IC]Nisoxetine, a Norepinephrine Uptake Inhibitor

Synthesis and Regional Mouse Brain Distribution of [‘I C]Nisoxetine, a Norepinephrine Uptake Inhibitor

MICHAEL S. HAKA and MICHAEL R. KILBOURN*

Division of Nuclear Medicine, Department of Internal Medicine, University of Michigan Medical Center, Ann Arbor, MI 48109, U.S.A.

(Received 26 Mq 1989)

Nisoxetine, a selective and high affinity (IC, = I nM) inhibitor of NE reuptake, has been radiolabeled in high specific activity (>600 Ci/mmol) by the alkylation of the nor-methyl precursor with [“C]CH,I. Synthetic yields are good (4&60% from [” Clmethyl iodide, corrected for decay, 20 min synthesis), with the product purified by HPLC. In vivo studies of regional brain distribution in CD-1 mice show uptake and retention of tracer in the cortex, striatum, hypothalamus and thalamus, with the highest levels in the hypothalamus and cortex. Specific binding in the cortex and hypothalamus can be reduced by preadministration of 7 mg/kg i.v. unlabeled nisoxetine. The possible value of [“Clnisoxetine as a PET imaging agent is discussed.

Introduction prepared by the alkylation of the nor-methyl derivative (1 mg, free base) in 25OpL DMF/DMSO (3:l) Nisoxetine [DL-N-methyl-3-(o-methoxyphenoxy)3- for 5 min at 100°C. Purification and isolation was by phenylpropylamine; Fig. I] is a high affinity HPLC using an alumina column (Maxsil 5 pm) and (IC,, = 1 nM) inhibitor of the presynaptic nor- a gradient of ethanol in chloroform (lo%, O-7 min, epinephrine (NE) reuptake system (Wong and linear gradient lO-20%; 7-9 min, hold at 20% Bymaster, 1976; Wong et al., 1982). It has good for 4min; flow rate 2 mL/min: R, nisoxetine = selectivity for the NE system (serotonin uptake, 10-12 min; R, nomisoxetine > 30 min). The organic IC, = 1000; dopamine uptake, I& = 360) and has solvents were removed by rotary evaporation and the little affinity for neurotransmitter receptors. The product formulated in saline containing a trace characteristics make nisoxetine an excellent candidate amount of HCI. Decay-corrected yields of for use in in vivo studies of the NE uptake system. As [I’ Clnisoxetine (based on [‘I Clmethyl iodide) were part of our program to prepare new radiopharmaceu- 40-60%, with a radiochemical purity of >95% ticals for positron emission tomography (PET), based (TLC, silica gel, 95:5 chloroform/ethanol, R,= 0.3) on the neurotransmitter uptake systems (Kilbourn and sp. act. > 600Ci/mmol (at EOS). et al., 1989), we have prepared (fti2 = 20.4 min)“C- labeled nisoxetine and have begun the evaluation of Animal studies the time course, regional selectivity and specificity of CD-I mice (Charles River; 25-30 g) were anes- this new radiotracer in rodent brain. thetized (ether) and injected with 10-300 PCi of [lLC]nisoxetine in saline-HCl solution via the tail Materials and Methods vein. At designated times the animals were sacrificed Chemicals by decapitation and the brain rapidly removed and dissected [modifications of the method of Glowinski Synthesis of I” Cjnisoxetine. Carbon-l 1 was pro- and Iversen (1966)]. Tissue samples were weighed, duced by proton irradiation of a nitrogen gas target. then counted in an automatic y-counter. The radioactivity obtained in the cyclotron target For blocking studies animals were injected via the (mainly l’COZ) was converted to [“Clmethyl iodide tail vein with nisoxetine hydrochloride (7 mg/kg, by modifications of established procedures (Mara- saline solution), 20 min prior to radiotraer injection. zano et al., 1977; Jewett, 1987). [“C]Nisoxetine was Control animals were injected with vehicle alone. Data were analyzed using a two-tailed Students f- *Author for correspondence. test: P-values are given in Table 1. 112 MICHAELS. HAKAand MICHAELR. KILBOURN

centrations in the cortex and hypothalamus, with little effect in other areas (Table 1). Whole brain OCH, radioactivity (%ID/organ) was decreased by 20%.

Discussion Nisoxetine (Fig. 1) is a simple chemical structure with remarkable selectivity for the NE reuptake system. Although it has been previously prepared in Fig. 1. Structure of [“Clnisoxetine. carbon-14 form (Wong and Bymaster, 1976), the low specific activity obtained with that radionuclide Results (3.48 pCi/mg) did not allow for in uivo study of this compound at true tracer levels. The use of carbon-l 1 Synthesis of [I’ Clnisoxetine allows the study of nisoxetine in vivo at a much higher The synthesis of [“Clnisoxetine is very simple, specific activity (> 600 Ci/mmol). We have success- involving the alkylation of nor-methyl nisoxetine fully prepared this radiotracer by the simple N-alky- with high specific activity, no-carrier-added lation of the nor-methyl derivative with high specific [“Clmethyl iodide. Under these conditions, utilizing activity [“Clmethyl iodide, with the product isolated an excess of the primary amine and no added base, by HPLC. we observe clean mono-methylation at the no-carrier- The regional distribution of [“Qrisoxetine in added level. The product is separated from the pre- mouse brain is consistent with the rank order cursors (nor-nisoxetine, [I’ CJmethyl iodide and previously reported for rat brain using in vitro [“Clmethanol) by an alumina HPLC column. autoradiography and [3H]desmethylimipramine (Lee et al., 1982; Biegon and Rainbow, 1983) or Time course and regional brain uptake [‘Hlmazindol (Javitch et al., 1985). Binding is higher The time course of [“Clnisoxetine distribution in in the cortex and hypothalamus than in the striatum, mouse brain is shown in Table 1. Initial uptake is cerebellum or thalamus. The gross dissection tech- nearly uniform through all brain tissues, and with nique utilized in this work does not allow finer increasing time radioactivity levels in all brain tissues determinations of regional radiotracer binding, and slowly decrease. Blood levels of carbon-l 1 are actu- thus we cannot examine if the binding of ally higher at 10min than at 2min, but then also [” Clnisoxetine will show the regional variation slowly decrease with time. [“C]Nisoxetine levels in apparent for noradrenaline concentrations in rodent brain are highest at 2 min and only decrease about brain (Versteeg et al., 1976) or in vitro binding of 30% within the 60-min experiment. At 60min the [‘Hldesmethylimipramine (Lee et al., 1982; Biegon rank order of radioactivity in different brain regions and Rainbow, 1983). Measurement of radioactivity in is hypothalamus = cortex > cerebellum = striatum = very small brain regions is difficult as methodological thalamus. High initial uptake is seen in the heart errors are introduced in the excision, weighing and followed by a rapid decrease, and considerable lung counting of such small samples of tissue. We have activity is seen at 60 min. previously utilized in oivo autoradiography with an ‘*F-labeled dopamine uptake inhibitor, [‘*FjGBR Blocking studies 13119 (Kilbourn et al., 1988) and have obtained a Pretreatment with nisoxetine (7 mg/kg i.v.) caused more precise definition of the regional brain uptake a statistically significant decrease in radiotracer con- patterns than was possible using dissection tech-

Table I. h uivo distribution of [“Clnisoxetine in mouse brain Time (min) 2 10 20 60 60 + nisoxetine % ID/g Striatum 1.46 f 0.20 1.21 +0.41 0.64 f 0.61 0.71 f 0.13 0.76fO.13 Cerebellum I .49 f 0.25 1.O f 0.36 1.01 k 0.23 0.92 f 0.12 0.76 f 0.07 Cortex 1.38 f 0.28 1.12kO.43 1.18 f0.35 1.13 kO.099 0.92 f 0.10’ Hypothalamus 1.4 f 0.29 1.04 f 0.28 1.34 + 0.75 1.19f0.31 0.59 + 0.24’. Thalamus 1.41 + 0.25 1.1 f0.15 0.79 i 0.12 0.81 + 0.29 0.64 ?r.0.23 BlOOd 1.28+0.17 I .72 + 0.5 1.51 It 0.20 0.84 f 0.25 0.82 f 0.23 Heart 12.14k 1.5 3.1+ 0.69 3.4 + 0.41 1.04 f 0.25 1.23 + 0.31 Lung - - - 5.16 + 2.9 7.2 -+ 1.5 % ID /organ 0.71 f 0.12 0.51 +0.15 0.53~0.11 0.513 + 0.08 0.403 * o.os*** Ratios Cortex/striatum 0.96 f 0.07 1.17 kO.27 1.72 + 0.69 1.62 It 0.41 1.21 f 0.19**** Hypo/striatum 0.96+0.16 1.41 + 0.79 1.96 + 1.4 1.51 k 0.67 0.87 * 0.09**** Data shown are time course in control animals, and distribution in mow brain 60 min after prior administration of 7 mg/kg nisoxetine. Data are given as mean + SD. lP = 0.009; **I’ = 0.04; ***P = 0.08; ****P = 0.06. Synthesis and brain uptake of [“Clnisoxetine 113 niques similar to those used here. Thus, more exact cross-reactivity with the other monoamine reuptake determinations of nisoxetine binding sites in brain systems (Fielding and Szewczak, 1984), which would will await autoradiographic studies using complicate in oiuo applications. For these reasons we 1” Clnisoxetine or an “F-labeled derivative. are continuing to examine the properties of nisoxetine Only a portion of the radioactivity in the target in viuo, and in particular its kinetic properties in tissues (cortex, hypothalamus) can be blocked by primates, as well as the changes of specific nisoxetine pretreatment with pharmacological doses of nisox- binding after chemical lesions (e.g. DSP4) or phar- etine (7mg/kg i.v.). The blocking study resulted in macological treatments [MAO inhibitors and reser- reductions of the cortex/striatum and hypothala- pine (Lee et al., 1983)] which should alter NE uptake mus/striatum ratios (Table 1) but not in the cor- sites. Such studies may be of importance in under- tex/cerebellum ratio (data not shown). Decreases in standing the in vivo behavior of NE uptake inhibitors, the specific binding of [“C]nisoxetine in the cortex and have not been attempted to date due to the lack and hypothalamus, but not the striatum, are consist- of a ligand of good selectivity and high specific ent with decreases in in vitro NE uptake or binding activity. of NE uptake inhibitors (13H]desipramine and Nisoxetine has also been reported as a high affinity [‘Hlmazindol) by brain synaptosomes after in uivo inhibitor of NE uptake by synaptosomes prepared treatment of rodents with 6-hydroxydopamine (Bi- from heart tissue (Wong et al., 1975). Although there geon and Rainbow, 1983) or DSP-4 [N-(2- is high initial heart uptake of [“Clnisoxetine, chloroethyl)-N-ethyl-2-bromobenzylamine (Jonsson washout of radioactivity is rapid. This reuptake et al., 1981; Lee et al., 1982; Logue et al., 1985)]. The inhibitor is thus different than 2-iodo- results obtained in this study suggest considerable desmethylimipramine, previously proposed as a (6@75%) levels of non-specific binding for “chemical microsphere” due to high extraction and [” Clnisoxetine (with non-specific binding assumed to retention in heart tissue (Little et al., 1986). The low include binding to other receptor or uptake sites as heart/lung ratio obtained with [‘I Clnisoxetine at well as low affinity high capacity sites). Such levels are 60 min does not support development of this radio- similar to those found with other lipophilic amines, tracer for the study of the noradrenergic innervation such as [“Clcyanoimipramine [serotonin uptake in- of the heart. hibitor (Hashimoto et al., 1987)] and [“Clpyrilamine [histamine H-l receptor antagonist (Yanai et al., 1988)] when studied in similar protocols in rodent Summary brain. In this study we have made the assumption We have prepared [“Clnisoxetine, a high affinity that the in vim binding characteristics of nisoxetine NE reuptake inhibitor, in high specific activity. In will be identical to those in vitro (low affinity for DA vivo studies in mice show adequate brain penetration or 5-HT uptake systems; little affinity for any recep- and good retention of radioactivity, although the tors); these assumptions need to be tested. Lipophilic greater proportion of such appears to be by non-spe- labeled metabolites are unlikely, but have not been cific binding. Regional distribution of radioactivity is ruled out for [” Clnisoxetine: the expected mode of consistent with the known distribution of noradrener- metabolism should be via N-demethylation to form gic nerve terminals. Future evaluation of this radio- [” Clmethanol, and in preliminary experiments (data tracer will include measurements of [‘I Clnisoxetine in not shown) the radioactivity in blood remains as DSP+treated rats and mice, kinetic studies of the lipophilic material throughout the study. regional brain distribution in primates using PET and The data reported here constitute the very first in the determination of possible metabolites in blood ciao study of a high specific activity radiolabeled NE and tissue. uptake inhibitor. Non-specific binding of [“Clnisoxetine appears high, which may diminish its Acknowledgements-This work was supported by Depart- ultimate value as a PET imaging agent. Unfortu- ment of Energy Grant DE-AC02-76EV0203 1, NIH Grant 2 nately, there are few other candidate compounds PO 1 NS 15655 and NIH Training Grant l-T-32-NS0722206 (to M.S.H.). The authors thank Dr David Robertson (Eli which might be radiolabeled. Tomoxetine [N-methyl- Lilly Co.) for samples of nor-nisoxetine and nisoxetine; Drs 3-(0-methylphenoxy)-3-phenylpropylamine (Wong et Kirk Frev and David Kuhl for heluful discussions: the staff ul., 1982)]; is structurally very similar and will likely of the cyclotron facility for production of the radionuclide; suffer the same problems in non-specific binding. and Phil Sherman and Teresa Pisani for performing the dissections. Modifications of either nisoxetine or tomoxetine will require extensive in vitro and in viva analyses to determine the effects of structural changes on affinity References and selectivity for the NE uptake system. Most Biegon, A.; Rainbow, T. C. Localization and characteriza- studies of the NE uptake system (all in vitro) have tion of [‘H]desmethylimipramine binding sites in rat brain utilized [3Hldesmethylimipramine, a compound by quantitative autoradiography. J. Neurosci. 3: 1069; 1983. in vitro which has been shown to bind to a second site Fielding, S.; Szewczak, M. R. Pharmacology of and whose selectivity in viva is unknown. Other nomifensine: a review of animal studies. J. Clin. Psychiat. candidates, such as nomifensine, show considerable 45: 12; 1984. 774 MICHAELS. HAKAand MICHAELR. KILBOURN

Glowinski, J.; Iversen, L. L. Regional studies of cate- Little, S. E.; Link, J. M.; Krohn, K. A.; Bassingthwaithe, cholamines in the rat brain-l. J. Neurochem. 13: 655; J. B. Myocardial extraction and retention of 2-iodo- 1966. desmethylimipramine: a novel flow marker. Am. J. Phys- Hashimoto, K.; Inoue, 0.; Suzuki, K.; Yamasaki, T.; iol. 250: H1060; 1986. Kojima, M. Synthesis and evaluation of [“Cl- Logue, M. P.; Growdon, J. H.; Coviella, I. L. G.; Wurtman, cyanoimipramine. Nucl. Med. Biol. 14: 587; 1987. R. J. Differential effects of DSP4 administration on Javitch, J. A.; Strittmatter, S. M.; Snyder, S. H. Differential regional brain norepinephrine turnover in rats. LiJe Sri. visualization of dopamine and norepinephrine uptake 37: 403; 1985. sites in rat brain using [‘H]mazindol autoradiography. J. Marazano, C.; Maziere, M.; Berger, G.; Comar, D. Synthe- Neurosci. 5: 1513; 1985. sis of methyl iodide-“C and formaldehyde-” C. ht. J. Jewett, D. M. Electrical heating with polyimide-insulated Appl. Radial. Isot. 28: 49; 1977. magnet wire. Rev. Sci. Instrum. 58: 1964; 1987. Versteeg, D. H. G.; Van der Gugten, J.; De Jong, W.; Jonsson, G.; Hallman, H.; Ponzio, F.; Ross, S. DSP4 (N- Palkovits M. Regional concentrations of noradrenaline (2-chloroethyl)-N-ethyl-2-bromobenzylamine)--a useful and dopamine in-rat brain, Brain Res. 113: 563; 1975. denervation tool for central and peripheral noradrenergic Wona._. D. T.: Bvmaster. F. P. Effect of nisoxetine on uotake. neurons. Eur. J. Pharmacol. 72: 173; 1981. of catecholamines in synaptosomes isolated from discrete Kilbourn, M. R.; Haka, M. S.; Ciliax, B. J.; Penney, J. B.; regions of rat brain. Biochem. Pharmacol. 25: 1979; Young, A. B. In vivo autoradiography of [‘*F]GBR 13119 1976. binding in rat brain. Sot. Neurosci. 18th Annu. Meet. Wong, D. T.; Horng, J. S.; Bymaster, F. P. DL-N-Methyl-3- Toronto; 1988: Abstr. 376.1 I. (o-methoxy-phenoxy)-3phenylpropylamine hydrochlo- Kilbourn, M. R.; Haka, M. S.; Mulholland, G. K.; Jewett, ride, Lilly 94939, a potent inhibitor for uptake of D. M.; Kuhl, D. E. Synthesis of radiolabeled inhibitors of norepinephrine into rat brain synaptosomes and heart. presynaptic monoamine uptake systems: [ 18F]GBR 13119 Life Sci. 17: 755; 1975. (DA), [“C]nisoxetine (NE), and [“CJfluoxetine (5-HT). J. Wong, D. T.; Threlkeld, P. G.; Best, K. L.; Bymaster, F. P. Labelled Compd. Radiopharm. 26: 412 (Abstr.); 1989. A new inhibitor of norepinephrine uptake devoid of Lee, C. M.; Javitch, J. A.; Snyder, S. H. Characterization of affinity for receptors in rat brain. J. Pharmacoi. Exp. [)H]desipramine binding associated with neuronal nore- Ther. 222: 61; 1982. pinephrine uptake sites in rat brain membranes. J. Neu- Yanai, K.; Dannals, R. F.; Wilson, A. A.; Ravert, H. T.; rosci. 2: 1515; 1982. Scheffel, U.; Tanada, S.; Wagner Jr, H. N. (N-Methyl- Lee, C. M.; Javitch, J. A.; Snyder, S. H. Recognition sites [“C])pyrilamine, a radiotracer for histamine H-l recep- for norepinephrine uptake: regulation by neurotransmit- tors: radiochemical synthesis and biodistribution study in ter. Science 220: 626; 1983. mice. Nucl. Med. Biol. 15: 605; 1988.