Original Articles
18F-labeled FECNT: a selective radioligand for PET imaging of brain dopamine transporters

https://doi.org/10.1016/S0969-8051(99)00080-3Get rights and content

Abstract

Fluorine-18 labeled 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(2-fluoroethyl)nortropane (FECNT) was synthesized in the development of a dopamine transporter (DAT) imaging ligand for positron emission tomography (PET). The methods of radiolabeling and ligand synthesis of FECNT, and the results of the in vitro characterization and in vivo tissue distribution in rats and in vivo PET imaging in rhesus monkeys of [18F]FECNT are described. Fluorine-18 was introduced into 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(2-fluoroethyl)nortropane (4) by preparation of 1-[18F]fluoro-2-tosyloxyethane (2) followed by alkylation of 2β-carbomethoxy-3β-(4-chlorophenyl)nortropane (3) in 21% radiochemical yield (decay corrected to end of bombardment [EOB]). Competition binding in cells stably expressing the transfected human DAT serotonin transporter (SERT) and norepinephrine transporter (NET) labeled by [3H]WIN 35428, [3H]citalopram, and [3H]nisoxetine, respectively, indicated the following order of DAT affinity: GBR 12909 > CIT >> 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(3-fluoropropyl)nortropane (FPCT) > FECNT. The affinity of FECNT for SERT and NET was 25- and 156-fold lower, respectively, than for DAT. Blocking studies were performed in rats with a series of transporter-specific agents and demonstrated that the brain uptake of [18F]FECNT was selective and specific for DAT-rich regions. PET brain imaging studies in monkeys demonstrated high [18F]FECNT uptake in the caudate and putamen that resulted in caudate-to-cerebellum and putamen-to-cerebellum ratios of 10.5 at 60 min. [18F]FECNT uptake in the caudate/putamen peaked in less than 75 min and exhibited higher caudate- and putamen-to-cerebellum ratios at transient equilibrium than reported for 11C-WIN 35,428, [11C]CIT/RTI-55, or [18F] β-CIT-FP. Analysis of monkey arterial plasma samples using high performance liquid chromatography determined that there was no detectable formation of lipophilic radiolabeled metabolites capable of entering the brain. In equilibrium displacement experiments with CIT in rhesus monkeys, radioactivity in the putamen was displaced with an average half-time of 10.2 min. These results indicate that [18F]FECNT is a radioligand that is superior to 11C-WIN 35,428, [11C]CIT/RTI-55, [18F]β-CIT-FP, and [18F]FPCT for mapping brain DAT in humans using PET.

Introduction

The brain dopamine transporter (DAT) is critical to dopamine neurotransmission and the pyschostimulant effects of cocaine 6, 12, 27 and is decreased by Parkinson’s disease 3, 17, 26. A major recent effort has focused on the development of radiolabeled DAT ligands to study the physiology, pharmacology, and pathophysiology of the brain DAT using positron emission tomography (PET). Initial attempts focused on nontropanes labeled with carbon-11 or fluorine-18. PET imaging studies of DAT labeling with [11C]nomifensine in nonhuman and human primates have been reported (1). [11C]Nomifensine rapidly entered the brain but exhibited a low DAT selectivity that resulted in relatively low striatum-to-cerebellum binding ratios (1.53 at 50 min). A fluorine-18 analogue of a class of DAT selective aryl 1,4-dialkylpiperazines has also been developed for PET studies 18, 19, 20. Imaging studies with [18F]GBR 13119 in nonhuman primates also indicated low DAT selectivity and resulted in low striatum-to-cerebellum ratios (1.51 at 50 min) comparable to those of [11C]nomifensine.

The tropane (-)-cocaine has been labeled with 11C 10, 21 and used to examine brain DAT distribution in nonhuman and human primates. Several shortcomings of this product as a DAT radioligand are its nonselectivity, low binding affinity and rapid metabolism, which result in rapid binding site dissociation (t1/2 = 25 min), and its low target (striatum) to nontarget (cerebellum) binding ratios (e.g., 2.0). Dynamic PET imaging studies in baboons of brain DAT binding by the low affinity tropane [O-11C-methyl]d-threo-methylphenidate ([11C]d-threo-MP) has recently been reported (9). Serial PET images demonstrated that [11C]d-threo-MP accumulates in the DAT-rich striatum, with peak striatal uptake (0.05 %dose/g) occurring rapidly within 5–15 min postinjection. Moreover, [11C]d-threo-MP exhibited a low striatum-to-cerebellum ratio of 2.4:1 at 75 min postinjection.

A number of higher affinity DAT ligands, in which the 3β-(benzoyl) group of (-)-cocaine was replaced by 3β-(4-substituted-phenyl) groups, have been radiolabeled with carbon-11 for in vivo PET studies of the striatal DAT in nonhuman and human primates. This group of DAT PET imaging agents includes [11C]2β-carbomethoxy-3β-(4-fluorophenyl)tropane ([11C]-WIN 35,428) 8, 11, 24 and [11C]2β-carbomethoxy-3β-(4-iodophenyl)tropane ([11C]CIT/RTI-55) (25) and exhibits increased striatal uptake with low nonspecific binding relative to earlier agents. Replacement of the methyl group at the 8-position in CIT by either a 3-fluoropropyl group or a 2-fluoroethyl group yielded the [11C]-labeled DAT ligands N-3-fluoropropyl-2β-carbomethoxy-3β-(4-iodophenyl)nortropane (O-methyl-[11C]FP-CIT) (22) and N-2-fluoroethyl-2β-carbomethoxy-3β-(4-iodophenyl)nortropane ([11C]β-CIT-FE) (16), respectively. Radiosynthesis of fluorine-18 labeled FP-CIT has recently been reported (7) though in only a 1% radiochemical yield. Replacement of the 2β-carbomethoxy group in 2β-carbomethoxy-3β-(4-chlorophenyl)tropane by a 2β-carbo-2′-fluoroethoxy group yielded the DAT PET ligand FECT radiolabeled with fluorine-18 (28).

Recently, we reported the synthesis of high specific activity (2 Ci/μmole) fluorine-18 labeled 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(3-fluoropropyl)nortropane (FPCT) as a new DAT PET imaging agent (15). PET imaging studies in rhesus monkeys with [18F]FPCT showed high radioligand uptake and retention in the putamen and caudate nucleus, brain regions that are rich in DAT, with low nonspecific binding in the DAT poor cerebellum and cerebral cortex. An undesired property accompanying DAT binding by [18F]FPCT, as well as by [11C]-WIN 35,428 and [11C]CIT/RTI-55, is the need for a prolonged (>2 h) period required to reach binding equilibrium with the DAT. Unlike the parent, [11C]CIT/RTI-55, O-methyl-[11C]FP-CIT, and [11C]β-CIT-FE DAT binding have been reported to reach equilibrium 16, 22. PET imaging studies with [11C]β-CIT-FE in nonhuman primates exhibited high (9:1) caudate-to-cerebellum values at the time of transient equilibrium, which occurred 60 min postinjection. PET imaging with [18F]FP-CIT in humans and nonhuman primates also has been reported 7, 23. [18F]FP-CIT in nonhuman primates exhibited high 5:1 caudate-to-cerebellum values at the time of transient equilibrium, which occurred 60–100 min postinjection.

For PET studies, fluorine-18 offers significant advantages in comparison to the short 20 min half-life of carbon-11. The 110 min half-life of fluorine-18 allows sufficient time (3 × 110 min) for isotope incorporation into the tracer molecule and its purification to a final product suitable for human administration and permits the preparation of multidose batches for distribution by a cyclotron based radiopharmacy. Fluorine-18 is also the lowest energy positron emitter (0.635 MeV, 2.4 mm positron range) and affords the highest PET resolution images. Thus, the development of DAT ligands labeled with fluorine-18 that achieve high (>10) caudate/putamen-to-cerebellum ratios and exhibit more favorable kinetics (e.g., peak caudate/putamen-to-cerebellum ratios achieved in less than 2 h) would be ideal candidates for quantitative in vivo mapping of brain DAT sites by PET.

During attempts 14, 15 to develop a fluorine-18 labeled PET probe that exhibits high brain uptake and high selectivity and affinity for the DAT, we identified an alternative N-substituted fluoroalkyl derivative, 2β-carbomethoxy-3β-(4-chlorophenyl)-8-(2-fluoroethyl)nortropane (FECNT) with lower nanomolar affinity than FPCT for DAT (15). A potential advantage of employing a lower affinity selective DAT radioligand such as [18F]FECNT is that the radioligand may have a higher rate of dissociation from the DAT binding site, attain binding equilibrium more rapidly, and thus allow measurement of the DAT density from a single-frame image, avoiding the need for more complicated multicompartment kinetic models. In this article, we report the in vitro characterization of FECNT as a DAT ligand, the radiosynthesis of high specific activity [18F]FECNT in 20% (corrected for decay at end of bombardment [EOB]) radiochemical yield, a comparison with [18F]FPCT of the kinetics of regional brain radioactivity using PET, and the in vivo displacement of brain 18F-FECNT binding by CIT in rhesus monkeys.

Section snippets

Experimental

All animal experiments were carried out according to protocols approved by the Institutional University Animal Care Committee and Radiation Safety Committees of Emory University.

Radiolabeling

Fluorine-18 labeled FECNT was prepared following target bombardment by a two step reaction sequence that first involved the preparation of the radiolabeled precursor 1-[18F]fluoro-2-tosyloxyethane (5) in 68% yield, corrected for decay at EOB (Fig. 1). Alkylation of 1-[18F]fluoro-2-tosyloxyethane with 2β-carbomethoxy-3β-(4-chlorophenyl)nortropane in DMF at 135°C afforded [18F]FECNT in 21% overall yield corrected for decay at EOB, in a synthesis time of 122 min including HPLC purification. Final

Discussion

[18F]FECNT, a high affinity ligand for brain DAT, was radiofluorinated by a two-step synthesis using no-carrier added [18F] fluoride (283 mCi K18F) and a commercially available precursor 1,2-ditosyloxyethane (Fig. 1). The synthesis is amenable to automation (13) and yielded 26.5 mCi of [18F]FECNT at end of synthesis (EOS) with a specific activity of 2 Ci/μmol.

Comparison of in vivo imaging studies in the same rhesus monkey demonstrated that both [18F]FECNT and [18F]FPCT showed high preferential

Acknowledgements

This research was sponsored by the Office of Health and Environmental Research, U.S. Department of Energy under Grant No. DE-FG05-93ER61737, and the National Institute of Mental Health Drug Screening Program. The authors thank Carolyn K. Malcolm for assistance in preparing the manuscript, and Delicia Votaw, Teresa Abak, and Elizabeth Smith for assistance in the data collection and analysis.

References (28)

  • J Billings et al.

    Characterization of radio-iodinated TISCHA high affinity and selective ligand for mapping CNS dopamine D-1 receptor

    J. Neurochem.

    (1991)
  • D Block et al.

    The N.C.A. nucleophilic 18F-fluorination of 1,N-disubstituted alkanes as fluoroalkylation agents

    J. Labelled Compd. Radiopharm.

    (1987)
  • F.I Carroll et al.

    Cocaine receptorBiochemical characterization and structure-activity relationships of cocaine analogues at the dopamine transporter

    J. Med. Chem.

    (1992)
  • R.F Dannals et al.

    Synthesis of a radiotracer for studying dopamine uptake site in vivo using PET2β-Carbomethoxy-3β-(P-fluorophenyl)-[N-11C-methyl]tropane ([11C]CFT or (11C-WIN 35,428)

    J. Labelled Compd. Radiopharm.

    (1993)
  • Cited by (124)

    • Microglia, inflammation and gut microbiota responses in a progressive monkey model of Parkinson's disease: A case series

      2020, Neurobiology of Disease
      Citation Excerpt :

      Radiochemical purity was confirmed to be >96% with specific activities greater than 1 Ci/μmol. The preparation of [18F]FECNT was conducted as previously described (Goodman et al., 2000; Masilamoni et al., 2010; Masilamoni et al., 2011). In the present study, we performed [18F]FECNT PET scans to determine the neuroprotective effects of XPro1595 on the nigrostriatal dopaminergic system during the course of escalating MPTP intoxication.

    • Quantification of dopamine transporter density with [<sup>18</sup>F]FECNT PET in healthy humans

      2014, Nuclear Medicine and Biology
      Citation Excerpt :

      Separating the grey and white matter regions could not be investigated in this study because volunteer-specific MRI scans were not available for segmentation. However in the presented analysis, a 2-tissue compartment model provided the best fits to the cerebellum time-activity curves suggesting the presence of a non-specific compartment that was not displaceable in earlier non-human primate studies [27]. Our simulations do address the possibility of a non-specific compartment in the cerebellum and caudate regions.

    View all citing articles on Scopus
    View full text