PII SOO24-3205(98)00478-O

Life Sciences, Vol. 63, No. 22, pp. 2001-2013, 1998

Copyright 0 1998 Ekvier Science Inc.

Printed in the USA. All rights reserved

om4-3205/98 $19.00 + .oo

1231-MSP~~F[11C]MSP: NEWSELECTIVE ~-HT~A RECEPTORRADIOPHAFWWEUTICALSFORIN VIVO STUDIESOFNEURONALM~T~ SEROTONIN RECEPTORS. S~NTHESIS,INV~~B~DINGST~DY

WlTHUNLAE3ELLEDANALoGUESANDPRELIMINARYINVIVOEVALUATIONINMICE

Samuel Samnickl, Nathalie Remy2, Simon Ametamey2, Jochen B. Baderl, Wolfgang Brandau3 and Carl-Martin Kirschl

Departments of Nuclear Medicine, 1 D-6642 1 Homburg/&r and 3University of Essen, D-45 122 Essen; 2Paul Schqrer Institut, Centre of Radiopharmacy, CH-5232 Villigen-PSI.

(Received in final form September 17, 1998)

Summary

In vine bindin [3H]spiperone,

study on bovine brain membranes using [3H]SCH23390, [ B HJprazosin and [3H]RP62203 as radioligands (for Dl, D2, ~1 and

5-HT~A receptors respectively) indicate that the new butyrophenones 8-[3-(4-fluoro- benzoyl)propyl]-l-methyl-1,3,8-triazaspiro[4,5]d~~-one (FMSP) and 8-[3-(4- iodobenzoyl)propyl]- 1 -methyl- 1,3,8-triazaspiro[4,5]decan-4-one (IMSP) exhibit a significantly higher selectivity for the 5-I-IT A over Dl, D2 and al receptors. Consequently, the radiolabelled analogues F[ If C]MSP and 1231-MSP were prepared in attempt to obtain potential radiopharmaceuticals for in viva imaging of neuronal S-I-IT~A receptors with positron emission tomography (PET) and single photon emission tomography (SPET). F[ 1 lC]MSP was synthesized by reaction of [ 1 1C]CH31 with 8-[3-(4-fluorobenzoyl)propyl]-1,3,8-triazaspiro[4,5]decan-4-one (DMSP) in 12 + 3 % radiochemical yield, whereas 1231-MSP was obtained in 82 + 8 % radiochemical yield by a no-carrier-added Cu(I)-assisted [ 123Iliododebromination of 8-[3-(4-bromo-benzoyl)propyl]-l-methyl-l,3,8-triazaspiro[4,5]decan-4-one (BrMSP). /n viw pharmacokinetic and brain binding characterization of 1231-MSP assessed in mice following intravenous injection, showed a fast clearance of l231-MSP from blood and relatively high initial uptakes in the liver, kidneys and in the lung. Significant uptake and long retention were observed in the brain (up to 1.64 % i.d., 60 min p.i.), with a regional accumulation of radioactivity consistent with the reported ~-HT~A receptors distribution in the brain. Frontal cortex to cerebellum ratio of 3.5 was calculated at 60 min p.i. Furthermore, the initial brain uptake was significantly reduced after pretreatment of the animals with ritanserin, a selective 5-HT2 antagonist, and by preinjection of the non-radiolabelled anal0 IMSP, thus indicating the specificity of the brain uptake. These data suggest that ?23I-MSP may be a promising compound for studying the serotoninergic 5-HT2 receptors with SPET. Due to the low specific activity of F[l lC]MSP currently obtained by the [llC]methylation reaction, systematic in viva investigation of F[ 1 lC]MSP are as yet not feasable.

Key Work: S-HIT, receptors, 1”I-labellin~“C-labellin~SPETradiopharmaceuticals,PETradiopha~aceuticals

Corrcqor&nt~ to: Dr. Samuel Samnick, Department of Nuclear Medicine, University Medical School, D- 66421 Homburg/Saar.Tel.: (+49)6841_164682;Fax. (+49) 6841-164692; e-Mail: rassam@,med-rz.uni-sb.de

2002 New Seleche Radiotraccrs for 5-HT, Receptors Vol. 63. No. 22, 1998

Alterations in serotoninergic system have been implicated in various neuropsychiatric disorders such as depression, anxiety, schizophrenia and Alzheimer’s disease (1, 2, 3). More specifically, results of in vifru studies performed on human brain specimens obtained postmortem and few PET studies could demonstrate pathological changes in the density of the 5-HT2 receptors (4, 5, 6).

Thus, in viva imaging of brain 5-HT2 receptors by positron emission tomography (PET) and

single-photon emission tomography (SPET) has been pursued for several years (7, 8, 9, 10). Unfortunately, the number of S-HT2 radioligands that firlfil the requirements for PET or SPET is

limited. Most of the currently available 5-HT2 ligands display high in vivo nonspecific binding,

because of the lack of selectivity for S-HT2 mainly due to their affinity for other neuroreceptors

including Dr. DZ and al receptors, the high lipophilicity and /or too rapid dissociation constants. Nevertheless, interest in the development of radiopharma-ceuticals labelled with single photon emitting radionuclides, suitable for a widespread clinical application with SPET is considerably increased since potential S-HT2 receptor imaging agents for SPET imaging have been reported

(13, 14, 15). However, evaluation of their clinical potential has not been encouraging or remains to be investigated. The starting point for the present work was the finding that substitution of the anilino phenyl group on the N-l nitrogen of spiperone with other alkyl moities, might result in an increase of selectivity of the resulted spiperone derivatives for ~-HT~A versus dopamine and other

neuro-receptors (16, 17). Therefore, we prepared FMSP, BrMSP and IMSP (Fig. I), which could be rapidly labelled with positron or single-photon emitters without being so lipophilic as to give high non-specific binding after a possible penetration of the blood-brain barrier. This report describes the m vitro receptor binding study with the new compounds on bovine brain membranes,

the simple one-step preparation of the radiolabelled analogues F[“C]MSP and 1231-MSP, the

pharmacokinetic and the in viva brain distribution of 1231-MSP in mice, with a view to determine their potential as possible radioligands for PET and SPET investigation of 5-HTzA receptors.

SPIPERONE X-F. Y=CHgFMSP

X = Br, Y = CH 3: BrMSP

x=1, Y = CH?: IMSP

Fig. 1 Structures of spiperone and the new N-l alkylated derivatives

Materials and methods

General 8-[3-(4-Fluorobenzoyl)propyl]-l-methyl-l,3,8-triazaspiro[4,S]decan-4-one (FMSP), 8-[3-(4-

fluorobenzoyl)propyl]-1,3,8-triazaspiro[4,5]decan-4-one (DMSP), 8-[3-(4-bromobenzoyl)propyl]- 1 -methyl- 1,3,8-triazaspiro[4,5]decan-4-one (BrMSP), 8-[3-(4-iodobenzoyl)propyl]-l-methyl- 1,3,8-triazaspiro[4,5]decan-4-one (IMSP) and 8-[3-(4-tributylstannylbenzoyl)propyl]-l-methyl-

Vol. 63, No. 22, 1998 New Selective Radiotracers for 5-HT2 Receptors 2003

1,3,8-triazaspiro[4,5]decan-4-one (TBMSP) were prepared as described (17) or according to the method prleviously described (18, 19). Ritanserin and ketanserin, used for blocking experiment

were purc:hased commercially from RBI (Rahn AG, Zurich). [3H]RP62203, [3H]SCH23390,

[3H]spiperone, [3H]prazosin and other [3H]radioligands including, [3H]MK801, [3H]NBTI,

[3H]AMPA, [3H]paroxetine, [3H]pirenzepine, [3H]cytisine and [3H]CPDPX were purchased via Biotrend Chemicals (Kiiln, Germany), NEN Research Products (Dreieich, Germany) or C.E.A (France) and were routinely used as radioligands for in vitro receptor screening according to standard ml vitro techniques (32). The initial specific activities were determined to be 50 - 100

Ci/mmol. [ IIC]methyl iodide was prepared from [11C]C02 obtained via 14N(p,a)llC reaction in a

sequence of reactions involving the reduction of [llC]CO2 with lithium aluminium hydride and subsequent hydrolysis with hydriodic acid (57%) according to the standard procedure described by

Crouzel et al. 1987 (20). [ 1 IC]methyl iodide was obtained on average in more than 80% yield.

Sodium [1:!3I]iodide was obtained carrier free in alkaline solution in the highest obtainable radiochemic;al purity (FZ Karhuhe, Germany). Isocratic HFLC separation of the radiolabelled compounds was performed on a Latek system (Latek, Eppelheim, Germany) using an Cl8

BonclonelO column (300x8 mm; Phenomenex) and 1% aqueous triethylammonium phosphate (pH

= 7.1) / MeCN (70:30 v/v) as eluent at 3 mL/min and an Cl8 p-Bondapak column (300 x 3.9 mm;

Waters) and O.l%(v/v) H3P04:EtOH (85:15) as eluent at 1.5 mL/min for the isolation of

F[“C]MSP and 1231-MSP respectively. The eluents were monitored with an ultraviolet-detector (SPDdAV, Shimadzu, Japan) at 254 nm and a sodium iodide scintillation detector (Berthold,

Wildbad, Germany).

Membrane preparation and in vitro binding studj In vitro binding experiments were carried out to determine the affinities of FMSP, BrMSP and

IMSP for different neuronal receptors, more specifically for 5-HT2A, Dl, D2 and CL~ receptors

using [3H]RP62203, [3H]SCH23390, [3H]spiperone and [3H]prazosin as radioligands and bovine brain membranes.

Vanities for al and 5-HTz* receptors were determined on bovine frontal cortex tissue homogenates. In detail: Tissues obtained from a local slaughterhouse were homogenized in 10 volumes of 0.32 M sucrose and the homogenate centrifuged at 700 g for 10 min. at 4 “C followed by centrifbgation of the supernatant at 50 000 g for 10 min. The resulting pellet was resuspended in 10 volumes of 50 nih4 Tris-HCl buffer (pH = 7.4). After an incubation period of 10 min at 37 “C the suspension was centritiged once again at 50 000 g. The final pellet was resuspended in 40 volumes of buffer containing 50 mM Tris-HCl (pH = 7.7), 4 mM CaC12, 10 PM pargylin, 0.1 % ascorbic acid and stored at -80 “C. For the binding assay, 100 ~1 of the membrane suspension

(2.5 mg fres.h weight), 50 pL of [3H]prazosin or [3H]RP62203 ligands (1 nM) and 50 pL of the test compou:nd were added to a well of a micro titer plate. The plates were incubated for 30 min. at 37 “C and rapidly filtered through Whatman GF/B filters. Subsequent, filters were washed rapidly with 10 mL ice-cold Tris buffer and transferred to plastic T-trays. After addition of scintillant, the

filters were counted in a Betaplate TM liquid scintillation counter (Wallac). To determine unspecific binding 1 pb/i ketanserin or prazosin were used respectively. To determine the Dl I Dz receptor affinity, the corpora striata were prepared from bovine brain and the tissue homogenized in 50 volumes of 50 mM Tris-HCl (pH = 7.4). This homogenate was centrifuged at 48 000 g for 10 min at 4°C and the resulting pellet resuspended in 10 volumes of 50 niM Tris-HCl buffer. The suspension was centritiged once again at 48000 g for 10 min. and the final pellet resuspended in 30 volumes of Tris-HCl buffer. The protein concentration was adjusted to 1 mg/mL and the aliquots stored at -80 “C. For the binding assay 100 FL of the membrane suspension, 50 pL of

2034 New Selective Radiotracers for 5-HT2 Receptors Vol. 63, No. 22, 1998

[3H]SCH-23390 or [3H]spiperone (0.5 nM) ligand and 50 uL of the test compound were added to a well of a micro titer plate, followed by a similar treatment as described above. To determine unspecific binding 1 uM XXI23390 (Dl) and d-butaclamol (D2 receptor) were used. All

incubations were done in duplicate and data from preliminary screening experiments were analyzed to determine the concentration required to inhibit 50% of the binding of the radioligand (IC50)

followed by determination of the dissociation constants (Ki) using the method of Cheng and

Prusoff and as described previously (15, 21).

Radiosynthesis

1231-MSP was prepared by non-carrier-added Cu(I)-assisted, nonisotopic [123I]iodineibromine

exchange. For this purpose, a solution of sodium [ 1231]iodide (300-500 MBq in 50-100 pL 0.01 N

NaOH) and 5 uL aqueous Na2S205 (4.0 mg Na2S205/mL H20) was evaporated to dryness by

passing a stream of nitrogen through a reaction vessel at lOO’C, followed by addition of 50 pL of

the bromoderivative BrMSP (0.25 mg/O. 1 mL AcOH) and 5 pL of Cu(I)-solution. The reaction vessel was heated for 15 min at 160°C in a Heating Module (Pierce), cooled and‘ the residue

dissolved in up to 200 pL EtOH for the HPLC purification. Alternatively, 1231-MSP was prepared

by the reaction of the tributylstannyl precursor TBMSP with sodium [123I]iodide in hydrochloric

acid solution in the presence of chloramin-T (20 pL, 1 mg/ml water). After 10 min at room

temperature, 25 p.L aqueous Na2S205 (4.0 mg Na2S205/mL water) were added, followed by

neutralisation with sodium bicarbonate. In both cases, 1231-MSP was separated from unreacted starting materials and radioactive impurities by means of isocratic HPLC. The fraction containing

1231-MSP was collected into test tubes, buffered with 0.6 M phosphate buffer (Braun), diluted in saline (0.9%, Braun) and sterile-filterted (0.22 mm, Millipore) in a evacuated steril tubes for in vivo studies.

For the preparation of F[l k]MSP, 0.5 - 0.8 mg of the desmethyl precursor DMSP was dissolved

in 200 pL dry dimethyl formamide (DMF), and 200 uL of DMF containing 2 mg/mL NaH (95%) was added. The reaction mixture was stirred under N2 for 10 min at room temperature. The reacti-

vial8 containing the solution was then positioned in a quartz lamp-heated vessel and [l lC]methyl iodide was added to the solution via a slow stream of N2. The N2 flow was stopped, followed by

heating for 5 min at 60°C and separation of F[IlC]MSP from radioactive impurities by means of isocratic reversed-phase HPLC.

The in vitro stability and purity of the isolated products were determined by rechromatography of the sterile-filtered fractions by analytical reversed-phase HPLC 5, 60, 120 and 180 min

(F[llC]MSP), and additionaly 6 h and 24 h (1231-MSP) after isolation.

Determination of octanol-water partition coefficients Octanol - water partition coefficient was determined at pH = 7.4 as follows: 1 g of aqueous

buffer solution was added to approx. 0.5 - 1.0 MBq of the purified radiopharmaceuticals. Atter addition of 1 g n-octanol, the mixture was vortexed vigorously for 5 min. The layers were separated by centrifugation and 500 mL aliquots of the organic and aqueous phase were monitored for radioactivity. The procedure was repeated twice by addition of fresh buffer to the octanol and fresh octanol to the aqueous phases until constant partition coefficients were obtained. The partition coefficient was calculated as: log P = log (cpmo,tanol / cpmbuEer), and compared to the

1ogP values of the non-radiolabelled derivatives determined by HPLC.

Vol. 63, No. 22, 1998 New Selective Radiotracers for S-l-IT2 Receptors 2005

In vivo experiments Biodistribution study and blocking experiments were carried out in female ICR mice weighing

25 - 30 g, according to the regulations for animal research of the Veterinary Health Authorities of the Canton Aargau and the Saarland. After a single intravenous (i.v) administration of 1.5 - 3.0

MBq of 1231-MSP (in 0.1 - 0.2 mL solution). Animals were held in metabolic cages and were sacrified (3 mice per time point) 5, 30, 60, 120, 240 min and 24 h post injection (p.i). Samples of blood and ‘organs of interest were removed, blotted dry and weighed. Activity was measured on a Packard auto-gamma 500 scintillation counter after a reference sample of the injected dose was prepared and counted. AtIer correction for physical decay, percent injected dose per gram organ (% I.D./g) was calculated for each organ. The brain/blood ratios were calculated from the corresponding % ID/g organ values. The brain was dissected and the following regions were isolated: cerebellum, frontal cortex, temporal cortex, striatum and brain stem. In separate

experiments 1231-MSP was injected 30 min after pretreatment of the animals with the W-IT2

antagonists ritanserin and ketanserin as well as with the non-radiolabelled analog IMSP (2.5 mg/kg, respectively). The animals were sacrified 60 min pi. and tissue radioactivities assayed as described above.

Determination ofplasma metabolites, protein and blood cell binding

For the determination of protein and blood cell binding, the HPLC purified 1231-MSP (1 MBq) was mixed with 2 mL mice blood. After incubation for 10 min at ambient temperature cells were separated by centritbgation at 3000g for 20 min and the supematant ultrafiltered through membranes with a cut-off of 30000 Dalton (30 min at 2000 g, Centricon 30 membranes, Amicon - Grace & Co., Beverley, MA USA). Aliquots of whole blood, plasma and Ultrafiltrate were monitored for radioactivity in a well type counter and plasma protein binding (PB) was calculated from Eq. (1):

PB[%l = [cpm(plasma) - cpm(ultrtiltrate)I * IO0 1 cPm(plasma) (1)

The hematocrit was determined by centrifugation and the cell bound fraction (CBF) was calculated from Eq. (2):

CBF[%l =( cPm(whole blood) - [(l - hematmrit) * cPm(pIasma)Il * lOO/cPm(whole blood) (2)

Metabolite analysis in plasma samples obtained from drawn blood 30 and 60 min after tracer injection, was performed by methanol and acetonitrile extraction through SEP-PAR Cl8 cartridges,

followed by HPLC (p-Bondapak 3.9 x 300, 20/80 acetonitrW0.1 M PP buffer (PH 7.0)) and additionally TLC (Gelman ITLC-SG, 80/20 CH&/methanol).

Results and Discussion

The primary goal of this study was to identify spiperone based analogues that display an

improved selectivity for 5-HT2A vs al, Dl and D2 receptors in order to obtain potential

radiopharma.ceuticals for in vivo investigations of the 5-HT2A receptors by PET or SPET after

labelling with suitable radionuclides. Candidates as radioligands for PET or SPET should ideally satisfy many criteria including, good affinity for the target binding site, high selectivity as well as minimal metabolism. Radioligands for central sites must also show adequate blood-brain barrier (BBB) penetration without being so lipophilic as to give high non-specific binding (8, 10, 22).

2006 New Selective Radiotracers for S-HT, Receptors Vol. 63, No. 22, 1998

Spiperone and its reported derivatives are known to bind preferentially to dopamine D2 receptors, exhibiting a manifold lower affinity for 5-HT2A and aI receptor sites (23, 24, 25). In contrast,

results of our in vifro binding study indicate that the new N- 1 alkylated derivatives FMSP, BrMSP

and IMSP display higher selectivity for the serotoninergic 5-HT2A vs the adrenergic al and the

dopaminergic D1 and D2 receptors (Table 1). This result is in agreement with the finding that

substitution of the anilino phenyl group on the spiperone structure by other alkyl groups could increase the selectivity of the obtained spiperone derivative for 5-HT2 receptors (15, 16).

[3H]RP62203 was used as radioligand to determine the affmity for 5-HT2A receptors, since

RP62203 has proved to exhibit the highest aflinity and selectivity for 5-HT2A receptors (11, 12).

The affinity of FMSP and IMSP for other examined neural receptors, including NMDA, Mi, K’, A,, 5-I-IT-carrier, GABA-a and Ach receptors remained with I& > 10’ nM very low (data not shown).

TABLE 1 5-HTz receptor selectivities and octanol-water partitions (log P) of FMSP, BrMSP, IMSP,

Spiperone (SP) and N-Methylspiperone (NMSP).* Ref. 23; n.f not reported. 0 Indicates IMSP has at least 196-fold higher selectivity for 5-I-IT2 vs D2 receptors.

log P ccl Dl D2 5HT2 at/S-HT2 Dl/S-HT2 D2/5-HT2

FMSP 1.4OkOo.12 2780 350 f9.5 441+13 30.8f1.5 90.3 11.7 14.3

BrMSP I .62?0.20 1735 > IO5 3530f54 149 * 3 II.6 > 670 23.7

IMSP 1.94?0.20 1543 595ok24 > lo4 5lk3.2 30.2 116.7 > 196.1’

BP62203 nd 35 >I000 >I000 0.26 I35 >3800 a3800

sp* 2.67’ n.f nf 0.06* 0.45* - 0.13

NMSP 3.23 n.f n.f 0.12 0.55 - 0.22

F[ 1 lC]MSP was prepared by N-[llC]methylation of the desmethyl precursor DMSP with

[1IC]methyl iodide (Scheme 1). As expected, the methylation occurs mainly at the secondary amino group (N-l position). Substitution at the less reactive amidogroup (N-3) remains relatively

low (< 2 %), thus leading mainly to the [ 1 lC]dimethylated product, resulting from simultaneous methylation at the N-l and N-3 positions. The latter product was attested by chromatographic

comparison with the non-radioactive reference. F[l IC]MSP was obtained in 12 k 3 % radiochemical yield after semipreparative reversed-phase HPLC purification. The retention times of

DMSP, F[llC]MSP and of the [llC]dimethylated by-product were 8.2, 10.8 and 13.3 min

respectively. The identity and purity of F[l lC]MSP were assessed by analytical HPLC, TLC and

by coelution of the isolated F[l lC]MSP with its authentic reference FMSP (Fig. 2). The specific

activity of F[lIC]MSP determined by HPLC was relatively low (850 GBq / mmol). The non

isotopic [1231]iodine-bromine exchange labelling and the radioiododemetalation are versatile

Vol. 63, No. 22, 1998 New Selective Radiotracers for 5-HTz Receptors 2W7

methods for the preparation of radioiodinated radiopharmaceuticals, yielding iodinated products

with high specific activities (18, 26, 27). Therefore 1231-MSP was prepared by Cu(I)-assisted

[1231]iodine-bromine exchange (15, 28) at 16O’C in the presence of Na2S205 as reduction agent.

1231-MSP was obtained in 82 _+ 8 % radiochemical yield with a specific activity determined to be at

least 0.50 TBq /umol by HPLC and uv profiles of samples with known concentration of IMSP. As

shown in E’ig. 3, the differences of the retention times of 1231-MSP and BrMSP (15.6 and 10.9

min) allowed a good HPLC isolation of 1231-MSP. Alternatively, reaction of sodium [ 1*31]iodide with the trj~butylstannyl precursor TBMSP in acidified solution containing chloramine-T at room

temperature yielded 1231-MSP in 80 - 90 % radiochemical yield and a specific activity of 0.7-l 5

TBqtumol. The specific activity of 1231-MSP used for the present in vivo studies was determined

to be 0.70 - 0.80 TBq/umol abler a Cu(I)-assisted [ 1231]iododebromination preparation. As

attested by rechromatography of the HPLC isolated products, F[llC]MSP and 1231-MSP are stable up to 180 min and 24 h respectively in aqueous solution at room temperature. Moreover,

F[llC]MSP and 1231-MSP revealed with 1ogP of 1.40 and 1.94, appropriate lipophilicities for good brain penetration (22).

DMSP F[ “‘CjMSP

BrMSP

TBMSP 12j I-MSP

Scheme 1.

R,adiosynthesis of F[ 1 IC]MSP and 1231-MSP: (a) [ 1 1C]CH31, 5 min at 60°C;

(b) [123I]NaI, Cu(I), Na2$05,15 min, 16O“C; (c) [1*3I]NaI, chloramin-T, 10 min, r.t.

2a-B New Selective Radiotracers for SHT, Receptors Vol. 63, No. 22, 1998

F[“C]MSP (A)

(8)

Fig. 2.

HPLC chromatogram of the isolated F[ 1 lC]MSP (A) and identification of F[ 1 *C]MSP by co- elution of F[t k]MSP with its non-radiolabelled standart FMSP (B).

Vol. 63, No. :!2, 1998 New Selective Radiotracers for F&IT, Receptors 2009

‘231-MSP

Fig. 3

Semipreparative reversed phase HPLC ofthe crude reaction mixture of *231-MSP

The distribution of radioactivity in various tissues of female ICR mice following intravenous

administration of 1231-MSP is summarized in Table 2. The negligible activity in the thyroid (data not shown) indicates that no significant in vivo deiodination of the tracer might have occured.

1231-MSP shows a relatively fast blood clearance. Initial high uptakes were observed in the lung, kidneys, liver and in the heart, decreasing gradually by time. The significant accumulation of radioactivity in these organs could be in part explained by the good perfusion and by the

lipophilicity of 1231-MSP (log P = 1.9). Moreover, 1231-MSP showed good brain uptake (up to 1.64 % I.D./g , 60 min p.i.) indicative of good BBB penetration. In regional dissection study, the highest radioactivity concentrations were found in the frontal and temporal cortices, regions known to contain the highest density of S-HT2 receptors (6, 30). The highest brain uptake and frontal

cortex to cerebellum ratio (1.64% i.d./g and 3.53) were measured 60 min p.i.. In addition, 1231-MSP seems to be metabolically stable at least 60 min, since at 60 min p.i. more than 90% of the extracted1 plasma activity was shown to be unchanged ligand, thus confirming the suggested relative metabolically stability of the butyrophenone derivatives (10, 29). Therefore, blocking studies were performed at 60 min pi., using ritanserin, a central high selective 5-HT2 receptor

antagonist as well as ketanserin and the non-radiolabelled derivative IMSP.

2010 New .S&ctivc Radiotraccrs for S-HT, Receptors Vol. 63, No. 22, 1998

Pretreatment of the animals with ritanserin as well as with IMSP 30 min prior to i.v injection of

1231-MSP led to a reduction of brain uptake (up to 45 and 71 % respectively), with the highest

levels in the frontal and temporal cortices, and in brain stem, whereas the reduction of the initial radioactivity concentration in the cerebellum and striatum remains low (Table 3). This important

reduction suggests a significant specificity of the 1231-MSP uptake in mice brain. In contrast, uptake reduction in mice brain following ketanserin pretreatment was moderate (- 20 %). As possibly explanation, ketanserin in comparison to ritanserin is a less selective S-HTz antagonist

with a considerable ai-adrenoceptor antagonist activity, showing high non-specific binding in vivu and in addition acting more periphericahy (10, 3 1). Blocking experiment with RP62203 could not

be performed, since this drug is currently not commercially available. Finally, 1231-MSP shows a relatively low binding to the cell fraction of mice blood (CBF < 40%). In contrast, as experted for

a lipophilic compound a high plasma protein binding of 85 f 5 % was obtained.

TABLE 2

Biodistribution of 1231-MSP in mice and the corresponding brain/blood ratios. n.d = not determined, FCK: frontal cortex to cerebellum; * (% ID/g organ)/ (% I.D/g organ blood).

1231-MSP in % I.D. /g (n=3)

5 min 30 min 60 min 120 min 4h 24 h

Blood

Heart

Lung

Kidneys

Liver

Cerebellum

Brain stem

Front. cortex

Temp. cortex

Striatum

Whole brain

Brain/blood*

FCIC

0.71 f 0.12 0.51 kO.15 0.58 zk 0.10 0.40 + 0. I1

3.24 f 0.60 I .60 + 0.23 1.20f0.12 1.32 f 0.12

10.43 + 1.46 4.20 + 1.80 6.1 + 0.50 3.94 f 0.45

12.58 f 2.40 11.13 f. 1.12 6.83 k 1.10 6.00 + 1.2

18.50 +_ 2.12 15.31 f 1.23 15.34 + 0.90 17.0+3.12

n.d 0.19 zk 0.03 0.17 + 0.06 0.18 + 0.05

n.d 0.13 f 0.02 0.16+0.10 0.12 Lb 0.03

n.d 0.46 + 0.05 0.59 k 0.04 0.58 + 0.10

n.d 0.21 ?I 0.04 0.33 ?r 0.10 0.29 f 0.04

n.d 0.09 * 0.02 0.13 * 0.03 0.12 f 0.04

0.68 f 0.03 1.2 1 + 0.07 1.72 f0.10 1.38 f 0.31

0.96 f 0.12 2.37 f 0.16 2.96 k 0.10 3.45 f 0.33

n.d 2.42 f 0.06 3.47 f 0.10 3.22 f 0.15

0.30 f 0.10 0.25 f 0. IO

1.20 * 0.13 0.53 + 0.07

2.70 f 0.67 1.04-f0.13

5.20 f 0.32 1.86 + 0.21

13.85k1.58 2.85 f 0.65

0.07 * 0.02 n.d

0.11 f0.04 n.d

0.23 f 0.06 n.d

0.16 I!I 0.03 n.d

0.03 f 0.01 n.d

0.73 f 0.12 0.31 f0.14

2.43 f 0.16 1.24f0.17

3.28 k 0.08 n.d

Vol. 63, No. 22, 1998 New Selective Radiotracers for 5-HTz Receptors

TABLE 3

Biodistribution of 1231-MSP tier pretreatment with ritanserin, IMSP and ketanserin.

1231-MSP in % I.D. / g (n=3)

min 60

Controle ritanserin IMSP ketanserin

Blood

Heart

Lung

Kidneys

Liver

Cerebellum

Brain stem

Front. cortex

Temp. cortex

Striatum

Whole brain

Brain/blood

FCK

0.52 k 0.05

1.33 ix 0.16

5.21 + 0.62

6.44 _+ 0.91

16.34 + 0.47

0.17 f 0.04

0.1710.10

0.60 + 0.14

0.32 f 0.05

0.14 f 0.02

1.64 f 0.06

3.15 f 0.07

3.53 * 0.15

0.68 I+Z 0.14

1.69 f 0.44

6.40 f 0.27

4.58 + 0.32

8.42 f 3.6

0.18 zk 0.04

0.08 + 0.03

0.36 + 0.10

0.24 f 0.06

0.10 f 0.03

0.98 f 0.10

1.44 f 0.17

2.oIko.11

0.41 f 0.05

1. 16 CL 0.20

6.34 f 0.3 1

5. 60 f 0.54

17.44 a.75

0.12 + 0.03

0.05 * 0.02

0.16 5 0.08

0.10 f 0.02

0.05 f 0.02

0.47 f 0.07

1.12 kO.10

1.33 kO.14

0.47 + 0.06

1.20f0.18

5.80 + 0.54

5.48 k 0.32

16. 91 f 2.10

0.15 + 0.04

0.13 f 0.02

0.50 + 0.08

0.27 + 0.06

0.11 kO.05

1.39 -+ 0.21

2.72 k 0.22

3.33 + 0.10

Conclusions

As yet, the number of radioligands that tXti1 the requirements for investigation of S-HT2

receptors by !3PET is very limited. 1231-MSP exhibits moderate high affinity for 5-HT2A receptors,

based on in vitro experiment. However, considering the other important criteria required of candidates as radioligands for emission tomography, including high selectivity for the target receptors, adlequate liophilicity and labelling with a suitable radioisotop at high specific activity,

investigations of l231-MSP were undertaken. In our study of lz31-MSP pharmacokinetics in mice, relatively higln uptake and long retention of the radiotracer were observed in the brain, with a regional distribution of radioactivity consistent with the known distribution of 5-HT2_4 receptors in

the brain. The: specificity and selectivity of the 1231-MSP binding in mice brain were attested by the uptake reduction following ritanserin, ketanserin and IMSP pretreatment. These data indicate

therefore that lz31-MSP is a promising tracer for further investigations to determine, whether the new radioiodinated compound could be use as a sensitive, quantitative probe for studying the serotoninergic .~-HT~A receptors non-invasively in experimental animals and humans by SPET

imaging.

2012 New Selective Radiotracers for 5HT, Receptors Vol. 63, No. 22, 1998

Acknowledgements

The authors are grateful to Mrs Eichholzer and Mr Sven Richter for assistance at the in vivo studies and to Dr. Gerhard Nolke of Hoechst AG. Frank&t/Main, for performing in vitro binding experiments.

References

1. D.L. MURPHY, Neuropsychopharmacol. 3 457-471 (1990). 2. G.M. WHITFORT, Neuropsychobiology 15 133-142 (1986). 3 A.S. EISON, M.S. EISON, Prog Neuro-Psychopharmacol. & Biol. Psychiat. 18 63-75

(1994). 4. S.C. RISCH and C.B. NEMEROFF, J. Clin. Psych. 53 3 - 7 (1992). 5. H.S. MAYBERG, R.G. ROBINSON, D.F. WONG, R. PARIKH, P. BOLDUC, S.E.

STARKSTEIN, T. PRICE, R.F. DANNALS, J.M. LINKS, A.A. WILSON, H.T. RAVERT and H.N. WAGNER JR., Am. J. Psych. 145 937-943 (1988).

6. D.F. WONG, J.R. LEVER P.R. I-IARTIG, R.F. DANNALS, V. VILLEMAGNE, B.J. HOFFMANN, A.A. WILSON, H.T. RAVERT, J.M. LINKS, U. SCHEFFEL and H.N. WAGNER JR. Synapse 1 393 (1987).

7. F.B. GERALDO, Eur. J. Nucl. Med. 23 867-870 (1996). 8. C CROUZEL, M. GUILLAUME, L. BARRE, C. LEMAIRE and V.W. PIKE, Nucl. Med.

Biol. 19/8 857-870 (1992). 9. P.R. HARTIG and J. LEVER. Quantitative Imagine: Neuroreceptors, Neurotransmitters and

Enzymes, J.J. Frost and H.N. Wagner Jr (Eds), 153 - 165, Raven Press Ltd., New York (1990).

10. G. STGCKLIN, Eur. J. Nucl. Med. 19 527 - 551 (1992). 11. C. FAJOLLES, A. BOIREAU, M. PONCHANT and P.M. LADURON, Eur. J. Pharmacol.

216 53-57 (1992). 12. L. BESRET, F. DAUPHIN, C. HUARD, M.-C. LASNE, R.VIVET, P. MICKALA, A.

BARBELIVIEN and J.-C. BARON, Nucl. Med. Biol. 23 169-171 (1996). 13. J.R. LEVER J.L. MUSACHIO, U. SCHEFFEL, M. STATHIS, D.F. WONG, E. SHAYA,

Z. SZABO and H.N. WAGNER Jr., J. Nucl. Med. 32 964 (1991). 14. A. AI31 DARGHAM, Y. ZEA PONCE, D. TERRIER M. AL TIKRITI, R.M. BALDWIN,

P. HOFFER, D. CHARNEY, J.E. LEYSEN, M. LARUELLE, J. MERTENS, R.B. INNIS, Eur. J. Pharmacol. 321 285-293 (1997).

15. S. SAMNICK, W. BRANDAU, G. NGLKEN, H.J. GERHARDS and 0. SCHOBER, Nucl.Med. Biol. 24 295 - 303 (1997).

16. S. SAMNICK, W. BRANDAU, H.J. GERHARDS and 0. SCHOBER Eur. J. Nucl. Med. 22 770 (1995).

17. S. SAMNICK, W. BRANDAU, G. NGLKEN, H.J. GERHARDS and 0. SCHOBER, (Manuscript submitted for publication).

18. C.G. SWAI-IN, C. HALLDIN, I GUNTHER J. PATT and S. AMETAMEY, J. Labelled Compd. Radiopharm. 38 676 - 685 (1996)

19. R.M. BALDWIN, Y. ZEA-PONCE, M.S. AL-TIKRITI, S.S. ZOGHBI, J.P. SEIBYL, D.S. CHARNEY, P.B. HOFFER, S WANG, R.A. MILIUS, J.L. NEUMEYER and R.B. INNIS, Nucl. Med. Biol. 22 211 (1995).

20. C. CROUZEL, B. LANGSTROM, V.W. PIKE and H.H. COENEN, Int. J. Appl. Radiat. Isot. 38 601 - 604 (1987).

21. Y.-C. CHENG and W.H PRUSOFF, Biochem. Pharmacol. 22 3099-3 108 (1973).

Vol. 63, No. 212, 1998 New Selective Radiotracers for S-HT2 Receptors

22. M.R. KlLBOURN, Nucl. Med. Biol. 16 681-686 (1989) 23. R.H. MACH, J.R. JACKSON, R.R. LUEDTKE, J. IVINS, P.B. MOLINOFF and R.L.

EHRENKAUFER, J. Med. Chem. 35 423 - 430 (1992). 24. R.A. LYON, M. TITELER, J.J. FROST, P.J. WHITEHOUSE, D.F. WONG, H.N.

WAGNER Jr, R.F. DANNALS, J.M. LINKS and M.J. KUHAR, J. Neurosci. 6 2941-2949 (1986).

25. H.N. WAGNER Jr, H.D. BURNS, R.F. DANNALS, D.F. WONG, B. LANGSTROM, T. DUELFER, J.J. FROST, H.T. RAVERT, J.M. LINKS, S.B. ROSENBLOOM, SE. LUKAS, A.V. KRAMER and M.J. KUHAR, Sciences 221 1264-1266 (1983).

26. H.H. CGENEN, SM. MOERLEIN and G. STOCKLIN, Radio&m. Acta 34 47 - 68 (1983). 27. J. MERTENS, W. VANRYCKEGHEM, M. GYSEMANS, J. EERSELS, E. FINDA-

PANEK and L. CARLSEN, Eur. J. Nucl. Med. 13 380-381 (1987). 28. S.M. MOERLEIN, Radiochim. Acta 50 55 - 61 (1990). 29. C.D. ARNETT, J.S. FOWLER, A.P. WOLF, C.Y. SHIUE, D.W. McPERSON, Life Sci. 36

1359 - 1366 (1985). 30. B. SADZOT, C. LEMAIRE, P. MAQUET, E. SALMON, A. PLANEVAUX, C.

DEGUEILDRE, J.P. HERMANNE, M. GUILLAUME, R. CANTINEAU, D. COMAR and

G. FRANCK, J. Cereb. Blood Flow Metab. 15 787-797 (1995). 31. Y. WATANABE, H. USUI, S. KOBAYASHI, H. YOSHIWARA, T. SHIBANO, T.

TANAKA Y. MORISHIMA, M. YASUOKA and M. KANAO, J. Med. Chem. 35 189-194 (1992).

32. Dr. G. N’iilke, Hoechst AG, Frankfurt/Main (personal communication)