Peripheral type benzodiazepine binding sites are a sensitive indirect index of neuronal damage

Brain Research, 421 (1987) 167-172 Elsevier

167

BRE 12877

Peripheral type benzodiazepine binding sites are a sensitive indirect index of neuronal damage Jesus Benavides, Dominique Fage, Christopher Carter and Bernard Scatton Laboratoires d'Etudes et de RecherchesSynthdlabo, BiochemicalPharmacology Group, Bagneux (France) (Accepted 17 February 1987) Key words: Excitotoxin; Choline acetyltransferase; Glutamate decarboxylase; Peripheral type benzodiazepine binding site; Rat striatum

The effects of excitotoxic lesions on the neuronal marker enzymes choline acetyltransferase and glutamate decarboxylase and on the levels of 'peripheral type' benzodiazepine binding sites (PTBBS) (a putative glial marker) have been compared to see whether PTBBS provide a suitable if indirect quantitative index of neuronal damage. Intrastriatal injection of excitotoxic compounds provoked a dose-dependent increase in the levels of PTBBS. The potency order was the following: kainate > AMPA > N-methyl-D-aspartate (NMDA) > quisqualate. The maximal increases in this parameter were 400,470, 320 and 210% for kainate (12 nmol), AMPA (100 nmol), NMDA (500 nmol) and quisqualate (250 nmol), respectively. 2-Amino-5-phosphonovalerate (100 nmol) - - an antagonist of the NMDA receptor subtype - - completely blocked the increase in PTBBS induced by NMDA (250 nmol), but was without effect against the other excitotoxins. Increases in binding levels were in general mirrored by a decrease in choline acetyltransferase and glutamate decarboxylase activity. However, PTBBS were a more sensitive indirect index of neuronal damage than neuronal enzymes because the alterations in binding were statistically significant at doses of excitotoxins lower than those causing a loss of marker enzymes. It is concluded that PTBBS are a suitable and sensitive means of detecting discrete neurotoxic changes and that its measurement will help in the study of other pathological and experimental models.

INTRODUCTION Neurodegenerative p h e n o m e n a such as those produced by excitotoxins, ischaemia or other brain instilts 8'11'13'17'19'20'22'30'31can be evidenced histologically or by assay of specific markers known to be restricted to susceptible neurones; for instance choline acetyltransferase (CAT) or glutamate decarboxylase ( G A D ) have been used as indices of neuronal damage produced by neurotoxic lesions of the striatum 24. This approach is suitable provided the identity of the damaged neurone is already known and also if the damage is sufficiently large or uniform to produce significant changes in enzyme activity of selected brain regions. However, this is very often not the

case because neuronal damage to neurotoxins, or other insults may additionally be restricted to subpopulations of neurones in one area or to limited architectural layers 13,26. Given the constraints of dissection techniques, quite large neurochemical deficits may be masked by the bulk of tissue taken for assay, or simply not detected because of lack of knowledge of a suitable marker. There is clearly a need for a sensitive but general index of neuronal damage, and this may be provided indirectly by quantitative assessment of other factors such as reactive gliosis which follows brain damage 8'28. Reactive gliosis is characterized by changes in the number and shape of astrocytes 15 and by a microglial proliferation. Reactive astrocytosis, as

Correspondence: B. Scatton, Laboratoires d'Etudes et de Recherches Synth61abo, Biochemical Pharmacology Group, 31, avenue Paul Vaillant Couturier, 92220 Bagneux, France. 0006-8993/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)

168 indexed by glial fibrillary protein immunochemistry, has been observed after excitotoxic lesions with quinolinic acid 6 and in Huntington's disease 25. Glial cells particularly microglia are believed to possess 'peripheral type' benzodiazepine binding sites (PTBBS) 27 and this can explain the dramatic increase in these binding sites after excitoxic lesions of striatum and piriform cortex 12A7,21. PTBBS levels are also slightly augmented in human neurodegenerative states 17,21. PTBBS differ from the 'classical' (central type) benzodiazepine receptors in their pharmacological specificity4, subcellular 2 and anatomical 4,5 distribution and their preferential labelling of glial cells in culture 27. They can be specifically labelled by ligands such as [3H]PK 11195 (refs. 4, 5) and [3H]RoS-4864 (ref. 12) which are devoid of affinity for central type benzodiazepine receptors. In this study, we have examined the feasibility of using PTBBS levels as an index of neuronal damage by comparing the effects of various excitotoxin-induced striatal lesions on PTBBS and the more conventional neurochemical indices CAT and GAD activity. Because the effects of such toxins are receptor-mediated, dose-dependent and in some cases amenable to pharmacological blockade 16'23the quantitative effects of graded lesions in various circumstances can be compared. Our main intention was to see whether PTBBS levels quantitatively reflect lesion size as reflected by other currently available parameters, and thus to characterise the technique for future use in other instances of cerebral injury. MATERIALS AND METHODS Male Sprague-Dawley rats (200 g, COBS CD strain, Charles River, France) were used. The animals were housed under standard laboratory conditions (22 + 0.5 °C) and maintained under a 12-h light/dark cycle (lights on between 07.00 and 19.00 h) with free access to food and water. Kainic acid, Nmethyl-D-aspartic acid (NMDA), quisqualic acid, (RS)- a- amino- 3-hydroxy-5- methyl-4-isoxazolopropionic acid (AMPA), 2-aminophosphonovaleric acid (2-APV) and 2-amino phosphonoheptanoic acid (2APH) were all obtained from Tocris Neuramin (U.K.). Drug doses always refer to the free bases. For excitotoxic lesions rats were anaesthetised (chloral hydrate, 400 mg/kg, i.p.) and immobilised in

a Kopf stereotaxic frame. Unilateral striatai injection of the excitotoxic compounds was performed via a cannula (0.45/~m external diameter) implanted at the following coordinates: A: 8.2; V: 0; L: 2.7 (atlas of K6nig and Klippe114). The injection cannula was connected by polyethylene tubing to a motor-driven 10-~1 Hamilton syringe and the excitotoxins were injected in a volume of 4/A NaOH buffered saline (pH 7.4) at a perfusion rate of 1/A/min for 4 min. The injection cannula was kept in place for a further 5-min before removal. Controls received an equivalent volume of saline. One week after surgery the striata were dissected out and Teflon/glass homogenised in 250/~1 distilled water. 200/A aliquots were diluted to 500 ~1 with 300 /~1 of a 0.5% Triton X-100, 10 mM EDTA (pH 7.4) solution and further homogenised by sonication. CAT and GAD activities were measured in aliquots of this homogenate using the methods of Fonnum 1° and Albers and Brady 1, respectively. The remaining original 50-/~1 aliquots were diluted to a final volume of 1.5 ml with Tris-satine buffer (50 mM Tris HC1, 120 mM NaC1; pH 7.4) and polytron homogenised (setting 6, 10 s). Aliquots of 500/~1 of polytron-homogenised membranes were incubated with 1 nM [3H]Ro5-4864 (80.5 Ci/mmol, NEN) in a final volume of 1 ml Tris-saline buffer. Non-specific binding was defined by 1/~M PK 11195 (ref. 4). After 120 min incubation at 4 °C, bound radioactivity was recovered by vacuum filtration onto Whatman GF/B filters followed by 3 washes with 3.5 ml Tris-saline buffer21. The statistical analysis of the data was evaluated using the one-way analysis of variance followed by multiple comparison tests: the Duncan test when variances were homogenous, the Kruskal and Wallis test when variances were heterogenous. Correlation between the biochemical parameters investigated was examined by the least squares method of linear regression analysis. RESULTS Intrastriatal administration of all the excitotoxic compounds tested (kainate, quisqualate, AMPA, NMDA) induced a dose-dependent increase in the levels of PTBBS and a decrease in the activity of the neuronal marker enzymes CAT and GAD (Fig. 1).

169

KAINATE

AMPA

l

500 *

T

T

T

T

0 0

a 2

NMDA

QUISQUALATE

1

"I

2

4

,

T

T

8

T

I 12

T 10

• T

T 30

100

10

30

T

62.5 125 250 500 nmoles

t00 250

j.'"j.

~E ~so 100

Fig. 1. Effect of intrastriatal injection of kainate, quisqualate, AMPA or NMDA on PTBBS density and CAT and GAD activities in the striatum. Rats were sacrificed 7 days following excitotoxin injection. Results are means with S.E.M. of data obtained from at least 4 rats and are expressed as percent changes in respective control values ([3H]Ro5-4864 binding: 17.4 + 2.3 pmol/g tissue; CAT: 11.6 + 0.46 pmol/g tissue/h; GAD: 5.3 + 0.4/zmol CO2/g tissue/h). ÷P < 0.05; *P < 0.01 respective controls.

The order of potency for the increase in PTBBS density and the decrease in CAT or G A D activities were the same (kainate > AMPA > N M D A > quisqualate). In general the increases in PTBBS levels were statistically significant at doses lower than those causing a significant loss of neuronal marker enzymes. However, the relationship between the elevations in PTBBS and the loss of CAT and G A D was not the same for all the excitotoxins studied. It was

clearly linear in the case of N M D A but curvilinear for kainate (Fig. 2). Moreover, kainate effects on PTBBS appeared to reach a plateau at doses lower than those eliciting the maximal loss in neuronal marker enzymes (Fig. 1). Linear regression analysis of the relationship between the changes in binding and enzyme levels also indicates a better correlation for N M D A and AMPA than for kainate effects (Table I). No other regression (logarithmic, expo-

KAINATE

4(](3

NMDA

//13

3O0

///////I "°

//1

o "/

200

100

/

o//

iI I

2o

4'o

6'o

8o

I

I

I

o

,o

20

% DECREASE IN CAT { o--o) AND

GAD

3'o

L

go

(e-~eJ ACTIVITIES

Fig. 2. Comparison between the alterations of PTBBS density and CAT (O) and G A D (Q) activities in the striatum provoked by intrastriatal injection of several doses of kainate (left panel) and NMDA (right panel). Data plotted are from Fig. 1.

170 TABLE I

TABLE II

Correlation between the changes in PTBBS and CA T or GAD activity after excitotoxic lesions o f the rat striatum

Protection by 2-APV and 2-APH against the neurotoxic effect~ of agonists acting on the 3 subtypes of excitatory aminoacid receptors

Correlation coefficients were computed by linear regression analysis from data presented in Fig. 1.

Excitotoxin

Correlation coefficient

Kainate Quisqualate AMPA NMDA

PTBBS vs CA T

PTBBS vs GAD

0.89* 0.92" * 0.99** 0.99"*

0.86* 0.74 0.98** 0.99"*

Compounds were administered intrastriatally in a volume of 4 /d. Rats were sacrificed at 7 days postinjection. Results are means + S.E.M. of data obtained from at least 5 rats and are expressed as percentage of respective controls (saline-injected rats). The effects of the excitotoxins alone on the parameters investigated were significant at P < 0.001. Intrastriatal administration of 2-APV or of 2-APH alone did not provoke significant alteration in any parameter.

Drug

%[3H]Ro54864binding

% Enzyme activities CAT C~-D "

NMDA (250 nmol) N M D A (250 nmol) + 2-APV (100 nmol) N M D A (250 nmol) + 2-APH (30 nmol) Quisqualate (250 nmol) Quisqualate (250 nmol) + 2-APV (100 nmol) Kainate (4 nmol) Kainate (4 nmol) + 2-APV (100 nmol)

525 ___70

60 + 6

61 _+ 7

146 _+ 30*

97 + 2*

101 + 2*

195 + 18" 409 + 28

92 + 5" 86 + 4

94 _+ 2" 81 + 4

464 + 38 321 + 41

80 + 3 64 + 6

81 + 5 70 + 5

382 + 30

64 _ 1

68 _+ 1

*P < 0.05, **P < 0.01.

nential) analysis of kainate effects gave a better fit than the linear analysis. The low neurotoxic potency of quisqualate precluded an accurate analysis. The effects of an intrastriatal injection of N M D A (250 nmol) on CAT and GAD activities and on PTBBS levels were antagonised in a dose-dependent manner by co-injection of 2-APV (Fig. 3). 2-APV

*P < 0.01 vs the excitotoxin alone,

T

I

400+

T m

T

o

~

2-APV NMDA

0 j0

0

7_. -,-

250

3

10

30

100

250

250

250

250

nmoles

r~

~

20

z/Z z/" L

alone had no effect on these parameters (data not shown) but at 3 nmol significantly protected against the effects of NMDA on PTBBS levels and produced near total protection at 100 nmol. The antagonism of the NMDA-provoked loss of neuronal markers only became statistically significant at 10 nmol for GAD and 30 nmol for CAT. 2-APH (another NMDA receptor antagonist 29) also blocked the effects of NMDA on these parameters (Table II). 2-APV at a dose which fully antagonized the excitotoxic effects of NMDA (100 nmol) failed to prevent the alterations of PTBBS and enzyme levels provoked by kainate and quisqualate (Table 1I).

• DISCUSSION

x

~

//

40 i

Fig. 3. Blockade by 2-APV of the neurotoxic effects of N M D A . 2-APV at different doses (0-100 nmol) was coinjected with N M D A (250 nmol) in the striatum. Rats were sacrificed at 7 days postinjection. Results are means + S.E.M. of data obtained from at least 4 rats. +P < 0.05; *P < 0.01 vs N M D A alone.

In normal rat brain, PTBBS are mainly concentrated in the choroid plexus and ependyma, suggesting a predominantly extraneuronal localisation 4. This view is supported by binding studies in cultured glial cells27. Our studies confirm that there is a dramatic increase in PTBBS levels after neurotoxic lesions of the striatum. Previous studies 21 have established that this elevation of [3H]Ro5-4864 binding is

171 due to a change in the maximal capacity but not in the affinity for this radioligand. A glial proliferation accompanying neuronal degeneration 6A5'25'28 may be responsible for this increase in PTBBS binding. Although further characterisation is needed of the localisation of these sites to the different types of glia a possible candidate is the astrocytic cell which proliferates after excitotoxic lesions 15. On the other hand, the presence of PTBBS in proliferating microglia would be consistent with the high levels of binding sites found in cells of monocytic lineage 18'32(to which microglia are related). Moreover, the modulation of monocytic cell activity by ligands of PTBBS 1s'33 suggests that these binding sites may be involved in controlling the defensive or scavenging reactions to brain injury in which glial cells play a pivotal role. Whatever their role, it is clear that PTBBS are markedly elevated following neurotoxic lesions of the rat striatum, and that they can be used to indirectly index the degree of neuronal damage produced by such lesions. This was shown by the good correlation between the increase in PTBBS density and the decrease in CAT or GAD activities produced by each neurotoxin. It also seems that PTBBS levels, because of their higher reactivity are a more sensitive index of neural damage, as changes in PTBBS were statistically significant for each neurotoxin at doses below those producing significant reductions in CAT and GAD activity. This was also evident when examining the protective effects of 2-APV vs NMDA toxicity, i.e. significant protective effects using the PTBBS index were observed before noting significant modifications of CAT and GAD activity. CAT and GAD only index damage to cholinergic and GABAergic neurones within the striatum and different neurotoxins may be able to destroy other striatal neuronal populations 3. This may explain the observed large increases in PTBBS at doses of kainate which failed to decrease CAT and GAD activities. Indeed, using autoradiographic studies with PTBBS we have observed differing patterns of extrastriatal damage, some of which are anatomically localised to areas projecting to, or receiving inputs from, the striatum with the different neurotoxins (in

preparation). Analysis of the mechanism of action of the various neurotoxins was not central to these experiments, but the results using CAT or GAD and PTBBS binding as an index support the receptor-mediated nature of excitotoxicity. Thus the effects of NMDA on both indices were antagoniseable by NMDA antagonists (2-APV, 2-APH) 29, while the effects of kainate or quisqualate were not blocked by these compounds. Little conclusion can be drawn from the effects of the quisqualate/kainate antagonist ~,-D-glutamyl-aminomethyl sulphonic acid (GAMS) 9 as this compound was neurotoxic in itself. In conclusion, PTBBS provide a convenient and sensitive indirect index of cerebral neuronal damage, in this case that produced by neurotoxic agents. Increases in PTBBS mirror the reduction in CAT or GAD activity and clearly show when a lesion is produced, or when the effects of a neurotoxic substance can be antagonised. The use of this assay has obvious advantages in terms of facility, sensitivity and quantification. This technique can also be applied in autoradiographic studies where the secondary lesions provoked by neurotoxic compounds such as kainate (in preparation), and the regional distribution of focal ischaemic lesions in rats 5 can be clearly visualized. This marker may also be of use in defining the anatomy of brain lesions in clinical samples. We have indeed observed recently in autoradiographic studies performed on postmortem samples from patients with Parkinson's or Huntington's disease, dramatic increases in PTBBS levels in those brain areas that are known to undergo neuronal degeneration in these diseases (in preparation). These observed increases in PTBBS levels in human neurodegenerative states predict the usefulness of this marker in postmortem studies and PET scanning diagnosis (by using [11C]PK 11195) 7 of neurological diseases.

ACKNOWLEDGEMENTS We thank V. Eninger and S. Bleasdale for their secretarial help in preparing this manuscript.

172 REFERENCES 1 Albers, R.W. and Brady, R.O., The distribution of glutamate decarboxylase in the nervous system of the rhesus monkey, J. Biol. Chem., 234 (1959) 926-928. 2 Basile, A.S. and Skolnick, P., Subcellular localization of 'peripheral type' binding sites for benzodiazepine in rat brain, J. Neurochem., 46 (1986) 305-308. 3 Beal, M.F., Kowall, N.W., Ellison, D.W., Mazurek, M.F., Swartz, K.J. and Martin, J.B., Replication of the neurochemical characteristics of Huntington's disease with quinolinic acid, Nature (London), 321 (1986) 168-170. 4 Benavides, J., Quarteronet, D., Imbault, F., Malgouris, C., Uzan, A., Renault, C., Dubroeucq, M.C., Gueremy, C. and Le Fur, G., Labelling of 'peripheral type' benzodiazepine binding sites in the rat brain. Kinetic studies and autoradiographic localization, J. Neurochem., 41 (1983) 1744-1750. 5 Benavides, J., Duverger, D., MacKenzie, E.T., Peny, B. and Scatton, B., Peripheral type benzodiazepine binding sites as a marker of focal brain infarction in the rat, Xlllth International Symposium on Cerebral Blood Flow and Metabolism, Montr6al, 1987. 6 Bj6rklund, H., Olson, L., Dahl, D. and Schwarcz, R., Short- and long-term consequences of intracranial injections of the excitotoxin, quinolinic acid, as evidenced by GFA immunohistochemistry of astrocytes, Brain Research, 371 (1986) 267-277. 7 Charbonneau, P., Syrota, A., Crouzel, C., Prennat, C. and Crouzel, M., Peripheral-type benzodiazepine receptors in the living heart characterized by position emission tomography, Circulation, 73 (1986) 476-483. 8 Coyle, J.T. and Schwarcz, R., Lesion of striatal neurons with kainic acid provides a model for Huntington's chorea, Nature (London), 263 (1976) 244-246. 9 Davies, J. and Watkins, J.C., Depressant actions of y-Dglutamyl-aminomethyl sulfonate (GAMS) on amino acidinduced and synaptic excitation in the cat spinal cord, Brain Research, 337 (1985) 113-120. 10 Fonnum, F., Radiochemical assay for choline acetyltransferase and acetylcholinesterase. In N. Mark and R. Rodnight (Eds.), Research Methods in Neurochemistry, Vol. 3, Plenum, New York, 1975, pp. 253-275. 11 Foster, P.C., Whetsell, W.O., Bird, E.D. and Schwarcz, R., Quinolinic acid phosphoribosyl transferase in human and rat brain activity in Huntington's disease and quinolinate-lesioned rat striatum, Brain Research, 336 (1985) 207-214. 12 Gehlert, D.R., Yamamura, H.I. and Wamsley, J.K., Autoradiographic localization of 'peripheral type' benzodiazepine binding sites in the rat brain, heart and kidney, Naunyn-Schmiedeberg's Arch. Pharmacol., 328 (1985) 454-460. 13 Johansen, F.F., J~rgensen, M.B., Egstr6m, D.K. and Diemer, N.H., Selective dendrite damage in hippocampal CA 1 stratum radiatum with unchanged axon ultrastructure and glutamate uptake after transient cerebral ischemia in the rat, Brain Research, 291 (1984) 373-377. 14 Kfnig, J.F.R. and Klippel, R.A., The Rat Brain: a Stereotaxic Atlas of the Forebrain and Lower Parts of the Brain Stem, Williams and Wilkins, Baltimore, 1963. 15 Miller, R.H., Abney, E.R., David, D., Ffrench-Contant, C., Lindsay, R., Patel, R., Stone, G. and Raft, M.C., Is

reactive gliosis a property of a distinct subpopulation of astrocytes?, J. Neurosci., 6 (1986) 22-29. 16 Olney, J.W., Labruyere, J., Collins, J.F. and Curry, K., oAminophosphonovalerate is 100-fold more powerful than D-alpha-aminoadipate in blocking N-methylaspartate neurotoxicity, Brain Research, 221 (1981) 207-210. 17 Owen, F., Poulter, M., Waddington, J.L., Mashal, R.D. and Crow, T.J., 3H-Ro5-4864 and [3H]flunitrazepam binding in kainate-lesioned rat striatum and temporal cortex of brains from patients with senile dementia of the Alzheimer subtype, Brain Research, 278 (1983) 373-375. 18 Pert, C.B., Ruff, M.R., Weber, R.J. and Herkenham, M., Neuropeptides and their receptors: a psychosomatic network, J. lmmunol., 135 (1985) 8231-8261. 19 Peterson, D.W., Collins, J.F. and Bradford, H.F., The kindled model of epilepsy: anticonvulsant action of amino acid antagonists, Brain Research, 275 (1983) 169-172. 20 Sanberg, P.R. and Johnston, G.A., Glutamate and Huntington's disease, Med. J. Aast., 2 (1981) 460-465. 21 Schoemaker, H., Morelli, M., Desh, M.P. and Yamamura, H.I., [3H]Ro5-4864 benzodiazepine binding in the kainate lesioned striatum and Huntington's diseased basal ganglia, Brain Research, 248 (1982) 396-401. 22 Schwarcz, R. and Coyle, J., Striatal lesions with kainic acid: neurochemical characteristics, Brain Research, 127 (1977) 235-249. 23 Schwarcz, R. and Meldrum, B., Excitatory amino acid antagonists provide a therapeutic approach to neurological disorders, Lancet, ii (1985) 140-143. 24 Schwarcz, R., Scholz, D. and Coyle, J.T., Structure-activity relations for the neurotoxicity of kainic acid derivatives and glutamate analogues, Neuropharmacology, 17 (1978) 147-151. 25 Selkoe, D.J., Salazar, F.J., Abraham, C. and Kosik, K.S., Huntington's disease: changes in striatal proteins reflect astrocytic gliosis, Brain Research, 245 (1982) 117-125. 26 Simon, R.P., Swan, J.H., Griffiths, T. and Meldrum, B.S., Blockade of N-methyl-D-aspartate receptors may protect against ischemic damage in the brain, Science, 226 (1984) 850-852. 27 Syapin, P.J. and Skolnick, P., Characterization of benzodiazepine binding sites in cultured cells of neuronal origin, J. Neurochem., 32 (1979)1047-1051. 28 Treip, C.S., A Colour Atlas of Neuropathology, Wolfe, 1978, pp. 23-27. 29 Watkins, J.C. and Evans, R.H., Excitatory amino acid neurotransmitters, Annu. Rev. Pharmacol. Toxicol., 21 (1981) 165-204. 30 Watson, M., Vickroy, T.W., Fibiger, H.C., Roeske, W.R. and Yamamura, H.I., Effects of bilateral ibotenate-induced lesions of the nucleus basalis magnocellularis upon selective cholinergic biochemical markers in the rat anterior cerebral cortex, Brain Research, 346 (1985) 387-391. 31 Wielloch, T., Hypoglycemia-induced neuronal damage prevented by an N-methyl-D-aspartate antagonist, Science, 230 (1985) 681-683. 32 Zavala, F., Haumont, J. and Lenfant, M., Interaction of benzodiazepines with mouse macrophages, Eur. J. Pharmacol., 106 (1984) 561-566. 33 Zavala, F. and Lenfant, M., A functional receptor for peripheral benzodiazepines in macrophages. Symposium on Purification, Biosynthesis and Regulation of Membrane Bound Receptors, Cap d'Agde, France, 1986.