Selective visualization of rat brain 5-HT2A receptors by autoradiography with [3H]MDL 100,907

Abstract

The recently developed 5-HT2A receptor selective antagonist [3H]MDL100,907 ((+/-)2,3-dimethoxyphenyl-1-[2-(4-piperidine)-methanol]) has been characterized as a radioligand for the autoradiographic visualization of these receptors. [3H]MDL100,907 binding to rat brain tissue sections was saturable, had sub-nanomolar affinity (Kd = 0.2-0.3 nM), and presented a pharmacological profile consistent with its binding to 5-HT2A receptors (rank order of affinity for [3H]MDL100,907-labelled receptors: MDL100,907 > spiperone > ketanserin > mesulergine). The distribution of receptors labelled by [3H]MDL100,907 was compared to the autoradiographical patterns obtained with [3H]Ketanserin, [3H]Mesulergine, and [3H]RP62203 (N-[3-[4-(4-fluorophenyl)piperazin-1-y1]propyl]-1,8-naphtalenes ultam) and to the distribution of 5-HT2A receptor mRNA as determined by in situ hybridization. As opposed to the other radioligands, [3H]MDL100,907 labelled a single population of sites (5-HT2A receptors) and presented extremely low levels of non-specific binding. The close similarity of the distributions of [3H]MDL100,907-labelled receptors and 5-HT2A mRNA further supports the selectivity of this radioligand for 5-HT2A receptors and suggests a predominant somatodendritic localization of these receptors. The present results point to [3H]MDL100,907 as the ligand of choice for the autoradiographic visualization of 5-HT2A receptors.

Full text

Naunyn-Schmiedeberg’s Arch Pharmacol (1997) 356:446–454 © Springer-Verlag 1997
Abstract The recently developed 5-HT
2A
receptor selec-
tive antagonist [
3
H]MDL100,907 ((+/–)2,3-dimethoxy-
phenyl-1-[2-(4-piperidine)-methanol]) has been charac-
terized as a radioligand for the autoradiographic visual-
ization of these receptors. [
3
H]MDL100,907 binding to
rat brain tissue sections was saturable, had sub-nanomo-
lar affinity (Kd = 0.2–0.3 nM), and presented a pharma-
cological profile consistent with its binding to 5-HT
2A
receptors (rank order of affinity for [
3
H]MDL100,907-la-
belled receptors: MDL100,907 > spiperone > ketanserin
> mesulergine). The distribution of receptors labelled by
[
3
H]MDL100,907 was compared to the autoradiographi-
cal patterns obtained with [
3
H]Ketanserin, [
3
H]Mesuler-
gine, and [
3
H]RP62203 (N-[3-[4-(4-fluorophenyl)-
piperazin-1-y1]propyl]-1,8-naphtalenesultam) and to the
distribution of 5-HT
2A
receptor mRNA as determined by
in situ hybridization. As opposed to the other radio-
ligands, [
3
H]MDL100,907 labelled a single population of
sites (5-HT
2A
receptors) and presented extremely low
levels of non-specific binding. The close similarity of the
distributions of [
3
H]MDL100,907-labelled receptors and
5-HT
2A
mRNA further supports the selectivity of this ra-
dioligand for 5-HT
2A
receptors and suggests a predomi-
nant somatodendritic localization of these receptors. The
present results point to [
3
H]MDL100,907 as the ligand of
choice for the autoradiographic visualization of 5-HT
2A
receptors.
Key words [
3
H]MDL100,907 · 5-HT
2A
receptors ·
Rat brain · Autoradiography · in situ hybridization ·
[
3
H]ketanserin · [
3
H]mesulergine · [
3
H]RP62203
Introduction
Receptors for serotonin (5-hydroxytryptamine, 5-HT) are
currently classified in seven families, some of which in-
clude several receptor subtypes. The 5-HT
2
receptor
family includes 5-HT
2A
(previously called 5-HT
2
, seroto-
nin 2 or S2 receptors), 5-HT
2B
(previously termed SRL,
(Serotonin Receptor Like) or 5-HT
2F
receptors), and 5-
HT
2C
receptors (previously included into the
5-HT
1
family and termed 5-HT
1C
) (Hoyer et al. 1994,
Baxter et al. 1995). Within this family, 5-HT
2A
receptors
are involved in many peripheral and central actions of
5-HT (Hoyer et al. 1994) and have been implicated in
disorders such as schizophrenia (Schmidt et al. 1995).
The characterization and visualization of 5-HT
2A
recep-
tors has relied on compounds which, although originally
described as being selective for this receptor subtype,
were later shown to have limited selectivity (see Discus-
sion). The recent development of MDL100,907 ((±)2,3-
dimethoxyphenyl-1-[2-(4-piperidine)-methanol]) (Dud-
ley et al. 1990) which appears to be a highly selective
antagonist for 5-HT
2A
receptors has provided a new tool
for the characterization of these receptors. This com-
pound has been used in several biochemical, electro-
physiological and behavioural studies (Sorensen et al.
1993; Marek and Aghajanian 1994; Schmidt and Fadayel
1995) and subsequently proposed as a putative atypical
antipsychotic with a favorable CNS safety profile
(Schmidt et al. 1995, Kehne et al. 1996). Its labelled
form [
11
C]MDL100,907 was used successfully as a ra-
dioligand for PET studies in primate brain (Lundkvist et
al. 1996). The tritiated form of the ligand, available only
recently, has been used in membrane binding studies
with rat brain cortex confirming its selectivity for 5-HT
2A
receptors (Johnson et al. 1996). In the present study we
report on the characterization of [
3
H]MDL100,907 as a
J.F. López-Giménez · G. Mengod ()) · J. M. Palacios
1
M. Teresa Vilaró
Department of Neurochemistry,
Instituto de Investigaciones Biomédicas de Barcelona,
Consejo Superior de Investigaciones Científicas (CSIC),
Jordi Girona 18–26, E-08034 Barcelona, Spain
1
Permanent address: Research Institute, Laboratorios Almirall,
Cardoner 68–74, E-08024 Barcelona, Spain
ORIGINAL ARTICLE
Juan F. López-Giménez · Guadalupe Mengod
José M. Palacios · M. Teresa Vilaró
Selective visualization of rat brain 5-HT
2A
receptors
by autoradiography with [
3
H]MDL 100,907
Received: 7 April / Accepted: 18 May 1997

radioligand for the selective autoradiographic visualiza-
tion of 5-HT
2A
receptors in rat brain.
Methods
Animal brains. Male Wistar rats (200–300 g) were purchased from
Iffa Credo (Lyon, France). The animals were killed by decapita-
tion and the brains were removed quickly and frozen on dry ice.
Animal care followed the spanish legislation on “Protection of An-
imals Used in Experimental and Other Scientific Purposes”, in
agreement with the E.E.C. regulations on the care of laboratory
animals (O.J. of E.C. L358/1 18/12/1986). Tissue sections, 14 µm
thick, were cut using a microtome-cryostat (Leitz 1720), thaw-
mounted onto APTS (3-aminopropyltriethoxysilane, Sigma, St.
Louis, Mo, USA) coated slides, and kept at –20ºC until use.
Receptor autoradiography. [
3
H]MDL100,907 (82 Ci/mmol) was a
generous gift from Amersham (UK). [
3
H]Mesulergine (89
Ci/mmol) and [
3
H]RP62203 (66 Ci/mmol) were purchased from
Amersham (UK) and [
3
H]Ketanserin (80.9 Ci/mmol) from
DuPontNEN (Boston, Mass., USA). Ketanserin, mesulergine,
mianserin, ritanserin and spiperone were purchased from Research
Biochemicals International (Natick, Mass., USA) and Gpp(NH)p
(5’-Guanylylimidodiphosphate) from Sigma. MDL100,907 was a
gift from Dr. M. Galvan (Marion Merrell Dow, Strasbourg,
France). The experimental incubation conditions for each radio-
ligand are summarized in Table 1. Conditions for incubation were
as described previously with only minor modifications, except for
[
3
H]MDL100,907, which were determined in the present study. In
saturation studies, [
3
H]MDL100,907 was used at increasing con-
centrations of 0.12, 0.22, 0.33, 0.49, 0.60, 0.78, 1.6, 1.9, and 4 nM
to obtain values of total binding. Non-specific binding was deter-
mined at 0.17, 0.52 and 3.2 nM radioligand concentration, and the
values obtained were used to generate a linear regression curve.To
obtain specific binding for each radioligand concentration, the cal-
culated value of non-specific binding was substracted from the
corresponding total binding value. In competition studies, [
3
H]-
MDL100,907 was used at a concentration of 0.4 nM. In both types
of experiments, non-specific binding was defined as that remain-
ing in the presence of 10
–5
M spiperone. In experiments with
Gpp(NH)p the following incubation buffer (Branchek et al. 1990)
was used: 50 mM Tris-HCl pH 7.4, 0.5 mM EDTA, 10 mM
MgSO
4
, 0.1% ascorbic acid, 10 µM pargyline. After incubation
and washing, tissue sections were dipped in distilled water at 4ºC
and dried rapidly under a cold air stream. Tissues were exposed to
tritium-sensitive film (Hyperfilm-
3
H, Amersham, UK) together
with plastic
3
H-standards (
3
H-Microscales, Amersham, UK).
Quantitative analysis of the autoradiograms was done with a com-
puterized image analysis system (MCID, St. Catharines, Ontario,
Canada).
447
Table 1 Incubation conditions for the radioligands used in the present study
Ligand [Ligand] Preincubation Buffer
(a)
Incubation Washing Exposure Blank References
nM protocol temperature Time protocol generation
[
3
H]MDL100,907 0.4 30 min at room 50 mM Tris-HCl Room temp. 60 min 2×10 min 15 days 10
-5
M Johnson
temperature (pH 7,4) (4°C) spiperone et al. 1996;
present work
[
3
H]ketanserin 2 15 min at room 170 mM Tris-HCl Room temp. 120 min 2×10 min 30 days 10
-6
M Pazos et al.
temperature (pH 7.7) (4°C) mianserin 1985; Hoyer
et al. 1986
[
3
H]mesulergine 4.8 Same as above Same as above Room temp. 120 min 2×10 min 44 days 10
-6
M Pazos et al.
(4°C) mianserin 1985; Hoyer et
al. 1986
[
3
H]RP62203 0.3 30 min at room 10 mM PIPES, Room temp. 120 min 2×5 min 13 days 10
-6
M Malgouris et
temperature 200 mM NaCl, (4°C) ritanserin al. 1993 with
10 mM MgCl2, modifications(
b
)
1 mM EGTA (pH 6.5)
a
For each radioligand, the same buffer was used for pre-incubation, incubation and washing
b
See results
Table 2 Binding parameters of [
3
H]MDL100,907 obtained from saturation experiments at two coronal levels of the rat brain
Bregma -0.30 mm Bregma -1.30 mm
n B
max
± SEM K
d
± SEM n B
max
± SEM K
d
± SEM
(fmol/ (nM) (fmol/ (nM)
mg tissue) mg tissue)
Frontoparietal motor cortex
lamina I 4 96.00 ± 4.25 0.20 ± 0.04 4 71.58 ± 2.54 0.19 ± 0.03
lamina II-III 4 93.51 ± 2.52 0.30 ± 0.03 4 65.33 ± 2.35 0.24 ± 0.03
lamina V superficial 4 166.20 ± 4.62 0.22 ± 0.02 4 140.20 ± 5.77 0.20 ± 0.03
lamina V deeper 4 110.20 ± 3.04 0.26 ± 0.03 4 83.25 ± 4.13 0.21 ± 0.04
Caudate-Putamen 4 71.90 ± 2.44 0.29 ± 0.03 2 72.55 ± 3.79 0.20 ± 0.04
Claustrum 4 168.80 ± 5.59 0.22 ± 0.03
n = number of animals

In situ hybridization histochemistry. Oligodeoxyribonucleotide
S2/CL complementary to bases 1482–1520 of the rat 5-HT
2A
re-
ceptor sequence (Pritchett et al. 1988) was labelled with
[
32
P]dATP and terminal deoxynucleotidyltransferase as previously
described (Pompeiano et al. 1994) to specific activities of
1–4×10
4
Ci/mmol. Tissue sections were treated and hybridized as
described (Pompeiano et al. 1994) with the labelled oligonucle-
otide probe at a final concentration of 0.4–0.8 pmol/ml in 40%
formamide/600 mM NaCl/10 mM Tris-HCl, pH 7.5/1 mM ED-
TA/0.02% ficoll/0.02% bovine serum albumin/0.02% polyvi-
nylpyrrolidone/0.5 mg/ml yeast tRNA. Hybridization was carried
out for 18 h under a nescofilm coverslip in a humid chamber at
42ºC. Hybridized sections were washed four times (1 h each) in
600 mM NaCl/10 mM Tris-HCl, pH 7.5/1 mM EDTA at 60ºC and
dehydrated by serial passages in ethanol containing 0.3 M ammo-
nium acetate pH 7.0. Autoradiograms were generated by apposi-
tion of the hybridized sections to β-max film (Amersham, UK) for
two weeks at –70ºC.
Data analysis. Estimates of binding parameters (K
d
, B
max
, IC
50
and
Hill slope) were obtained from the data of autoradiographic quan-
tification of saturation and competition experiments with the
GraphPad Prism software (GraphPad Software Inc., San Diego,
Calif., USA). The program calculated K
i
values according to the
Cheng-Prusoff equation using the IC
50
previously determined in
competition experiments, the mean of the K
d
for the radioligand in
the different brain regions obtained in saturation experiments (see
Table 2), and the concentration of free radioligand used in compe-
tition experiments.
Results
Incubation conditions for [
3
H]MDL100,907
In order to optimize the incubation conditions to visual-
ize [
3
H]MDL100,907 binding sites, two incubation times
(1 and 2 h) at room temperature and different washing
times (2×2, 2×5 and 2×10 min) at 4ºC were examined in
consecutive rat brain sections by quantification of the
autoradiograms. The results obtained are shown in
Fig. 1. There were no significant differences in specific
binding between both incubation times as well as for the
different washing times; non-specific binding was very
low in all conditions and showed a slight decrease with
the longest washing time. We therefore chose an incuba-
tion time of 60 min and 2×10 min washes for further ex-
periments.
Saturation studies with [
3
H]MDL100,907
and regional distribution
The analysis of saturation curves suggested an homoge-
nous and saturable population of binding sites that pre-
sented high affinity for [
3
H]MDL100,907 in every rat
brain region exhibiting [
3
H]MDL100,907 binding. Satu-
ration curves for some selected brain regions are shown
in Fig. 2, and the corresponding K
d
and B
max
values are
listed in Table 2. The affinity constant values were com-
parable in the various rat brain regions and ranged be-
tween 0.20–0.30 nM. Subsequent competition experi-
ments were performed at a concentration of 0.4 nM
[
3
H]MDL100,907, this ligand concentration resulting in
high receptor occupancy and presenting a favorable ratio
of total to non-specific binding.
The highest densities of [
3
H]MDL100,907-labelled
receptors were found in lamina V (frontoparietal motor
cortex), and in the claustrum. Other regions enriched in
receptors include different areas of the olfactory system,
caudate-putamen, nucleus accumbens, CA3 subfield of
the hippocampus and several brainstem nuclei such as
the pontine nuclei, motor trigeminal nucleus and facial
nucleus. (Table 2, Fig. 3A)
448
Fig. 1 Effect of incubation and washing times on the binding of
[
3
H]MDL100,907 to rat brain tissue sections. Each point repre-
sents the mean of 3 animals (error bars = SEM). Solid lines: 1 h
incubation; broken lines: 2 h incubation. ■: frontal cortex (layer
V), specific binding; : frontal cortex (layer V), non-specific
binding. ●: caudate-putamen, specific binding; : caudate-puta-
men, non-specific binding
Fig. 2 Saturation analysis of [
3
H]MDL100,907 binding to several
regions of the rat brain determined by quantification of the autora-
diograms. Each point represents the mean of 4 animals. Filled
symbols: specific binding; open symbols: non-specific binding. ■:
Frontal cortex (layer I); ▼: Frontal cortex (layer V); ●: claustrum;
♦: caudate-putamen

449
Fig. 3 Comparison of the distribution of sites labelled by several
radioligands with high affinity for 5-HT
2A
receptors in parasagittal
sections of the rat brain. Pictures are photomicrographs from film
autoradiograms. A [
3
H]MDL100,907 (0.4 nM), A’ non-specific
binding determined with 10
–5
M spiperone. B [
3
H]ketanserin
(2nM), B’ non-specific binding with 10
–6
M mianserin. C [
3
H]me-
sulergine (4.8 nM), C’ non-specific binding with 10
–6
M mian-
serin. D [
3
H]RP62203 (0.3 nM), D’ non-specific binding with
10
–6
M ritanserin. E distribution of 5-HT
2A
mRNA as visualized
by in situ hybridization with an oligonucleotide probe. E’ Sche-
matic drawing of the level shown in A and E (simplified from
Paxinos and Watson 1982). Abbreviations: CA3, field CA3 of hip-
pocampus; ChP: choroid plexus (indicated by arrowheads); CPu:
caudate-putamen; LC: locus coeruleus; Mo5: motor trigeminal nu-
cleus; Pn: pontine nuclei; SN: substantia nigra; V: layer V of cor-
tex; 7: facial nucleus. Bar: 3 mm
Competition studies with [
3
H]MDL100,907
The compounds MDL100,907, spiperone, ketanserin and
mesulergine, all having high affinity for 5-HT
2A
recep-
tors in rat brain, were used to compete at increasing con-
centrations against [
3
H]MDL100,907 and their IC
50
and
K
i
values were calculated from the data obtained after
quantifying the autoradiograms in different layers of the
frontoparietal motor cortex (lamina II-III, lamina V su-
perficial and lamina V deeper) and in the caudate-puta-
men. logIC
50
, K
i
, and Hill values resulting from these
analyses are summarized in Table 3, and Fig. 4 A,B
shows competition curves generated by the GraphPad

450
Prism software for two representative regions. In all the
regions and for all the compounds, the experimental
points were best fitted to a one-site model. The rank or-
der of affinity for these ligands was the same in all the
rat brain regions examined (MDL100,907 > spiperone >
ketanserin > mesulergine). The affinities of these com-
pounds for [
3
H]MDL100,907-labelled receptors were
compared in different brain regions and a high correla-
tion coefficient (r = 0.99) was obtained (Fig. 4 C,D).
The possible influence of GTP analogues on the bind-
ing of [
3
H]MDL100,907 and on the displacement of
[
3
H]MDL100,907 by 5-HT was studied with Gpp(NH)p
(5’-Guanylylimidodiphosphate), a non-hydrolizable GTP
analogue. Tissue sections were incubated in the presence
of increasing concentrations of Gpp(NH)p and a fixed
concentration (0.4 nM) of [
3
H]MDL100,907. After this
first incubation, tissue sections were removed, 5-HT was
added at a final concentration of 10
–7
M to each incuba-
tion container, and a new series of tissue sections were
incubated. In the first part of the experiment, no effect of
Gpp(NH)p on [
3
H]MDL100,907 specific binding was
observed. Addition of 10
–7
M 5-HT resulted in a partial
inhibition of [
3
H]MDL100,907 binding which was partly
Fig. 4 Competition curves of [
3
H]MDL100,907 binding in rat
frontal cortex, layer V (A) and caudate-putamen (B) by ■:
MDL100,907, ●: spiperone, : ketanserin, : mesulergine. Each
point represents the mean of 3–4 animals. C,D Correlation be-
tween the Ki values of the different competitor drugs in different
brain regions; 1: MDL100,907, 2: spiperone, 3: ketanserin, 4: me-
sulergine
Fig. 5 Effect of increasing concentrations of Gpp(NH)p on the
binding of [
3
H]MDL100,907 (■) and on reversing the agonist
(10
-7
M 5-HT) inhibition of [
3
H]MDL100,907 binding ( ) in rat
frontal cortex, layer V (A) and caudate-putamen (B). Points repre-
sent the mean of 3–4 animals. Values are reported as percentage of
specific binding in the absence of competing drugs

n p IC
50
± SEM K
i
(nM) nH ± SEM
Frontoparietal motor cortex
lamina II-III
MDL100,907 3 9.17 ± 0.03 0.22 0.89 ± 0.05
Spiperone 3 8.69 ± 0.10 0.63 0.85 ± 0.17
Ketanserin 4 8.49 ± 0.05 1.05 0.96 ± 0.09
Mesulergine 4 7.92 ± 0.06 3.85 1.30 ± 0.21
lamina V superficial
MDL100,907 3 9.13 ± 0.02 0.24 1.00 ± 0.07
Spiperone 3 8.67 ± 0.09 0.67 0.85 ± 0.16
Ketanserin 4 8.49 ± 0.07 1.04 1.04 ± 0.14
Mesulergine 4 8.10 ± 0.02 2.48 0.93 ± 0.05
lamina V deeper
MDL100,907 3 9.08 ± 0.02 0.27 1.04 ± 0.05
Spiperone 3 8.69 ± 0.11 0.64 0.84 ± 0.19
Ketanserin 4 8.47 ± 0.06 1.10 1.07 ± 0.11
Mesulergine 4 8.16 ± 0.04 2.21 0.88 ± 0.05
Caudate-Putamen
MDL100,907 3 9.22 ± 0.04 0.20 0.89 ± 0.09
Spiperone 3 8.71 ± 0.10 0.60 0.77 ± 0.14
Ketanserin 4 8.53 ± 0.03 0.96 0.93 ± 0.03
Mesulergine 4 8.08 ± 0.03 2.56 1.02 ± 0.09
reverted by increasing concentrations of Gpp(NH)p, pre-
sumably due to the conversion of high-affinity agonist
state to low-affinity agonist state by the GTP analogue
(Fig. 5).
Visualization of 5-HT
2A
receptor in rat brain
with different radioligands
Figure 3 shows the autoradiographical signal obtained
with [
3
H]MDL100,907 (A) compared to those obtained
with [
3
H]ketanserin (B), [
3
H]mesulergine (C) and
[
3
H]RP62203 (D) at similar sagittal levels of the rat
brain. Also shown are the corresponding non-specific
signals remaining in the presence of the blank-generating
drug. It should be mentioned that a modification was in-
troduced in the previously reported incubation protocol
for [
3
H]RP62203 (Malgouris et al. 1993b). In prelimi-
nary experiments (data not shown) with this radioligand,
the reported conditions (1 h incubation at 37°C) were
compared to incubation at room temperature for 1 h or
2 h. The results of incubation for 1 h at 37°C were fully
comparable to incubation for 2 h at room temperature.
Therefore we adopted this latter protocol.
The patterns of signal distribution for total binding
obtained with the four radioligands present some similar-
ities and some notable differences. Thus, the pattern in
the neocortex was similar for all four radioligands in that
they all label high densities of receptors in layer V,
which correspond to 5-HT
2A
receptors. Other nuclei
which are also labelled by all four radioligands are the
pontine nuclei and the facial and motor trigeminal nuclei
(shown in Fig. 3 only for [
3
H]MDL100,907). In addition
to the sites already mentioned, [
3
H]ketanserin also la-
belled significant densities of sites in locus coeruleus,
substantia nigra, caudate-putamen and nucleus accumb-
ens. These sites were not displaced by the blank-generat-
ing drug (10
–6
M mianserin) and represent other non-se-
rotonergic [
3
H]ketanserin-binding sites (see discussion).
The choroid plexus, which is not labelled by
[
3
H]MDL100,907, is very heavily labelled by [
3
H]Me-
sulergine and [
3
H]RP62203 and moderately by
[
3
H]Ketanserin. Striking differences were observed in
the non-specific signal obtained with the four radio-
ligands. In the case of [
3
H]MDL100,907, non-specific
signal generated by addition of 10
–5
M spiperone was ho-
mogeneus and practically undistinguishable from film
background (Fig. 3A’). In contrast, for the other three ra-
dioligands, much stronger non-specific signals were gen-
erated by the corresponding drugs (10
–6
M mianserin for
[
3
H]ketanserin and [
3
H]mesulergine; 10
–6
M ritanserin
for [
3
H]RP62203) (Fig. 3B’,C’,D’).
Comparison of [
3
H]MDL100,907 binding sites
and 5-HT
2A
receptor mRNA in rat brain
In situ hybridization histochemistry experiments were
carried out to compare the distribution of [
3
H]-
MDL100,907-labelled receptors (Fig. 3A) with that of
5-HT
2A
receptor mRNA (Fig. 3E). A high degree of simi-
larity was observed for both distributions in neocortex
(strong labelling in lamina V), caudate-putamen (stron-
ger signal in caudal parts of the nucleus), olfactory tu-
bercle, and in several brainstem nuclei, e.g. pontine nu-
clei, motor trigeminal nucleus and facial nucleus.
451
n, number of animals
K
i
values were calculated from
IC
50
values according to Cheng
and Prusoff (1973), considering
a mean K
d
of 0.21 nM and a
free radioligand concentration
of 0.4 nM
Table 3 Affinity constants of
several compounds for [
3
H]-
MDL100,907-labelled receptors
in different rat brain regions

Discussion
The main finding of these investigations is that
[
3
H]MDL100,907 used in autoradiographic studies binds to
a single, high affinity, saturable population of sites in the
rat brain that presents the pharmacological and regional
characteristics of 5-HT
2A
receptors. When compared to
other 5-HT
2A
radioligands examined directly in this work
or previously described in the literature ([
3
H]LSD (lyser-
gic acid diethylamide), [
3
H]spiperone, [
3
H]mesulergine,
[
3
H]ketanserin, [
125
I]DOI ((±)-1-(2,5-dimethoxy-4-iodo-
phenyl)-2-aminopropane), [
3
H]DOB (4-bromo-2,5-dim-
ethoxyphenylisopropylamine), [
125
I]7-amino-8-iodoketan-
serin ([
125
I]AMIK), [
3
H]SR46349B (1(Z)-[2-(dimethy-
lamino)ethoxyimino]-1-(2-fluorophenyl)-3-(4-hydroxy-
phenyl)-2(E)-proprene) and [
3
H]RP62203), [
3
H]MDL100,
907 appears to be the first truly selective 5-HT
2A
recep-
tor ligand. The other radioligands mentioned above label
at least one additional site besides the 5-HT
2A
receptor.
[
3
H]LSD (Peroutka and Snyder 1979; Meibach et al.
1980) and [
3
H]spiperone (Palacios et al. 1981) were the
first radioligands used to label or to visualize these re-
ceptors, but it was shown that LSD had affinity for
dopamine D
2
receptors and for 5-HT
1
receptors and spip-
erone for D
2
dopamime receptors. The agonist com-
pounds [
125
I]DOI and [
3
H]DOB have been used to label
5-HT
2A
receptors (McKenna and Saavedra 1987; Teitler
et al. 1990) but they also present high affinity for 5-HT
2C
receptors (Appel et al. 1990; Nazarali et al. 1989), there-
fore it is necessary to block 5-HT
2C
receptors to visualize
only 5-HT
2A
receptors. However this is not possible at
the present moment due to the lack of ligands with high
selectivity for 5-HT
2C
versus 5-HT
2A
receptors. In addi-
tion both radioligands label preferentially the agonist
high affinity state of 5-HT
2A
receptors (Appel et al.
1990; Lyon et al. 1987); therefore, when used as tools to
visualize 5-HT
2A
receptors they provide partial informa-
tion on the total densities of these receptors. The antago-
nist [
3
H]mesulergine was initially described as selective
for serotonin-2 receptors in binding studies with rat cere-
bral cortex membranes (Closse 1983). However it was
shown later that it also labels with high affinity 5-HT
2C
receptors both in binding and in autoradiographic experi-
ments (Pazos et al. 1984, 1985a; Pazos and Palacios
1985). In addition it was demonstrated that this radio-
ligand labelled 5-HT
2A
receptors (at that moment termed
5-HT
2
) in rat brain but not in porcine and human brain
(Pazos et al. 1985b). For all these reasons [
3
H]mesuler-
gine has been more extensively used to visualize 5-HT
2C
than 5-HT
2A
receptors in several species (Pazos et al.
1985a; Hoyer et al. 1986; Mengod et al. 1990). The com-
pound SR46349B has been described as a potent and se-
lective 5-HT
2A
receptor antagonist although it also dis-
played moderate affinity for the 5-HT
2C
receptors in pig
brain cortical membranes (Rinaldi-Carmona et al. 1992).
The tritiated form of this ligand has been used both in in
vitro and in vivo binding experiments (Rinaldi-Carmona
et al. 1993) but not as radioligand for autoradiographical
visualization. Another recently available antagonist,
RP62203, relatively selective for 5-HT
2A
receptors (Do-
ble et al. 1992) has been used in its tritiated form for the
autoradiographical visualization of these receptors in rat
brain (Malgouris et al. 1993b). In addition to labelling
5-HT
2A
receptors the radioligand labels other serotonin
receptor sites wich are especially prominent in choroid
plexus. The authors of that study do not favour the hy-
pothesis that the [
3
H]RP62203 binding sites in this struc-
ture represent 5-HT
2C
receptors. However, given the very
high density of 5-HT
2C
receptors present in choroid plex-
us, and the relatively high affinity of RP62203 for
[
3
H]mesulergine binding sites in this area (IC
50
=
29.2 nM; Malgouris et al. 1993a), the possibility that
[
3
H]RP62203 is labelling 5-HT
2C
sites in the rat choroid
plexus, and by extension in the rat brain, cannot be com-
pletely ruled out. In addition this radioligand presents
very high levels of non-specific binding (Malgouris et al.
1993b; present study) which makes it of limited utility in
autoradiographic studies. It has also been recently de-
scribed that RP62203 presents high affinity for cloned
human D
4
dopamine receptors expressed in Chinese
hamster ovary cells (Heuillet et al. 1996).
[
3
H]ketanserin has been the radioligand most exten-
sively used to study or to visualize 5-HT
2A
receptors
since its description (Leysen et al. 1982). However, in
the initial autoradiographical studies (Pazos et al.
1985a), it was already observed that in some areas of the
rat brain containing very high densities of [
3
H]ketan-
serin-binding sites, mainly the caudate-putamen and the
dorsal raphe, most of the binding was not inhibited by
any serotonergic drug. These sites were considered as
“ketanserin chemical recognition sites” rather than sero-
tonergic receptors. Subsequently, the results of mem-
brane binding and autoradiographic studies with
[
3
H]ketanserin (Leysen et al.,1987; Roth et al. 1987)
showed that these ketanserin recognition sites were relat-
ed to a DOPAC (dihydroxyphenylacetic acid) release
system on dopaminergic terminals in rat striatum, and
were blocked by tetrabenazine. Furthermore, ketanserin
was also shown to present relatively high affinities for
H1 histamine receptors in guinea pig cerebellum and for
α
1
adrenoceptors in rat forebrain (Leysen et al. 1981).
Further studies with [
125
I]AMIK ([
125
I]7-amino-8-iodo-
ketanserin) also identified the non-serotonergic binding
sites of this radioligand as α
1
adrenoceptors and amine
release sites in rat brain (Schotte and Leysen 1988;
1989). These non-5-HT
2A
sites for which [
3
H]ketanserin
has relatively high affinities can account for the high
densities of [
3
H]ketanserin-binding sites that remain in
certain regions of the rat brain after the addition of
10
–6
M mianserin to generate the blank: i.e. α
1
ad-
renoceptors in the locus coeruleus and sites involved in
DOPAC release in the caudate-putamen (Fig. 5B’).
In view of all this data, it seems evident that
[
3
H]MDL100,907 is superior to other radioligands hith-
erto used for the visualization of 5-HT
2A
receptors be-
cause it does not label any additional receptor sites. Fur-
thermore, the extremely low non-specific binding shown
452

In conclusion, [
3
H]MDL100,907 appears to be the
first truly selective 5-HT
2A
radioligand allowing for a di-
rect visualization, quantification and characterization of
these receptor sites without the need of blocking the dif-
ferent additional sites which are labelled by the other ra-
dioligands used until now to label these receptors. Con-
sequently, [
3
H]MDL100,907 is at present the radioligand
of choice for autoradiographical studies of 5-HT
2A
recep-
tors.
Acknowledgements This work was supported by a grant from
CICYT (SAF96-0336). J.F.L-G. is a recipient of a fellowship
“Beca de Formació de Personal Investigador Institut d’Investigac-
ions Biomèdiques August Pi i Sunyer”. Support from the CIRIT
(Generalitat de Catalunya) to the Department of Neurochemistry
(IIBB/CSIC) as Grup de Recerca de Qualitat (1995SGR00445) is
acknowledged.
References
Appel NM, Mitchell WM, Garlick RK, Glennon RA, Teitler M,
De Souza EB (1990) Autoradiographic characterization of (±)-
1-(2,5-dimethoxy-4-[
125
I]iodophenyl)-2-aminopropane ([
125
I]-
DOI) binding to 5-HT
2
and 5-HT
1C
receptors in rat brain. J
Pharmacol Exp Ther 255:843–857
Battaglia G, Shannon M, Titeler M (1984) Guanyl nucleotide and
divalent cation regulation of cortical S
2
serotonin receptors. J
Neurochem 43:1213–1219
Baxter G, Kennet G, Blaney F, Blackburn T (1995) 5-HT
2
receptor
subtypes: a family re-united? Trends Pharmacol Sci
16:105–110
Blue ME, Yagaloff KE, Mamounas LA, Hartig PR, Molliver ME
(1988) Correspondence between 5-HT
2
receptors and sero-
toninergic axons in rat neocortex. Brain Res 455:315–328
Branchek T, Adham N, Macchi M, Kao HT, Hartig PR (1990)
[
3
H]-DOB(4-bromo-2,5-dimethoxyphenylisopropylamine) and
[
3
H]Ketanserine label two affinity states of the cloned human
5-Hydroxytryptamine
2
receptor. Mol Pharmacol 38:604–609
Cheng YC, Prusoff WH (1973) Relationship between the inhibi-
tion constant (K
I
) and the concentration of the inhibitor wich
causes 50 per cent inhibition (IC
50
) of an enzymatic reaction.
Biochem Pharmacol 22:2099–3108
Closse A (1983) [
3
H]Mesulergine, a selective ligand for serotonin-
2 receptors. Life Sci 32:2485–2495A
Doble A, Girdlestone D, Piot O, Allam D, Betschart J, Boireau A,
Dupuy A, Guérémy C, Ménager J, Zundel JL, Blanchard JC
(1992) Pharmacological characterization of RP 62203, a novel
5-Hydroxytryptamine 5-HT
2
receptor antagonist. Br J Pharma-
col 105:27–36
Dudley M, Ogden A, Carr A, Nieduzak T, Kehne J (1990) (±-(2,3-
Dimethoxyphenyil)-1-[2-(4-Fluorophenylethyl)]-4-Piperidine-
methanol), a potent, chiral, 5-HT
2A
receptor antagonist. Ab-
stract 427.19, Society for Neuroscience, St. Louis
Heuillet E, Petitet F, Mignani S, Malleron JL, Lavayre J, Néilat G,
Doble A, Blanchard JC (1996) The naphtosulfam derivative
RP 62203 (fananserine) has high affinity for the dopamine D
4
receptor. Eur J Pharmacol 314:229–233
Hoyer D, Pazos A, Probst A, Palacios JM (1986) Serotonin recep-
tors in the human brain. II. Characterization and autoradio-
graphic localization of 5-HT
1C
and 5-HT
2
recognition sites.
Brain Res 376:97–107
Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylech-
arane EJ, Saxena PR, Humphrey PPA (1994) VII. International
union of pharmacology cassification of receptors for 5-Hy-
droxytryptamine (serotonin). Pharmacol Rev 46:157–203
Johnson MP, Siegel BW, Carr AA (1996) [
3
H]MDL 100,907:
a novel selective 5-HT
2A
receptor ligand. Naunyn-Schmiede-
berg’s Arch Pharmacol 354:205–209
453
by this radioligand in membrane (Johnson et al. 1996) as
well as in autoradiographic studies (present results) rep-
resents an additional advantage.
Several compounds with high affinity for 5-HT
2A
re-
ceptors have been used here to establish the pharmaco-
logical profile of the receptors labelled by
[
3
H]MDL100,907 in various regions of the rat brain. The
results are consistent with the radioligand labelling
5-HT
2A
receptors in all the regions and are in good
agreement with results from membrane binding experi-
ments with this radioligand (Johnson et al. 1996). Fur-
thermore the affinities of these compounds in different
regions show a very good correlation, suggesting identi-
cal populations of receptors.
As expected for an antagonist ligand (Dudley et al.
1990; Sorensen et al. 1993), the binding of
[
3
H]MDL100,907 was not affected by the non-hydroliz-
able GTP analogue Gpp(NH)p. A similar lack of effect of
GTP analogues on the binding of the antagonist
[
3
H]Ketanserin has been described in cell lines transfected
with the cloned rat 5-HT
2A
receptor (Teitler et al. 1990). In
contrast, increasing concentrations of Gpp(NH)p inhibit-
ed agonist (5-HT) binding to 5-HT
2A
receptors labelled
by [
3
H]MDL100,907. The same situation has been ob-
served for the binding of 5-HT to [
3
H]ketanserin-labelled
receptors, both in rat brain membranes (Battaglia et al.
1984) and in cell lines expressing cloned 5-HT
2A
recep-
tors (Branchek et al. 1990). This is a general effect ob-
served for many receptor systems (Lefkowitz et al. 1983)
due to an uncoupling of the receptor-G-protein complex
which results in a decrease in agonist affinity.
The close relationship between the distribution of
[
3
H]MDL100,907-labelled receptors and that of 5-HT
2A
mRNA as seen by in situ hybridization histochemistry
(Pompeiano et al. 1994; present results) further supports
that the sites labelled by the radioligand correspond to
5-HT
2A
receptors and that these receptors are most prob-
ably localized preferentially on the cell bodies and den-
drites of the cells that contain the mRNA. This subcellu-
lar localization is also supported by immunocytochemi-
cal studies in which a polyclonal antibody against
5-HT
2A
receptors labelled perykaria and dendritic ar-
borizations (Morilak et al. 1993).
The laminar distribution of 5-HT
2A
receptors in rat
neocortex has been differently described in the literature.
In several autoradiographical studies, the highest densi-
ties of neocortical 5-HT
2A
receptors labelled with a vari-
ety of radioligands (Palacios et al. 1981; Pazos and Pal-
acios 1985; Appel et al. 1990; Malgouris et al. 1993a,b)
were assigned to layer IV. However, a detailed study
(Blue et al. 1988) where a systematic comparison of the
distribution of 5-HT
2A
binding sites with the correspond-
ing Nissl-stained tissue sections was performed, demon-
strated that the highest densities of receptors correspond
to the superficial half of layer V (also termed layer Va).
In the present study, we have followed the criteria of
Blue and collaborators for the description of the laminar
distribution of neocortical 5-HT
2A
receptors in rat brain.

Kehne JH, Baron BH, Carr AA, Chaney SF, Elands J, Feldman DJ,
Frank RA, Van Giersbergen PLM, McCloskey TC, Johnson
MP, McCarty DB, Poirot M, Senyah Y, Siegel BW, Widmaier
C (1996) Preclinical characterization of the potential of the pu-
tative antipsychotic MDL 100,907 as a potent 5-HT
2A
antago-
nist with a favorable CNS safety profile. J Pharmacol Exp
Ther 277:968–981
Lefkowitz RJ, Stadel JM, Caron MG (1983) Adenylate cyclase-
coupled beta-adrenergic receptors: structure and mechanisms
of activation and desensitization. Annu Rev Biochem
52:159–186
Leysen JE, Awouters F, Kennis L, Laduron PM, Vandenberk J,
Janssen PAJ (1981) Receptor binding profile of R 41 468, a
novel antagonist at 5-HT
2
receptors. Life Sci 28:1015–1022
Leysen JE, Niemegeers CJE, Van Neuten JM, Laduron PM (1982)
[
3
H]Ketanserine (R 41 468), a selective
3
H-ligand for seroto-
nin
2
receptor binding sites. Mol Pharmacol 21:301–314
Leysen JE, Eens A, Gommeren W, Van Gompel P, Wyants J, Jan-
ssen AJ (1987) Non-serotoninergic [
3
H]ketanserine binding
sites in striatal membranes are associated with a dopac release
system on dopaminergic nerve endings. Eur J Pharmacol
134:373–375
Lundkvist C, Halldin C, Ginovart N, Nyberg S, Swahn C-G, Carr
AA, Brunner F, Farde L (1996) [
11
C]MDL 100907, a radio-
ligand for selective imaging of 5-HT
2A
receptors with positron
emission tomography. Life Sci 58:187–192
Lyon RA, Davis KH, Titeler M (1987) [
3
H]DOB (4-bromo-2,5-
dimethoxyphenylisopropylamine) labels guanyl nucleotide-
sensitive state of cortical 5-HT
2
receptors. Mol Pharmacol
31:194–199
Malgouris C, Flamand F, Doble A (1993a) Autoradiographic stud-
ies of RP 62203, a potent 5-HT
2
receptor antagonist. In vitro
and ex vivo selectivity profile. Eur J Pharmacol 233:29–35
Malgouris C, Flamand F, Doble A (1993b) Autoradiographic stud-
ies of RP 62203, a potent 5-HT
2
receptor antagonist. Pharma-
cological characterization of [
3
H]RP 62203 binding in the rat
brain. Eur J Pharmacol 233:37–45
Marek GJ, Aghajanian GK (1994) Excitation of interneurons in
piriform cortex by 5-hydroxytryptamine: blockade by MDL
100,907, a highly selective 5-HT
2A
receptor antagonist. Eur J
Pharmacol 259:137–141
McKenna DJ, Saavedra JM (1987) Autoradiography of LSD and
2,5-dimethoxyphenylisopropylamine psychomimetics demon-
strates regional, specific cross-displacement in the rat brain.
Eur J Pharmacol 142:313–315
Meibach RC, Maayani S, Green JP (1980) Characterization and
radioautography of [
3
H]LSD binding by rat brain slices in vit-
ro: the effect of 5-hydroxytryptamine. Eur J Pharmacol
67:371–382
Mengod G, Nguyen H, Le H, Waeber C, Lübbert H, Palacios JM
(1990) The distribution and cellular localization of the seroto-
nin 1C receptor mRNA in the rodent brain examined by in situ
hybridization histochemistry. Comparison with receptor bind-
ing distribution. Neuroscience 35:577–591
Morilak DA, Garlow SJ, Ciaranello RD (1993) Immunocytochem-
ical localization and description of neurons expressing Seroto-
nin
2
receptors in the rat brain. Neuroscience 54:701–717
Nazarali AJ, McKenna DJ, Saavedra JM (1989) Autoradiographic
localization of 5HT2 receptors in rat brain using [125I]-DOI, a
selective psychomimetic radioligand. Prog Neuropsychophar-
macol Biol Psychiatry 13:573–581
Palacios JM, Niehoff DL, Kuhar MJ (1981) [
3
H]Spiperone bind-
ing sites in brain: autoradiographic localization of multiple re-
ceptors. Brain Res 213:277–289
454
Paxinos G, Watson C (1982) The rat brain in stereotaxic coordi-
nates. Academic Press, New York
Pazos A, Hoyer D, Palacios JM (1984) The binding of serotonin-
ergic ligands to the porcine choroid plexus: characterization of
a new tipe of recognition site. Eur J Pharmacol 106:539–546
Pazos A, Palacios JM (1985) Quantitative autoradiographic map-
ping of serotonin receptors in the rat brain. I. Serotonin-1 re-
ceptors. Brain Res 346:205–230
Pazos A, Cortés R, Palacios JM (1985a) Quantitative autoradio-
graphic mapping of serotonin receptors in the rat brain. II. Se-
rotonin-2 receptors. Brain Res 346:231–249
Pazos A, Hoyer D, Palacios JM (1985b) Mesulergine, a selective
serotonin-2 ligand in the rat cortex, does not label these recep-
tors in porcine and human cortex: evidence for species differ-
ences in brain serotonin-2 receptors. Eur J Pharmacol
106:531–538
Peroutka SJ, Snyder SH (1979) Multiple serotonin receptors: dif-
ferential binding of [
3
H]5-hydroxytryptamine, [
3
H]lysergic ac-
id diethylamide and [
3
H]spiroperidol. Mol Pharmacol
16:687–699
Pompeiano M, Palacios JM, Mengod G (1994) Distribution of the
serotonin 5-HT
2
receptor family mRNAs: comparison between
5-HT
2A
and 5-HT
2b
receptors. Mol Brain Res 23:163–178
Pritchett DB, Bach AWJ, Wozny M, Taleb O, Dal Toso R, Shih J,
Seeburg PH (1988) Structure and functional expression of
cloned rat serotonin 5HT-2 receptor. EMBO J 7:4135–4140
Rinaldi-Carmona M, Congy C, Santucci V, Simiand J, Gautret B,
Neliat G, Labeeuw B, Le Fur G, Soubrie P, Breliere JC (1992)
Biochemical and pharmacological properties of SR 46349B, a
new potent and selective 5-Hydroxytryptamine2 receptor an-
tagonist. J Pharmacol Exp Ther 262:759–768
Rinaldi-Carmona M, Congy C, Pointeau P, Vidal H, Brelière JC,
Le Fur G (1993) Identification of binding sites for SR 46349B,
a 5-Hydroxytryptamine2 receptor antagonist, in rodent brain.
Life Sci 54:119–127
Roth BL, McLean S, Zhu XZ, Chuang DM (1987) Characteriza-
tion of two [
3
H]ketanserine recognition sites in rat striatum. J
Neurochem 49:1833–1838
Schmidt CJ, Fadayel GM (1995) The selective 5-HT
2A
receptor
antagonist, MDL 100,907, increases dopamine efflux in the
prefrontal cortex of the rat. Eur J Pharmacol 273:273–279
Schmidt CJ, Sorensen SM, Kehne JH, Carr AA, Palfreyman MG
(1995) The role of 5-HT
2A
receptors in antipsychotic activity.
Life Sci 56:2209–2222
Schotte A, Leysen JE (1988) Distinct autoradiographic labelling of
serotonin 5-HT
2
receptors, 1-adrenoceptors and histamine-H1
receptors and of tetrabenazine-displaceable ketanserine bind-
ing sites in rodent brain with [
125
I]7-amino-8-iodo-ketanserine.
Eur J Pharmacol 145:213–216A
Schottte A, Leysen JE (1989) Identification of 5-HT
2
receptors,
1
-adrenoceptors and amine release sites in rat brain by autorad-
iography with [
125
I]7-amino-8-iodo-ketanserine. Eur J Pharma-
col 172:99–106
Sorensen SM, Kehne JH, Fadayel GM, Humphreys TM, Ketteler
HJ, Sullivan CK, Taylor VL, Schmidt CJ (1993) Characteriza-
tion of the 5-HT
2
receptor antagonist MDL 100907 as a puta-
tive atypical antipsycotic: behavioral, electrophysiological and
neurochemical studies. J Pharmacol Exp Ther 266:684–691
Teitler M, Leonhardt S, Weisberg EL, Hoffman BJ (1990) 4-[
125
I]-
Iodo-(2,5-dimethoxy)phenylisopropylamine and [
3
H]Ket-
anserine labeling of 5-Hydroxytryptamine2 (5HT
2
) receptors
in mammalian cells transfected with a rat 5HT
2
cDNA: evi-
dence for multiple states and not multiple 5HT
2
receptor sub-
types. Mol Pharmacol 38:594–598