Amphetamine-induced Fos expression in globus pallidus is altered by frontal cortex injury

BRAIN RESEARCH ELSEVIER

Brain Research7 16(I 996) 207-Z 12

Short communication

Amphetamine-induced

Fos expression in globus pallidus is altered by frontal cortex injury

Janet M. Vargo, Michelle Dcput-tment

of Psychobiology.

S. Estrick, John F. Marshall

Uninii ersity of Califomic~,

Accepted 27 December

Ircine,

CA 92717-4550,

* USA

199.5

Abstract

Functional recovery from cortical injury may result from subcortical compensatory processes. This study examined basal gangliar expression of the immediate early gene cYfr,s after unilateral medial agranular cortex (AGm) ablation. In the ipsilateral dorsal globus pallidus of rats demonstrating neglect of contralateral stimuli (sacrificed 5 days postinjury), the numbers of amphetamine-induced Fos-positive nuclei were reduced 37% compared to intact hemisphere values. These reductions were no longer apparent in recovered AGm-ablated rats (sacrificed 21 + days postinjury). These findings minor in timing and direction the changes in Fos seen in dorsolateral striatum after AGm ablation. Keywor-dstGlobus pallidus: Glutamate: Striatum; early gene

Fos; Recovery;

Frontal

Neglect, characterized by a lack of response to contralateral stimulation. is a potentially devastating clinical disorder often seen after injury to cortical brain regions following stroke. While patients typically show recovery from neglect over a matter of months. this recovery may be incomplete [IO]. Animal models of neglect resulting from cortical injury indicate that striatal dopaminergic function may play an important role in cortical neglect and its spontaneous recovery, and evidence gained from these models may lead to the eventual formulation of biologically based therapies for neglect in humans. Evidence linking dopamine (DA) transmission to cortical neglect includes the finding that DA agonists produce acute recovery [5,14], while DA antagonists reinstate neglect in rats spontaneously recovered [25,28]. Involvement of striatal receptor mechanisms in recovery from neglect is suggested by the finding that DA agonist-induced expression of Fos, the protein product of the immediate early gene (IEG) c-fos. and an indicator of cellular activity, has been found to change over time in conjunction with the behavioral state of the cortically lesioned rat. Five days after unilat-

* Corresponding [email protected]

author.

Fax:

0006-8993/96/$15.00 8 1996 Elsewer PII SOOOS-8993(96)000273

(1)

(714)

824-2447;

Science B.V.

E-mail:

All rights reserved

cortex:

Medial

agranular cortex; Dopamine: Neglect: Amphetamine; Immedmte

era1 aspiration of medial agranular cortex (AGm) [6], in rats still demonstrating neglect, decreases in DA agonistinduced Fos are seen in dorsolateral striatum. However, by 21 or more days after AGm injury, in rats recovered from neglect. Fos expression is normalized [27]. DA modulates activity within two distinct striatal projections: the striatonigral (direct) and striatopallidal (indirect) pathways (see [s]). Both types of cells receive afferents from cerebral cortex and they are partially denervated by AGm ablation. Under conditions of DA agonist administration, Fos is induced in striatonigral cells [ 19,201. Thus, the reductions in striatal DA agonist-induced Fos seen in rats showing neglect after AGm ablation suggest that striatonigral output is reversibly compromised by this injury. Since striatopallidal neurons express Fos in animals receiving DA antugonist, but not DA agonist, administration [20], the effects of AGm ablation on striatopallidal output under conditions of DA agonist treatment cannot be determined by direct examination of Fos within these cells. However. the effects of cortical injury on striatopallidal output might be indirectly assessed through quantification of DA agonist-induced Fos in globus pallidus (GP) itself. This study examines amphetamine-induced Fos in the GP of rats unrecovered or recovered from neglect produced by unilateral AGm ablation to determine the relationship be-

208

J.M. Vmy~ et d/Brain

Research 716 (19%~ 207-212

tween the behavioral state of the cortically lesioned rat and cellular function within this important output nucleus of striatum. Twenty-six Long Evans hooded rats. weighing 250-350 g at the start of the experiment were kept under a 12 h light/l2 h dark cycle. Food and water were available ad libitum. Surgical procedures were identical to those used previously 124,271. Subjects were anesthetized with ketamine/xylazine (SO/4 mg/kg, i.p.>, and received either left AGm ablation (n = 14), skull removal only ( IZ= 6), or incision only (n = 6). Since behavioral and immunocytochemical findings were similar between skull removal and incision only groups, their data were combined for comparisons to lesioned rats. Skull was removed over the left hemisphere from 4.0 mm anterior and 2.0 mm posterior to bregma, and from the sagittal suture to 2.5 mm laterally. In

0

lesioned animals gray matter was aspirated down to white matter. The testing procedures were identical to those reported in previous publications [5,24,27]. Briefly, degree of head turn toward visual, tactile, and auditory stimuli was determined based upon demarcations on the test platform. The maximum possible score for appropriate responding on each body side was 6.0 for each modality, 18.0 in total. Orientation tests were done blindly in regards to subject group. All subjects were tested on postoperative days 2 and 4. Long-term survivors were tested 3 times per week beginning in postoperative week 3 until recovery was demonstrated (see below). The orientation scores, summed across the three modalities tested, were used to derive the following neglect ratio: (contralateral total minus ipsilateral total)/(contralateral

lpsllateral (or Left) Body Side

m

Contralateral Body Side

Week

Postoperative

(or Right)

1

I IControl 1

*

i

ACIII

VISUAL

Postoperative

C

AGm

AUDITORY

TACTILE

Week

3-1

7T

I IdI ontr01

VISCAL

AGm

TACTILE

Control

AGn1

AIJDITOR~

Fig. I. Ipsilateral (left body aide) and contralateral (right body side) orientation scores of control and left AGm-ablated rats in the first postoperative week (Week 1) compared to scores obtained 3 or more weeks after surgery when recovery was demonstrated (Week 3 + ). Values represent means and S.E.M. Week I data represent 12 control and I3 AGm-ablated rats. Week 3 + data represent 7 control and 8 AGm-ablated rata. ’ Compared to ipsilateral scores of AGm-ablated group. t-values z 5.33, P < 0.001 for each modality; compared to contralateral scores of control group, ~-values > 2.10, P 5 0.047 for each modality.

total plus ipsilateral total) [24]. A neglect ratio of - 1.OO signifies a complete lack of response to contralateral stimulation. and a ratio of zero signifies symmetrical response to stimulation on either side of the body. An animal demonstrated severe neglect by obtaining a neglect ratio < -0.40. All AGm-ablated rats used in this study displayed severe neglect. An animal demonstrated recovery by obtaining a neglect ratio between 10.25 on two consecutive test days. Short-term surviving subjects were sacrificed on postoperative day 5. Long-term surviving subjects were sacrificed 21 or more days postoperatively, and only after meeting the recovery criteria. The mean postoperative survival time for this group was 50 days. After receiving d-amphetamine sulfate (5 mg/kg. i.p.> 2 h prior to perfusion, subjects were overdosed with sodium pentobarbital (65 mg), then transcardially perfused with 25-50 ml 4% paraformaldehyde in phosphate buffered saline (PBS). Brains were postfixed I h in 4% paraformaldehyde and then placed in 30% sucrose for at least I8 h. Immunocytochemical procedures were identical to those described in detail in Vargo and Marshall [27]. Forty Frn sections were incubated in polyclonal antibodies generated in rabbit

~Lesioned tlemrsphcre 5

5

E

IG

12::

-

_

(Left)

5 D;iys

Control

AGrn

DORSAL

C‘olltrol

AGni

DORSAL

Ulntact

against a synthetic peptide corresponding to amino acid residues 4- 17 of the human Fos protein (dilution 1:4000 from Santa Cruz Biotechnology or 1: IO00 from Oncogene Science). The secondary antibody was biotinylated goat anti-rabbit IgG (1:200). Immunoreactivity was visualized with an avidinbiotinylated peroxidase complex using 3,3’diaminobenzidine (DAB) as the chromogen. Fos-like immunoreactive (IR) nuclei were counted via a computer-assisted image analysis program. Four regions of GP (left and right hemispheres. dorsal and ventral regions) in each of three coronal sections posterior to the crossing of the anterior commissure and anterior to hippocampus were examined (see Fig. 2). The primary criterion used for targeting IR nuclei was optical density. The relative optical density (ROD) criterion was finely adjusted until an optimal match was obtained between computer-identified targets on the video screen and the microscopically verified IR nuclei [27]. Sections approximately 200 pm apart throughout the extent of the lesion were slide-mounted and stained with cresyl violet to determine lesion location and extent. Parametric statistical analyses were conducted with an alpha level of 0.05 adopted in all cases. For analyses of the

(Right) Hemisphere

Postoperative

Control

AGm

VENTRAL

Control

AGm

VENTRAL

Fig. 2. Number\ (per mm’) of amphetamine-induced Fos-like immunoreactive nuclei found in dorsal and ventral globus pallidua on postoperative day 5 (short-term survi\,al group) and on postoperative da>1 21 + (long-term survival group). There were 5 control and 6 ACin-ablated ratb (unrecovered) in the short-term group. and 7 control and 8 AGm-ablated rats (recovered) in the long-term group. Values represent mean? and S.E.M. D-amphetamine sulfate (5 mg/kg, i.p.) was given 2 h prior to perfusion. All AGm ablations were in the left hemisphere. The coronal sections (right side of the figure) show dorsal and ventral regions analyzed [ 171. ’ Compared to intact (right) hemisphere of AGm-ablated group, f = 3.17. P = 0.025; compared to left hemisphere of control group. t = 2.21, P = 0.055.

210

J.M. VWgo et al./ Brairt Research 716 (19961207-212

numbers of Fos-like IR nuclei, four separate ANOVAs were conducted. Data from animals sacrificed 5 days and 21 + days postsurgery were analyzed separately, as were data from dorsal and ventral GP regions. Main effects included subject group (control and lesioned) and hemisphere (left [or lesioned], and right [or intact]) for each analysis. In the first postoperative week, rats with left AGm ablation demonstrated impaired orientation to contralateral stimuli only (see Fig. 1). The mean (+S.E.M.) neglect ratio in the first postoperative week was -0.43 f 0.05 for lesioned animals and -0.04 i: 0.04 for controls. Lesioned animals demonstrated recovery from neglect 3-5 weeks postoperatively (see Fig. 1). Statistical analyses showed no significant differences in the extent of brain injury between AGm-ablated animals surviving 5 days or 21 + days after surgery. AGm ablation produced a 37% reduction in the numbers of Fos-like IR nuclei induced by amphetamine in dorsal GP. This lesion effect was restricted to ipsilateral GP and was evident only in rats still demonstrating neglect (5 days postoperative survival). In contrast, amphetamineinduced Fos in ipsilateral GP of recovered rats (21 + days postoperative survival) was found to be normal when compared to that seen either in the intact hemisphere or in unlesioned controls (Fig. 2). For rats surviving 5 days postoperatively. ANOVA of the numbers of amphetamine-induced Fos-like IR nuclei found in dorsal (but not ventral) GP revealed a significant hemisphere effect (F,,, = 5.96, P = 0.037) and hemisphere by group interaction (F,,, = 8.77, P = 0.016). Posthoc tests (see legend for Fig. 2) revealed that the AGm ablation reduced Fos expression only in ipsilateral dorsal GP. ANOVAs of the numbers of Fos-like IR nuclei in rats surviving 21 + days postoperatively revealed no significant effects involving lesion or hemisphere for either dorsal or ventral GP (P > 0.304, see Fig. 2). Overall, the numbers of Fos-like IR nuclei were greater for the animals (both lesioned and control) surviving 21 + days versus those surviving only 5 days postoperatively (see Fig. 2). This may be due to different antibody batches having been used for the two survival conditions. When the tissue from one AGm-ablated rat surviving 5 days was incubated in the antibody used for long-term survivors, the numbers of IR nuclei in dorsal GP were twice as high compared to those seen under the original short-term survivor incubation conditions. Despite this batch-specific effect on overall numbers of IR nuclei, the lesion effect in the short-term survivor was not batch-specific: compared to intact hemisphere values, dorsal GP Fos in the lesioned hemisphere was decreased 40% in the original incubation conditions and 39% in the long-term survivor incubation conditions. Thus, the lack of significant asymmetries in GP Fos expression in recovered (long-term surviving) rats was unlikely to have been the result of a ceiling effect due to intense reactivity. In addition, in the GP of AGm-ablated

rats, the degree of hemispheric asymmetry in the numbers of Fos-like IR nuclei was unrelated to the general intensity of immunoreactivity. Findings from this study indicate that frontal cortex injury produces alterations in pallidal DA agonist-induced Fos which are evident 5 days, but not 21 -t days, after surgery. Contralateral neglect produced by unilateral AGm ablation is accompanied by decreases in amphetamine-induced Fos expression in dorsal GP, while amphetamine-induced Fos expression is normalized weeks after AGm injury in the GP of rats recovered from neglect. Electrophysiological studies indicate that systemic administration of nonselective DA agonists increases the firing rates of many GP neurons [1,2.11]; therefore, GP Fos expression seen after amphetamine administration is likely the result of pallidal excitation. The current study’s finding that fewer GP cells express amphetamine-induced Fos after AGm injury suggests that dorsal pallidal function is initially compromised by this cortical damage, but compensatory processes occurring over a matter of weeks permit normalization of GP response to DA agonists. In a number of electrophysiological experiments, Walters and her colleagues have found that blocking glutamate receptor activation decreases pallidal response to DA agonists. They report that Type II GP neurons are activated by systemic apomorphine. and that this activation is less after pretreatment with the N-methyl-D-aspartate (NMDA) antagonist MK-801 [ 1l-1 31. MK-801 alone has little effect on the firing rates of the same pallidal cell population [ 121. Changes in pallidal firing rates following systemic apomorphine are likely to be mediated at least partly through striatum since they are attenuated by striatal lesions [ 161. Findings from the current study indicate that, like NMDA receptor blockade via MK-801 administration, loss of glutamatergic transmission from frontal cortex attenuates DA agonist activation of GP cells. While AGm has no known direct projections to GP. there are at least two indirect pathways through which injury to this cortical region could affect GP function. First, AGm sends a dense excitatory projection to dorsolatera1 caudate-putamen (CPU) [6:9], which in turn sends an inhibitory projection to dorsal GP [9]. Previous studies have indicated that AGm ablation affects cellular function in dorsolateral CPU. Constitutive as well as DA agoniststimulated immediate early gene expression is temporarily reduced in ipsilateral dorsolateral CPU following unilateral AGm injury [26,27]. S’mce DA agonists induce Fos in striatonigral (but not striatopallidal) cells, and this expression is dependent upon glutamatergic NMDA receptor activation [3,15,22.23], AGm ablation likely results in reduced DA agonist-induced Fos within striatonigral cells by decreasing corticostriatal glutamatergic transmission. Because striatopallidal cells are also deprived of glutamatergic input after AGm ablation, it might be argued that the decreases in amphetamine-induced Fos in dorsal GP of AGm-ablated rats found in the present study result indi-

rectly from compromised striatopallidal function. However, under the assumption that the pallidal Fos induced by systemic amphetamine occurs via striatal DA D,-receptormediated disinhibition, removal of excitatory corticostriatal glutamatergic input should augment this effect, leading to greater Fos expression in the lesioned hemisphere (see [ 1.51).In contrast, our current findings indicate that AGm ablation temporarily reduces amphetamine-stimulated Fos production in GP. Thus, other explanations for the effect of AGm ablation on GP Fos response to DA agonists should be considered. The cortico-striato-pallidal connection is not the only pathway linking AGm to GP. AGm sends an excitatory projection to caudal subthalamic nucleus (STN) [4], which then projects to dorsal GP [ 181. Under conditions of dopamine receptor activation, STN neurons increase their firing rates [7] and show Fos expression [21]. Possibly, reductions in excitatory input from cortex after AGm injury limit the extent of STN response to DA agonists, and subsequently limit STN excitatory output to GP under DA agonist conditions. However, further studies are needed to determine the time course and extent of alterations in subthalamic DA agonist activation after cortical injury before we can understand STN’s influence on GP Fos expression under DA-stimulated conditions. The current study provides novel information concerning the extent and time-course of subcortical cellular dysfunction after cortical injury. Previous studies have demonstrated that AGm injury produces deficits in DA agonistinduced Fos expression within striatonigral cells. and that this expression normalizes over time in conjunction with behavioral recovery. The current study indicates that DA agonist-induction of Fos is temporarily reduced by cortical injury not only in striatum, a primary efferent of AGm. but also in GP, a major target nucleus of striatum. The initial decreases in pallidal Fos expression occurring as a result of cortical injury are not easily explained by current models of basal ganglia function, and the mechanisms responsible for normalization of pallidal Fos in recovered AGm-lesioned rats are unknown. The present findings underscore the importance of understanding cortical interactions with the basal ganglia since they indicate that following cortical injury, dynamic changes involving both glutamate and DA occur, and that recovery of behavioral function may be associated with normalizations in subcortical neuronal processing.

References [II Bergstrom, D.A., Bromley, S.D. and Walterr. J.R.. Apomorphine increases the activity of rat globus pallidua neurons, Brcrin Rrs.. 238 (I 982) 266-27 I, [2] Bergstrom, D.A., Bromley, S.D. and Walters, J.R., Dopamine agonists increase pallidal unit activity: attenuation by agonist pretreatment and anesthesia. Eur. ,I. Phur-rnacol.. 100 (1984) 3- 12. [3] Bronstein. D.M., Ye, H.. Pennypacker, K.R.. Hudson, P.M. and

Hong, J.S., Role of a 35 kDa Fos-related antigen (FRA) in the long-term induction of striatal dynorphin expression in the 6-hydroxydopamine lesioned rat, Mol. Bruin Rec.. 23 (1993) I91 -203. [Al Canteras. N.S., Shammag-Lagnado, S.J., Silva, B.A. and Ricardo. J.A.. Afferent connections of the subthalamic nucleus: a combined retrograde and anterograde horseradish peroxidase study in the rat, B/-&I Res., 513 (IYYO) 43-59. bl Corwin, J.V.. Kanter. S., Watson, R.T., Heilman, K.M., Valenstein. E. and Hashimoto, A., Apomorphine has a therapeutic effect on neglect produced by unilateral dorhomedial prefrontal cortex lesions in rats. Ex/J. ~liruml., 94 (1986) 683-689. [61 Donoghue. J.P. and Wise. S.P., The motor cortex of the rat: cytoarchitecture and microstimulation mapping. J. Camp. Newel., 212 (I 982) 76-88. [71 Feger, J.. Robledo, P. and Retwart, X.. The wbthalamic nucleus: new data, new questions. In G. Bernardi. M.B. Carpenter. G. Di Chiara, M. Morelli and P. StanLione (Eds.). Bnsal GmLq/iu II/, Plenum Press, New York. 198’). pp. 99-108. [81 Gerfen. C.R., Dopamine receptor function m the basal ganglia, Clinicml 1veuro~karr,2nco/.. I8 (1995) S 162-S 177. [91 Groenewegen. H.J., Berendae, H.W., Woltera, J.G. and Lohman. A.H.M., The anatomical relationship of the prefrontal cortex with the striatopallidal system. the thalamus and the amygdala: evidence for a parallel organization. In H.B.M. Uylings. C.G. Van Eden. J.P.C. De Brum. M.A. Corner and M.G.P. Feenstra (Eda.). Progress h? Btuin R~.wurch, Elsevier, Amsterdam. 1990, pp. 95-l 18. [lOI Heilman. KM.. Watson. R.T. and Valenstein, E., Neglect and related disorders. In C/in&r/ ~lieumlo~~. Oxford University Press, New York, 1985. pp. 233-794. [I 11 Kelland. M.D. and Walters, J.R.. Apomorphine-induced changes in striatal and pallidal neuronal activity are modified by NMDA and muscarinic receptor blockade, L(fe Sci.. SO ( 1992) PL 179-l 84. [I?] Kelland. M.D.. Soltis, R.P., Boldry. R.C. and Walters. J.R., Behavioral and electrophysiological comparison of ketaminc to diaocilpine in the rat. P/z@/. BPALW.. 54 (1993) 547-554. [I31 Kelland, M.D., Soltis, R.P.. Anderson. L.A.. Bergstrom, D.A. and Walters. J.R., In viva characterization of twu cell types in the rat globus pallidus which have opposite responsea to dopamine receptor stimulation: comparison of electrophysiological properties and responses to apomorphine. dirocilpine, and ketamine anesthesia. Synupse. 20 (199% 338-350. [I41 King. V. and Corwin. J.V., Neglect following unilateral ablation of the caudal but not the rostra1 portion of medial agranular cortex of the rat and the therapeutic effect of apomorphine, Behap. Brain Rrs., 37 (1990) 169%184. [I51 Paul, M.L., Graybiel. A.M., David, J.-C. and Robertson, H.A.. Dl-like and DZ-like dopamine receptors synergistically activate rotation and c-fi1.7expression in the dopamine-depleted striatum in a rat model of Parkinson’s disease, J. Nwru,sc~i.. I2 (1992) 3729-3742. [I61 Pan, H.S.. Engber, TM., Chase. T.N. and Walters, J.R., The effects of striatal lesion on turning behavior and globus pallidus single unit response to dopamine agonist administration, L$i Sci., 46 (I 990) 73-80. [I71 Paxinos, G. and Watson. C., T/w Rat Bmin in Sfereofa.xic Coordnutes, Academic Preys Australia. North Ryde, 1986. [I81 Ricardo, J.A., Efferent connections of the subthalamic region in the rat. I. The subthalamic nucleus of Luys, Bmirz Res., 202 (1980) 2.57-271. [I91 Robertson, G.S.. Vincent. S.R. and Fibiger. H.C.. Striatonigral projection neurons contain D I dopamine receptor-activated c-fo.r, Brain Rrs.. 52.1 (I 990) 288%290. DO1 Robertson, G.S.. Vincent. S.R. and Flbiger. H.C.. DI and D? dopamine receptors differentially regulate c+s expression in striatonigral and striatopallidal neurons. Nrur-oscirncr, 4Y (1992) 28% 296. 1211 Ruskin, D.N. and -Marshall. J.F.. D, dopamine receptors influence Fos immunorcactivity in the globus palhdus and subthalamic nu-

212

J.M. Vargo et ~1./Brain

clew of intact and nigrostriatdl-lesioned rats. Bourn Res., 703 ( 1995) 156&161. [22] Snyder-Keller, A., Striatal c-fos induction by drugs and stress in neonatally dopamine-depleted rats given nigral transplants: importance of NMDA activation and relevance to sensitization phenomena. Ex[I. Nuurol., 113 (1991) 1555165. 1-31 Torres, G. and Rivier. C., Cocaine-induced expression of striatal c&s in the rat is inhibited by NMDA receptor antagonists. Brain Res. BdI.. 30 (1993) l73- 176. [24] Vargo. J.M.. Corwin. J.V., King, V. and Reep. R.L., Hemispheric asymmetry in neglect produced by unilateral lesions of dorsomedial prefrontal cortex in rats, Exp. Ne~ud.. I02 (19X8) 199-209. [25] Vargo, J.M., Richard, S.M. and Corwin, J.V., Spiroperldol reinstates asymmetries in neglecl in rats recovered from left or right dorsome-

Resrarch 716 (1996) 207-212 dial prefrontal cortex lesions. Be/w/,. ~V~ur~~.osci., 103 (1989) IO 171027. [26] Vargo, J.M. and Marshall, J.F., Frontal cortex ablation reversibly decreases striutal zif‘/268 and junB expression: temporal corw spondencc with sensory neglect and its spontaneous recovery, Swupse, (in press). [27] Vargo, J.M. and Marshall, J.F.. Time-dependent changes in dopamine agonist-induced striatal Fos immunoreactivity are related to sensory neglect and its recovery after unilateral prefrontal cortex injury. Synupse. 20 (1995) 305~315. [2X] Vargo. J.M., Bromberg, B.B., Best, P.J., Corwin, J.V. and Marshall. J.F., D,-class dopamine receptor involvement in the behaworal recovery from prefrontal cortical injury, B&w. Bruin Res.. 77 (1996) 39-38.