Chapter 25 Immunolesion by 192IgG-saporin of rat basal forebrain cholinergic system: a useful tool to produce cortical cholinergic dysfunction

J. Klein and K. Lijffelholz (Eds.) Progress in Brain Research, Vol. 109 0 1996 Elsevier Science B.V. All rights reserved.

CHAPTER 25

Immunolesion by 192IgG-saporin of rat basal forebrain cholinergic system: a useful tool to produce cortical cholinergic dysfunction Reinhard Schliebs, Steffen RoBner and Volker Bigl Paul Flechsig Institute f o r Brain Research, Medical Faculiy, University of Leipzig. 0-04109 Leipzig. Germany

Introduction The basal forebrain cholinergic system is known to play an important role in cortical arousal and normal cognitive function. Cortical cholinergic dysfunction has been implicated in cognitive deficits that occur in Alzheimer’s disease, and the cholinergic projection from the nucleus basalis of Meynert to areas of the cerebral cortex is the pathway that is earliest and severely affected in brains from Alzheimer patients (for review, see e.g. Bigl et al., 1989). In the rat this nucleus is not yet developed into a delineated nuclear structure but corresponds to a more heterogeneous region of cholinergic neurons which often is referred to as nucleus basalis magnocellularis (Nbm) providing the main source of cholinergic terminals to the cerebral cortex (Wenk et al., 1980; Bigl et al., 1982). The changes in markers of the neocortical cholinergic system found in brains of Alzheimer patients are complemented by alterations in other cortical transmitter systems like glutamate, GABA, noradrenaline or serotonin receptors (for reviews, see Nordberg, 1992; Carlson et al., 1993; Greenamyre and Maragos, 1993), suggesting (i) an important influence of the cholinergic basal forebrain system on cortical neurotransmission, and (ii) a cholinergic role in cortical reorganization and in adaptive processes following injury. This is consistent with the hypothesis that the basal forebrain cholinergic system is not directly involved in the formation of learning and memory but acts as a modul-

atory system to control cortical information processing. Characterization of the mechanisms underlying this adaptive response might be of particular importance (i) to elucidate the cascade of events initiated by decreased cortical cholinergic activity and (ii) to further derive rationales to pharmacologically intervene in this process with respect e.g. to find a therapeutic strategy to treat Alzheimer’s disease or to characterize the role of the cholinergic system in cortical information processing, learning and memory and in cognitive behaviour in greater detail. Such investigations on the functions of the central cholinergic system require adequate animal models to produce specific cholinergic deficits in vivo. This would allow for a detailed evaluation of the neurochemical, neuropathological, and behavioural sequela as well as functional implications of plastic repair mechanisms following cholinergic hypofunction, and provide information that cannot or only partially be obtained in humans. At present there is no adequate animal model available which could mimic all the biochemical, behavioural, and histopathological abnormalities as observed in patients with Alzheimer’s disease. However, partial success can be achieved with so called “isomorphic models” (Fisher and Hanin, 1986) representing partial parallelism between model and some human conditions. The value of such models is to delineate mechanisms underlying the pathological processes as well as to test for new potential thera-

254

peutic strategies. In the last decades a number of different paradigms have been introduced to produce cortical cholinergic dysfunction which are briefly discussed in the following chapter.

Paradigms to produce cholinergic lesions in the basal forebrain Mechanical lesion of the Nbm

Mechanical lesion of the Nbm (e.g. by radiofrequency or electrolysis) results in damage to all neural tissue at the lesion site including cell populations residing in the basal forebrain as well as passing (e.g. noradrenergic and dopaminergic) fibre bundles. Thus, besides a loss of cortical cholinergic input a considerable reduction of dopaminergic and noradrenergic innervation of the cortex as well as degeneration of non-cholinergic neurons which are present in varying proportions depending on the lesion site must be taken into account. Furthermore, if the size of the lesion is not kept small enough, it encroaches on the neighbouring nuclei and might damage, e.g. the globus pallidus, the nucleus caudate-putamen or hypothalamic nuclei and even the capsula interna which might also considerably affect the results obtained. Although the lesion is relatively non-specific for the cholinergic system destroying all the neural tissue at the lesion site including all passing fibres, the destruction of cholinergic cells is relatively massive as revealed by considerable decrease in cortical choline acetyltransferase (ChAT) and acetylcholinesterase (AChE). Lesion of basal forebrain nuclei by excitotoxins

Excitotoxins are conformationally restricted analogues of the excitatory amino acid neurotransmitter glutamate (e.g. ibotenic acid, quisqualic acid, kainic acid, N-methyl-D-aSpartiC acid (NMDA), a-amino-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA)). These compounds act as glutamate receptor agonists and exert their toxic action by prolonged activation of the receptors resulting in increased influx of chloride and calcium ions, excess water entry and osmotic lysis of

the cell. The excitotoxins cited have differential affinities to distinct glutamate receptor subtypes (e.g. quisqualate acts as agonist of the AMPA-type receptor, whereas ibotenic acid preferentially binds to NMDA receptors), thus the cytotoxicity of each glutamate analogue is dependent on the presence of a particular glutamate receptor subtype on the neuron. This might partly explain the differential cytotoxic effects of the known excitotoxins in different regions of the brain. Quinolinic, ibotenic, and quisqualic acid destroy cholinergic cells in the ventral pallidum and substantia innominatacomplex, whereas quinolinic and quisqualic acid do not degenerate cholinergic neurons in the medial septum (see e.g. Wenk et al., 1992). Ibotenate and quisqualate induce loss of neuronal cells throughout the Nbm complex but they produce different behavioural impairments in a variety of tasks indicating the presence of heterogeneous cell populations with differential sensitivity to a certain excitotoxin. This might be partly due to the fact that the various excitotoxins differentially affect cholinergic neurons in the basal forebrain. Quisqualate has been seen to produce a greater destruction of Nbm-cholinergic neurons than ibotenic acid. Cortical ChAT depletion by ibotenic acid infusion into the Nbm ranges between 27 and 46%, whereas quisqualate lesions of the Nbm result in cortical ChAT depletions by 41-74%. AMPA is even more effective in destroying cholinergic neurons of the Nbm (greater than 70%), but sparing dorsal pallidum and other non-cholinergic neurons in the basal forebrain (Boegman et al. 1992; for overview and references, see Dunnett et al., 1991). Lesion of basal forebrain by ethylcholine aziridinium ion (AF64A)

Ethylcholine aziridinium ion (AF64A) is a neurotoxic analog of choline and exerts its toxic action by disrupting the high affhity choline transport system that regulates the rate and extent of acetylcholine (ACh) synthesis. At higher concentrations AF64A also inhibits AChE activity in vitro. Several authors have suggested that AF64A completely lacks selectivity for cholinergic markers (see e.g. McGurk et al., 1987). However, in further

255

studies it was demonstrated that the specificity of AF64A depends on both the dosage applied and the site of injection. Local administration of AF64A at concentrations higher than 0.02 nmol into various brain regions was shown to produce considerable non-specific tissue destruction at the site of injection (McGurk et al., 1987). However, intraventricular injection of AF64A at low concentrations (less than 5 nmol) produces a relatively specific loss of cholinergic neurons restricted to the medial septal nucleus and the vertical limb of the diagonal band, but sparing cholinergic neurons in the Nbm, and without inducing histological damage to overlying cortex, fimbria fornix, or adjacent structures (Chrobak et al., 1988, 1989; Johnson et al., 1988; Gower et al., 1989; Potter et al., 1989; Lorens et al., 1991; Hortnagl et al., 1992). These AF64A-induced degenerations are accompanied by decreased choline uptake, ChAT, and ACh synthesis in the hippocampus (Lorens et al., 1991). Therefore low doses of AF64A allow for a selective lesion of medial septal cholinergic neurons, which might be useful when separately studying the role of the septo-hippocampal cholinergic system. Lesion of basal forebrain cholinergic system by 192IgG-saporin, a novel cholinergic immunotoxin Cholinergic neurons of the basal forebrain possess nerve growth factor (NGF) receptors (Chapter 33) whereas other neurons in this region including the cholinergic cells in the nearby striatum do not express detectable levels of NGF receptors (Gage et al., 1989; Yan and Johnson, 1989). It was demonstrated that a well-characterized monoclonal antibody to the low-affinity NGF receptor, 192IgG, accumulates bilaterally exclusively in cholinergic neurons of the basal forebrain following intracerebroventricular administration (see e.g. Thomas et al., 1991). Employing these properties of 192IgG, a cholinergic immunotoxin was developed by chemical linking of 192IgG via a disulfide bond to the ribosome inactivating protein saporin (192IgGsaporin; see Wiley et al., 1991; Wiley, 1992; for details of preparation, see Wiley and Lappi, 1993). The immunotoxin can be applied both systemically

and by intraventricular as well as parenchymal injection. The most promising results, however, have been obtained by intracerebroventricular applications. Intracerebroventricularadministration of 4,ug of 192IgG-saporin conjugate (at concentrations of 0.3-0.4 mg/mI) results in substantial reductions in ChAT activity in widespread areas of the cortex and hippocampus and in a nearly complete disappearance of ChAT-positive, NGF receptor immunoreactive neurons in the medial septum, in both the vertical and horizontal limbs of the nucleus of the diagonal band of Broca and in the Nbm, whereas cholinergic interneurons in the striatum are not affected (Book et al., 1992; Berger-Sweeney et al., 1994; Heckers et al., 1994; RoBner et al., 1995b). Seven days following injection of the immunotoxin there was a dramatic loss of AChE staining in frontal, parietal, piriform, temporal and occipital cortices, hippocampus and olfactory bulb, but not in the striatum and cerebellum (Heckers et al., 1994; RoBner et al., 1994a, 1995a). Non-cholinergic septal neurons containing parvalbumin and non-cholinergic substantia innominata neurons containing calbindin-DzsKor NADPH-diaphorase were not affected by 192IgGsaporin (Heckers et al., 1994). The number of parvalbumin-containing GABAergic projection neurons in the septum-diagonal band of Broca complex and Nbm was not reduced following intraventricular 192IgG-saporin application (Lee et al., 1994; Leanza et al., 1995; RoBner et al., 1995b). Moreover, 192IgG-saporin did not destroy neurotensin, galanin, somatostatin, or neuropeptide neurons within the Nbm (Wenk et al., 1994). Corresponding to the topographic location of cholinergic neurons in the basal forebrain a dramatic increase in microglia has been demonstrated (RoBner et al., 1995b), suggesting that the immunotoxin is lethal to cholinergic cells in the Nbm rather than suppressing the expression of cholinergic markers (e.g. ChAT) in these cells (Book et al., 1994). It was found that 192IgG-saporin affects two neuronal groups outside of the basal forebrain which express p75NGF receptors: NGF-reactive cerebellar Purkinje cells after intraventricular injection and cholinergic striatal interneurons after injections into the substantia innominata (Heckers

256

et al., 1994). There are ChAT-positive, but NGFreceptor negative neurons in the rat Nbmsubstantia innominata complex innervating the amygdala and parts of the rhinal paralimbic areas (see e.g. Woolf et al., 1989; Bickel and Kewitz, 1990) which are spared or only partially affected by the immunotoxin (Heckers et al., 1994). Similarly, cholinergic neurons in the ventral pallidum and sublenticular substantia innominata not expressing p75NGF receptors are not affected by the immunotoxin. Complete cholinergic lesion by 192IgG-saporin did not produce any deficit in the Morris water maze task (Torres et al., 1994). Despite the high depletion in cortical ChAT activity by 192IgGsaporin acquisition, performance of the delayed alternation or passive avoidance tasks were not impaired by the lesions suggesting that selective loss of cholinergic cells is not sufficient to produce functional impairments (Wenk et al., 1994). In contrast, other authors reported that intracerebral administration of 192IgG-saporin induced dosedependent (ranging between 1 and 1Opg) impairments in the water maze task and passive avoidance retention, but only weak effects on locomotor activity (Leanza et al., 1995; Waite et al., 1995). Intracerebroventricular injections of 192IgGsaporin severely affected spatial and cued navigation (Nilsson et al., 1992; Berger-Sweeney et al., 1994). However, an almost 90% reduction in ChAT activity is needed to produce substantial behavioural deficits (Waite et al., 1995).

Effect of different cholinergic lesion procedures on cortical cholinergic markers As outlined in the previous chapter the procedures to lesion cholinergic nuclei in rat basal forebrain differ in selectivity and specificity to degenerate cholinergic cells. A comparison of the effects of the various lesion procedures on cholinergic markers in cholinoceptive cortical target regions should therefore allow a further valuation of the different lesion techniques. Therefore, the same experimental design was applied to both electrolytic, ibotenic acid and immunolesion of basal forebrain cholinergic nuclei. Seven days after lesion receptor au-

toradiography and in situ hybridization were performed in adjacent coronal brain sections at six selected distances from the bregma ranging from +2.7 to -5.3 mm according to the atlas of Zilles (1985). The levels of cryocutting were selected to include for data analysis all cortical areas which receive a prominent cholinergic innervation from the basal forebrain. To prove the efficiency of the lesion histochemistry for AChE and ChAT were performed in adjacent brain sections (RoBner et al., 1994a,b, 1995a; Schliebs et al., 1994). The data obtained are summarized in Table 1. Seven days following unilateral electrolytic Nbm lesion we found a small reduction in M2muscarinic ACh receptor (mAChR) binding restricted to frontal and parietal cortices, but no change in MI-receptor binding sites in any of the cortical regions studied as compared to the unlesioned brain side. These alterations in cortical M2mAChR binding are complemented by corresponding changes in the m2-and m-mRNA transcripts (Schliebs et al., 1994; Table 1). Ibotenic acid lesion resulted in a striking loss of AChE-staining in the lesioned Nbm which is associated with a 60% decrease in AChE staining and a 30% reduction in [3H]hemicholinium-3 binding in frontal and parietal cortical regions as well as forehindlimb areas ipsilateral to the lesion, being more prominent in the more rostra1 cortical regions. MI-mAChR binding was not changed in any of the cortical regions studied 1 week after lesion. M2-mAChR binding levels are slightly increased in the parietal cortex only. The lesioninduced increase in parietal cortical M2-mAChR binding is complemented by an increase in the hybridization signal for the corresponding m-mRNA transcript (RoBner et al., 1994b; Table 1). Seven days following an intracerebroventricular injection of the cholinergic immunotoxin 192IgGsaporin hemicholinium-3 binding to high-affinity choline uptake sites was considerably decreased by up to 45% in all cortical regions and in the hippocampus as compared to the corresponding control values. In contrast, MI-mAChR sites were increased over the corresponding control values in the anterior parts of cingulate, frontal, and piriform cortex by about 20%, in the hindlimb/forelimb

257 TABLE I

TABLE I (continued)

Pattern of changes in cholinergic markers in selected rat brain regions 1 week after electrolytic and ibotenic acid lesion of the nucleus basalis magnocellularis as well as immunolesion of rat basal forebrain cholinergic system by 192IgG-saporin

Region

Region

Electrolytic

Cholinergic lesion procedure (significant relative changes over control in %) Electrolytic

Acetylcholineste,rase -12 cg -40 Fr Par -18 Pir occ Temp -

Ibotenic acid

1921ssaporin

-

-19 -80 -80 -46

-50 -50

-25 -

High-afinity choline uptake sites n.d. cg Fr n.d. -22 Par n.d. -20 Pir n.d. occ n.d. Temp n.d. Nicotinic acetylcholine receptor cg n.d. Fr n.d. Par n.d. Pir n.d. occ n.d. Temp n.d.

M2-muscarinic acetylcholine receptor cg Fr -10 -8 +15 Par Pir occ Temp -

-

-

-

-

-

Ibotenic acid

192IgGsaporin

-85 -79

-40 -40

-38 -30 -40 -38

n.d. n.d. n.d. n.d. n.d. n.d.

MI-muscarinic acetylcholine receptor cg Fr Par Pir occ Temp

ml-mAChR cg Fr Par occ Pir

Cholinergic lesion procedure (significant relative changes over control in %)

-18 -18

-35 -18 -17 -25

+22 +20 -

The alterations in various neurotransmitter receptors and AChE staining in selected brain regions 1 week after lesion are summarized and given as relative changes over the corresponding control value ( P < 0.05 or higher, two-tailed Student’s t-test) obtained from vehicle-injected control animals (immunolesion)or from the unlesioned brain side (electrolytic and ibotenic acid lesion). Unilateral electrolytic and ibotenic acid lesion of the nucleus basalis magnocellularis were performed as previously described (Schliebs et al., 1994; RoSner et al., 1994b). Immunolesion of rat basal forebrain cholinergic system by intracerebroventricularinjection of 192IgG-saporin was carried out as described by RoBner et al. (1994a). AChE was measured by histochemical staing of adjacent brain sections and quantified by image analysis; high-affinity choline uptake sites, M1-and MZ-mAChR were assayed by receptor autoradiography and quantitative image analysis (Schliebs et al., 1994; RoSner et al., 1994a,b; RoBner et al., 1995a). mlm4-mAChR subtypes were determined by in situ hybridization using 35S-labeledoligonucleotide probes (RoBner et al., 1993, 1994~). No significant change over control; n.d., not determined; Cg, cingulate cortex; Fr, frontal cortex; Par, parietal cortex; Pir: piriform cortex; Occ, occipital cortex; Temp, temporal cortex.

-.

25 8

areas (18%), in the parietal cortex (35%), in the occipital cortex (17%) as well as in the temporal cortex (25%) following immunolesion. M2mAChR levels were found to be significantly enhanced in the posterior part of the parietal cortex (by about 22%) and in the occipital cortex area (20%) only (RoSner et al., 1995a; Table 1). The increase in MI-mAChR binding in the temporal and occipital cortex as a consequence of immunolesion is complemented by an increase in the amount of ml and m3 mAChR mRNA by about 20% in these regions. The elevated levels of M2mAChR sites in the occipital and temporal cortex following immunolesion are accompanied by an increase in the m4 (by 25%) but not m2 mAChR mRNA. There was no effect of immunolesion on the ml-m4 mAChR mRNA in frontal cortical regions. In the basal forebrain, however, immunolesioning resulted in a considerable decrease in the level of m2 mAChR mRNA in the medial and lateral septum as well as in the vertical and horizontal limb of the diagonal band by about 4 0 8 , whereas MI-and M2-mAChR binding and the levels of ml, m3, and m4 mAChR mRNA were not affected by immunolesion in any of the basal forebrain nuclei studied. Seven days following a single dosage of the 192IgG-saporin no change in the level of cortical nicotinic acetylcholine receptor sites in any of the regions studied was observed as compared to the corresponding control values (RoBner et al., 1995a). One week after lesion a reduced high-affinity uptake of [3H]choline into cholinergic nerve terminals in the cerebral cortex and hippocampus was observed, which was accompanied by a decreased K+-stimulated release of r3H]ACh from cortical and hippocampal slices of immunolesioned rats (RoSner et al., 199%). Cholinergic immunolesion led to enhanced cortical MI-mAChR numbers, but did not alter mAChR sensitivity as measured by carbachol-stimulated inositol phosphate production or phorbol ester binding to membrane-bound protein kinase C (RoBner et al., 1995~).In the h i p pocampal formation differential enhancements in binding levels of MI-mAChR sites in the CA1 region and in the dentate gyrus were observed, whereas the nicotinic and M2-mAChR subtype are

seemingly not affected by the immunotoxin in either of the subfields studied. Cholinergic immunolesioning did not result in any alterations in the hybridization signals for ml-m4 mAChR mRNA in any region or layer of the hippocampus (RoSner et al., 199%). In summary, electrolytic Nbm lesions which destroy both cells, nerve terminals and passing fibres, did not change MI-mAChR but resulted in reduced M2-mAChR in frontal and parietal cortices 1 week after lesion. Nbm ibotenic acid lesion which likely affects both cholinergic and GABAergic cells but spares crossing fibres, did not alter MI-mAChR in any cortical region but resulted in enhanced M2-mAChR binding in the parietal cortex only. When applying the cholinergic immunotoxin 192IgG-saporin which specifically and selectively destroys basal forebrain cholinergic cells only, both MI-and M2-mAChR binding sites were increased in a number of cortical areas 1 week after lesion. From this comparison it can be suggested that the various lesion procedures differentially affect populations of mAChR localized on distinct cortical cholinoceptive cell populations. In particular, the different effects of the specific and selective cholinergic immunolesion and the less specific ibotenic acid lesion on MI-mAChR suggest that other transmitter systems, probably GABAergic projection neurons, contribute to the different cortical effects.

Effect of cholinergic immunolesion by 1921gGsaporin on cortical glutamate and GABA neurotransmission Glutamate is used as an excitatory transmitter in corticofugal as well as cortico-cortical systems and plays an important role in realizing cortico-cortical information transfer. GABA represents the major inhibitory transmitter in the cerebral cortex. It is generally accepted that the precise interaction of excitatory and inhibitory signals seems to be a major step in efficient processing of cortical information transfer. In a current study alterations in dendritic morphology of cortical neurons after basal forebrain lesions have been described (Wellman and Sengelaub, 1995) suggesting that

259

the cholinergic input plays an important modulatory role in cortical function and plasticity. To study the impact of reduced cortical cholinergic activity on glutamatergic und GABAergic transmission in the cerebral cortex and to elucidate possible adaptive responses, glutamate and GABA receptor subtypes were assayed by quantitative receptor autoradiography 1 week after a single intracerebroventricular injection of 4 p g of 192IgG-saporin. Receptor autoradiography and AChE staining were performed in adjacent brain sections, which allows simultanous detection of the consequences of lesions on various parameters in a distinct cortical area, and thus provides an appropriate tool to reveal correlations between cortical cholinergic hypoactivity and lesioninduced adaptive response in distinct cholinoceptive target regions. One week after cholinergic lesion by 1921gG-saporin, NMDA receptor binding was markedly reduced in cortical regions displaying a reduced activity of AChE and highaffinity choline uptake sites as a consequence of cholinergic lesion, whereas AMPA and kainate binding sites were significantly increased in these regions (RoBner et al., 1995d; Table 2). Muscimol binding to GABA, receptors was increased in the caudal portions of frontal and parietal cortices as well as occipital and temporal cortex as compared to the corresponding brain regions from vehicleinjected control rats (Table 2). Binding levels of benzodiazepine receptors were not affected by the lesion in any of the cortical regions studied (RoBner et al., 1995d). Equivalent changes in cortical glutamate and GABA receptor subtype levels have been observed 7 days after electrolytic (Schliebs et al., 1994) or ibotenic acid lesion (RoBner et al., 1994b) of the Nbm applying the same experimental design. In Table 2 the pattern of changes in selected cortical rat brain regions due to various cholinergic lesion procedures are summarized. Despite some differences in the specificity of the various lesion procedures applied, the data support the view that the alterations in cortical glutamate and GABA receptor subtypes following immunolesion are mainly due to the loss of cortical cholinergic input originating preferentially in the Nbm. To study whether the lesion-

induced alterations in glutamate and GABA transmission are consequences of reduced cortical cholinergic input, the parietal cortex with its wide rostral-caudal extension was used as an appropriate cortical model region displaying a gradient in the lesion-induced decreases in AChE activity and choline uptake sites from rostra1 to the caudal extension. The loss of cholinergic input as assayed by choline uptake sites is significantly correlated with the changes in binding levels of glutamate subtype and GABAA receptors (RoBner et al., 1995d) suggesting that the receptor changes might be the consequence of the imbalance between cortical cholinergic innervation and intracortical glutamatergic and GABAergic neurotransmission. This is supported by a recent report demonstrating that primate cortical M1-and M2-mAChR are associated with asymmetric synapses thus providing morphological evidence for cholinergic modulation of excitatory transmission via M,-and M2 receptors (Mrzljak et al., 1993). In the rat, cholinergic fibres from the basal forebrain terminate preferentially in cerebral cortical layers I and V, and MI-mAChR are mainly concentrated in layers IVnI and VI (Eckenstein et al., 1988; Schliebs and RoBner, 1995). Glutamate-containing neurons are concentrated in cortical layer V and the deep part of layer VI, whereas glutamate-containing axon terminals show the highest density in layers I-IV (Zilles et al., 1990). Each glutamate receptor subtype exhibits a distinct cortical laminar pattern (Kumar et al., 1993) suggesting that glutamate exerts a different influence in each particular cortical layer by inducing different cellular responses through distinct receptor subtypes. The decrease in NMDA receptor binding following immunolesion could be explained when assuming that at least some of the cortical NMDA receptors are located on cholinergic terminals originating in the basal forebrain. However, there is no evidence that cortical Nh4DA receptors may exist on presynaptic terminals of cholinergic neurons originating in the Nbm (Maragos et al., 1991). But the basal forebrain magnocellular complex receives among others also a strong glutamatergic innervation from the cortex (Martin et al., 1993) suggesting that the cholinergic immunolesion of the Nbm should have

260 TABLE 2 Pattern of changes in glutamatergic and GABAergic markers in selected rat brain regions one week after electrolytic and ibotenic acid lesion of the nucleus basalis magnocellularis as well as immunolesion of rat basal forebrain cholinergic system by 19218-saporin Region

Cholinergic lesion procedure (significant relative changes as compared to controls in %) Electrolytic

Ibotenic acid

192IgGsaporin

NMDA receptor cg -8 Fr -15 Par -20 Pir occ Temp -

-18 -18 -

-15 -20 -20 -15

AMPA receptor cg Fr +25 Par Pir occ Temp -

+20 +18 -

+I2 +15 +12

+18 +20 -

+20 +25 +20

-

-

+16 +16 -

+18 +20 +I8 +18

Kainate receptor cg Fr +30 Par +40 Pir +12 OCC +18 Temp GABAA receptor n.d. cg Fr n.d. Par n.d. Pir n.d. occ n.d. Temp n.d. Benzodiazepine receptor cg n.d. Fr n.d. Par n.d. Pir n.d. occ n.d. Temp n.d.

-

--

-

-

-

-

-

The alterations in various neurotransmitter receptors in selected brain regions 1 week after lesion are summarized and given as relative changes over the corresponding control value ( P < 0.05 or higher, two-tailed Student’s t-test) obtained from vehicle-injected control animals (immunolesion)or from the unlesioned brain side (electrolytic and ibotenic acid lesion). Unilateral electrolytic and ibotenic acid lesion of the nucleus basalis magnocellularis were performed as previously described (Schliebs et al., 1994; RoSner et al., 1994b). Immunolesion of rat basal forebrain cholinergic system by intracerebroventricular injection of 19218-saporin was carried out as described by RoSner et al. (1994a). Glutamate receptor subtypes like NMDA, AMPA, and kainate as well as GABA and benmdiazepine receptors were assayed by receptor autoradiography and quantitative image analysis (Schliebs et al., 1994; RoSner et al., 1994a,b, 199%). -, No significant change over control; n.d., not determined; Cg, cingulate cortex; Fr, frontal cortex; Par, parietal cortex; Pir, piriform cortex; Occ, occipital cortex; Temp, temporal cortex.

functional consequences also on cortical glutamatergic neurons. Therefore, the changes in the number of cortical NMDA receptors following lesion could be considered as a loss and/or downregulation of NMDA receptor sites. In contrast, the increased kainate and AMPA binding following lesion should be considered as up-regulation of receptor sites. Up-regulation of postsynaptic glutamate receptors is assumed to compensate for reduced presynaptic input. This would suggest that cholinergic terminals directly affect glutamate transmission on presynaptic glutamatergic elements. However, MK-801 is assumed to bind to a site within the NMDA receptor ion-channel and thus can also be considered as a marker of the agonist-bound, open state of the channel (Seeburg, 1993). Therefore, the immunotoxin-induced decline in MK-801 binding also indicates a lower amount of glutamate bound to the NMDA receptor channel. This supports the suggestion that cholinergic hypofunction reduces cortical glutamatergic activity by less release of glutamate from presynaptic elements presumably due to enhanced inhibition by GABA. Seven days after immunolesion we found significantly increased GABAA but not benzodiazepine binding sites in the frontal and parietal cortices (Table 2) suggesting an upregulation of postsynaptically localized GABAA receptors as an adaptive response to the reduced GABAergic input

26 1

as measured by Gomeza et al. (1992). In a recent study it was suggested that cholinergic excitation of GABAergic interneurons is mediated via M2mAChR (Mrzljak et al., 1993). But whether the immunolesion-induced increase in GABAA recep tor binding is a direct consequence of the enhanced M2-mAChR level observed in these regions 1 week after immunolesion (RoBner et al., 1995a) cannot be concluded from these data and must await further analysis. It is interesting to note that GABAAreceptors can up-regulate while benzodiazepine receptors remain unchanged 1 week after immunolesion, although both receptors should exist within the same protein receptor complex. However, from in vitro studies it is well known that a functional coupling exists between GABAA and benzodiazepine receptors: benzodiazepines can affect GABA binding and vice-versa by altering binding affinities (see e.g. Bureau and Olsen, 1993). Thus immunolesion-induced changes in GABA release could affect the binding states for benzodiazepines by altering the binding affinity and this could cover some changes in benzodiazepine binding. However, regardless of possible interpretations the immunotoxin-induced differential changes in glutamate and GABA receptor subtypes in cortical regions displaying reduced cholinergic activity clearly demonstrate that cortical glutamatergic and GABAergic markers are partially driven by cholinergic activity. Moreover, it is interesting to note that the same sort of alterations in glutamate and GABA receptor subtypes observed in rat cortex following basal forebrain cholinergic immunolesion have been detected in cortical brain areas from patients with Alzheimer’s disease (Nordberg, 1992). This supports the suggestion that the receptor changes observed might indicate compensatory mechanisms due to presumably cholinergic degenerative events. However, these data further support a glutamatergic strategy which might be therapeutically potential in treating Alzheimer’s disease (Advokat and Pelligrini, 1992; Carlson et al., 1993; Burney, 1994). Moreover, they suggest that cholinergic immunolesion by 192IgG-saporin exhibits a valuable tool to produce specific cholinergic deficits in rats, which can be used as a

model to study the effect of treatment with various drugs.

Summary Cholinergic lesion paradigms have been used to study the role of the cholinergic system in cortical arousal and cognitive function, and its implication in cognitive deficits that occur in Alzheimer’s disease. In the last few years an increasing number of studies have applied neurotoxins including excitotoxins or cholinotoxins (e.g. AF64A) by stereotaxic injection into the Nbm to produce reductions in cortical cholinergic activity. One of the most serious limitations of these lesion paradigms is the fact that basal forebrain cholinergic neurons are always intermingled with populations of noncholinergic cells and that the cytotoxins used are far from being selective to cholinergic cells. Excitoxins when infused directly into the Nbm destroy non-specifically cell bodies but spare axons passing the injection site, whereas the specificity of AF64A to destroy cholinergic neurons depends on both the dosage applied and the site of injection. Recently, a monoclonal antibody to the lowaffinity nerve growth factor (NGF) receptor, 192IgG, coupled to a cytotoxin, saporin, has been described as an efficient and selective immunotoxin for the NGF-receptor bearing cholinergic neurons in rat basal forebrain. Intraventricular administration of the 192IgG-saporin conjugate appears to induce a nearly complete and specific lesion of neocortical and hippocampal cholinergic afferents. Other neuronal systems in the basal forebrain are spared by the immunotoxin. Electrolytic, ibotenic acid, and cholinergic immunotoxic lesions of cholinergic basal forebrain nuclei resulted in slightly different effects on cortical cholinergic markers: Electrolytic lesion of the Nbm did not change MI-mAChR but resulted in reduced M2-mAChR in frontal and parietal cortices 1 week after lesion. Ibotenic acid lesion of the nucleus basalis did not alter MI-mAChR in any cortical region but led to enhanced M2-mAChR binding in the parietal cortex only. When applying the cholinergic immunotoxin 1921gG-saporin, both MI-and M,-mAChR binding sites were increased

262

in a number of cortical areas 1 week after lesion. This comparison suggests that possibly the destruction of non-cholinergic basal forebrain cells by ibotenic acid and electrolytic lesion, might partly contribute to these different cortical effects. NMDA receptor binding was markedly reduced and AMPA, kainate, and GABAA receptor binding has been significantly increased in cortical regions displaying a reduced activity of AChE and decreased levels of high-affinity choline uptake sites due to immunolesion of the basal forebrain cholinergic system. Equivalent changes in cortical glutamate and GABA receptor subtype levels have been observed 7 days after electrolytic or ibotenic acid lesion of the Nbm. The data suggest that cholinergic immunolesion by 192IgG-saporin exhibits a valuable tool to produce specific cholinergic deficits in rats, which can be used as a model to study the effect of treatment with various drugs for compensating the impaired cortical cholinergic input.

Acknowledgements This work was partly supported by a grant of the Bundesministerium fur Forschung und Technik to R.S.,no. FKZ 01 ZZ 9103/2.8.

References Advokat, C. and Pelligrini, A.I. (1992) Excitatory amino acids and memory, evidence from research on Alzheimer’s disease and behavioral pharmacology. Neurosci. Behav. Rev., 16: 13-24. Berger-Sweeney, J., Heckers, S., Mesulam, M.-M., Wiley, R.G.,Lappi, D.A. and Sharma, M. (1994) Differential effects on spatial navigation of immunotoxin-inducedcholinergic lesions of the medial septa1 area and nucleus basalis magnocellularis. J. Neurosci., 14: 450745 19. Bickel, U. and Kewitz, H. (1990) Colocalization of choline acetyltransferase and nerve growth factor receptor in the rat basal forebrain. Dementia, 1: 146-150. Bigl, V., Woolf, N.J. and Butcher, L.L. (1982) Cholinergic projections from the basal forebrain to frontal, parietal, temporal, occipital, and cingulate cortices: a combined fluorescent tracer and acetylcholinesterase analysis. Brain Res. Bull., 8: 727-749. Bigl, V., Arendt, T. and Biesold, D. (1989) The nucleus basalis Meynert during aging and in dementing disorders. In: M. Steriade and D. Biesold (Eds.), Cholinergic Systems of

the Basal Forebrain, Oxford University Press, Oxford, pp. 364-386. Boegman, R.J., Cockhill, J., Jhamandas, K. and Beninger, R.J. (1992) Excitoxic lesions of rat basal forebrain: differential effects on choline acetyltransferase in the cortex and amygdala. Neuroscience, 5 1: 129-1 35. Book, A.A., Wiley, R.G. and Schweitzer, J.B. (1992) Specificity of 192 IgG-saporin for NGF receptor-positive cholinergic basal forebrain neurons in the rat. Brain Res., 590: 350-355. Book, A.A., Wiley, R.G. and Schweitzer, J.B. (1994) 192 IgG-saporin: I. Specific lethality for cholinergic neurons in the basal forebrain of the rat. J. Neuropathol. Exp. Neurol., 53: 95-102. Bureau, M.H. and Olsen, R.W. (1993) GABAA receptor subtypes: Ligand binding heterogeneity demonstrated by photoaffhity labeling and autoradiography. J. Neurochem., 61: 1479-149 1. Burney, R.N. (1994) Therapeutic potential of NMDA antagonists in neurodegenerative diseases. Neurobiol. Aging, 15: 27 1-273. Carlson, M.D., Penney, Jr., J.B. and Young, A.B. (1993) NMDA, AMPA, and benzodiazepine binding site changes in Alzheimer’s disease visual cortex. Neurobiol. Aging, 14: 343-352. Chrobak, J.J., Hanin, I., Schmechel, D.E. and Walsh, T.J. (1988) AF64A-induced working memory impairment: Behavioural, neurochemical and histological correlates. Brain Res., 463: 107-117. Chrobak, J.J., Spates, M.J., Stackman, R.W. and Walsh, T.J. (1989) Hemicholinium-3prevents the working memory impairments and the cholinergic hypofunction induced by ethylcholine aziridinium ion (AF64A). Bruin Res., 504: 269-275. Dunnett, S.B., Everitt, B.J. and Robbins, T.W. (1991) The basal forebrain-cortical cholinergic system: interpreting the functional consequences of excitotoxic lesions. Trends Neurosci., 14: 494-501. Eckenstein, F.P., Baughman, R.W. and Quinn, J. (1988) An anatomical study of cholinergic innervation in rat cerebral cortex. Neuroscience, 25: 457474. Fisher, A. and Hanin, I. (1986) Potential animal models for senile dementia of Alzheimer’s type, with emphasis on AF64A-induced cholinotoxicity. Annu. Rev. Pharmacol. Toxicol., 26: 161-181. Gage, F.H., Batchelor, P., Chen, K.S., Chin, D., Deputy, S., Rosenberg, M.B., Higgins, G.A., Koh, S., Fischer, W. and Bjorklund, A. (1989) NGF-receptor expression and NGFmediated cholinergic neuronal hypertrophy in the damaged adult neostriatum. Neuron, 2: 1177-1 184. Gomeza, J., Arag6n, C. and Gimhez, C. (1992) High-affinity transport of choline and amino acid neurotransmitters in synaptosomesfrom brain regions after lesioning the nucleus basalis magnocellularisof young and aged rats. Neurochem. Res., 17: 345-350. Gower, A.J., Rousseau, D., Jamsin, P., Gobert, J., Hanin, I.

263 and Wulfert, E. (1989) Behavioral and histological effects of low concentrations of intraventricular AF64A. Eur. J. Pharmacol., 166: 271-281. Greenamyre, J.T. and Maragos, W.F. (1993) Neurotransmitter receptors in Alzheimer disease. Cerebrovasc. Brain Met. Rev., 5 : 61-94. Heckers, S., Ohtake, T., Wiley, R.G., Lappi, D.A., Geula, C. and Mesulam, M.M. (1994) Complete and selective denervation of rat neocortex and hippocampus but not amygdala by an immunotoxin against the p75 NGF receptor. J. Neurosci., 14: 1271-1289. Hortnagl, H., Sperk, G., Sobal, G. and Maas, D. (1990) Cholinergic deficit induced by ethylcholine aziridinium (AF64A) transiently affects somatostatin and neuropeptide Y levels in rat brain. J. Neurochem., 54: 1608-1613. Johnson, G.V.W., Simonato, M. and Jope, R.S. (1988) Doseand time dependent hippocampal cholinergic lesions induced by ethylcholine mustard aziridinium ion: effects of nerve growth factor, GM1 ganglioside, and vitamin E. Neurochem. Res., 13: 685-692. Kumar, A,, Schliebs, R. and Bigl, V. (1993) Development of NMDA, AMPA, and kainate receptors in individual layers of rat visual cortex and the effect of monocular deprivation. Int. J. Dev. Neurosci., 12: 31-41. Leanza, G., Nilsson, O.G., Wiley, R.G. and Bjorklund, A. (1995) Selective lesioning of the basal forebrain cholinergic system by intraventricular 192 IgG-saporin: behavioural, biochemical and stereological studies in the rat. Eur. J. Neurosci., 7: 329-343. Re, M.G., Chrobak, J.J., Sik, A,, Wiley, R.G. and BuzsBki, G. (1994) Hippocampal theta activity following selective lesion of the septa1 cholinergic system. Neuroscience, 62: 1033-1047. .orens, S.K., Kindel, G., Dong, X.W., Lee, J.M. and Hanin, I. (1991) Septa1 choline acetyltransferase immunoreactive neurons: dose-dependent effects of AF64A. Brain Res. Bull., 26: 965-97 1. Maragos, W.F., Greenamyre, J.T., Chu, D.C.M., Penney, J.B. and Young, A.B. (1991) A study of cortical and hippocampal NMDA and PCP receptors following selective cortical and subcortical lesions. Brain Res., 538: 36-45. Martin, L.J., Blackstone, C.D., Levey, A.I., Huganir, R.L. and Price, D.L. (1987) Cellular localization of AMPA glutamate receptors within the forebrain magnocellular complex of rat and monkey. J. Neurosci., 13: 2249-2263. McGurk, S.R., Hartgraves, S.L., Kelly, P.H., Gordon, M.N. and Butcher, L.L. (1987) Is ethylcholine mustard aziridinium ion a specific cholinergic neurotoxin? Neuroscience, 22: 215-224. Mrzljak, L., Levey, A.1. and Goldman-Rakic, P.S. (1993) Association of ml and m2 muscarinic receptor proteins with asymmetric synapses in the primate cerebral cortex:morphological evidence for cholinergic modulation of excitatory neurotransmission. Proc. Natl. Acad. Sci. USA, 90: 5194-5198. Nilsson, O.G., Ixanza, G., Rosenblad, C., Lappi, D.A., Wiley,

R.G. and Bjorklund A (1992) Spatial learning impairments in rats with selective immunolesion of the forebrain cholinergic system. NeuroReport, 3: 1005-1008. Nordberg, A. (1992) Neuroreceptor changes in Alzheimer disease. Cerebrovasc. Brain Met. Rev., 4: 303-328. Potter, P.E., Tedford, C.E., Kindel, G. and Hanin, I. (1989) Inhibition of high affinity choline transport attenuates both cholinergic and non-cholinergic effects of ethylcholine aziridinium (AF64A). Brain Res., 487: 238-244. RoBner, S., Perez-Polo, J.R., Wiley, R.G., Schliebs, R. and Bigl, V. (1994a) Differential expression of immediate early genes in distinct layers of rat cerebral cortex after selective immunolesion of the forebrain cholinergic system. J. Neurosci. Res., 38: 282-293. RoRner, S., Schliebs, R. and Bigl, V. (1994b) Ibotenic acid lesion of nucleus basalis magnocellularis differentially affects cholinergic, glutamatergic and GABAergic markers in cortical rat brain regions. Brain Res., 668: 85-99. RoBner, S., Schliebs, R., Perez-Polo, J.R., Wiley, R.G. and Bigl, V. (1995a) Differential changes in cholinergic markers from selected brain regions after specific immunolesion of rat cholinergic basal forebrain system. J. Neurosci. Res., 40:31-43. RoBner, S.,Hiirtig, W., Schliebs, R., Briickner, G., Brauer, K., Perez-Polo, J.R., Wiley, R.G. and Bigl, V. (1995b) 192IgGsaporin immunotoxin-induced loss of cholinergic cells differentially activates microglia in rat basal forebrain nuclei. J. Neurosci. Res., 41: 335-346. RoBner, S., Schliebs, R., H W g , W. and Bigl, V. (199%) 1921s-saporin-induced selective lesion of cholinergic basal forebrain system: neurochemical effects on cholinergic neurotransmission in rat cerebral cortex and hippocampus. Brain Res. Bull., 38: 371-381. RoBner, S., Schliebs, R. and Bigl, V. (1995d) 192IgG-saporininduced immunotoxic lesions of cholinergic basal forebrain system differentially affect glutamatergic and GABAergic markers in cortical rat brain regions. Brain Res., 696: 165176. Schliebs, R. and RoBner, S. (1995) Distribution of muscarinic acetylcholine receptors in the CNS. In: T.W. Stone (Ed.), CNS Neurotransmitters and Neuromodulators. AcetylchoIin, CRC Press, Boca Raton, FL, pp. 67-83. Schliebs, R., Feist, T., RoBner, S. and Bigl, V. (1994) Receptor function in cortical rat brain regions after lesion of nucleus basalis. J. Neural Transm., 44(Suppl.): 195-208. Seeburg, P.H. (1993) The molecular biology of mammalian glutamate receptor channels. Trends Pharmacol. Sci., 14: 297-303. Thomas, L.B., Book, A.A. and Schweitzer, J.B. (1991) Immunohistochemical detection of a monoclonal antibody directed against the NGF receptor in basal forebrain neurons following intraventricular injection. J. Neurosci. Methods, 37: 37-45. Torres, E.M., Perry, T.A., Blokland, A,, Wilkinson, L.S., Wiley, R.G., Lappi, D.A. and Dunnett, S.B. (1994) Behavioural, histochemical and biochemical consequences of se-

264 lective immunolesions in discrete regions of the basal forebrain cholinergic system. Neuroscience, 63: 95-122. Waite, J.J., Chen, A.D., Wardlow, M.L., Wiley, R.G., Lappi, D.A. and Thal, L.J. (1995) 192 immunoglobulin G-saporin produces graded behavioral and biochemical changes accompanying the loss of cholinergic neurons in the basal forebrain and cerebellar h k i n j e cells. Neuroscience, 65: 463-476. Wellman, C.L. and Sengelaub, D.R. (1995) Alterations in dendritic morphology of frontal cortical neurons after basal forebrain lesions in adult and aged rats. Brain Res., 669: 48-58. Wenk, H., Bigl, V. and Meyer, U. (1980) Cholinergic projections from magnocellular nuclei of the basal forebrain to cortical areas in rats. Brain Res. Rev., 2 295-316. Wenk, G.L., Harrington, C.A., Tucker, D.A., Rance, N.E. and Walker, L.C. (1992) Basal forebrain neurons and memory: a biochemical, histological, and behavioral study of differential vulnerability to ibotenate and quisqualate. Behav. Neurosci., 106: 909-923. Wenk, G.L., Stoehr, J.D., Quintana, G., Mobley, S. and Wiley, R.G. (1994) Behavioral, biochemical, histological, and electrophysiological effects of 192 IgG-saporin injections into the basal forebrain of rats. J. Neurosci., 14: 5986-5995.

Wiley, R.G. (1992) Neural lesioning with ribosomeinactivating proteins: suicide transport and immunolesioning. Trends Neurosci., 15: 285-290. Wiley, R.G. and Lappi, D.A. (1993) Preparation of antineuronal immunotoxins for selective neural immunolesioning. Neurosci. Prot., 93-02b-02-01-12. Wiley, R.G., Oeltmann, T.N. and Lappi, D.A. (1991) Immunolesioning: selective destruction of neurons using immunotoxin to rat NGF receptor. Brain Res., 562: 149-153. Woolf, N.J., Gould, E. and Butcher, L.L. (1989) Nerve growth factor receptor is associated with cholinergic neurons of the basal forebrain but not the pontomesencephalon.Neuroscience, 3 0 143-152. Yan, Q. and Johnson, Jr., E.M. (1989) Immunohistochemical localization and biochemical characterization of nerve growth factor receptor in adult rat brain. J. Comp. Neurol., 290: 585-598. Zilles, K. (1985) The Cortex of the Rat. A Stereotaxic Atlas. Springer, Berlin. Zilles, K., Wree, A. and Dausch, N.D. (1990) Anatomy of the neocortex: neurochemical organization. In: B. Kolb and R.C. Tees (Eds.), The Cerebral Cortex of the Rat, MIT Press, Cambridge, MA, pp. 113-150.