ELSEVIER Behavioural Brain Research 60 (1994) 211-215 BEHAVIOURAL BRAIN RESEARCH Short Communication Differential effects of lesions of the dorsomedial and dorsolateral caudate-putamen on reaction time performance in rats Wolfgang Hauber b'*, Werner J. Schmidt ~' ~' Universio' ~?/' Tfibingen. Zoological Institute, Department 0[' Neuropharmacology, Mohlstr. 54/1, D-72074 Tiihin£,en. Germwo' h University 0[" Stuttgart. Biological Institute. Department Of Animal Physiology. Pfl~fJbnwaldring 57. D-70550 Stuttgart. Germum' (Received 16 June 1993: revised 5 November 1993: accepted 5 Novcmber 1993) Abstract In order to investigate the role of the dorsomedial and dorsolateral caudate-putamen (CPu) in movement initiation of rats, we ex- amined the effects of quinolinic acid lesions (30 nmol in 1 #1) in these striatal subregions in a simple reaction time task. Results show that lesions of the dorsomedial, but not of the dorsolateral CPu increased reaction times. These findings provide further evidence for a functional heterogenity of the CPu and demonstrate an involvement of the dorsomedial CPu in processes related to rapid initiation of responses. Key words: Reaction time; Locomotor initiation; Dorsomedial caudate-putamen; Dorsolateral caudate-putamen; Striatal heterogene- ity: Rat There is now considerable anatomical, biochemical and behavioural evidence for a functional differentiation of the striatum [8] and distinct cortico-striato-thalamo-cortical loops may exist which process different aspects of behav- iour [1]. These findings are primarily based on primate studies, however in the rat brain there may exist at least in part a similar organization. Several rodent studies showed that the medio-laterat organization of the caudate- putamen (CPu) is behaviourally relevant, since lesions of the lateral CPu specifically produced motor deficits [ 17,18] while lesions of the medial CPu impaired elayed alterna- tion and discrimination learning [7,10,18,22 ]. Lateral and medial CPu seem to mediate different behavioural func- tions and these functions are closely related to those me- diated by the anatomically inked regions of the cortex: the lateral CPu and the anatomically related sensorimotor cortical regions have been suggested to form a loop me- diating motor functions. In contrast, the medial CPu re- ceives prominent glutamatergic projections from the me- dial prefrontal cortex. This loop has been implicated in processing cognitive, i.e. complex, non-motor-functions [201. In the present stud}' we investigated the effects of lesions in the dorsomedia] and dorsolateral CPu on reaction time * Corresponding author. Fax: (49) (711)685-5096. performance of rats in a task demanding rapid locomotor initiation. According to the loop model it was expected that lesions in these subregions hould differentially affect reaction time performance. In fact it was recently shown that the relative contribution of the medial and lateral CPu to performance of a visual reaction time task is distinct [3]. We used lesion placements in the dorsomedial and dorsolateral CPu which are not exactly in the same anterio-posterior plane. According to anatomical studies the medial prefontal cortex projects primarily to the more rostral parts of the medial CPu, while the sensorimotor cortex mainly projects more caudally to the lateral CPu [2,14]. Therefore, the dorsomedial lesion was placed more rostrally than the lateral lesion. Male Sprague-Dawley rats (Interfauna, Tuttlingen, FRG) were used weighing 250-275 g at the beginning of the experiment. They were housed in groups of four or five animals with free access to water. Food was restricted to 12 g standard maintenance diet (Altromin, Lage, FRG) per animal and day. Rats were housed in a room main- tained at a constant emperature (22_+ 3 °C) on a 12:12 light-dark cycle (lights on 6.00 h). The experiments were perfumed in a modified runway made of transparent perspex as described previously [9,11]. Briefly,, the apparatus consisted or" a start box (18 x 9 x 9 cm) and a runway (100 x 9 x 9 cm) terminating 0166-4328/94,$7.00 © 1994 Elsevier Science Publishers B.V. All rights rescrved SSDI 016(~-4328(Y4)EO 163-U 212 H'. Hauher and H,'.,I. Schnddt, Behavimo'al Brain Research 60 f 1994, 2 t 1-2 ]- in a goal box (22 x 9 x 9 cm). The entrance to the runway was lockable by a remote controlled guillotine door situ- ated between start box and runway. The entrance to the runway was monitored by an infrared photocell beam (IDEC, FRG; resolution: < 10 ms) horizontally mounted directly behind the guillotine door. A combined stimulus light (10 W) and tone (8 kHz, 40 dB) signalled the simul- taneous opening of the front door (duration of the open- ing: < 80 ms). The stimulus lights, one on each side, and a loudspeaker were located in front of the start box. A control unit synchronized stimulus presentation and open- ing of the guillotine door. Control unit and photoelectronic switch were connected with an oscilloscope (ITT-metrix OX 750/2, Nttrnberg, FRG) and a printer for evaluating the latencies between stimulus presentation and photo- beam interruption. The rats were trained for rapid initiation of locomotion in response to the stimulus to receive a food reward (food pellet 45 mg; Noyes, Lancaster, UK): a food-deprived rat was placed into the start box facing the closed guillotine door blocking the entrance to the runway. After a varia- ble delay (3-10 s) the stimulus signalled the simultaneous £ *a 300 250 200 150 100 i11111111 111111111 pre post dorsomedial CPu & L 300 250 dorsolaterol CPu 200 150 I I I I t I I 100 pre post Fig. 1. Mean pre- and postoperat ive react ion times ( _+ S.E.M.) of ani- mals with quinotinic acid lesions in the dorsomedia l (N= 7. np,.~,,p = 270: npo~,, p = 270) or dorsolateral (N= 7: np,~,,p = 257: n~,o,~o p = 279) CPu. *P < 0.001 (Student's't-test). opening of the front door. A trained rat rapidly initiated locomotion, moved through the rtmway to the goal box and received the food reward in a baited cup. When the pellet was eaten or 10 s passed the rat was placed again into the start box for a new trial. The reaction time of locomotor initiation was defined as latency fi'om stimulus presentation up to photobeam inten'uption. The effect of an animals' prestart position relative to the guillotine door on reaction time nleasures was negligible because ,,f the narrow dimensions of the start box and the fact that trained rats always took a prestart position close to the guillotine door. First of all, rats were habituated individually to the baited apparatus for 10 min. From the following clay om animals were trained in one session per day. Each session comprised 10 successive trials. Aiier the rats had learned the task (10 correct rials, i.e. wiH~ reaction times between 100-1000 ms) the preoperative performance was tested in 4 sessions. Thereafter. animais received stereotaxic lesions. The postoperative p rformance was tested after a recovery period of 7 days. The reaction times tYom correct trials in pre- and postoperative s ssions were pooled and subjected to Student's t-test. The data are presented as means and standard error of the mean ( ± S.E.M.), Sur- gery was performed under sodium pentobarbitai naes- thesia (50 mg/kg., i.p.) with atropinsulfate pretreatment (0.25 mg/kg, i.p.; Serva, Heidelberg, FRG). Rats were placed in a Trent Wells stereotaxic nstrument.. Quinolinic acid (Sigma, Deissenhofen, FRG) (30 nmol in I ~1 ft. 1 M phosphate buffer, pH 7.0) or vehicle ( 1 ill 0.1 M phosphate buffer, pH 7.0) was infused via ;, 26 gauge cannula at a constant rate over 10 min. The cannula was left in place for 2 min after the infusion, l..esions in the dorsomedial or dorsolateral subregion of the CPu were made at the fol- lowing placements (with reference to bregma): dorsome- dial CPu, anterio-posterior: ~ 2 ram, dorsolateral 2.0 ram, ventral 5.1) lnm; dorsolateral CPu, anterio-posterior ~ 0.9 ram, dorsotateral 3.5 ram, ventral 5.5 mm (according to the atlas of Paxinos and Watson [16]). At the end of testing all animals were anaesthetized with sodium pen- tobarbital (50 mg/kg, i.p. I and transcardially perfused with 0.9'% saline followed by a 4", solution of phosphate buff ered formalin (pH 7.4). Brains were removed, postfixed and thereafter stored in 30",, .~ucrose solution. Coronal sections of 40 itm were cut with a cryostat (Reichert&J ung, NuBloch, FRG) and stained with Cresyl violet. The sec- tions were examined for qualitative assessment oi neu- ronal damage. The areas of neuronal oss IYom different anterio-posterior planes were mapped on brain sections according to the atlas of Paxinos and Watson [10]. Quinolinic acid lesions within thc dorsomedial CPu produced a small but significant increase of reaction time ( t - 3.27, d f - 538; P<0.001) (Fig. 1). In contrast, lesions of the dorsolateral CPu did not affect reaction time per- H:. Hauber and W.J. Schmidt / Behavioural Brain Research 60 (1994) 211-215 Table 1 Mean correct reaction times ( + S.E.M.) of animals with vehicle infusions into the dorsomedial or dorsolateral caudate-putamen 213 Site Preoperative (N, n) Postoperative (N, n) t-vahw P Medial 207.4 + 6.8 ms (5: 197) 211.4_+ 4.3 ms (5: 194) 0.50 0.62 Latcral 191.5 + 3.8 ms (4; 159) 199.7 _+ 4.0 ms (4: 155) 1.48 0.14 formance (t = 0.223, df= 534, P = 0.82) as shown in Fig. 1. Vehicle infusions also did not change reaction times post- operatively (Table 1). The histological examination re- vealed that quinolinic acid infusion produced a circum- scribed neuronal cell loss (Figs. 2 and 3). Within this area a moderate gliosis was found while ventricular alterations or a striatal shrinkage were not detected by gross inspec- tion. Furthermore the overlap of lesions within the dor- solateral and dorsomedial CPu was minimal. Control brains with vehicle infusions showed no degenerative al- terations except a gliosis along the canula track. In accordance with a previous report [3] these results demonstrate hat reaction time was affected selectively by lesions of the dorsomedial CPu. Quinolinic acid is a po- tent neuroexcitotoxin acting via N-methyl-D-aspartate re- ceptors. The dose of quinolinic acid used here was shown to be the treshold dose for producing consistent neuronal degeneration of a defined size without affecting fibres of passage [4,5]. Numerous neurochemical nd histochemi- cal studies showed that quinolinic acid predominantly 07U< 1.2 / 17 22 Fig. 2. Location and extcnl of quinolinic acid-induced lesions of the dorsolateral (right) and dorsomedial (left) caudate-putamen in different anterio-posterior planes of slriatmn (according to the atlas of Paxinos and Watson [16]). Abscissae are scaled in ram. Numbers represent the antcrio-posterior planes with reference to bregma. destroys medium spiny neurons with gamma-amino- butyric acid (GABA) as neurotransmitter l aving aspiny neurons with somatostatin unaffected [6]. Thercfore quinolinic acid-induced impairments een here are most probably due to the loss of striatal GABAergic projection neurons bearing NMDA receptors. In a recent study [3] rats with unilateral ibotenic a id lesions were tested in a reaction time task demanding head movements in response to visual stimuli. In this task rats with lesions of the medial CPu showed a significant slow- ing of initiation latency to a similar extent as observed here, while lesions of the lateral CPu had no effects on reaction time. The present data corroborate and extend these findings in that irrespective of the kind of movement and the type of neuroexcitotoxin used, and, despite of minor differences in the lesions placements and different lesion modes (bilateral versus unilateral), lesions of the dorsomedial CPu impaired rapid initiation of responses triggered by a stimulus. An involvement of the caudate nucleus and the putamen in externally and internally gen- erated processes which are related to the initiation of be- havioural acts was also shown in primate studies [ 15,21 ]. However, our data give no indications about the under- lying mechanisms resulting in an increase of reaction time. A disrupted attentional control as previously suggested [3] may also account for the slowing of reaction time seen here. Evidence for an impaired attentional control came, among others, from studies using maze tasks. For in- stance, pharmacological blockade of output neurons in the dorsomedial CPu impaired acquisition of delayed alterna- tion in a T-maze possibly due to attentional deficits [ 10]. This and numerous other findings are in line with a role of the medial CPu in processing complex behavioural functions. In accordance with the loop model it was fur- ther found that 6-hydroxydopamine lesions in the medial prefrontal cortex, the region which sends prominent pro- jections to the medial CPu, produced a marked increase of reaction times in the task used here [12]. Thus the cognitive cortico-striato-cortical loop may play a role in monitoring reaction time performance. On the other hand the failure to detect an involvement of the dorsolateral CPu in control of reaction time was unexpected in view of the motor functions of this striatal subarea [ 13 ]. However the detection of a reaction time impairment is dependent on the task and stimulus configuration used and other 214 W. Hauber and W.J. Schmicfl / Behavhmral Brain Research 60 r1994; 21t-2U Fig. 3. Micrographs showing an example of the quinolinic acid injection site in tile dorsolateral striatum. A: locatiotl ,>l'lhe injection as i dicated b3 the cannula track (magnification 8 x ). B shows at higher magnificatkm (32 x ) a part of the borderline between intact (left side) and damaged tissue (right side) illustrated in A. Note the numerous cell bodies stained with Cresyl violet on the lef side which are absev~ m the right side. factors as e.g. the dose of the neurotoxin. In fact in a re- cent stt~dy using a visual spatial discrimmination task it was suggested that the lateral striatum is the principle area concerned with control of reaction time although the me- dial CPu also plays a role [ 19]. It seems therefore that both striatal subareas contribute to intact reaction time performance. With regard to the different functional loops of which both subregions are part of, the underlying mechanisms disrupted after lesions in the dorsomedial and dorsolateral CPu leading to a slowing of reaction time are probably different, but they are not very well charac- terized. In summary the present study provides further evidence for a functional heterogenity of the CPu and demonstrates an involvement of the dorsomedial CPu in processes re- lated to the rapid initiation of responses. We thank E. Wacker and S.Schmidt for excellent tech- nical assistance. The study was supported by the Deut- sche Forschungsgemeinschaft (SFB 307/A4). References [ l] Alexander, G.E., DeLong, M.R. and Strick, P.I., Parallel organi- zation of functionally segregated circuits linking basal ganglia and cortex, lmm Rev. Neurosci.. 9(U)~6) 357-381. W. Hauber amt W.J. Schmidt Bekavioural Brain Research 60 (1994) 211-215 215 [2] Berendsc, H.W., Galis-de Graaf, Y. and Grocnwegen, H.J., Topographical organization and relationship with ventral striatal compartments of prefrontal corticostriatal projections in thc rat, J. Comp. Neurol., 316 (1992) 314-347. [3] Brown, V.J. and Robbins, T.W., Elementary processes of response selection mediated by distinct regions of the striatum, J. Neurosci.. 9 (1989) 3180-3186. [4] Churchill, L.. Dilts, R.P. and Kalivas, P.W., Changes in gamma- aminobutyric acid, /~-opioid and neurotensin receptors in the accumbens-pallidal projection after discrete quinolinic acid lesions in the nucleus accumbens, Brain Res., 511 (1990)41-54. [5] Davies, S.W. and Roberts, P.J.. Sparing of cholinergic neurons following quinolinic acid lesions in the rat striatum. Neuroscience, 26 (1988) 387-393. [6] DiFiglia, M., Excitotoxic injury of the neostriatum: a model for Huntington's disease, Trends Neurosci., 13 (1990) 286-289. [7] Divac. I., Markowitsch, H.J. and Pritzel, M., Behavioural and anatomical consequences of small intrastriatal injections f kainic acid in the rat, Bra#7 Res., I51 (1978) 523-532. [8] Goldman-Rakic, P.S. and Selemon, L.D., New frontiers in basal ganglia research, Trends ;v~,urosci., 13 (1990) 241-243. [ 9] Haubcr, W., A novel reaction time task for investigating force and time parameters of locomotor initiation in rats, Experient#*, 46 (1990) 1084-1087. [ 10] Hauber. W. and Schmidt, W.J., Effects ofintrastriatal b ockade of glutamatcrgic transmission on the acquisition ofT-maze and radial maze tasks, d. Neural. Transm., 78 (1989) 29-41. [ 11 ] Hauber, W. and Schmidt, W.J., The NMDA antagonist dizocilpine (MK-801) reverses haloperidol-induced movement initiation defi- cits, Behav. Bra#~ Res., 41 (1990) 161-166. [12] Hauber, W.. Bubscr, M. and Schmidt W.J., Cortico-striatal loops and reaction time perormancc in rats, Eur. J. Neurosci., 4 (1991) 167. [ 13] Hauber, W. and Schmidt, W.J., Discrete quinolinic acid lesions of the lateral but not of the medial caudate-putamen reversed haloperidol-induced catalepsy in rats, J. Neural. Transm., in press. [14] McGeorge, A.J. and Faull, R i .M . , The organization of the projection from the cerebral cortex to the striatum in the rat, Neu- roscience, 29 (1989) 503-537. [15] Montgomery, E.B. and Buchholz, S.R., The striatum and motor cortex in motor initiation and execution. Brain Res.. 549 (1991) 222-229. [ 16] Paxinos, G. and Watson, C., The Rat Braitl in Stereotaxic Coordi- nates, Academic Press, Sydney, 1986. [17] Pisa, M., Motor somatotopv in the striatum of rat: manipulation, biting and gait, Behav. Brain Res., 27 (1988) 21-35. [ 18] Pisa, M. and Cyr, J., Regionally. selective roles of rat's striatum in modality-specific discrimination learning and forelimb reaching, Beha~. Brain Re.s'., 37 (1990) 281-292. [ 19] Pretsell, D.O. and Robbins, T.W., Simple and choice reaction time performance following unilateral ibotenic acid lesions of dorsome- dial or dorsolateral striatum in the rat, Soc. :\'eurosci. A bstr., (1991) 483.7. [20] Robbins, T.W. and Brown, V.J., The role of the striatum in the mental chronometry of action: a theoretical revicw, Rev. Nearosci., 2 (1990) 181-213. [21] Schultz, W. and Romo, R., Neuronal actix.ity in the monkc 3 stria- tuna during initiation of movements, E.,cp. Brain Res., 71 (1988) 431-436. [22] Wiknlark, R.G.E., Divac, I. and Weiss, R., Retention of spatial delay.ed alternation in rats with lesions in the frontal lobes, Brain Behav. Evol., 8 (1973) 329-339.