- Email: [email protected]
[11C]MDL 100907, a radioligand for selective imaging of 5-HT2A receptors with positron emission tomography
Life sciences, Vol. 58, No. 10, pp. PL 187-192,1!3% CopyTight0 1996 Fuscvierscience Inc. Printed in the USA. AU rights rcservcd aat-32&s/96 SlsOO + .OO
PIIsOO24-32@5(%W13-6
ELSEVIER
PHARMACOLOGY LElTERS Accekrated Communi~adon
[llC]MDL 100907, A RADIOLIGAND FOR SELECTIVE IMAGING OF ~-HT~A RECEPTORS WITH POSITRON EMISSION TOMOGRAPHY Camilla Lundkvistl, Christer Halldin], Nathalie Ginovartl, Svante Nybergl, Carl-Gunnar Swahnl, Albert A. Carr2, Francoise Brunner3 and Lars Fardel
‘Karolinska
Institute& Department of Clinical Neuroscience, Section of Psychiatry, Karolinska Hospital, Stockholm, Sweden 2Hoechst Marion Roussel, Inc., 2110 E. Galbraith Road, Cincinnati (OH), USA 3Hoechst Marion Roussel, Clinical Pharmacology, Allmendsstrasse 6, CH-8320 Fehraltorf, Switzerland (Submitted October 31, 1995; accepted November 15, 1995; received in final form November 26, 1995)
Abstract The highly selective ~-HT~A receptor antagonist, MDL 100907 ((R)-(+)-4-( I-hydroxy- l(2,3-dimethoxyphenyl)methyl)-N-2-(4-fluorophenylethyl)piperidine), was labeled with 1‘C for Positron Emission Tomography (PET) studies. After iv. injection of (R)-(+)-[3-OCH3-“C]MDL 100907 ([“CIMDL 100907) in Cynomolgus monkeys a marked accumulation in the ~-I-IT~A receptor rich neocortical regions was obtained with a neocortex to cerebellum ratio of 3.5-4.5 after 60-80 minutes. In the neocortical regions a transient equilibrium occured within 40-60 minutes. Radioactivity in the neocortex, but not in the cerebellum, was reduced after injection of ketanserin, indicating that neocortical radioactivity following injection of [“C]MDL 100907 represents specific binding to ~-HT~A receptors. There was no evident effect on neocortical binding after pretreatment with raclopride or SCH 23390. [“C]MDL 100907 has potential to become the first selective radioligand for PET-quantitation of ~-HT~A receptors in the human brain in vivo. Key Wimk: serotonin receptors, MDL lCKl!?O7,positron emission tomography, monkey brain
Introduction Several methods for receptor binding studies with Positron Emission Tomography (PET) have been developed for quantitative determination of D2 dopamine receptors in the major basal ganglia (1). The dopamine hypothesis of antipsychotic drug action has been supported by consistent PET findings of a high degree of central D2 receptor occupancy. However, support for an important role of D2 receptor antagonism does not preclude the possibility that antipsychotic effects may be induced or modulated by drugs acting on other neurotransmitter systems. Serotonin (5-HT2) receptor antagonists have been suggested to improve efficacy and reduce the risk of sideeffects induced by D2 receptor antagonists (2-4). Moreover, the 5-HT&5-HTzc receptor antagonist ritanserin alone has been suggested to have antipsychotic efficacy (5). Suitable radioligands must be developed to explore the role of ~-HT~A receptor antagonists in the treatment of schizophrenia. Corresponding author: Camilla Lundkvist, Karolinska Institutet, Department of Clinical Neuroscience, Psychiatry Section, Karolinska Hospital, S- 17 176 Stockholm, Sweden (Fax: 46-8346563; email: [email protected]).
[“C]MDL
PL-188
lCKI907Bindingto
5-HTzA
Vol. 58, No. 10,1996
Several PET radioligands such as [18F]altanserin and [18F]setoperone have been proposed for quantitation of ~-I-IT~A receptor occupancy (6-8). None of these proposed agents are ideal since the rather low total-to-nonspecific binding ratio limits their potential for an accurate quantitative analysis. Another disadvantage with these ligands is their relatively high affinity for cc-adrenergic receptors (9,lO). In addition, the longer half-life of fluorine-18 (110 min) limits their utility in studies where several PET experiments are repeated on the same day. As an alternative, the nonselective radioligand [’ ‘CINMSP has been used for PET examination of ~-HT~A receptor binding in the human frontal cortex (11). The rather low ratio of neocortical to cerebellar binding, and the high affinity for D2 receptors has limited its use for defined quantitative analysis. An optimal carbon- 11 radioligand should have high affinity and selectivity for ~-HT~A receptor, enter the brain with a low nonspecific binding and give a transient equilibrium within the time of the PET study. MDL 100907 ((R)-(+)-4-(l-hydroxy-l-(2,3-dimethoxyphenyl)methyl)-N-2-(4flurophenylethyl)piperidine) is a potent and a highly selective ~-I-IT~A receptor antagonist in vitro (12,13), exhibiting the pharmacological profile of an atypical antipsychotic in behavioral, neurochemical and electrophysiological animal models (14). MDL 100907 binds with high affinity to the ~-HT~A receptor (Ki = 0.2 nM) and is approximately 300-fold less potent at the 5-HTzc, 5-I-IT6 and 5-HT7 receptors (15). Its affinity is more than lOOO-fold lower for the dopamine Dl, D:! and D4 receptors as compared to ~-I-IT~A receptors. The potency at the a-adrenergic receptor level is almost 100 times less than for ketanserin or ritanserin. MDL 100907 has a dose dependant and low degree of binding (< 60 %) with human plasma proteins (Internal communication). Furthermore, a 1ogP of 2.7 combined with a molecular structure enabling selective llC-methylation on either methoxy groups from the corresponding 2or 3-hydroxyl precursor, identifies MDL 100907 as a potential PET radioligand for examination of ~-HT~A receptors. In the present communication we report on the preparation of (R)-(+)-[3-OCH3-“CIMDL 100907 ([“CIMDL 100907) by 0-methylation of the 3-hydroxy precursor (MDL 105725) with [* lC]methyl iodide (Fig. 1). Metabolite studies in monkey plasma and a pharmacological characterization of [ 1lC]MDL 100907 binding was performed by PET in Cynomolgus monkeys. Methods Radioligand svnthesis. The precursor (MDL 105725) and standard of MDL 100907 was obtained from Marion Merrell Dow (Dr. A.A. Carr). High specific radioactivity [ 1‘Clmethyl iodide was synthesized from [ 1lC]carbon dioxide utilizing a one-pot reaction set-up similar to that reported previously (16). [ 1lC]Methyl iodide was trapped at room temperature in a reaction vessel (1 .O mL mini-vial, Alltech), containing the 3-hydroxy precursor (MDL 105725,0.5 mg), acetone (400 pL) and sodium hydroxide (5 M, 2 pL). The vessel was sealed and heated at 800C for 3 minutes. Mobile phase (600 pL) was added before injection into the semi-preparative normalphase HPLC column. [“C]MDL 100907 eluted after 14 minutes with a retention time identical to a reference sample. After evaporation of the mobile phase, the residue was dissolved in 8 mL sterile phosphate buffered saline (pH=7.4) and filtered through a Millipore filter (0.22 pm), yielding a solution which was sterile and free from pyrogens. Under these experimental conditions stereoisomerism is conserved during methylation.
[“C]CH,I Acetone, NaOH
Fig. 1 Methylation
of MDL 105725 with [l’C]methyl
iodide to obtain [ 1‘C]MDL 100907.
2
A
FL-189
[“CJMDL loo907Biidingto 5-HTti
Vol. 58,No. lo,1996
-+--
Control
frontal cortex
visuelcortex
B
Specific
binding
c
01
20
40
60
80
0
100
I
20
Time (min)
C
Binding
ratio
---A-
frontal cortex visual cortex
D
temporallobes
;r
SB
-
I
I
/
40 60 Time (min)
80
100
Total brain radioactivity _rc.total
I
‘0
I
frontalcortex SBvisuatcortex SBtemporallobes cerebelium
------t
20
,
,
40 60 Time (min)
,
80
brain
,
100
Fig. 2 Time course for regional radioactivity (nCi/ml) in the brain of a Cynomolgus monkey after intravenous administration of 100 MBq 11lC]MDL 100907 in a control experiment (A), specific binding (B), binding ratio (C) and total brain radioactivity in % of injected dose (D). The PET-camera system was a Siemens ECAT EXACT HR.
PET studies in monkeys. The PET-camera used was the Siemens ECAT EXACT HR which measures radioactivity in 47 slices with a separation of 3.1 mm and a spatial resolution of about 3.8 mm FWHM (Full Width Half Maximum) (17). Three male Cynomolgus monkeys weighing about 4 kg were supplied by the National Laboratory for Bacteriology. Anaesthesia was induced and maintained by repeated i.m. injection of ketamine (Ketalarm 5-10 mg kg-l h-l). A head fixation system was used to secure a fixed position of the monkey head during the PET experiments (18). The positioning was parallel to the canto-meatal line. Body temperature was controlled by a heating pad with thermostat. In monkey A a control experiment was performed in which 100 MBq [l lC]MDL 100907 was injected as a bolus during two seconds into the sural vein. In monkey B three PET experiments were performed. The first was a control experiment and the second was a displacement experiment performed 2.5 hours later in which ketanserin (1.5 mg/kg) was injected i.v. 20 minutes after the injection of [ * tC]MDL 100907. The third experiment was a pretreatment ex eriment in which ketanserin (1.5 mg/kg) was injected i.v. 13 minutes before the injection of [l PC]MDL 100907. Three experiments were also performed with monkey C. The first was a control and the second was a displacement experiment performed 2.5 hours later in which raclopride (1 mg/kg) was injected i.v. 30 minutes before the injection of [llC]MDL 100907. The third was a displacement experiment performed 2.5 hours later in which SCH 23390 (0.5 mg/kg) was injected i.v. 30 minutes before the injection of [“CIMDL 100907. Radioactivity in brain was measured according to a pre-programmed sequence of frames during 93 minutes.
PL190
Vol. 58,No. 10,19%
[“C]MDL 100!307 Bindingto 5-HTzA
Regions of interest (ROTS) were drawn on the PET-images which were reconstructed using data from 9 to 93 minutes after injection of [I tC]MDL 100907, The 5-HT2 receptors are widely distributed in the brain, with the highest density in the frontal cortex, moderate density in the striatum and lowest density in the cerebellum (10,19,20). The cerebellum, frontal cortex, visual cortex, temporal lobes and the whole brain contour were defined on each slice according to an atlas of a cryosected Cynomolgus monkey brain. The ROI’s were pooled for the two hemispheres. Radioactivity was calculated for the sequence of time frames, corrected for the radioactive decay and plotted versus time. Specific binding was defined as the difference in radioactivity concentration between a ROI and the cerebellum, a reference region with low density of 5HT2 receptors. The method to differentiate the radioactivity that corresponds to unchanged [l lC]MDL 100907 from labeled metabolites, measured in plasma, during the PET study was a slight modification of an HPLC method which has been used for other PET ligands (2 1,22).
Results The incorporation of a 11C-methyl group (via the iodide) to [ltC]MDL 100907 was 6080%. The total radiochemical yield of I* ‘ClMDL 100907, calculated from end of bombardment (EOB) and decay-corrected, @as 40-Z&% with a total synthesis time of 30 minutes. Purification [tlC]MDL 100907 with a was performed by semi-preparative normal-phase HPLC yieldin radiochemical purity better than 99%. The specific radioactivity of [KlC]MDL 100907 obtained at time of injection was 500-1000 Ci/mmol(18-37 GBq&mol) corresponding to a dose injected of 0.5-1.0 ltg in the monkey experiments. After i.v. injection of [ttC]MDL 100907 in the three control experiments there was a rapid accumulation of radioactivity in the brain. Four minutes after injection, 3.5% of the total radioactivity injected was present in the monkey brain (Fig. 2D). The highest uptake of radioactivity was observed in the neocortical regions whereas radioactivity was lower in the cerebellum region with a lower density of 5-HT2A receptors (Fig. 2A). The striatum could not be visually delineated from the background. The ratio of radioactivity in neocortex to cerebellum on peak level was 3.5-4.5 after 60-80 minutes (Fig. 2C). Specific binding in the neocortical regions was obtained within 40-60 minutes (Fig. 2B). In displacement or pretreatment experiments, radioactivity in the neocortex but not in the cerebellum was reduced after injection of ketanserin (Fig. 3A and 3B). There was no evident effect on neocortical binding after pretreatment with raclopride or SCH 23390 (data not shown). g ~5000
A
Control + displacement
with ketanserin
_
frontal
cortex
visual cortex tern ral lobes cereE llum frontal cortex + ketanserine visual cortex + ketansaine tern ml lobes + ketanserine cereE llum + ketanserine
40 60 Time (min)
80
100
OO_1
loo Time (min)
Fig. 3. Time course for regional radioactivity (nCi/ml) in the brain of a Cynomolgus monkey after administration of [IlC]MDL 100907. Control and displacement ex riment in which ketanserin (1.5 mg/kg) was given 20 minutes after injection of [t p” C]MDL 100907 (A). Control and pretreatment experiment in which ketanserin (1.5 mg/kg) was given 13 minutes before injection of [rrC]MDL 100907 (B). The PET-camera system was a Siemens ECAT EXACT HR.
Vol. 58,No. 10,1996
In minutes fraction minutes
[“CJMDL loo907Bindingto 5-HT=
PL191
the plasma metabolite studies all radioactivity eluted from the HPLC column within 7 with a good resolution of unchanged radioligand from the labelled metabolites. The of the radioactivity in plasma representing unchanged [l lC]MDL 100907 was 75% at 4 and 2530% at 40 minutes.
The molecular structure of MDL 100907 allows labelling in two different positions. It has been demonstrated that MDL 100907 is partly metabolized in humans to its 3-OH-analog (MDL 105725), a molecule for which the characteristics remains to be clarified. In the present study labelling of [“C]MDL 100907 was performed in the 3-OCH3-position. Labelling in the alternative 2-OCHs-position would lead to extensive formation of labelled 3-OH-analog, which could be expected to enter the brain and thus interfere with the interpretation of [“CIMDL 100907 uptake. The use of a normal-phase HPLC column in the preparative purification resulted in the labelled product being eluted before the unlabelled precursor, with a good resolution and showing no interference of precursor in the final product. The regional distribution of radioactivity was in accordance with the known distribution of 5-HT2A serotonin receptors, with highest density in the neocortical regions such as frontal cortex, visual cortex and temporal lobes (Fig. 2A). The neocortex to cerebellum ratio of 3.5-4.5 after 6080 minutes is the highest so far obtained in any PET study of the 5HT2A serotonin receptors in the monkey brain. Specific binding in the neocortical regions was on peak level within 40-60 minutes, indicating that [‘lC]MDL 100907 is a suitable radioligand for quantitation of 5-HTzA serotonin receptors in the human brain using an equilibrium analysis. The marked effect in the displacement or pretreatment experiments, where radioactivity in the neocortex, but not in the cerebellum, was reduced after injection of the 5-HT2~/5-HTzc rece tor antagonist ketanserin, clearly indicates that neocortical radioactivity following injection of [ 1PC]MDL 100907 represents to a large extent specific binding to 5-HTzA receptor sites and that this binding is reversible (Fig. 3A and 3B). The lack of evident effect on neocortical binding after pretreatment with either the selective dopamine D2 receptor antagonist raclopride, or the selective dopamine D1 receptor antagonist SCH 23390 is consistent with in vivo observations that [“C]MDL 100907 binds selectively to 5-HT2A receptors. In summary, the high specific binding of [“CIMDL 100907 to 5-HT2A demonstrated in the Cynomolgus monkey brain motivates its further investigation radioligand for the quantitation of 5-HT2A receptors in the human brain in vivo.
receptors as a PET
Acknowledeements This work was supported by grants from the Swedish Medical Research Council (091140613), the Swedish Natural Science Research Council (K-KU 9973-306), the National Institute of Mental Health USA (NIMH, 41205-9), Karolinska Institutet and Marion Merrell Dow. The assistance of the members of the Karolinska PET group involved in the PET experiments is also gratefully acknowledged. References 1. 2. 3. 4. 5.
6. 7.
L. FARDE, H. HALL, E. EHRIN and G. SEDVALL. Science 231258-261 (1986). J. GERLACH. Schizophr. Bull. 17 289-309 (1991). J.E. LEYSEN, P.M.F. JANSSEN, A. SCHOTTE, W.H.M.L. LUYTEN and A.A.H.P. MEGENS. Psychopharmacology 112 (suppl): S40-S54 (1993). H.Y. MELTZER. Schizophr. Bull..D 263-287 (1991). F.-A. WIESEL, A.-L. NORDSTROM, L. FARDE and B. ERIKSSON. Psychopharmacology 114 3 l-38 ( 1994). F. BIVER, S. GOLDMAN, A. LUXEN, M. MONCLUS, M. FORESTRINI, J. MENDLEWICZ and F. LOTSTRA. Eur. J. Nucl. Med. U 937-946 (1994). J. BLIN, J.C. BARON, B. DUBOIS, C. CROUZEL, M. FIORELLI, D. ATTARLEVY, B. PILLON, D. FOURNIER, M. VIDAILHET and Y. AGID. Brain 116 497-5 10 (1993).
PLl!z
8. 9. 10. 11. 12. 13.
14.
15.
16. 17. 18.
::: 21. 22.
[“C]MDL XXI907 Bindingto 5-HTw
Vol. 58,No. 10,19%
C. CROUZEL, M. GUILLAUME, L. BARRE, C. LEMAIRE and V.W. PIKE. Int. J. Radiat. Appl. Inst. Part B 19 857-870 (1992). J.E. LEYSEN and W. GOMMEREN. Drug Dev. Res. 8 119-131 (1986). R. LYON, M. TITELER, J. FROST, P. WHITEHOUSE, D. WONG, H.N. WAGNER, R. DANNALS, J. LINKS and M. KUHAR. J. Neurosci. 6 2941-2949 (1986) S. NYBERG, L. FARDE, L. ERIKSSON and C. HALLDIN. Psychopharmacology 110 265-272 (1993). M. DUDLEY, A. OGDEN, A. CARR, and J. KEHNE. Abstract 427.19, Society for Neuroscience, St. Luis, (1990). J.H. KEHNE, B.M. BARON, A.A. CARR, S.F. CHANEY, J. ELANDS, D.J. FELDMAN, R.A. FRANK, P.L.M. van GIERSBERGEN, T.C. McCLOSKEY, M.P. JOHNSON, D.R. MCCARTY, M. POIROT, Y. SENYAH, B.W. SIEGEL and C. WIDMAIER. J. Pharmacol. Exper. Therap. (in press) S.M. SORENSEN, J.H. KEHNE, G.M. FADAYEL, T.M. HUMPHREYS, H.J. KETTELER, C.K. SULLIVAN, V.L. TAYLOR and C.J. SCHMIDT. J. Pharmacol. Exper. Therap. m 684-691 (1993). B.L. ROTH, S.C. CRAIGO, M.S. CHOUDHARY, A. ULUER, F.J. MONSMA, Jr.,Y. SHEN, H.Y. MELTZER and D.R. SIBLEY. J. Pharmacol. Exper. Therap. 268 14031410 (1994). C. HALLDIN, L. FARDE, T. HGGBERG, H. HALL and G. SEDVALL. Appl. Radiat. Isot. 41669-674 (1990). K. WIENHARD, M. DAHLBOM, L. ERIKSSON, C. MICHEL, T. BRUCKBAUER, U. PIETRZYK and W.-D. HE1SS.J. Comput. Assist. Tomogr. 18 110-l 18 (1994). P. KARLSSON, L. FARDE, C. HALLDIN, C.-G. SWAHN, G. SEDVALL, C. FOGED, K.T. HANSEN and B. SKRUMSAGER. Psychopharmacology 113 149-156 (1993). A. PAZOS, A. PROBST and J. PALACIOS. Neuroscience 2 123-139 (1987). A. SCHOTTE, J.M. MALOTEAUX, and P.M. LADURON. Brain Res. 276 231-235 (1983) C. HALLDIN, K. NAGREN, C.-G. SWAHN, B. LANGSTRGM and H. NYBACK. Nucl. Med. Biol. 19 871-880 (1992). C.-G. SWAHN, C. HALLDIN, L. FARDE and G. SEDVALL. Human Psychopharmacol. 9 25-3 1 (1994).
PIIsOO24-32@5(%W13-6
ELSEVIER
PHARMACOLOGY LElTERS Accekrated Communi~adon
[llC]MDL 100907, A RADIOLIGAND FOR SELECTIVE IMAGING OF ~-HT~A RECEPTORS WITH POSITRON EMISSION TOMOGRAPHY Camilla Lundkvistl, Christer Halldin], Nathalie Ginovartl, Svante Nybergl, Carl-Gunnar Swahnl, Albert A. Carr2, Francoise Brunner3 and Lars Fardel
‘Karolinska
Institute& Department of Clinical Neuroscience, Section of Psychiatry, Karolinska Hospital, Stockholm, Sweden 2Hoechst Marion Roussel, Inc., 2110 E. Galbraith Road, Cincinnati (OH), USA 3Hoechst Marion Roussel, Clinical Pharmacology, Allmendsstrasse 6, CH-8320 Fehraltorf, Switzerland (Submitted October 31, 1995; accepted November 15, 1995; received in final form November 26, 1995)
Abstract The highly selective ~-HT~A receptor antagonist, MDL 100907 ((R)-(+)-4-( I-hydroxy- l(2,3-dimethoxyphenyl)methyl)-N-2-(4-fluorophenylethyl)piperidine), was labeled with 1‘C for Positron Emission Tomography (PET) studies. After iv. injection of (R)-(+)-[3-OCH3-“C]MDL 100907 ([“CIMDL 100907) in Cynomolgus monkeys a marked accumulation in the ~-I-IT~A receptor rich neocortical regions was obtained with a neocortex to cerebellum ratio of 3.5-4.5 after 60-80 minutes. In the neocortical regions a transient equilibrium occured within 40-60 minutes. Radioactivity in the neocortex, but not in the cerebellum, was reduced after injection of ketanserin, indicating that neocortical radioactivity following injection of [“C]MDL 100907 represents specific binding to ~-HT~A receptors. There was no evident effect on neocortical binding after pretreatment with raclopride or SCH 23390. [“C]MDL 100907 has potential to become the first selective radioligand for PET-quantitation of ~-HT~A receptors in the human brain in vivo. Key Wimk: serotonin receptors, MDL lCKl!?O7,positron emission tomography, monkey brain
Introduction Several methods for receptor binding studies with Positron Emission Tomography (PET) have been developed for quantitative determination of D2 dopamine receptors in the major basal ganglia (1). The dopamine hypothesis of antipsychotic drug action has been supported by consistent PET findings of a high degree of central D2 receptor occupancy. However, support for an important role of D2 receptor antagonism does not preclude the possibility that antipsychotic effects may be induced or modulated by drugs acting on other neurotransmitter systems. Serotonin (5-HT2) receptor antagonists have been suggested to improve efficacy and reduce the risk of sideeffects induced by D2 receptor antagonists (2-4). Moreover, the 5-HT&5-HTzc receptor antagonist ritanserin alone has been suggested to have antipsychotic efficacy (5). Suitable radioligands must be developed to explore the role of ~-HT~A receptor antagonists in the treatment of schizophrenia. Corresponding author: Camilla Lundkvist, Karolinska Institutet, Department of Clinical Neuroscience, Psychiatry Section, Karolinska Hospital, S- 17 176 Stockholm, Sweden (Fax: 46-8346563; email: [email protected]).
[“C]MDL
PL-188
lCKI907Bindingto
5-HTzA
Vol. 58, No. 10,1996
Several PET radioligands such as [18F]altanserin and [18F]setoperone have been proposed for quantitation of ~-I-IT~A receptor occupancy (6-8). None of these proposed agents are ideal since the rather low total-to-nonspecific binding ratio limits their potential for an accurate quantitative analysis. Another disadvantage with these ligands is their relatively high affinity for cc-adrenergic receptors (9,lO). In addition, the longer half-life of fluorine-18 (110 min) limits their utility in studies where several PET experiments are repeated on the same day. As an alternative, the nonselective radioligand [’ ‘CINMSP has been used for PET examination of ~-HT~A receptor binding in the human frontal cortex (11). The rather low ratio of neocortical to cerebellar binding, and the high affinity for D2 receptors has limited its use for defined quantitative analysis. An optimal carbon- 11 radioligand should have high affinity and selectivity for ~-HT~A receptor, enter the brain with a low nonspecific binding and give a transient equilibrium within the time of the PET study. MDL 100907 ((R)-(+)-4-(l-hydroxy-l-(2,3-dimethoxyphenyl)methyl)-N-2-(4flurophenylethyl)piperidine) is a potent and a highly selective ~-I-IT~A receptor antagonist in vitro (12,13), exhibiting the pharmacological profile of an atypical antipsychotic in behavioral, neurochemical and electrophysiological animal models (14). MDL 100907 binds with high affinity to the ~-HT~A receptor (Ki = 0.2 nM) and is approximately 300-fold less potent at the 5-HTzc, 5-I-IT6 and 5-HT7 receptors (15). Its affinity is more than lOOO-fold lower for the dopamine Dl, D:! and D4 receptors as compared to ~-I-IT~A receptors. The potency at the a-adrenergic receptor level is almost 100 times less than for ketanserin or ritanserin. MDL 100907 has a dose dependant and low degree of binding (< 60 %) with human plasma proteins (Internal communication). Furthermore, a 1ogP of 2.7 combined with a molecular structure enabling selective llC-methylation on either methoxy groups from the corresponding 2or 3-hydroxyl precursor, identifies MDL 100907 as a potential PET radioligand for examination of ~-HT~A receptors. In the present communication we report on the preparation of (R)-(+)-[3-OCH3-“CIMDL 100907 ([“CIMDL 100907) by 0-methylation of the 3-hydroxy precursor (MDL 105725) with [* lC]methyl iodide (Fig. 1). Metabolite studies in monkey plasma and a pharmacological characterization of [ 1lC]MDL 100907 binding was performed by PET in Cynomolgus monkeys. Methods Radioligand svnthesis. The precursor (MDL 105725) and standard of MDL 100907 was obtained from Marion Merrell Dow (Dr. A.A. Carr). High specific radioactivity [ 1‘Clmethyl iodide was synthesized from [ 1lC]carbon dioxide utilizing a one-pot reaction set-up similar to that reported previously (16). [ 1lC]Methyl iodide was trapped at room temperature in a reaction vessel (1 .O mL mini-vial, Alltech), containing the 3-hydroxy precursor (MDL 105725,0.5 mg), acetone (400 pL) and sodium hydroxide (5 M, 2 pL). The vessel was sealed and heated at 800C for 3 minutes. Mobile phase (600 pL) was added before injection into the semi-preparative normalphase HPLC column. [“C]MDL 100907 eluted after 14 minutes with a retention time identical to a reference sample. After evaporation of the mobile phase, the residue was dissolved in 8 mL sterile phosphate buffered saline (pH=7.4) and filtered through a Millipore filter (0.22 pm), yielding a solution which was sterile and free from pyrogens. Under these experimental conditions stereoisomerism is conserved during methylation.
[“C]CH,I Acetone, NaOH
Fig. 1 Methylation
of MDL 105725 with [l’C]methyl
iodide to obtain [ 1‘C]MDL 100907.
2
A
FL-189
[“CJMDL loo907Biidingto 5-HTti
Vol. 58,No. lo,1996
-+--
Control
frontal cortex
visuelcortex
B
Specific
binding
c
01
20
40
60
80
0
100
I
20
Time (min)
C
Binding
ratio
---A-
frontal cortex visual cortex
D
temporallobes
;r
SB
-
I
I
/
40 60 Time (min)
80
100
Total brain radioactivity _rc.total
I
‘0
I
frontalcortex SBvisuatcortex SBtemporallobes cerebelium
------t
20
,
,
40 60 Time (min)
,
80
brain
,
100
Fig. 2 Time course for regional radioactivity (nCi/ml) in the brain of a Cynomolgus monkey after intravenous administration of 100 MBq 11lC]MDL 100907 in a control experiment (A), specific binding (B), binding ratio (C) and total brain radioactivity in % of injected dose (D). The PET-camera system was a Siemens ECAT EXACT HR.
PET studies in monkeys. The PET-camera used was the Siemens ECAT EXACT HR which measures radioactivity in 47 slices with a separation of 3.1 mm and a spatial resolution of about 3.8 mm FWHM (Full Width Half Maximum) (17). Three male Cynomolgus monkeys weighing about 4 kg were supplied by the National Laboratory for Bacteriology. Anaesthesia was induced and maintained by repeated i.m. injection of ketamine (Ketalarm 5-10 mg kg-l h-l). A head fixation system was used to secure a fixed position of the monkey head during the PET experiments (18). The positioning was parallel to the canto-meatal line. Body temperature was controlled by a heating pad with thermostat. In monkey A a control experiment was performed in which 100 MBq [l lC]MDL 100907 was injected as a bolus during two seconds into the sural vein. In monkey B three PET experiments were performed. The first was a control experiment and the second was a displacement experiment performed 2.5 hours later in which ketanserin (1.5 mg/kg) was injected i.v. 20 minutes after the injection of [ * tC]MDL 100907. The third experiment was a pretreatment ex eriment in which ketanserin (1.5 mg/kg) was injected i.v. 13 minutes before the injection of [l PC]MDL 100907. Three experiments were also performed with monkey C. The first was a control and the second was a displacement experiment performed 2.5 hours later in which raclopride (1 mg/kg) was injected i.v. 30 minutes before the injection of [llC]MDL 100907. The third was a displacement experiment performed 2.5 hours later in which SCH 23390 (0.5 mg/kg) was injected i.v. 30 minutes before the injection of [“CIMDL 100907. Radioactivity in brain was measured according to a pre-programmed sequence of frames during 93 minutes.
PL190
Vol. 58,No. 10,19%
[“C]MDL 100!307 Bindingto 5-HTzA
Regions of interest (ROTS) were drawn on the PET-images which were reconstructed using data from 9 to 93 minutes after injection of [I tC]MDL 100907, The 5-HT2 receptors are widely distributed in the brain, with the highest density in the frontal cortex, moderate density in the striatum and lowest density in the cerebellum (10,19,20). The cerebellum, frontal cortex, visual cortex, temporal lobes and the whole brain contour were defined on each slice according to an atlas of a cryosected Cynomolgus monkey brain. The ROI’s were pooled for the two hemispheres. Radioactivity was calculated for the sequence of time frames, corrected for the radioactive decay and plotted versus time. Specific binding was defined as the difference in radioactivity concentration between a ROI and the cerebellum, a reference region with low density of 5HT2 receptors. The method to differentiate the radioactivity that corresponds to unchanged [l lC]MDL 100907 from labeled metabolites, measured in plasma, during the PET study was a slight modification of an HPLC method which has been used for other PET ligands (2 1,22).
Results The incorporation of a 11C-methyl group (via the iodide) to [ltC]MDL 100907 was 6080%. The total radiochemical yield of I* ‘ClMDL 100907, calculated from end of bombardment (EOB) and decay-corrected, @as 40-Z&% with a total synthesis time of 30 minutes. Purification [tlC]MDL 100907 with a was performed by semi-preparative normal-phase HPLC yieldin radiochemical purity better than 99%. The specific radioactivity of [KlC]MDL 100907 obtained at time of injection was 500-1000 Ci/mmol(18-37 GBq&mol) corresponding to a dose injected of 0.5-1.0 ltg in the monkey experiments. After i.v. injection of [ttC]MDL 100907 in the three control experiments there was a rapid accumulation of radioactivity in the brain. Four minutes after injection, 3.5% of the total radioactivity injected was present in the monkey brain (Fig. 2D). The highest uptake of radioactivity was observed in the neocortical regions whereas radioactivity was lower in the cerebellum region with a lower density of 5-HT2A receptors (Fig. 2A). The striatum could not be visually delineated from the background. The ratio of radioactivity in neocortex to cerebellum on peak level was 3.5-4.5 after 60-80 minutes (Fig. 2C). Specific binding in the neocortical regions was obtained within 40-60 minutes (Fig. 2B). In displacement or pretreatment experiments, radioactivity in the neocortex but not in the cerebellum was reduced after injection of ketanserin (Fig. 3A and 3B). There was no evident effect on neocortical binding after pretreatment with raclopride or SCH 23390 (data not shown). g ~5000
A
Control + displacement
with ketanserin
_
frontal
cortex
visual cortex tern ral lobes cereE llum frontal cortex + ketanserine visual cortex + ketansaine tern ml lobes + ketanserine cereE llum + ketanserine
40 60 Time (min)
80
100
OO_1
loo Time (min)
Fig. 3. Time course for regional radioactivity (nCi/ml) in the brain of a Cynomolgus monkey after administration of [IlC]MDL 100907. Control and displacement ex riment in which ketanserin (1.5 mg/kg) was given 20 minutes after injection of [t p” C]MDL 100907 (A). Control and pretreatment experiment in which ketanserin (1.5 mg/kg) was given 13 minutes before injection of [rrC]MDL 100907 (B). The PET-camera system was a Siemens ECAT EXACT HR.
Vol. 58,No. 10,1996
In minutes fraction minutes
[“CJMDL loo907Bindingto 5-HT=
PL191
the plasma metabolite studies all radioactivity eluted from the HPLC column within 7 with a good resolution of unchanged radioligand from the labelled metabolites. The of the radioactivity in plasma representing unchanged [l lC]MDL 100907 was 75% at 4 and 2530% at 40 minutes.
The molecular structure of MDL 100907 allows labelling in two different positions. It has been demonstrated that MDL 100907 is partly metabolized in humans to its 3-OH-analog (MDL 105725), a molecule for which the characteristics remains to be clarified. In the present study labelling of [“C]MDL 100907 was performed in the 3-OCH3-position. Labelling in the alternative 2-OCHs-position would lead to extensive formation of labelled 3-OH-analog, which could be expected to enter the brain and thus interfere with the interpretation of [“CIMDL 100907 uptake. The use of a normal-phase HPLC column in the preparative purification resulted in the labelled product being eluted before the unlabelled precursor, with a good resolution and showing no interference of precursor in the final product. The regional distribution of radioactivity was in accordance with the known distribution of 5-HT2A serotonin receptors, with highest density in the neocortical regions such as frontal cortex, visual cortex and temporal lobes (Fig. 2A). The neocortex to cerebellum ratio of 3.5-4.5 after 6080 minutes is the highest so far obtained in any PET study of the 5HT2A serotonin receptors in the monkey brain. Specific binding in the neocortical regions was on peak level within 40-60 minutes, indicating that [‘lC]MDL 100907 is a suitable radioligand for quantitation of 5-HTzA serotonin receptors in the human brain using an equilibrium analysis. The marked effect in the displacement or pretreatment experiments, where radioactivity in the neocortex, but not in the cerebellum, was reduced after injection of the 5-HT2~/5-HTzc rece tor antagonist ketanserin, clearly indicates that neocortical radioactivity following injection of [ 1PC]MDL 100907 represents to a large extent specific binding to 5-HTzA receptor sites and that this binding is reversible (Fig. 3A and 3B). The lack of evident effect on neocortical binding after pretreatment with either the selective dopamine D2 receptor antagonist raclopride, or the selective dopamine D1 receptor antagonist SCH 23390 is consistent with in vivo observations that [“C]MDL 100907 binds selectively to 5-HT2A receptors. In summary, the high specific binding of [“CIMDL 100907 to 5-HT2A demonstrated in the Cynomolgus monkey brain motivates its further investigation radioligand for the quantitation of 5-HT2A receptors in the human brain in vivo.
receptors as a PET
Acknowledeements This work was supported by grants from the Swedish Medical Research Council (091140613), the Swedish Natural Science Research Council (K-KU 9973-306), the National Institute of Mental Health USA (NIMH, 41205-9), Karolinska Institutet and Marion Merrell Dow. The assistance of the members of the Karolinska PET group involved in the PET experiments is also gratefully acknowledged. References 1. 2. 3. 4. 5.
6. 7.
L. FARDE, H. HALL, E. EHRIN and G. SEDVALL. Science 231258-261 (1986). J. GERLACH. Schizophr. Bull. 17 289-309 (1991). J.E. LEYSEN, P.M.F. JANSSEN, A. SCHOTTE, W.H.M.L. LUYTEN and A.A.H.P. MEGENS. Psychopharmacology 112 (suppl): S40-S54 (1993). H.Y. MELTZER. Schizophr. Bull..D 263-287 (1991). F.-A. WIESEL, A.-L. NORDSTROM, L. FARDE and B. ERIKSSON. Psychopharmacology 114 3 l-38 ( 1994). F. BIVER, S. GOLDMAN, A. LUXEN, M. MONCLUS, M. FORESTRINI, J. MENDLEWICZ and F. LOTSTRA. Eur. J. Nucl. Med. U 937-946 (1994). J. BLIN, J.C. BARON, B. DUBOIS, C. CROUZEL, M. FIORELLI, D. ATTARLEVY, B. PILLON, D. FOURNIER, M. VIDAILHET and Y. AGID. Brain 116 497-5 10 (1993).
PLl!z
8. 9. 10. 11. 12. 13.
14.
15.
16. 17. 18.
::: 21. 22.
[“C]MDL XXI907 Bindingto 5-HTw
Vol. 58,No. 10,19%
C. CROUZEL, M. GUILLAUME, L. BARRE, C. LEMAIRE and V.W. PIKE. Int. J. Radiat. Appl. Inst. Part B 19 857-870 (1992). J.E. LEYSEN and W. GOMMEREN. Drug Dev. Res. 8 119-131 (1986). R. LYON, M. TITELER, J. FROST, P. WHITEHOUSE, D. WONG, H.N. WAGNER, R. DANNALS, J. LINKS and M. KUHAR. J. Neurosci. 6 2941-2949 (1986) S. NYBERG, L. FARDE, L. ERIKSSON and C. HALLDIN. Psychopharmacology 110 265-272 (1993). M. DUDLEY, A. OGDEN, A. CARR, and J. KEHNE. Abstract 427.19, Society for Neuroscience, St. Luis, (1990). J.H. KEHNE, B.M. BARON, A.A. CARR, S.F. CHANEY, J. ELANDS, D.J. FELDMAN, R.A. FRANK, P.L.M. van GIERSBERGEN, T.C. McCLOSKEY, M.P. JOHNSON, D.R. MCCARTY, M. POIROT, Y. SENYAH, B.W. SIEGEL and C. WIDMAIER. J. Pharmacol. Exper. Therap. (in press) S.M. SORENSEN, J.H. KEHNE, G.M. FADAYEL, T.M. HUMPHREYS, H.J. KETTELER, C.K. SULLIVAN, V.L. TAYLOR and C.J. SCHMIDT. J. Pharmacol. Exper. Therap. m 684-691 (1993). B.L. ROTH, S.C. CRAIGO, M.S. CHOUDHARY, A. ULUER, F.J. MONSMA, Jr.,Y. SHEN, H.Y. MELTZER and D.R. SIBLEY. J. Pharmacol. Exper. Therap. 268 14031410 (1994). C. HALLDIN, L. FARDE, T. HGGBERG, H. HALL and G. SEDVALL. Appl. Radiat. Isot. 41669-674 (1990). K. WIENHARD, M. DAHLBOM, L. ERIKSSON, C. MICHEL, T. BRUCKBAUER, U. PIETRZYK and W.-D. HE1SS.J. Comput. Assist. Tomogr. 18 110-l 18 (1994). P. KARLSSON, L. FARDE, C. HALLDIN, C.-G. SWAHN, G. SEDVALL, C. FOGED, K.T. HANSEN and B. SKRUMSAGER. Psychopharmacology 113 149-156 (1993). A. PAZOS, A. PROBST and J. PALACIOS. Neuroscience 2 123-139 (1987). A. SCHOTTE, J.M. MALOTEAUX, and P.M. LADURON. Brain Res. 276 231-235 (1983) C. HALLDIN, K. NAGREN, C.-G. SWAHN, B. LANGSTRGM and H. NYBACK. Nucl. Med. Biol. 19 871-880 (1992). C.-G. SWAHN, C. HALLDIN, L. FARDE and G. SEDVALL. Human Psychopharmacol. 9 25-3 1 (1994).