Introduction According to this hypothesis, synaptic contacts would be

Brain Research 888 2001 203–211 www.elsevier.com locate bres Research report Decreased benzodiazepine receptor density in the cerebellum of early blind human subjects a b a b ´ Sandra M. Sanabria-Bohorquez , Anne G. De Volder , Patricia Arno , Merence Sibomana , b b a , Ann Coppens , Christian Michel , Claude Veraart a ´ Neural Rehabilitation Engineering Laboratory , GREN-5446, Universite Catholique de Louvain, Avenue Hippocrate, 54, B-1200 Brussels, Belgium b ´ Positron Tomography Laboratory , Universite Catholique de Louvain, Brussels, Belgium Accepted 26 September 2000 Abstract As a first approach to study the effect of early visual deprivation in the GABA-ergic inhibitory system, the distribution of 11 11 benzodiazepine receptors BZR was accurately estimated using [ C]flumazenil [ C]FMZ. Measurements were carried out in five 11 subjects who became blind early in life and in five sighted control subjects. The interactions between [ C]FMZ and BZR were described using a non-linear compartmental analysis which permitted to estimate the BZR synaptic density independently of other model parameters. The distribution of BZR in the visual areas and other cortical regions of blind subjects was qualitatively and quantitatively similar to that of controls. However, the BZR density in the cerebellum was significantly lower in blind than in control subjects P,0.01. Our findings suggest that modifications of the cerebellar neural circuitry may be concomitant to the already observed compensatory reorganization in cerebral areas of blind subjects.  2001 Elsevier Science B.V. All rights reserved. Theme : Neurotransmitters, modulators, transporters, and receptors Topic : GABA receptors 11 Keywords : [ C]Flumazenil; Benzodiazepine receptors; Early blindness; Visual cortex; Cerebellum; Plasticity

1. Introduction According to this hypothesis, synaptic contacts would be

more numerous and or hyperactive in EB visual areas Positron emission tomography PET studies have [13,58]. Besides, the decreased glucose metabolism level shown that the glucose metabolism in primary and associa- in the EB cerebellum suggests that early visual deprivation tive visual areas of early blind EB subjects at rest is may also alter the cerebellar neural organization. comparable to that measured in sighted control SC The synaptogenesis process and the later phase of subjects with the eyes open, i.e. to the normally active synaptic elimination observed in primate cerebral cortex visual cortex [58,59]. Brain energy metabolism studies are accompanied by similar changes in several neuro- have demonstrated that the metabolic state of the EB visual transmitter receptor densities dopaminergic, adrenergic, areas is related to neuronal activity [13]. Furthermore, the serotonergic, cholinergic, and GABAergic receptors glucose utilization in the cerebellum of EB subjects was [36,37]. Moreover, since the interaction between neuro- significantly lower than in blindfolded SC subjects con- transmitters affects the neuronal cytoarchitecture and corti- trasting the findings in the visual areas [13]. cal developmental processes [17,21,30,31], the combina- The increased metabolism in EB visual cortex has been tion of morphological and neurochemical studies may attributed to the alteration of the phase of synaptic provide further insight on the effects of sensory depriva- elimination following the synaptogenesis process during tion. brain development, due to the lack of visual stimuli [58]. As a first approach to study in vivo the chemical circuitry of the human brain in the case of early blindness, 11 11 we have used [ C]flumazenil [ C]FMZ to quantify the Corresponding author. Tel.: 132-2-764-5446; fax: 132-2-764-9422. E-mail address : veraartgren.ucl.ac.be C. Veraart. benzodiazepine BZ receptor density of EB and SC 0006-8993 01 – see front matter  2001 Elsevier Science B.V. All rights reserved. P I I : S 0 0 0 6 - 8 9 9 3 0 0 0 3 0 4 9 - 3 204 S 11 Table 1 subjects. [ C]FMZ is a BZ antagonist widely used in Characteristics of the early-blind subjects human brain studies using positron emission tomography Subjects Age at test Age of blindness Etiology PET. BZ receptors BZR are functionally associated to years onset GABA receptors which are made up of pentameric A assemblies of subunits a, b, g, d, r that form a chloride EB1 69 Birth Unknown 2 EB2 45 18 months Retinoblastoma enucleated channel. The Cl conductance increases when GABA EB3 23 Birth Retrolental fibroplasia binds to the b-subunit and the effect is facilitated when BZ a EB4 54 Birth–10 years Congenital glaucoma binds to the ag subunit complex in the same channel EB5 57 ,1 year Unknown enucleated [11,53]. FMZ binds to most of the GABA receptors A a EB4 had very poor vision from birth. Loss of the right eye at the age of resulting from different subunit combinations in mam- 6 months, and definitive loss of the left eye at 10 years. malian brain, except to those containing d subunit [53]. The FMZ affinity for receptors containing a subunit is 6 lower than the affinity for receptors with a subunit proved by the Medical Ethics Committee of the School of 1 ´ [27,53]. GABA receptors containing a subunit are only Medicine at the Universite Catholique de Louvain. A 6 expressed in cerebellar granule cells and represent 45 of 11 all GABA receptors in the rat cerebellum [26]. 2.2. Synthesis of [ C]FMZ A The kinetics of in vivo ligand binding of FMZ to central 11 11 BZR was described using a nonlinear three compartment [ C]FMZ [Methyl- C] Ro 15-1788 was labelled by model [15]. The model parameters were accurately esti- N-alkylation of the desmethyl compound Ro 15-5528 11 mated by using a three-injection protocol which guarantees with anhydrous [ C]methyl iodide in acetone by use of 11 a unique solution and small parameter estimation un- sodium hydroxide as base [20]. [ C]FMZ was purified by certainties [15,16]. The input curve to the compartmental semi-preparative reverse phase using high-performance model was obtained by applying a metabolite correction liquid chromatography HPLC. The HPLC columns used 11 method which estimates the relative fraction of [ C]FMZ were Alltech Econosil with 0.01M H PO acetonitrile and 3 4 in the total radioactivity in plasma curve from the tissue 70 30 as eluent. The radiopharmaceutical formulation has kinetic data without the need for actual FMZ metabolite been performed following a method that uses a C Sep- 18 measurements [52]. Pak Plus cartridge from Waters [7]. 11 Labeling and quantification of BZR using [ C]FMZ may provide information on the effects of early blindness 2.3. Experimental protocol on the distribution of GABA receptors in the cerebral and A cerebellar cortex of human subjects. The effect of visual The experimental protocol lasted 90 min and consisted 11 deprivation on the distribution of GABA neurons and its of three injections of [ C]FMZ and or unlabeled FMZ 11 receptors in visual brain areas has been mainly studied in [15]. Firstly, an injection of about 370 MBq of [ C]FMZ cases of monocular deprivation in adult monkeys [8,21– with high specific activity was performed and 30 min later 23]. However, the effect of early binocular deprivation on a displacement injection of 0.7 mg of unlabeled FMZ was GABA distribution, to our knowledge, has not been performed. At 60 min, a coinjection of approximately 160 11 reported, neither in animals nor in humans. This study may MBq of [ C]FMZ and 1.2 mg of unlabeled FMZ was provide insight on the contribution of such inhibitory made. The specific activity of the first injection ranged synapses to the measured abnormal metabolism in the EB from 5300 to 16200 MBq mmol. Each time, the FMZ was visual cortex and cerebellum [13,58]. injected as a bolus of 30 s through a 22-gauge catheter Abbocath E in a forearm vein. A 24-gauge catheter Abbocath E was inserted in the radial artery of the other arm under local anesthesia with bipuvacaine for blood

2. Materials and methods sampling. After the labeled tracer injections at 0 and 60