Search

Introduction Significant discrepancies between in vitro and in vivo binding of the muscarinic receptor ligand – 3H-labeled Quinuclidinyl Benzilate (QNB) – have been well documented. Discernable in vivo cerebellar [3H]QNB binding has been observed in mouse brain, despite the maximum number of binding sites (Bmax) being low. In order to understand this unique in vivo binding phenomenon, the binding of two muscarinic receptor ligands – [3H]QNB and N-[11C]methylpiperidyl Benzilate ([11C]NMPB) – were compared in vivo and in vitro in 3- and 8-week-old mice. Method In vitro binding parameters of [3H]QNB were determined using brain homogenates. The time course of radioactivity concentration (TACs) in the cerebral cortex and cerebellum was measured following injection of [3H]QNB and [11C]NMPB with or without 3 mg/kg of carrier QNB in 3- and 8 week old mice using a dual tracer administration technique. A graphical method was employed for the quantitative analysis of in vivo binding of these radioligands. Results In vitro, the available number of binding sites for cerebral cortical muscarinic receptors increased by 17% during the developmental period studied. Paradoxically, in vivo, we observed a decrease of [3H]QNB binding in the cerebral cortex, while [11C]NMPB binding was markedly increased. In vivo saturation analysis of [3H]QNB in 3-week-old mice revealed an apparent positive cooperativity of binding in the cerebral cortex. Conclusions Our results support the hypothesis that microenvironmental factors proximal to muscarinic receptors cause a local decrease in the cortical free-ligand concentration of [3H]QNB and that this ‘ligand barrier’ is modulated during brain development. Advances in knowledge The present study demonstrates that the combined use of radiolabeled QNB and NMPB has the potential to reveal the important effects of receptor microenvironmental factors on receptor function in the living brain.

Cyclooxygenase-2 (COX-2) plays crucial roles in progressive neuronal death in ischemic brain injury. In the present study, we evaluated two radiolabeled COX-2 selective inhibitors, [11C]celecoxib and [11C]rofecoxib, as positron emission tomography (PET) tracers for COX-2 imaging in normal and ischemic mouse brains. We also took advantage of our newly-generated antibody highly selective for mouse COX-2 to prove accumulation of the radioligands in regions enriched with COX-2. In vitro autoradiography demonstrated specific binding of high-concentration [11C]rofecoxib but not [11C]celecoxib to the cerebellum and brain stem of normal brains wherein COX-2 immunoreactivity in neurons was most abundantly observed. Meanwhile, both of these radioligands failed to detect COX-2 expression in PET assays despite their excellent brain permeability. Hypoperfusion-induced ischemia caused marked necrotic neuron death accompanied by gliosis and enhancement of neuronal COX-2 immunoreactivity in the hippocampus. Correspondingly, in vitro autoradiographic binding of [11C]rofecoxib was increased in the injured hippocampus compared to the uninjured contralateral region, but failed in living brains of ischemia model likewise. Our work provides the rationale for monitoring COX-2 as a biomarker reflecting ischemic brain injuries and demonstrates that [11C]rofecoxib, not [11C]celecoxib, is useful for in vitro assays of COX-2, but its affinity would be insufficient for in vivo PET visualization.