Evaluation of radioiodo-4-amino-N-[1-[3-(4-fluorophenoxy)-propyl]-4-methyl-4-piperidinyl]-5-IODO-2-methoxybenzamide as a potential 5HT2 receptor tracer for SPE(C)T

Pergamon

Copyright 0 199s &e\ ET Science Inc Printed in Great Britain. All rights rewrved 0969-X05 1/9s $9 so + u.uu

EVALUATION OF RADIOIODO-4~AMINO-N-[l-[3-(4-FLUOROPHENOXY)-PROPYL]-4-METHYL-4-PIPERIDINYL]-5-IODO-2-METJ!IOXYBENZAMIDE AS A POTENTIAL 5HT2 RECEPTOR TRACER FOR SPE(C)T. D. Terriere*, P.M.F. Janssen**, W. Gommeren**, M. Gysemans*. J.J.R. Mertens*, J.E. Leysen**. VUB-Cyclotron *, 1090 Brussel, Belgium, Janssen Research Foundation**, 2340 Beerse, Belgium.

INTRODUCTION. 5-HT, Receptors seem to play an important role in psychiatric disorders. In order to investigate the role of these receptors in pathology there is an increasing interest in obtaining a selective and high affinity rabiolabelled ligand suitable for in vitro and in vivo receptor binding studies. For in vitro and rodent studies “‘1 labelled tracers are preferred as their specific radioactivity is generally at least 50 times higher than that offered by tritiated compounds. Labelled with 1231a suitable ligand offers the opportunity to perform SPE(C)T. At the end of the 1980’s 2-radioiodoketanserin was proposed as a potential 5 HT, receptor tracer for SPECT (1) and this compound was applied for SPECT imaging of 5-HT, receptors in depressed patients (2). The iodinated compound showed properties almost comparable to those of ketanserin (poorly selective for 5-HT, vis-a-vis H, and cr,-receptors) and high binding to serum proteins resulting in rather poor brain receptor to blood ratio. Recently 4-amino-N-[1-[3-(4-fluorophenoxy)propyl]-4-methyl-4-piperidinyl]2-methoxybenzamide derivatives have been discovered as a new class of 5-HT2 receptor antagonists. The parent compound shows very high affinity and selectivity for 5-HT, receptors (PI& for inhibition of [3H]ketanserin binding to rat frontal cortex membranes is 9.2 ; selectivity with regard to other neurotransmitter receptor sites such as 5-HT,,, 5-HTiB, 5-HT,n, 5-HT2c, 5-HT3, CY~and CY*adrenergic, histamine-H, and dopamine-D, is at least a factor of 50). It was decided to label this compound with radioiodine in order to develop a new radioiodinated ligand with high affinity for 5-HT, receptors. Earlier experiments have shown that substitution of radioiodine / iodine on the 5-position of the benzamide group yielded the better compound (3). This compound is named 5Radioiodo-R91150. In view of a potential increased brain uptake, due to an increase of lipophilicity, it was decided to substitute a methyl-group on the 4amino function of the benzamide side.

1006

EXPERIMENTAL. Synthesis. The synthesis of 5-I-R91150 and 5-Radioiodo-R91150 has been described earlier (3). N-methylation of R91150 was performed with CH,I in DMF at 55°C (4). Radioiodination of N-methyl-R91 150 is analogous to the radioiodination of R91150. Radioligand

binding in vitro.

Binding of the radioligands to membrane preparations of rat frontal cortex and inhibition by ketanserin and ritanserin as well as the screening of the nonradioactive analogues for selectivity was essentially carried out as described by Leysen et al. (5,6). The in vitro stability of the radioiodinated compound was checked for the compound in solution as well as for the compound bound to the membranes. After filtration of the membrane suspension in buffer containing the radioligand, an aliquot of the filtrate was injected for HPLC analysis. The membranes were resuspended in buffer by sonification of the filter and incubation in the presence of 10e6M of ritanserin was performed over the course of two hours. After filtration an aliquot of the filtrate was submitted for analytical HPLC analysis. Radioactivity detection was carried out with a Radiomatic A500 which allows one to detect very low amounts of ‘25I radioactivity ( 2 - 3 Bq can easily be measured above a background of 0,3 Bq, the counting efficiency amounts to 80%). After a total incubation time of six hours no free radioiodide could be detected and only a peak corresponding to the radioligand was present in the chromatogram. Radioligand binding in vivo. In vivo experiments were performed in the laboratories of Janssen Research Foundation licenced for this purpose. Radioiodinated ligand (740 kBq) was injected in the tail vein of male Wistar rats. After 0.5, 1.0, 1.5, 2.0 and 3.0 hours the rats were sacrificed, blood was collected and the brain areas were rapidly dissected. Radioactivity in the tissues was counted using a Cobra Packard autogamma counter. Samples of rat serum were treated with an equal amount of a 5% trichloroacetic acid/methanol : 50/50 and after centrifugation an aliquot was injected for analytical HPLC using the high sensitive detector as mentioned before.

Potential

5HT2

receptor

tracer for SPE(C)T

I007

RESULTS. 5-I-R91150 In vitro binding properties to the rat brain membranes. The affinity binding in vitro of 5-I-R91150 for 5-HT, receptor sites evaluated by the inhibition of [3H]ketanserin binding to rat frontal cortex membranes shows a pIC50 value of 9.45 (Ki: 0.2nM). The selectivity with regard to other 5-HT,, CY,,02, D, and D, receptors was at least a factor of 50. Iodination on the site mentioned thus yields a non radioactive ligand with a binding affinity and selectivity analogous to the parent compound. The binding of the radioiodinated compound to rat frontal cortex membranes is saturable. A linear Scatchard plot yields a Kd value of 0.11 + 0.01 nM and a Bmax value of 38.0 &- 0.7 fmoles/mg. Inhibition of the binding of the radioactive compound by ketanserin and ritanserin and the displacement by ritanserin followed by HPLC analysis, showing unmodified radioactive compound to dissociate almost quantitatively from the receptor sites, proved that the binding was reversible and 5- ‘2SI-R91150 to be stable and not to be internalized in in vitro conditions. In vivo binding properties in rat brain. As shown in Fig.1. for the new tracer the FC to CER and FC to Blood ratio increases from about 3 at 30 minutes p. injection to a steady value of about 10 for FC / CER and 6 for FC / Blood up from 60 minutes to 3 hours.This favourable tissue / Blood ratio is very important in view of SPECT application. Two hours post injection the %ID/g in FC was 0.24 and the ratio radioactivity in the FC to the rest of the brain was at least 4. When injecting i.v. an increasing amount of the radioactive tracer and plotting the specific bound SB (total bound - bound in cerebellum, expressed as fmol/mg of target tissue) in the frontal cortex (FC) and striatum (STR) after 2 hours as function of the injected amount of compound (non-radioactive compound was added to the radioactive to modify the specific activity), a nice saturation curve is obtained for both the FC and STR (Fig.2.) . This means that in vivo the binding of the new tracer in the FC and STR of rat is saturable, totally comparable to in vitro conditions. Within the large amount of radioactive tracers (labelled with 3H, i4C or ‘251)evaluated at Janssen Research Foundation, the new 5-HT, receptor tracer is one of the few showing an in vivo saturability of the receptors of interest almost approaching a theoretical behaviour.

0.3

%ID/g 0.2

0

1

2 Time(h)

3

Fig. 1. Distribution of 5- ‘zI-R91 150 in the frontal cortex,striatum, cerebellum and blood (fg / mg tissue) after i.v. injection of 5 ng (740 kBq) of the radioactive compound in function of time (hours).

F50 z2

n

n

Fi =I

100 fi El

n

50-

m FC ’ STR

l

n l l l :

04 0

I

100

I

200

I

1

300

400 F(nmol)

Fig.2. Specific binding (fmol/mg) of 5-‘251-R91150 after 2 hours in frontal cortex and striatum as a function of the injected dose (nmol). As shown in Fig.3., displacement by ketanserin, a ligand showing high affinity for 5-HT, receptors, which was i.v. injected 60 minutes after the administration of the radioactive compound, proves that the in the in vivo brain the binding of 5-‘“I-R91150 was reversible and occured at 5-HT, receptor sites. Injecting ritanserin 1 hour before the administration of the radioactive dose inhibits the specific binding of the tracer almost quantitativly in both the FC and STR. As ritanserin is also a ligand showing high affinity 5-HT, receptors, it can be concluded that the new tracer allows to measure 5-HT, receptor sites in the striatum.

Potential

5HT2 receptor

tracer for SPE(C)T

IO09

Fig. 3. Displacement of 5- 12’I-R91 150 by ketanserin (0.63 mg/kg) and inhibition of the tracer binding by ritanserin (2.5 mg/kg) in the rat brain.

5-I-N-methyl-R9

1150.

In vitro binding. The binding of 5- *SI-N-methyl-R91 150 to rat frontal cortex membranes is saturable and the linear Scatchard plot yields a Kd of 0.12 nM and a Bmax of 38 fmoles/mg of tissue.The specific binding was displaced by ketanserin and ritanserin. Those results are comparable to those obtained with the desmethyl compound.

In vivo binding. The amount of tracer bound in the brain (k 0.13 % of the injected dose) is not significantly increased as compared to the desmethyl compound. Table 1. compares the results obtained with both 5-‘=I-R91150 and 5-151-Nmethyl-R91 150. A slight decrease of the FCKER and STRKER activity ratios is observed for the methylated tracer as compared to 5-izI-R91 150 . This can be due to a higher non-specific binding. The increase of lipophilicity expressed as the ratio of k’ values of RP-HPLC at pH 4.8 due to N-methylation is about 2. The fact that increasing the lipophilicity does not result in a

1010

I>. ‘I‘FKRIEKL

t’l tr/.

higher brain uptake can be explained by an increased binding to the external membranes. From the product that crossed the BBB, a larger part is bound in the cerebellum. Ratio

5-1251-R91150

N-CH3- 1251-R91150

FC / CER

10.8

6.9

FC / Blood

4.8

4.2

STR / CER

4.8

3.6

STR / Blood

2.2

2.2

Table 1. Ratio of the amount of radioactivity in brain tissue (FC : frontal cortex, STR : striatum, CER : cerebellum) and blood, 3 hrs post I.V.injection.

CONCLUSION. 5-Radioiodo-R91150 could be evaluated as a promising potential 5-HT, receptor tracer for SPECT. As N-methylation does not significantly increase the specific brain uptake, so that for SPECT application 5-Radioiodo-R91150 is preferred. Altough N-methylation can yield an interesting PET tracer for 5-HT, receptors.

REFERENCES. 1. Mertens J., Gysemans M., Piron C., Thomas M., J. Labelled Comp. Radiopharm. 2&731-739 (1990). 2. D’Haenen H., Bossuyt A., Mertens J., Piron C., Gysemans M., Kaufman L. Psychiatry Reserach, Neuroimaging, 45, 227-237 (1992). 3. Mertens J., Terriere D., Sipido V., Gommeren W., Janssen P.M.F., Leysen J.E., J. Labelled Comp. Radiopharm., 34, 795-806 (1994). 4. Gysemans M., Leysen J.E., Sipido V., Gommeren W., Janssen P.M.F., Terriere D., Mertens J., Synthesis and Application of Isotopes and Isotopically Labelled Compounds, ed. Dr. J. Allen and Dr. R. Voges; John Wiley and Sons 1995, in press. 5. Leysen J.E., Niemegeers C.J.E., Van Nueten J.M., Laduron P.M., Mol. Pharmacol., 21, 301-314 (1982). 6. Leysen J.E., Gommeren W., Mertens J., Luyten W.H.M.L., et al., Psychopharmacology, l& 27-36 (1993).