Kainic acid: Enduring alterations in cerebellar morphology and in cerebral catecholamine and GABA concentrations after cerebellar injection in the rat

Neuroscience Letters,12 (1979) 339--342

339

© Elsevier/North-Holland Scientific Publishers Ltd.

K ~ I N I C ACID: E N D U R I N G A L T E R A T I O N S IN C E R E B E L L A R M O R P H O L O G Y AND IN C E R E B R A L CATECHOLAMINE AND GABA CONCENTRATIONS A F T E R C E R E B E L L A R INJECTION IN THE R A T

S T U A R T R. S N I D E R and R A Y S. S N I D E R Department of Neurology, College of Physicians and Surgeons of Columbia University, 630 W. 168 th,N. Y. 10032 and Center for Brain Research, University of Rochester Medical Center, Rochester, N.Y. 14642 (U.S.A.) (Received December 8th, 1978) (Accepted December 22nd, 1978)

SUMMARY

Kainic acid injected locally in the vermian cortex produces a focal lesion with severe cellular loss. Microscopic changes in 3-week-old preparations are given for both central and peripheral segments. Biochemical studies indicate that norepinephrine, dopamine and 7-aminobutyric acid (GABA) concentrations in the forebrain are consistently higher on the side of the lesion and remain elevated for at least 3 weeks. It is postulated that disinhibition of cerebellar activity traversing the uncrossed pathway from cerebellar nuclei to catecholamine cell bodies was a major mechanism causing increased catecholamine metabolism in the ipsilateral forebrain.

Previous research [ 5] has demonstrated a small fibered efferent pathway from the cerebellum to the locus coeruleus and ventral tegmental area. The present study is an extension of this work to determine some functional relationships with the catecholamine systems. Since Eccles and associates [1] have shown that Purkinje cells exert an inhibitory function on nuclear cells, it seemed appropriate to selectively eliminate cortical neurons and their localized inhibitory influences on nuclear cells. We postulated that the result might be an elevation of forebraln catecholamine concentrations due to release of nuclear activity from cortical inhibition. The use of kainic acid was proposed since this neurotoxin [3] acts on glutamate mechanisms and preferentially destroys glutamate-containing endings or glutamate-activated neurons. Purkinje (P), Golgi II (GII), basket (B) and steUate (S) cells are more vulnerable than granule (G) cells. These authors, following the studies of Hudson et al. [4], present evidence that granule cells exert a powerful excitatory glutamate-mediated input to other cortical neurons. These studies suggest dual effects on reduced cortical inhibition of nuclear cells (a) the removal of glutamate-mediated granule cell excitation of

340

Purkinie and basket cells and (b) direct toxicity on P, GII, B and S cells. Sprague--Dawley rats weighing 300 g were anesthetized and immobilized in a stereotaxic apparatus. Under direct visualization 1 ~g kainic acid in 1 ~g of saline was injected with stereotaxic guidance 1 mm deep into the right vermis and paravermis just anterior to the primary fissure. Control rats were sham-operated. Three weeks after surgery the rats were killed by decapitation and the brains were quickly removed from the calvarium, placed on ice and divided into right and left forebrain, and brainstem which included midbrain and cerebellum. The latter was used for light microscopic study. Forebrain halves were homogenized in 0.4 N perchloric acid and the resultant extract passed through a strong cation exchange column to separate 7-aminobutyric acid (GABA), norepinephrine and dopmmne. Standard fluonmetric assays were performed on the eluates as described in our previous study [6]. The light microscopic observations on the lesion site were made from serial paraffin e m b e d d e d sections stained either with 1% aqueous cresyl violet for Nissl studies or with gold-toned protargol-stained nerve fiber preparations[5]. Following focal injection of kainic acid and upon recovery from anesthesia a dramatic behavioral alteration appeared, consisting of (1) moderate to severe right scoliosis, (2) tight rotation to the right toward the lesion, (3) increased startle response and hyperactivity and (4) irregular myoclonic jerks, predominantly in the right limbs. This was usually accompanied by nystagmus to the right with a shift of the optical rest point to the left. This effect may be explained by the rapid onset of neurotoxic effects of kainic acid since Herndon and Coyle [3] reported cytological changes in P cells as early as one hour after local injection and describe them 'at 1 h it (chromatolysis) was clearly present in basket, stellate and Golgi II cells'. After 4--8 h the abnormal behavior disappeared, although rats showed a persisting preference for left-turning, e.g. when meeting an obstacle. In none of the animals reported was there evidence of kainic acid-induced damage to any part of the brain other than the local area of the injection. Therefore, we conclude that the alteration in m o t o r behavior was caused by the lesion. The rapid and nearly complete recovery is compatible with the well-known rapid physmlogm compensation for m o t o r deficits due to small cortical lesions. The lesion resulted in a right-left asymmetry of cerebral neurotransmitters, the concentratmns on the side ot~the lesion being higher (Fig. 1). This result is in agreement with data from recent experiments in which similar cerebellar cortical lesions were made surgically [6]. It is consistent with (but does n o t prove) the cerebellar disinhibition hypothesis presented above. Pharmacological evidence for the possibility that enhanced concentration of dopamine was due to increased dopaminergic activity was obtained by administering the dopamine agonist, apomorphine, intraperitoneally (2 mg/kg) at 1 week post-treatment. This caused pronounced turning to the right, indicating a relative supersensitivity of left sided dopamine receptors. The results suggest that functional compensation could be related in part to increased activity of the nigrostriatal dopamine system [ 7].

341

nm/g 2 . 4 - Norepinephrine

9.4

2.0 -

9.0

1.6

~ *

Dopomine

8.6

-

1.2-

nm/

~

_'9

pmlg 2.70 -

~'

2.

250~

8.2

T

2.40*-

o

?,

+i

0s-

GABA

7.8

o

2 30(,N

;;

I--I I--I LEFT RIGHT

r-~ LEFT

r~

RIGHT

¢

LEFT RIGHT

Fig. 1. Norepinephrine, dopamine and GABA concentrations in left and right forebrain three weeks following kainic acid injection of right cerebellum. Shown are the means + S.E.M. of 6 values. Control values for each side (numbers in bar) are means + S.E.M. of 5 values. StatmtlCS by 2-tailea t-test on matched pairs. Differences vs. control were not significant. *P < 0.02; **P < 0.01.

The histological studies were necessary to obtain information on the extent of the lesion as well as changes which occur in cell populations. Since Hemdon and Coyle [3] have given an accurate description of cellular changes occurring during the first week after a single injection of kainic acid, we will confine our description of changes to later stages with focus on the third week.

Fig. 2. Photomicrographs of cerebellar folium 3 weeks after local injection of kainic acid. M, molecular layer; P, Purkinje cell layer; G, granule cell layer; WM, white matter. A: junction of central (right) and peripheral (left) lesion (L) showing total loss of P cells and G cells at extreme right with considerable thinning of the cortex and white matter. Nissl stain × 150. B: peripheral lesion with degenerating fibers in molecular layer tentatively identified as climbing fibers. Ghost P and Golgi II cells can be found. G cells and mossy fibers are numerous. Gold tone protargol x 375.

342 Following a single dose (1 #g in 1 ~1) of kainic acid injected 1 mm deep into the cortex of the anterior lobe there was a thinning of all cortical layers and a total loss of all neurons in the central area of the lesion (Fig. 2A). In its severest form, only thin strands of connective tissue, glial fibers and a few capillaries remain. These fibers were separated by irregular-shaped vacuoles of varying sizes from 4 #m to 12/~m in the transverse dimension. Characteristically, total neuronal destruction occurred in an area 1.5--2.5 mm in diameter. At the edge of the lesion there were severe losses of P cells which extended for an additional millimeter. Cellular mitoses were n o t observed and neurophagia was rarely seen. Granule cells were reduced in numbers in fewer than 50% of the animals, but there were reduced numbers of basket cells within the 1 m m fringe. It was not possible to evaluate Golgi II and stellate cell loss on the basis of the Nissl stained tissues. Study of the silver stained fiber preparations of 3-week-old lesions (Fig. 2B) confirmed the above observations on severe loss of all cortical cells and fibers in the central area. In addition, it showed t h a t the peripheral part of the lesion which surrounded the central zone, had a well-organized granule cell layer with numerous small fibers resembling mossy fibers. Other fibers ascenaed into the molecular layer as small clusters between vacuoles which were the former sites of Purkinje cells. The distribution of m a n y of these fibers in the molecular layer indicated the presence of both climbing fibers and G cell axons in these clusters but clearly there were m a n y more fibers which could n o t be identified. Basket fibers and cells were n o t present in the central lesion and were usually also absent in the peripheral lesion where P cells had disappeared. Dark, shrunken B and P cells appeared more numerous at the margins of the peripheral lesions. These findings in older lesions are consistent with Herndon and Coyle's [3] observations t h a t fibers are more resistant than cell bodies to the toxic effects of kainic acid. The recent report of Hall et al. [2] questions the specificity of kainic acid binding to L-glutamic acid receptors but does not negate the postulate that lesions produced by these agents are the result of an 'excitotoxic' action. ~EFERENCES 1 Eccles, J., Operational features of the cerebellar cortex. In J. Eccles, M. Ito and J. Szentagothai (Eds.), The Cerebellum as a Neuronal Machine, Chap. XII, SpringerVerlag, New York, 1967, pp. 205--226. 2 Hall, J.G., Hicks, T.P. and McLennon, H., Kainic acid and the glutamate receptor,

Neurosci. Lett. 8 (1978) 171--175. 3 Herndon, R.M. and Coyle, J.T. Selective destruction of neurons by a transmitter, Science, 198 (1977) 71--72. 4 Hudson, D.B., Valcaca, T., Bean, G. and Tirniras, P.S. Glutamic acid: a strong candidate as the neurotransmitter of the cerebellar granule cell, Neurochem. Res., 1 (1976)

73--83. 5 Snider, R.S., Maiti, A. and Snider S.R., Cerebellar pathways to ventral midbrain and

nigra, Exp. Neurol., 53 (1976) 714--728. 6 Snider, S.R. and Snider, R.S., Alterations in forebrain catecholamine metabolism produced by cerebellar lesions in the rat, J. Neurol. Trans. 40 (1977) 115--128. 7 Snider, S.R., Snider, R.S., Glassgold, J., Tellerman, K., Astrow, A. and Fahn, S., Cerebellar control of catecholaminergic activities, Int. J. Neurol., (1978) in press.