J., Koshiya K., Usuda S., 31) Rao T. S., Kim H. S., Lehmann J., Martin L. L., Wood P. L., J. Neu- Murase K., J. Pharmacol. Exp. Ther., 276, 84—92 (1996).