Direct visualization of serotonin1D receptors in the human brain using a new iodinated radioligand

Molecular Brain Research, 13 (1992) 175-179 (~ 1992 Elsevier Science Publishers B.V. All rights reserved. 0169-328X/92/$05.00

175

BRESM 80115

Direct visualization of serotoninlo receptors in the human brain using a new iodinated radioligand Jos6 M. Palacios, Christian Waeber, Anne T. Bruinvels and Daniel Hoyer Preclinical Research Sandoz Pharma Ltd., Basel (Switzerland)

(Accepted 5 November 1991) Key words: Serotonin-5-O-carboxymethyl-glycyl[125I]tyrosinamide;Substantia nigra; Globus pallidus; [3H]Serotonin binding site

The development of the new ligand serotonin-5-O-carboxymethyl-glycyl[125I]tyrosinamide (abbreviated [125I]GTI) allows for the direct visualization of serotoninxa and serotoninlo (5-HTm/1D) sites. Autoradiographic techniques were used to demonstrate the selective binding of this ligand to 5-HT1D sites in human post-mortem brain materials. The distribution of [125I]GTIbinding sites was compared to [3H]5-HT sites in the presence of different displacers. The results show the selective binding of [125I]GTIto sites in the basal ganglia and substantia nigra which corresponds to 5-HT1D receptors. Pharmacological and molecular biological results have demonstrated the existence of a large family of serotonin (5-hydroxytryptamine, 5-HT) receptors which have been grouped in different families. In the human brain, receptors belonging to the 5-HT1, 5-HT 2 and 5-HT 3 families have been characterized by receptor binding techniques 8 and their anatomical and cellular localization 6 has been established autoradiographically. One of the 5-HT receptor subtypes which is present at highest densities in the human brain is the so-called 5-HT1D receptor 9. Until recently, this receptor has been defined by exclusion, i.e. as those [3H]5-HT binding sites which are neither of the 5-HT1A nor of the 5-HT~c subtype and therefore visualized in the presence of saturating concentrations of blocking agents for these subtypes such as 8-hydroxy-2N-dipropylamino tetralin (8-OH-DPAT) and mesulergine 9,1°. The recent development of the compound serotonin5-O-carboxymethyl-glycyl tyrosinamide (GTI) as a serotonergic ligand by Segu and collaborators 1 has led us to examine the characteristics of the binding of this compound in human post-mortem material. Here we report that in human brain this ligand selectively recognizes a 5-HT~D receptor binding site, allowing for the first time the direct visualization of 5-HTID receptors. Brains were obtained at autopsy from 18 different subjects (8 males, 10 females, aged 70 + 16 years, postmortem delay 12.3 + 4.5 h; data given as mean + S.E.M.) deceased without a reported history of neurological or psychiatric disease. Brains were promptly dissected and 1-cm-thick tissue blocks were excised and placed at -80°C. Tissue sections were cut in 10-mm-

thick slices with a microtome-cryostat, thaw-mounted onto gelatinized microscope slides and stored at -20°C until use. The incubations were made according to the following procedure, after two 15 min preincubations in 170 mM Tris-HC1 (pH 7.5) containing 4 mM CaC12, the slides were incubated for 1 h at room temperature in the same medium supplemented with 10 mM pargyline and 0.01% ascorbic acid and 2 nM [3H]5-HT (26.3 Ci/mmol, DuPont-NEN, Boston, MA). The binding to the 5-HT1D subtype was defined as that remaining in the presence of 100 nM 8-OH-DPAT and mesulergine. Non-specific binding was determined in the presence of 10 mM 5-HT. Alternate sections to those incubated with [3H]5-HT were labelled with 20 pM [125I]GTI (2175 Ci/mmol). G T I was kindly synthesized by Dr. J. Nozulak (Sandoz, Pharma Ltd. Basel, Switzerland) and custom labelled by ANAWA (Wangen, Switzerland). GppN(H)p (5'-guanylylimidophosphate) was obtained from Sigma (St. Louis, U.S.A.). The washing of labelled sections was carried out as follows: a brief dipping in preincubation buffer followed by two 5 min washes in the same buffer and a brief dipping in distilled water to remove the salts. Finally the sections were quickly dried under a stream of cold air. Autoradiograms were generated by apposing the labelled tissues to 3H-Hyperfilms (Amersham, U.K.) along with Amersham tritiated or iodinated polymer standards. For [3H]5-HT-labelled tissue sections, films were developed after 2 months exposure at 4°C, while 10 days were sufficient for the [125I]GTI-labelled slides. Autoradiograms were quantified with a computerized image-analysis system (MCID, Imaging Research, Ontario, Canada) 11.

Correspondence: J.M. Palacios, Laboratorios Almirall, Cardoner 68-74 Barcelona, 08024, Spain.

176

-

p

k

•

m i"

Fig. 1. Distribution of 5-HT binding sites in some areas of the human midbrain. The pictures are photo-micrographs from autoradiograms obtained by incubating human midbrain tissue sections with [3H]5-HT (A-C) or [125I]GTI (D,E). A illustrates the distribution of [3H]5-HT binding sites (note the high densities of binding, dark areas, in the substantia nigra (SN), the dorsal raph6 nucleus (DRN) and the periaquaductal grey (PAG)). B: the distribution of non 5-HT1A/5-HTlc binding sites by blocking with 100 nM 8-OH-DPAT and mesulergine (note the displacement of binding sites in the dorsal raph6 nucleus but not in the substantia nigra). C: image obtained in the presence of 10 mM 5-HT to determine non-specific binding. D depicts density of [125I]GTI binding sites in the substantia nigra while the dorsal raph6 nucleus and the periaquaductal grey remain unlabelled. E: amount of non-specific [lZSI]GTI binding that was obtained in the presence of 10 mM 5-HT. Exposure time of A-C was 2 months and for D and E I0 days, both on 3H-Hyperfilms. Bar = 5 mm.

In agreement with previous findings 9, highest densities of [3H]5-HT binding sites, corresponding to the 5-HTID subtype were enriched in the pars reticulata of the substantia nigra (Fig. 1) and in the basal ganglia (Fig. 2), in both parts (lateral and medial) of the globus pallidus, the putamen, the caudate and the accumbens nuclei. [125I]GTI binding sites were also observed in the substantia nigra, both pars compacta and reticulata and in the globus pallidus, both medialis and lateralis (Figs. 1 and 2). High levels of non-specific binding (i.e. insensitive to 10 m M 5-HT) were observed in the substantia ni-

gra pars compacta, probably corresponding to the neuromelanin-rich cell bodies of the dopaminergic neurons. Similar accumulations of non-specific binding on these cells have been previously observed with other radioligands, [3H]ketanserin for example 7. W h e n c o m p a r e d , the pattern of distribution of 5 - H T l o sites visualized with [3H]5-HT and [125I]GTI showed a good anatomical correspondence. It must be mentioned that the receptor occupancy of [3H]5-HT (in the presence of 100 nM 8 - O H - D P A T and mesulergine) is much higher than that of [125I]GTI. Moreover, the large variability between the

177 .

I

t"

;X.

i: i

;1

, ..

-t

C

"x e

Fig. 2. Distribution of 5-HT binding sites of some areas of the human basal ganglia. As in Fig. 1, A-C were obtained by using [3H]5-HT as a radioligand. High densities of these sites were seen in the globus pallidus (GP), both lateralis and medialis, the cortical areas (Cx) and the claustrum (Cl). Of a somewhat lower density was the binding in the putamen (Pu) (see A). The binding sites in the cortex and the claustrum were sensitive to 100 nM 8-OH-DPAT and mesulergine, but not the sites in the globus pallidus and the putamen (B). The non-specific binding was determined in the presence of 10 mM 5-HT (C). Similar to Fig. 1, D and E depict tissue sections labelled with [125I]GTI where high densities of binding sites can be observed in the globus pallidus (both medialis and lateralis) and lower densities in the putamen, claustrum and cortex (D). All these sites were completely displaced by 10 mM 5°HT (E). Bar = 5 mm.

different subjects and the limited number of cases used prohibit a direct and exact comparison of the number of sites labelled with both radioligands. Nevertheless, these preliminary results might suggest that smaller numbers of binding sites can be found in the claustrum, hippocampal formations, amygdala, accumbens nuclei and pos-

sibly the putamen and caudate. But a complete mapping study (in preparation) will have to provide more exact information about the quantitative distribution of this new radioligand. Pharmacological analysis of the characteristics of [~25I]GTI binding sites in the human substantia nigra

178 iii

100

FL~ CJ

8O

directly labels 5-HTID receptors in the human brain. Moreover, the sensitivity of [125I]GTI to guanylnucleotides suggests that this ligand is an agonist at the 5-HTID receptor. In contrast to [3H]5-HT binding sites, which show biphasic competition curves for several ligands (e.g. 5-CT) 12, [125I]GTI binding sites were identified as a ho-

6O .i "0 c

.o

coupling of the 5-HTID receptor to GTP-binding proteins. Similar sensitivity has been r e p o r t e d of [3H]5-HT labelled 5 - H T I o sites 3. The main results of this investigation is that [leSI]GTI

40 20 0

I JllllllJ

10

9

I I IIIIli~

8

I I Iltllll

7

I I IIIIIlI

6

I I[llJll[

5

I PIIII

i IIIII

4

3

concentration GppN(H)p (-log mol/I)

Fig. 3. Competition curves of increasing concentrations GppN(H)p versus [125I]GTI in the substantia nigra (pars reticulata, O), the medial globus pallidus (11) and the caudate (A) obtained by microdensitometric quantification of autoradiograms generated as described in the text (n = 2).

m e m b r a n e s 2 and the tissue sections revealed that these sites represent a profile corresponding to that of a 5-HTID receptor. Thus [125I]GTI binding was m o n o p h a sically displaced by low concentrations of 5-carboxamidotryptamine (5-CT) and sumatriptan (3-[2-(dimethylamino ) ethyl] - N- methylindole - 5 - methanesulfonamide) 2, amongst others, in the substantia nigra and the globus pallidus. F u r t h e r m o r e , in order to analyze the agonist or antagonist nature of this ligand we examined the effects of guanylnucleotides on the density of [125I]GTI binding. The presence of 10 -9 to 10 -4 M G p p N ( H ) p concentrations in the incubation media resulted in a dose dependent decrease of [lzSI]GTI binding, with an approximate ICs0 of 1 m M and a m a x i m u m displacement of 60-80% of the specific binding (Fig. 3). The sensitivity of [125I]GTI binding to guanylnucleotides is in concordance with the

1 Boulenguez, P., Chauveau, J., Segu, L., Morel, A., Lanoir, J. and Delaage, M., A new 5-hydroxy-indole derivative with preferential affinity for 5-HT1B sites, Eur. J. Pharmacol., 194 (1991) 91-95. 2 Bruinvels, A.T., Landwehrmeyer, B., Waeber, C., Palacios, J.M. and Hoyer, D., Homogeneous 5-HT1D recognition sites in the human substantia nigra identified with a new iodinated radioligand, Eur. J. Pharmacol., 202 (1991) 89-91. 3 Harrington, M.A. and Peroutka, S.J., Differential modulation of 5-hydroxytryptamine 1D binding sites by GTP and GTPyS, Brain Res., 506 (1990) 172-174. 4 Leonhardt, S., Herrick-Davis, K. and Titeler, M., Detection of a novel serotonin receptor subtype (5-HTtE) in human brain: interaction with a GTP-binding protein, J. Neurochem., 53 (1989) 465-471. 5 Lowther, S., De Parmentier, F., Crompton, M.R. and Horton, R.W., 5-HT~D and 5-HT~E binding sites in post-mortem human brain, Br. J. Pharmacol., 102 (1991) 246P. 6 Palacios, J.M., Waeber, C., Hoyer, D. and Mengod, G., Distribution of serotonin receptors. In P.M. Whitaker-Azmitia and S.J. Peroutka (Eds.), The Neuropharmacology of Serotonin, Ann. N.Y. Acad. Sci., Vol. 600, 1990, pp. 37-52. 7 Pazos, A., Probst, A. and Palacios, J.M., Serotonin receptors

mogeneous population by these ligands (ref. 2 and autoradiographic results). Therefore, it is possible that the sites recognized by [125I]GTI correspond to the high-affinity c o m p o n e n t of the receptor. H o w e v e r , at present time the existence of non-5-HTIA/5-HTIc/5-HTID [3H]5H T binding sites in human brain cannot be excluded 4'5. The d e v e l o p m e n t of this new specific ligand represents an important advance in the study of serotonin receptors. Studies in our laboratory as well as in that of Segu and collaborators I have shown that this ligand labels both 5-HT m and 5-HT m receptors, depending on the species studied. Thus, in rat, mouse (and very likely also in hamster and opossum) brain, this ligand recognizes 5-HT m sites, while in the guinea-pig, calf, pig, cat, dog, m o n k e y (unpublished results) and as shown in this paper, human brain [*25I]GTI binds to 5 - H T l o sites. The fact that this ligand is labelled with [~25I]iodine is also convenient because exposure times are now reduced from 2 months ([3H]5-HT) to only 10 days ([leSI]GTI). Even shorter exposure times could be achieved using higher concentrations of the radioligand. Alterations of 5-HTtD sites have already been r e p o r t e d in neurodegenerative diseases 1~. The availability of this new ligand will certainly simplify the analysis of the influence of disease and drug t r e a t m e n t on 5-HTID receptors in the human brain.

8

9

10 11

12

in the human brain-IV. Autoradiographic mapping of serotonin-2 receptors, Neuroscience, 21 (1987) 123-129. Peroutka, S.J., Schmidt, A.W., Sleight, A.J. and Harrington, M.A., Serotonin receptor 'families' in the central nervous system: an overview. In P.M. Whitaker-Azmitia and S.J. Peroutka (Eds.), The Neuropharmacology o f Serotonin, Ann. N.Y. Acad. Sci., Vol. 600, 1990, pp. 104-113. Waeber, C., Dietl, M.M., Hoyer, D., Probst, A. and Palacios, J.M., Visualisation of a novel serotonin recognition site (5HTlo ) in the human brain by autoradiography, Neurosci. Lett., 88 (1988) 11-16. Waeber, C., Schoeffter, P., Hoyer, D. and Palacios, J.M., The serotonin 5-HTID receptor: a progress review, Neurochem. Res.. 15 (1990) 567. Waeber, C. and Palacios, J.M., Serotonin-I receptor binding sites in the human basal ganglia are decreased in Huntington's chorea but not in Parkinson's disease: a quantitative in vitro autoradiography study, Neuroscience, 32 (1989) 337-347. Waeber, C. and Palacios, J.M., 5-HTlc, 5-HT~o and 5-HT 2 receptors in the mammalian brain: multiple affinity states with a different regional distribution. In J.R. Fozard and P.R. Saxena (Eds.), Serotonin: Molecular Biology, Receptors and Functional Effects, Birkha0ser, Basel, 1991, pp. 107-116.