Syntheses and preclinical evaluations of ¹¹ C-labeled radioligands for imaging brain orexin-1 and orexin-2 receptors with positron emission tomography

Despite their importance in regulating several functions in the brain, there is no effective radioligand for <i>in vivo</i> imaging of brain orexin-1 (OX₁R) or orexin-2 receptors (OX₂R) with positron emission tomography (PET). In a search for radioligand candidates, we identified GSK1059865 (1) as a highly potent and selective inhibitor for OX₁R (<i>K</i> _i = 5.0 nM for OX₁R, ∼80-fold selective over OX₂R) and similarly ET1 (2) for OX₂R (IC₅₀ = 0.8 nM for OX₂R, ∼3000-fold selective over OX₁R) with each possessing many physicochemical properties conducive for good brain permeability. We labeled compound 1 and compound 2 with carbon-11 (<i>t</i> _1/2 = 20.4 min) in high isolated yields (∼10-20%), radiochemical purities (≥99.5%), and molar activities (100-340 GBq μmol^-1) and assessed their potential as PET radioligands for <i>in vivo</i> imaging of brain OX₁R and OX₂R in healthy rodents and non-human primates. [¹¹C]1 and [¹¹C]2 showed excellent <i>in vitro</i> stability and also lipophilicity in a desirable range with measured log<i>D</i> _7.4 values of 3.69 and 2.90, respectively. After intravenous administration to mouse or monkey, both [¹¹C]1 and [¹¹C]2 gave moderately high peak radioactivity in brain (∼1.0-1.6 SUV). Unexpectedly, both [¹¹C]1 and [¹¹C]2 showed slightly lower monkey brain uptakes and distribution volumes at baseline than under blocking with suvorexant (a dual OX₁R/OX₂R antagonist), indicating a lack of specific binding to the target receptors in healthy animals. We infer that both OXRs exist in healthy mouse and monkey brain at very low density. Animal models, where OX₁R and OX₂R levels might be elevated, are desirable for candidate PET radioligand development, as are candidates with higher affinity.