Synthesis and evaluation of 6-(3-[18F]fluoro-2-hydroxypropyl)-substituted 2-pyridylbenzothiophenes and 2-pyridylbenzothiazoles as potential PET tracers for imaging Aβ plaques

https://doi.org/10.1016/j.bmc.2016.03.034Get rights and content

Abstract

3-[18F]Fluoro-2-hydroxypropyl substituted compounds were synthesized and evaluated as novel 18F-labeled PET tracers for imaging Aβ plaque in a living brain. All compounds exhibited high binding affinities toward the synthetic Aβ1–42 aggregate and/or Alzheimer’s disease brain homogenate. In the microPET study with normal mice, the 3-[18F]fluoro-2-hydroxypropyl substituted compounds resulted in fast brain washout by reducing the lipophilicities of the compounds. Intriguingly, (S)-configured PET tracers, (S)-[18F]1b and (S)-[18F]1c, exhibited a 2.8 and 4.0-fold faster brain washout rate at a peak/30 min in the mouse brain than the corresponding (R)-configured PET tracers despite there being no meaningful difference in binding affinities toward Aβ plaque. A further evaluation of (S)-[18F]1c with healthy rhesus monkeys also revealed excellent clearance from the frontal cortex with ratios of 7.0, 16.0, 30.0 and 49.0 at a peak/30, 60, 90, and 120 min, respectively. These results suggest that (S)-[18F]1c may be a potential PET tracer for imaging Aβ plaque in a living brain.

Introduction

The accumulation of Aβ plaque in the brain is believed to be the main origin of Alzheimer’s disease (AD), which is the leading cause of dementia, accounting for 60–70% of all cases.1 A positron emitting tracer that binds to Aβ plaque selectively offers an opportunity of the early detection of AD,2 and will be useful for identifying new therapeutic drugs and monitoring the treatment results as a surrogate marker.3 In 2002, 11C-labeled [11C]PIB (Pittsburgh compound B) was developed for imaging Aβ plaque in the brain of living subjects by modifying Thioflavin T, which has been used as a fluorescent dye for staining Aβ plaque in postmortem brains.4 [11C]PIB exhibited good Aβ plaque binding in a living brain with an appropriate brain accumulation and washout ratio. On the other hand, it is of limited use in clinical trials because the C-11 isotope has a half-life of only 20 min, which is too short to exhibit wide availability to AD patients from a single automated production. In this regard, many researchers have searched for longer lived F-18 (110 min half-life) labeled PET tracers for clinical use in the diagnosis of AD.5 As a result, three F-18 labeled PET tracers have been approved by the U.S. Food and Drug Administration (FDA) in 2012 (Amyvid™, Eli Lilly),6 2013 (Vizamyl™, GE Healthcare),7 and 2014 (Neuraceq™, Piramal)8 as Aβ imaging agents and a secondary tool for the diagnosis of AD.

To be a good Aβ imaging PET tracer, it should meet two essential criteria in the stage of candidate discovery; selective binding toward Aβ plaque with high affinity and rapid clearance of non-specific binding from the brain.9 Most biaryl and conjugated biaryl compounds have good binding affinities toward Aβ plaque in vitro.10 On the other hand, many F-18 labeled biaryl candidates have failed to meet the desired washout criterion in vivo. The nucleophilic aliphatic [18F]fluorination of alkyl tosylate or mesylate precursors is a common way to label the molecule of interest with F-18.11 The resulting [18F]fluoroalkyl substituents, however, play a key role in increasing the lipophilicity of the molecules, giving rise to poor pharmacokinetic properties as well as undesirable non-specific interactions.

Our attempts have been made to find novel candidates suited to PET imaging of Aβ deposition considering these requirements. As an early our attempt (unpublished), 6-(3-[18F]fluoropropyloxy)-2-(4-methylamino)phenyl benzo[b]thiophene ([18F]FC028) was synthesized and evaluated by in vitro binding assay and in vivo biodistribution. Although it showed an excellent binding affinity of 0.27 nM (Ki) to Aβ aggregates compared to [11C]PIB (0.78 nM), a biodistribution study with normal mice revealed low initial uptake (2.02%ID at 2 min) and a poor washout effect (2.18%ID at 30 min, and 3.14%ID at 60 min), probably due to its high lipophilicity (C Log P = 5.31). The desired Log P for BBB penetration is believed to range from 1 to 3.12 Subsequently, an effort was made to reduce the lipophilicity of the compound for better brain pharmacokinetics by modifying the core structure and functional groups. This paper describes how the introduction of a hydroxyl group and nitrogen atoms to the molecule affects the brain pharmacokinetics. More intriguingly, the large difference in brain pharmacokinetics between (R)- and (S)-enantiomers is also described.

Section snippets

Chemistry

A biaryl compound, 2-phenylbenzo[b]thiophene was chosen as the starting molecule, and the N-methylamino group on C4 position of the right ring was fixed. Two components, a nitrogen atom and hydroxyl group, were considered as polar modifiers to reduce the lipophilicity of the compounds (Fig. 1). Nitrogen atoms were added to the phenyl and benzo[b]thiophene rings to become pyridine and benzo[d]thiazole rings, respectively. The pyridine ring was reported to reduce the level of non-specific white

Conclusion

The introduction of a 3-[18F]fluoropropyl group to the molecules of interest has been a common strategy in the development of F-18 labeled PET tracers until now. The 3-[18F]fluoropropyl group, however, plays a role in increasing the lipophilicity of the molecules, resulting in poor pharmacokinetic properties as well as undesired non-specific interactions. To reduce the lipophilicity of the molecules, a hydroxyl group and a nitrogen atom to 3-[18F]fluoropropyl and aromatic ring were considered,

General methods

All chemicals were obtained from commercial suppliers and were used without further purification unless otherwise stated. Analytical thin layer chromatography (TLC) was performed with Merck silica gel F-254 glass-backed plates. Visualization on TLC was monitored by UV light (254 nm and 365 nm) and phosphomolybdic acid indicator. Column chromatography was carried out using Merck silica gel 60 (230–400 mesh) and eluting with ethyl acetate/hexanes or methanol/dichloromethane. 1H and 13C NMR spectra

Acknowledgement

The research was supported by the Converging Research Center Program through the Ministry of Science, ICT and Future Planning, Korea (2013K000326, 2014M2A2A7066472, and 2014M3C1A8066306).

References and notes (21)

  • J. Hardy et al.

    Trends Pharmacol. Sci.

    (1991)
    D.J. Selkoe

    Neuron

    (1991)
    J. Hardy et al.

    Science

    (2002)
  • A. Nordberg et al.

    Nat. Rev. Neurol.

    (2010)
    H. Quigley et al.

    Int. J. Geriatr. Psychiatry

    (2011)
  • S.R. Choi et al.

    J. Nucl. Med.

    (2009)
  • C.C. Rowe et al.

    Lancet Neurol.

    (2008)
  • C.C. Rowe et al.

    J. Nucl. Med.

    (2011)
  • C.A. Mathis et al.

    Nucl. Med. Biol.

    (2007)
    K.A. Johnson et al.

    J. Nucl. Med.

    (2013)
  • C.A. Mathis et al.

    J. Med. Chem.

    (2003)
    W.E. Klunk et al.

    Ann. Neurol.

    (2004)
  • W.E. Klunk et al.

    Curr. Opin. Neurol.

    (2008)
  • M. Koole et al.

    J. Nucl. Med.

    (2009)

    J. Nucl. Med.

    (2013)
  • W. Zhang et al.

    Nucl. Med. Biol.

    (2005)
There are more references available in the full text version of this article.

Cited by (0)

Current address: Department of Chemistry and Center for Innovation in Chemistry, Faculty of Science, Burapha University, Sangsook, Chonburi 20131, Thailand.

View full text