Introduction: Reactive oxygen and nitrogen species (RONS) contribute to the pathogenesis of neurodegenerative diseases, but inabilities detecting RONS in the central nervous system (CNS) have complicated the interpretation of antioxidant clinical trial results.^1-4 To address this challenge, we introduce the novel radiopharmaceutical [¹⁸F]Fluoroedaravone ([¹⁸F]FEDV), a radiopharmaceutical derived from the FDA-approved antioxidant, Edaravone. [¹⁸F]FEDV is stable in human plasma and interacts with an array of RONS in vitro. [¹⁸F]FEDV-PET identified RONS in the CNS of mice post intrastriatal sodium nitroprusside(SNP) injection, which induces peroxynitrite and lipid peroxidation. [¹⁸F]FEDV also detected RONS after middle cerebral artery photothrombosis (PT) in mice. In this study, we report the radiosynthesis of [¹⁸F]FEDV, its stability and reactivity in vitro, and its utility as a PET-imaging probe of CNS oxidative stress in vivo. 

Methods: RONS chemical-reactivity assays were conducted as previously described;^5,6 HPLC quantified [¹⁸F]FEDV concentrations over time. In vitro assays utilized doxorubicin-treated EMT6 cells to generate RONS (10,000 cells/well)^7,8 and were incubated with [¹⁸F]FEDV (5.6MBq) at 37°C. Radioactivity was measured using a Hidex AMG automated gamma counter. Oxidative stress was induced in mice in vivo by intrastriatal SNP injection, followed by 60 minute dynamic [¹⁸F]FEDV-PET/CT. The PT stroke model was performed as previously described,⁹ with 60 minute dynamic [¹⁸F]FEDV-PET/CT occurring at 4 and 24 hours post-PT. [¹⁸F]FEDV-PET signal was correlated with RONS formation by intravenous hydroethidine injection 30 minutes prior to PET completion, and quantifying oxidized ethidium signal in brain tissue ex vivo using fluorescence microscopy (590 nm). Brain lipid peroxidation post-PT was confirmed ex vivo by malondialdehyde (MDA) assay following manufacturer instructions.

Results/Discussion: [¹⁸F]FEDV was synthesized in a two-pot, three-step reaction and purified by HPLC in 60 minutes (12±1% activity yield (n=8), >99% radiochemical purity). [^18/19F]FEDV reaction with RONS in vitro rapidly converted to [^18/19F]F-OPB within 5 minutes, consistent with Edaravone’s RONS reactivity. EMT6 cells, post-doxorubicin treatment, rapidly accumulated [¹⁸F]FEDV and 10 mM [¹⁹F]FEDV blocked uptake. [¹⁸F]FEDV was stable in 5% ethanol/saline over 8 hours and  >90% unchanged in human plasma after 3 hours. Significantly increased [¹⁸F]FEDV PET signal was observed in SNP vs saline-injected mice over 60 minutes (p<0.001, n=4). Similar experiments with [¹⁸F]FDG (p=0.168, n=4) and [¹⁸F]FN (p=0.921, n=4), a recently reported redox-tuned PET reporter, showed no significant differences between SNP and saline injections. PT model experiments showed a marked increase in [¹⁸F]FEDV signal at 24h post-PT in the ipsilateral cerebral cortex (p<0.001,n=3) over 60 minutes compared to sham or 4h post-PT brains. Imaging the PT model with [¹⁸F]FN showed no difference versus the sham (p=0.913, n=3)  Ex vivo MDA quantification revealed increased lipid peroxidation in the ipsilateral cortex at 24 hours post PT (n=4,p<0.01), but not at 4 hours post PT. Parametric mapping improved [¹⁸F]FEDV sensitivity to RONS at 4h post-PT, consistent with ex vivo fluorescence microscopy and autoradiography; revealing intensified oxHET labeling and radioactive signal, respectively, at 4h and 24h post-PT, compared to sham.

Conclusion: [¹⁸F]FEDV was synthesized in high radiochemical yield and purity, displays a wide reaction spectrum to RONS, and shows exceptional stability in vitro. Following intrastriatal injection of SNP, [¹⁸F]FEDV exhibited a significantly increased PET signal in the ipsilateral hemisphere compared to [¹⁸F]FDG and [¹⁸F]4FN. Following PT, [¹⁸F]FEDV was clearly able to delineate the increased production of RONS at the site of injury after 24 hours when compared contralaterally in the same brain and to sham. These findings suggest that [¹⁸F]FEDV-PET holds promise as an imaging tool for accurately quantifying RONS longitudinally in vivo and warrants further research in preclinical and clinical studies.

Image/Figure:

Click to view full size

Image/Figure Caption:

Figure. A. Scheme of the radiosynthesis of [¹⁸F]FEDV, a radiopharmaceutical analogue of Edaravone B. Quantification of [¹⁸F]FEDV-to-[¹⁸F]F-OPB conversion after exposure to various oxidants for 5, 10, 15, or 20 min. The tested oxidants included hydrogen peroxide (H₂O₂), tert-Butyl hydroperoxide (tBuOOH), hypochlorite (ClO^–), peroxynitrite (ONOO^–), hydroxide (OH^•), nitroxyl (NO^•), tert-butoxide (t-BuO^•), water-soluble lipid peroxyl radicals [LOO^•(aq)], and lipid-soluble lipid peroxyl radicals [LOO^•(lip)]. Shown are the mean of three replicates. C. Axial (top) and coronal (bottom) PET/CT images of [¹⁸F]FEDV in mice treated with saline (left) or rose bengal dye (middle, right), followed by 543-nm laser stimulation at the proximal MCA branch. Mice were imaged at 4 h (middle) or 24 h (right) post- PT. Note the robust [¹⁸F]FEDV signals in the ipsilateral cortex at 24 h, but not at 4 h, post-stroke. D. Representative confocal microscopic images of oxidized hydroethidine (oxHET, red) and DAPI (blue) in sham-treated versus stroke-injured mouse brains at 4 h post-PT. E. Representative autoradiography of [¹⁸F]FEDV-injected mouse brains collected after PET/MRI imaging at 4 or 24 h post-PT. Note the increase of [¹⁸F]FEDV signals (red arrows) in ex vivo autoradiography at both time points. F. Dynamic PET/MRI imaging with parametric mapping shows enhanced uptake of [¹⁸F]FEDV at the sites of T2-indicated stroke injury (yellow arrows) at both 4 (left) and 24 h (right) post-PT.

Author