[¹⁸F]AV1451 is a Weak Inhibitor of Monoamine Oxidase (MAO) A and B

Objectives: There is a high degree of structural similarity between some Tau radioligands (AV1451, THK5351, and THK51117) and the beta-carboline monoamine oxidase (MAO) inhibitors. Recently multiple groups have evaluated the possible off-target binding of these ligands on post-mortem human Alzheimer’s Disease (AD) tissue;¹ Ng et al., showed in human studies decreased [¹⁸F]THK5351 binding (36-51%) in patients who had taken 10 mg Selegiline (MAO-B irreversible inhibitor).² In the case of [¹⁸F]THK5351, this greatly confounded the PET images. [¹⁸F]AV1451 is currently being used in 47 clinical trials (clinicaltrials.gov); despite supposedly having off-target affinity for MAO-A and -B, the extent of this off-target effect has not been addressed.³ This work aimed to evaluate the extent of AV1451’s MAO inhibition through in vitro kinetic analysis, post-mortem competitive tissue binding, and small animal PET imaging.

Methods: [¹⁸F]AV1451 was synthesized as previously described.⁴ Inhibitory constants were determined by monitoring the appearance of a fluorescent product using a coumarin-derived substrate in human MAO-A/B Supersomes (ex 360, em 460). This tetrahydropyridine derivative was used instead of the standard substrate, kynuramine, because AV1451 has native fluorescence which overlapped with the kynuramine signal. The in vitro assay also evaluated known inhibitors of MAO (norharmine, harmine, and lazabemide) as standards. Autoradiography with Selegiline blocking was performed on cortex regions of AD human brain tissue and healthy, age matched controls. Plaque load was confirmed with silver staining. A blocking study was performed in Sprague-Dawley rats with a pre-dose of clorgyline (10 mg/kg; MAO-A inhibitor) or deprenyl (10 mg/kg; MAO-B inhibitor) before [¹⁸F]AV1451 injection. Results: In vitro kinetic studies demonstrated that AV1451 is a nonselective MAO inhibitor (A K_i= 10 µM; B K_i= 15 µM). These inhibitory constants are in a consistent range with known MAO inhibitor norharmane (A Ki= 2.2 µM; B Ki= 0.7-1.1 µM).⁵ [¹⁸F]AV1451 specific binding on AD human brain tissue was not significantly affected by deprenyl (1 µM) blocking. Clorygline blocking studies in rodents significantly increased the SUV­_max by 60%; deprenyl blocking studies did not significantly affect the SUV­_max. Conclusions: Although AV1451 demonstrates the in vitro capability of inhibiting MAO similar to that of norharmane, this effect is not translated into cortex tissue or into a healthy rodent model. On tissue, Selegiline blocking did not influence the specific binding of [¹⁸F]AV1451; however, it is important to consider the many unknowns associated with intracellular proteins in post-mortem tissue. In the rodent model, an MAO-A inhibitor dramatically increased the SUV­_max which is most likely attributed to peripheral inhibition changing the AV1451 input function. The MAO-B inhibitor, deprenyl, did not have any pronounced effect. The MAO-A blocking effect might be more obvious in the rodent brain because of the significantly increased abundance of rMAO-A in adult rats.⁶ urther study is required to fully evaluate the effects of MAO binding with [¹⁸F]AV1451 given species variation between rodents, primates and human subjects, both in identity and expression levels. Acknowledgments: We thank Timothy Desmond, Janna Arteaga and Jenelle Stauff for their help conducting autoradiography and rodent PET imaging. References 1. Lemoine et al. Alzheimer’s Research & Therapy (2017) 9:96. 2. Ng et al. Alzheimer’s Research & Therapy (2017) 9:25. 3. Vermerien et al. Alzheimer’s & Dementia (2015) 11:283. 4. Mossine et al. EJNMMI Radiopharm Chem (2017) 2:7. 5. J. van Amsterdam et al. Life Sciences (2006) 79:1969. 6. Nicotra et al. NeuroToxicology (2004) 25: 155. <!--EndFragment-->