The radiosynthesis of [18F]PK 14105 as an alternative radioligand for peripheral type benzodiazepine binding sites

Appl. Radial. hr. Vol. 41, No. 5, pp. 417482, hr. J. Radiar. Appl. Ins~um. Parr A

0883.2889/90 $3.00 + 0.00 Copyright % 1990 Pergamon Press plc

1990

Printed m Great Britain. All rights reserved

The Radiosynthesis of [‘*F]PK 14105 as an Alternative Radioligand for Peripheral Type Benzodiazepine Binding Sites C. PASCALI, S. K. LUTHRA, V. W. PIKE,* G. W. PRICE, R. G. AHIER, S. P. HUME, R. MYERS, L. MANJIL and J. E. CREMER MRC

Cyclotron

Unit, Hammersmith

Hospital,

Ducane

Road,

London

WI2 OHS, U.K

(Received 20 July 1989) A method has been developed for labelling PK 14105 [N-methyl-N-(l-methyl-propyl)-l(2-fluoro-5-nitrophenyI)isoquinoline-3-carboxamide], a ligand that has high affinity and selectivity for peripheral type benzodiazepine binding sites (PBBS), with NCA fluorine-18 (l,,* = 109.8 min, j + = 96.9%). The method involves treating the 2-chloro-analogue with cyclotron-produced NCA [‘*F]fluoride in dimethyl sulphoxide, with rubidium carbonate as base, at 140°C for 20min. Purification is achieved by separation on a reverse phase Sep-Pak followed by HPLC on a silica gel column, to give chemically and radiochemically pure product with a specific activity of ca 7.4 GBq/pmol (200 mCi/pmol), decay-corrected to the end of radionuclide production (EOB). The radiosynthesis requires 210min, giving a radiochemical yield of lo-20%. decay-corrected to EOB. [‘*F]PK 14105 was found to bind avidly to sites associated with kainic acid-induced unilateral lesions of rat striata. Such binding was blocked by pre-dosing the rat with PK I1 195, so providing evidence for specific binding to PBBS. These results suggest that [18F]PK 14105 has potential for studying phenomena associated with PBBS in man by PET.

autoradiographically a similar effect as shown (Dubois et al., 1988). PBBS radioligands have been shown to bind to gliomas in the rat (Starosta-Rubinstein et al., 1987) and in human post-mortem tissue (B&avid&s et al., 1988). [rH]Ro 5-4864 and [3H]PK 11195 have been used as specific radioligands. [‘H]PK 11195 is the ligand of choice for the study of PBBS in human tissue because of its greater affinity. With the development of carbon-l 1 labelled PK 11195 (Camsonne et al., 1984) it became possible to study PBBS in living man with positron emission tomography (PET). PK 14105, like PK 11195, has high affinity and selectivity for PBBS (Doble et al., 1986). Calculation (Hansch and Leo, 1979) suggests its lipophilicity to be only marginally different to that of PK 11195. Structurally PK 14105 (I) is a close analogue of PK 11195 (II), as it differs only in possessing an aromatic fluoro group in para relationship to a nitro group, instead of a single chloro group. This structural relationship suggested the possibility of labelling PK 14105 with no-carrier-added (NCA) ‘*F (tllZ = 109.8 min, fl+ = 96.9%), and thereby of providing a longer-lived radioligand for PET studies of PBBS. Here we report on the successful labelling of PK 14105 with NCA 18F for i.v. injection and also preliminary biological data on [‘*F]PK

Introduction Binding sites for benzodiazepines are now classified as either “central” or “peripheral”. Peripheral benzodiazepine binding sites (PBBS), unlike the central type, display no association with GABA (Richards et al., 1982). They display very low affinity for benzodiazepines that are specific for the central site, such as clonazepam. but high affinity for Ro 5-4864 (Schoemaker et al., 1981) and for PK 11195 (Le Fur et ~1.. 1983), ligands that are almost inactive at central type sites. PBBS are well represented in kidneys, liver and lung (Anholt et al., 1985) and in several cell types of the immune system (Taniguchi et al., 1980; Zavola ef al., 1984). Their distribution does not exclude the CNS, but in regions where they occur [e.g. olfactory bulb, neurohypophysis, chlorid plexus, ependyma and pineal gland (Richards and Mohler, 1984)] they are predominantly located on non-neuronal elements such as glial and ependymal cells (McCarthy and Harden, 1981). Intrastriatal injection of neurotoxins causes a marked increase in the number of PBBS in the striatum, possibly associated with the ensuing gliosis (Schoemaker et al., 1982; Owen et al., 1983; B&avid& et al., 1987). Brain ischaemic insults cause *Author for correspondence. 417

478

C. PASCALI L'Ia(.

14105 as a radioligand rats.

I

for PBBS in striatal

Preparution

lesioned

Large quantities (5 g) of RP 58271 were synthesized from I-(2-chlorophenyl)isoquinoline-3-carboxylic acid and N-methyl- 1-methyl-propylamine. by nitrating the acid to I-(2-chloro-5-nitrophenyl)isoquinoline-3-carboxylic acid, treating this with ethyl chloroformate plus triethylamine to give the ethyloxy carbonyl derivative of the acid, and finally treating this derivative with N-methyl- 1-methyl-propylamine (Gueremy, 1989). Product was crystallized from diisopropyl ether to give pale yellow crystals (m.p. 164 C). This product had the same mobility as reference RP 58271 on TLC (silica gel, EtOH :AcOEt 9: I v/v: silica gel-C 18. CH,CN : saturated NaHCO, solution: H,O 7 : 0.3 : 2.7 WV). No impurities were detected by U.V. irradiation or exposure to iodine. Mass spectrometry (E.I. 70 eV) gave: m ,‘-_= 397 (M * . 60%); 382 ([M-CH,]‘. 12%); 368 ([M-CH,CH,]-. 100%); 340 ([M-CH(CH&H,]+, 50%): 31 I 70%; 237 WI-NC, H,zl+, 25%); 283 (311-CO. (283-NO,, 28%).

LL

\

\

/

,N

1

w 1

/

\

%R2

(I)

R1 = F, FP = NO2

(II)

Rl = Cl, R2 = H

(Ill) R’ = Cl, R2 = NO2 (IV) Rl = 18F, R2 = NO2

[‘“F]Fluoride

Experimental

and Results

Muterials

Authentic samples of N-methyl-N-( I-methylpropyl) - I(2 - fluoro - 5 - nitrophenyl)isoquinoline -3carboxamide (PK 14105) and N-methyl-N (l-methylpropyl) - I(2 - chloro - 5 - nitrophenyl)isoquinoline -3carboxamide (RP 58271) (III) were kindly donated by Dr C. Gueremy of Rhone-Poulenc Sante, Vitry Sur Seine (France). I-( 2-Chlorophenyl)isoquinoline-3carboxylic acid and N-methyl- 1-methyl-propylamine were also gifts from Dr C. Gueremy. Acetonitrile (HPLC grade, FSA Laboratory Supplies) and dimethyl sulphoxide (DMSO) ( > 99% purity, Aldrich Chemical Company, Inc.) were dried over calcium hydride and distilled before use. Rubidium carbonate (99% purity) was obtained from BDH Chemicals Ltd. Table

qf RP 58271

production

[‘XF]Fluoride was produced with the MRC Scanditronix MC 40 (Mark II) cyclotron by the ‘80(p.n)‘XF reaction on 180-enriched water (20 atom%; 2.5 mL). Irradiation was carried out with a 20 PA beam of 19 MeV protons. Short-lived radionuclidic impurities (e.g. “N) were allowed to decay out before the activity of lXF was measured in a calibrated high pressure ionization chamber. Optimizcrtion

of‘ wnditions

JOr [“F]fluoro

for chloro

exchange in RP 58271 Displacement of the chloro group in RP 58271 by [‘XF]fluoride was attempted under a variety of reaction conditions (Table I). In each experiment 0.331.3 GBq of cyclotron-produced aqueous [‘“F]fluoride solution (0.551 .O mL) was taken to dryness in the stated vessel [preloaded with a magnetic follower

I, Radiochemtcal yield of crude [‘“F]PK 14105 from the reactmn of [‘“F]Ruortdr wth RP 58271 under different conditmns. Unless otherwise stated the reaction vessel was a platmum cruable (%) Radiochemlcal

Base

Solvent*

lBu,NOH tBu,NOH K?CO,

DMSO DMF M&N

R&CO, cs,co, Rb!CO, Rb,CO,

MeCN DMF DMSO

wt of plXWrS0r (Wi IO I0 5 IO 5 10 5 2 3.8 4 5 IO

Oil bath temperature 85

100

122

137

yleldt ( C)

140

I50

I60

170

3 2 7.8 5.4 0 I

35 9.x:

xc: (n = 2) 3.4 445 (n = 3) 395 (n = 5) 415 (n =4): 7:

48

44

44$ (n = 2) 29

‘Typical volume I .5 mL. tCrude product separated by Sep-Pak. TLC with autoradiography of this mi-rure shows mostly PK 14105. iGlass vessel. silylated with Surfasil slliconiring agent (Pierce Chemical Co.) in dichloromethane (10% Y,I\‘). QAverage value for n preparatmns.

28

Radiosynthesis

and an aqueous solution (100 pL) of the stated base (5 mg) in acetonitrile (I mL)] by heating (1 IOC) under nitrogen. Three more times acetonitrile (1.5 mL) was added and taken to dryness. A solution of the precursor (III) was then added and the reaction carried out under the described conditions (Table I). At the end of reaction, the solution was passed through a C 18 Sep-Pak (Waters Associates) that had been primed by elution with ethanol (5 mL) and then water (10mL). The Sep-Pak was then washed first with water (5 mL) and then ethanol (5 mL). Radiochemical yield, in each case, was calculated as the percentage of the initial radioactivity (decaycorrected) that eluted with the ethanol fraction (Table I). Preparation

of [“F]PK

14105

The following method was developed for the preparation of [18F]PK 14105 in a form suitable for i.v. injection. The radioactivity was recovered from the target and transferred to a platinum crucible containing a Teflon-coated stirrer bar. To this solution were added an aqueous solution (100 p L) of rubidium carbonate (5 mg) and then dry acetonitrile (0.5 mL). The solution was evaporated to dryness by heating the oil bath at 100-C under a slow stream of nitrogen. Acetonitrile (1 mL) was added and evaporated under nitrogen three more times. A solution of RP 5827 1 (4 mg) in dry DMSO (1.5 mL) was then added to the crucible, which was covered with a watch glass and moved to a second oil bath kept at 140°C. The solution was left to reflux for 20min. Then the crucible was removed from the oil bath and, after 2 min, distilled water (5 mL) was added. The resultant yellow solution was chromatographed on a Cl8 Sep-Pak cartridge as described in the optimization procedure. The ethanol fraction was evaporated to dryness, the residue dissolved in HPLC mobile phase (pentane : CHCI, : Et, N, 92.5 : 7.5 : 0.1 v/v; 1 mL) and loaded onto a silica gel column (30 x 0.7 cm id.; particle size 10 pm, “p-Porasil”, Waters Associates), eluted at 6 mL/min. The eluate was monitored continuously for radioactivity and for absorbance at 278 nm. The radioactive fraction having the same retention time as reference PK 14105 (35min) was collected. In order to separate the product from the starting material (retention time: 32min), the collected radioactive fraction was re-chromatographed under the same conditions. The collected radioactive fraction was finally evaporated to dryness and the residue dissolved in absolute ethanol (0.2 mL). This solution was then diluted with normal saline for injection (1.8 mL, 0.9% v/v NaCl. BP, Boots Ltd). *The animal studies were carried out by licensed investigators in accordance with the British Biological Council’s Guidelines on the USC oj” Living Animals in Scientific Research. 2nd edn.

of [‘*F]PK

14105

479

The radiochemical yield [decay-corrected to the end of radionuclide production (EOB)] of the formulated radioactive product averaged 14.7% over 10 preparations. The overall synthesis time was ca 210 min from EOB. Analysis

of [“F]PK

14105

(1) High performance liquid chromatography. A sample of the radioactive fraction from preparative HPLC was analysed by re-injection on the preparative HPLC system. A single radioactive peak, with the same retention time (35 min) as authentic PK 14105 was obtained. The precursor RP 58271 was the only stable compound occasionally detected but the amount was always less than 1 nmol (0.4 pg). The specific activity was estimated from measurement of the radioactive peak in a calibrated high pressure ionization chamber and calculation of the mass of compound represented by the U.V. absorbance peak (the chromatography system was precalibrated for this purpose). The highest specific activity achieved was 7.4 GBq/pmol (200 mCi/pmol), decay-corrected to EOB, corresponding to a specific activity at the end of synthesis of 1.38 GBq/pmol (37.2 mCi/~mol). The starting activity in this particular synthesis was 0.79 GBq (21.4 mCi) and the amount of stable PK 14105 was 23 nmol (8.6 ng). (2) Thin layer chromatography. A sample of the radioactive fraction from preparative HPLC was analysed by thin layer chromatography on silica gel (CAMLAB Polygram SIL GjUV 254) using AcOEt : heptane (8 : 2 v/v) as eluent. The stable compound was visualized under u.v.-light and radioactivity by autoradiography. The product was found to be chemically and radiochemically pure and to comigrate with the appropriate reference compound. (3) Mass spectrometry. After radioactive decay, the radioactive fraction from preparative HPLC was evaporated and the residue examined by mass spectrometry (E.I., 70 eV). The obtained spectrum was identical to that recorded for reference PK 14105, i.e. m/z =381 (M+. 90%); 352 ([M-CH$H3]+, 15%); +, 53%); 295 ([M324 ([M-CH(CH,)C,H,] NC,H,,]+, 30%); 267 (295CO,75%); 221 (267-NO?, 27%). Excitotoxic

lesion *

Adult male Sprague-Dawley rats (22&250 g) were anaesthetized with a fluorane/N,O/O, mixture and immobilized in a Kopf stereotactic frame. Unilateral striatal injection of kainic acid (4 nmol in 4pL phosphate buffered saline) was performed through a cannula implanted at co-ordinates AP: 0.7 mm, ML: 3.00mm from bregma according to the atlas of Paxinos and Watson (1986) with the tip 5.3 mm below the pial membrane. The cannula was connected with polyethylene tubing to a 10 p L Hamilton syringe, motor driven to give a perfusion rate of 1 pL/min. Following the injection, the cannula was

C.PASCALI et ul

480

Time

0

I

0

post

injection (minsl

I

1

I

,

I

I

10

20

30

40

50

60

I 70

time post injection (mms)

I 0

10

20

30

40

50

60

time post injection (mans)

Fig. I. The binding of [“F]PK 14105 in striatal lesioned rats at various times after i.v. injection. n . Left lesioned striaturn; 0, right non-lesioned striatum: 0, cerebellar vermis; 0, blood; x. plasma

Fig. 2. The binding of [“F]PK I4105 in striatal lesioned rats, pre-dosed with PK I II95 (3 mgikg), at various times after i.v. injection. n . Left lesioned striatum; 0. right non-lesioned striatum: 0. cerebellar vermis; 0. blood: x . plasma.

left in place for a further 5 min before being slowly withdrawn. Animals were sutured and allowed to recover for 6 days. Confirmation of lesion was observed by marked contralateral turning for the first 24 h and ipsilateral turning between 48 and 72 h post-operative.

ation as a potential radioligand for PET studies of PBBS receptors. The introduction of [‘XF]fluoride into aromatic rings through pyrolysis of aryl diazonium salts (Robbins rr al.. 1978; Strouphaver et cd., 1984) or the Wallach triazene decomposition reaction (Tewson and Welch, 1979) suffers from low radiochemical yields, gives low specific activity and lacks versatility. For these reasons these methods have been largely abandoned in favour of aromatic nucleophilic substitution by NCA [‘“Flfluoride (see Berridge and Tewson. 1986). This kind of reaction can occur when there is sufficient activation by one or more electronwithdrawing groups (nitro, cyano or carbonyl) in ortho or pura relationship to the leaving group on the aromatic ring (Shiue et al., 1984). Compound (III), containing a nitro group in paru relationship to a chloro group, therefore seemed a good candidate for nucleophilic substitution with NCA [‘*F]fluoride to yield [“F]PK 14105 (IV). Though the nitro group also acts as a strong leaving group in aromatic nucleophilic substitution (Attina et ul.. 1983; Shiue et al.. 1984) competing displacement by [‘8F]fluoride was considered unlikely because the puru-chloro group is only weakly activating. Among the halogens, the leaving group ability is generally F > I > Br - Cl (see Berridge and Tewson, 1986) for exchange with 18F Isotopic exchange (i.e. ‘“F for F) has the drawback’that a large quantity of carrier is inevitably introduced. Iodides are sometimes unsuitable for aromatic nucleophilic substitution by virtue of thermal or photo-instability. Though the nucleofugacity of chloro is similar to bromo, the use of bromo as a leaving group would be expected to confer an advantage of easier product purification by chromatography. Even so, this preliminary study was conducted with the chloro precursor because it was readily available to us (initially as a gift and then by synthesis). Reaction conditions for the exchange of [‘XF]fluoro for chloro were investigated under a wide range of conditions. Many of these gave very low yields. The

Lesioned rats were anaesthetized as above and a tail vein and artery cannulated. Animals were allowed to recover from the anaesthesia for 223 h under light restraint. [lXF]PK 14105 (3.7 MBq;rat) in saline (300 ~1L) was injected as a bolus via the tail vein. Arterial samples were taken throughout the experiment. Animals were guillotined at various time intervals post-injection, brains rapidly removed and regions of interest dissected out. Radioactivity in tissue. blood (I 5 JoL) and plasma (30 !tL) samples was determined immediately on a Wallac counter (LKB) with correction for physical decay. Tissue samples were then weighed. Radioactivity was normalized for differences in animal body weight and for injected dose, according to the expression: Normalized

activity =

Tissue radioactivity radioactivity

(MBq) x body weight (g)

injected (MBq) x wet weight of tissue (g)

The time courses of [lXF]PK 14105 binding in tissue (Fig. I), blood and plasma (Fig. I, inset) were then drawn from the normalized activities. Similar time courses of [lXF]PK 14105 binding were also determined in rats injected with PK II 195 (3 mg:kg i.v.) I5 min prior to the injection of radioactivity (Fig. 2).

Discussion The aim of the present work was to label PK 14105 with “F at high specific activity, for biological evalu-

Radiosynthesis of [“F]PK 14105 best conditions were found to be the use of DMSO as solvent and rubidium carbonate as base at 14O”C, which gave an average radiochemical yield of 44% (Table 1). We have always observed higher yields for nucleophilic fluorinations with [‘8F]fluoride in platinum vessels compared to glass and silylated glass. However, in some reactions the yields in glass are only slightly lower than in platinum (see e.g. Brady et al., 1989). A glass reaction vessel could perhaps be more convenient for use in an automated radiosynthesis. It was therefore of interest to assess the effect of a glass vessel on the yield of [‘“Flfluoro for chloro exchange In (III). However. a marked decrease in yield was observed (Table 1). The small structural difference between RP 58271 and PK 14105 combined with their ratio in the reaction mixture (mg vs pg respectively) made complete purification difficult. Preparative chromatography on normal phase (p-Porasil) and reverse phase columns (p-Bondapak C18, Partisil ODS II), using a large variety of eluents, failed to give good separations. Complete separation could only be achieved by re-cycle chromatography on a p-Porasil column, as described in the Experimental section. In the course of the study we progressively tried to reduce the amount of starting material in the reaction (see Table 1) in order to improve the separation of product from precursor. Nevertheless, an amount of RP 58271 (0.5-l nmol; 0.2-0.4pg), small compared to the amount of PK 14105 (ca 21 nmol; 8 pg), was occasionally found. Mass spectrometry on residual carrier in [lXF]PK 14105 revealed only PK 14105. Care was taken to minimize contamination by RP 58271 because this compound also has strong affinity (K, = 17.5 nM for crude rat olfactory bulb homogenates at 20 C. compared to a K, of 5.2 nM for [3H]PK 14105; Price, 1989) for PBBS and might therefore act as unwanted “pseudo-carrier”. For the described animal experiments it was not necessary to sterilize the formulated sample. However, it should be noted that sterilization of the formulated sample (vol. 2mL) by filtration through a Millex GS filter (pore size, 0.22 pm; Millipore Corp.) resulted in 90% retention of the activity on the filter. Use of a Millex FG filter (pore size, 0.2 pm; Millipore Corp.) preconditioned with ethanol then water gave a lower retention (57%) of activity. Camsonne et al. (1984) report a similar loss (50%) of radioactivity upon sterilization of the structural congener, [“C]PK I 1195. by Millipore (0.22 pm) filtration. For any routine use of [‘“F]PK 14105 as a radioligand for PET studies in man, further optimization of the sterilization procedure to avoid such losses would be desirable. The specific activity achieved for [‘*F]PK 14105 (7.4 GBq/pmol; 200 mCi/pmol. decay-corrected to EOB) proved to be adequate for the described animal experiments. Much higher specific activities would be expected from higher starting activities, such as can

481

be produced from irradiations of more highly 180enriched (95-99 atom %) water (Kilbourn et al., 1984). The results of in vivo binding experiments, in which [‘*F]PK 14105 was injected into rats with unilaterally lesioned striata, demonstrate that [‘*F]PK 14105 rapidly crosses the blood-brain-barrier and that there is a marked retention of radioactivity in the lesioned striatum not seen in the unlesioned striatum or cerebellar vermis (Fig. 1). The ratio of radioactivity in the lesioned striatum to that in the unlesioned striatum is maximally about 3 at 12-25 min postinjection. Clearance of radioactivity is quite rapid from all areas, and from blood and plasma (Fig. I, inset). For a similar set of rats, that had been dosed with PK 11195 to saturate PBBS before injection with [“F]PK 14105, no significant differences were seen between lesioned, unlesioned striatum or cerebellar vermis in either uptake or clearance of radioactivity (Fig. 2). These observations provide evidence that the markedly increased uptake of [‘*F]PK 14105 into the lesioned striatum of rats is due to specific binding to PBBS. In more detailed studies (Price et al., 1990) it has been shown that the ratio of specific PBBS binding to non-specific binding for [“F]PK 14105 is comparable to that for [3H]PK 11195. However, in accord with the somewhat lower receptor binding affinity (Doble eb al., 1986) and lipophilicity of PK 14105 compared to PK 11195, the signal is not so persistent as that with [3H]PK 11195. On the basis of these results in the rat, [“C]PK 11195 would be the preferred radioligand for PET studies of PBBS in man. However, recent saturation binding studies (Price, 1989) show a species difference in the order of binding affinities. Thus in human temporal cortex the KD at 20°C is 4.3 nM for [‘H]PK 14105 vs 6.6nM for [3H]PK 11195, whereas in rat olfactory bulb the K, at 20°C is 5.3 nM for 13H]PK 14105 vs 1.4 nM for 13H]PK 11195. On this basis [3H]PK 14105 might yet prove to be the preferred radioligand for human PET studies. The results described here demonstrate that PK 14105 can be labelled with NCA ‘*F in one step and suggest that [‘*F]PK 14105 could serve as a radioligand for PET studies of PBBS in man. The short half-life of “C (t,,? = 20 min) virtually necessitates that [“C]PK 11 I95 be prepared and studied “on site”. Since ‘*F can be produced in high initial activity (ca 35 GBq) and has a sufficiently long half-life (t,,2 = 1 IO min for transport over considerable distance, [18F]PK 14105 might be of value to the increasing number of satellite PET centres (centres which have no cyclotron but which possess a PET camera and rely on the import of radiopharmaceuticals) as an alternative to [“C]PK 11 I95 for the study of PBBS in man. Acknowledgement-The authors gratefully acknowledge Dr S. L. Waters for performing mass spectroscopy.

C.

482

PASCAL1

References Anholt R. R.. DeSouza E. B., Oster-Granite M. L. and Snyder S. H. (1985) Peripheral-type benzodiazepine receptors: autoradiographic localisation in whole-body sections of neonatal rats. J. Pharmacol. E.vp. Thrr. 233, 517.

Attina M., Cacace F. and Wolf A. P. (1983) Labeled aryl fluorides from the nucleophilic displacement of activated J. Luhelled Compd. Radiopharm. nitro groups by “F-F 20, 501.

B&avid& J.. Fage D.. Carter C. and Scatton B. (1987) Peripheral type benrodiazepine binding sites are a sensitive indirect index of neuronal damage. Brain Rec. 421, 167.

B&avid& J.. Cornu P.. Dennis 7.. Dubois A., Hauw J J.. MacKenzie E. T.. Sazdorvitch V. and Scatton B. (1988) Imaging of human brain lesions with ojl cite rddiohgand. Ann.

Neural.

24, 708.

Berridge M. S. and Tewson T. J. (1986) Chemistry fluorine-18 radiopharmaceuticals. Appl. Rodiar. Iwr.

of 37.

685.

Brady F., Luthra S. K. and Pike V. W. (1989) Nucleophihc substitution at cychc sulphamates-routes to NCA [‘XF]fuoro-analogues of MK 801, AppI. Radrar. I.rot. 40, 325.

Camsonne R.. Crouzel C., Comar D.. Maziere M., Prenant C., Sastre J., Moulin M. A. and Syrota A. (1984) Synthesis of N-[“Clmethyl. N-(methyl-I-propyl). (chloro-2phenyl)-I isoquinoline carboxamide-3 (PK I 1195): a new ligand for peripheral benzodiazepine receptors. J. Lahelled

Compd.

Radiopharm.

21, 985.

Doble A., Ferris O., Burgevin M. C.. Menager J.. Uzan A.. Dubroeucq M. C.. Renault C., Gueremy C. and Le Fur G. (1986) Photoaffinity labeling of peripheral-type benzodiazepine-binding sites. Mol. Pharmacol. 31, 42. Dubois A.. Benavidts J., Peny B., Duverger D.. Fage D.. Gotti B., Mackenzie E. T. and Scatton B. (1988) Imaging of primary and remote ischaemic and excitotoxic brain lesions. An autoradiographic study of peripheral type benzodiazepine binding sites in the rat and cat. Bruin Re.c. 445, 77.

G&r&my C. (1989) Personal communication. Hansch C. and Leo A. J. (1979) Substiruent Correlation

Anal~~sis

in

Chemistry

Consfanrs fiw cold Biology,. John

Wiley. New York. Kilbourn M. R., Hood J. T. and Welch M. J. (1984) A simple [‘“Olwater target for ‘“F production. Inr. J. Appl. Radial.

Isot.

35, 599.

Le Fur G.. Vaucher N., Perrier M. L.. Flamier A.. B&avid& J.. Renault C., Dubroeucq M. C., Guertmy C. and Uzan A. (I 983) Differentiation between two ligands for peripheral benzodiazepine binding sites [‘H]RO 5-4864 and [‘H]PK 11195by thermodynamic studies. L[/e Ser. 33, 449. McCarthy K. D. and Harden T. K. (1981) Identification of two benzodiazepine binding sites on cells cultured from rat cerebral cortex. J. Pharmacol. E.Y~. Ther. 216, 183.

Ci td.

Owen F.. Poulter M., Waddmgton J. L.. Mama! R. 1,. and Crow T. J. (1983) [‘H]R05-4864 and [‘Hltlunitrazepam binding in kainate-lesioned rat striatum and temporai cortex of brains from patients with semle dementia of the .4lzheimer subtype. Bwm R1j.r. 278, 373 Paxinos G. and Watson c‘. (19X6; T/IC Rot Brtrirr ~1 Srrreottrsic Coordimr/es. 2nd edn Academic Press. Australia. Phelps M . Mazziotta J. and Schelbert 1-I. (19X6) Pf~.vrrrw Em~xtior~ TomogruplI~~ rrrrd A rr!ortrtlir~~~,-rrl)i~_,~: I’rrwipks uml Applicrrfior~ fiw r/w Brtlm (rm! Hetrr:. Raven Press.

New York. Prtce G. W. (1989) Unpublished observations. Price G. W.. Ahier R. G.. Hume S. P.. Mjcrs R.. Manjil L.. Cremer J. E.. Luthra S. K.. Pascali C.. Pike V. W and Frackovviak R. S. J. (1990). 1,~ I ire bmding to peripheral henzodiazepine binding sites in lesioncd rat hrain: comparison between [‘H]PK I 1195 and [‘“F]PK 14105 as markers for neuronal damage. J. .Veurokwf. In press. Richards J. G. and Mohler H. (19x4) Benzodiarepine receptors. .VrrrroJplirrrr,itrt,~~/~~‘~~.23, 23;. Rtchards J. G.. Mohler H and Haefely W. (I%:!) Benzudiazepine binding sites: receptors or acceptors’! I‘rt,rtrl\ Phurmawl.

Sci. 3. 133.

Robbins P. J., Fortman D L.. Schol/ K. L.. Fusar~> G. A. and Sodd V T. (1978) Synthesis and kincttcs trl [“F]4’Buoroantipyrine in normal mice. J ,Ywl. Med. 19, 1346. Schoemaker H.. Bliss ‘24. and Yamamura H. I. I 1981 ) Specific high affinity saturable binding of [‘H]Ro5-4864 t(, bcnzodiazepine hinding sites in the rat cerebral cortex Eur. J. Pharmacol.

71, 173.

Schoemaker H.. Morelli M., Deshmukh P and Yamamura H. I. (1982) [‘H]R05-4X64 benzodiazepine binding in the kainate-lesioned striatum and Huntington’s diseased basal ganglia. Bruin Rev. 248, 396. Shiue C.-Y.. Wantanahe W . Wolf A P. Fowler J. S. and Salvadori P. (1984) Apphcatiorr of the nucleophilic substitution (0 the synthesis of NCA reaction [‘“Flfluorobenzene and [‘“F]labcled aryl fluorides. J. Lahelled

Compd.

Radiophurm

21, 533.

Starosta-Rubinstein S., Ciliax B. J.. Penney J. B.. McKeever P. and Young A. B. (1987) Imaging of a glioma using peripheral benzodiazeptne receptor ligands. P,V.4.S. 84, 891. Strouphaver A. D., Liotta C. L. and Fink R. W. (1984) Reaction of henzenediazonium-2.carboxylatc wtrh reactor-produced NCA ]“F]fluoride. A novel sythesis of 2-[1XF]fluorobenzoic acid. Int. J. App/. Rudror. Irot. 35, 787. Taniguchi T.. Wang J. K. T. and Spector S. (1980) Properties of [jH]diazepam bmding to rat peritoneal mast cells. L!/k Sci. 27, I7 1. Tewson T, J. and Welch M. J. (1979) Preparation ol fluorine-18 aryl fluorides: pyperidyl trtazenes as a source of diazonium salts. J. Chem Sot. Chm~. Commur~. 1148. Zavola F., Haumont J. and Lenfant M. (1984) Interaction of benzodiazepines wtth mouse maorophages. Eur. .I. Pharmacol. 106, 56 I.