Preclinical Evaluation of [11C]YC-72-AB85 for In Vivo Visualization of Heat Shock Protein 90 in Brain and Cancer with Positron Emission Tomography

A novel PET probe to selectively image heat shock protein 90α/β isoforms in the brain

BACKGROUND

Heat shock proteins (HSPs) are present throughout the brain. They function as molecular chaperones, meaning they help with the folding and unfolding of large protein complexes. These chaperones are vital in the development of neuropathological conditions such as Alzheimer's disease and Lewy body disease, with HSP90, a specific subtype of HSP, playing a key role. Many studies have shown that drugs that inhibit HSP90 activity have beneficial effects in the neurodegenerative diseases. Therefore, HSP90 PET imaging ligand can be used effectively to study HSP90 in neurodegenerative diseases. Among four HSP90 isoforms, two cytosolic isoforms (HSP90α and HSP90β) thought to be involved in the structural homeostasis of the proteins related to the neurodegenerative diseases. Currently, no useful PET imaging ligands selectively targeting the two cytosolic isoforms of HSP90 have been available yet.

RESULTS

In this study, we developed a novel positron emission tomography (PET) imaging ligand, [11C]BIIB021, by 11C-radiolabeling (a positron emitter with a half-life of 20.4 min) 6-Chloro-9-[(4-methoxy-3,5-dimethylpyridin-2-yl)methyl]-9H-purin-2-amine (BIIB021), an inhibitor with a high affinity for and selectivity to HSP90α and HSP90β. [11C]BIIB021 was synthesized with a high yield, molar activity and radiochemical purity. [11C]BIIB021 showed a high binding affinity for rat brain homogenate as well as human recombinant HSP90α and HSP90β proteins. Radioactivity was well detected in the rat brain (SUV 1.4). It showed clear specific binding in PET imaging of healthy rats and autoradiography of healthy rat and human brain sections. Radiometabolite was detected in the brain, however, total distribution volume was well quantified using dual-input graphical model. Inhibition of p-glycoprotein increased brain radioactivity concentrations. However, total distribution volume values with and without p-glycoprotein inhibition were nearly the same.

CONCLUSIONS

We have developed a new PET imaging agent, [11C]BIIB021, specifically targeting HSP90α/β. We have been successful in synthesizing [11C]BIIB021 and in vitro and in vivo imaging HSP90α/β. However, the quantification of HSP90α/β is complicated by the presence of radiometabolites in the brain and the potential to be a substrate for p-glycoprotein. Further efforts are needed to develop radioligand suitable for imaging of HSP90α/β.

Identification of (R)-[18F]YH134 for Monoacylglycerol Lipase Neuroimaging and Exploration of Its Use for Central Nervous System and Peripheral Drug Development

This study aimed to evaluate (R)-[18F]YH134 as a novel PET tracer for imaging monoacylglycerol lipase (MAGL). Considering the ubiquitous expression of MAGL throughout the whole body, the impact of various MAGL inhibitors on (R)-[18F]YH134 brain uptake and its application in brain-periphery crosstalk were explored. Methods: MAGL knockout and wild-type mice were used to evaluate (R)-[18F]YH134 in in vitro autoradiography and PET experiments. To explore the impact of peripheral MAGL occupancy on (R)-[18F]YH134 brain uptake, PET kinetics with an arterial input function were studied in male Wistar rats under baseline and blocking conditions. Results: In in vitro autoradiography, (R)-[18F]YH134 revealed a heterogeneous distribution pattern with high binding to MAGL-rich brain regions in wild-type mouse brain slices, whereas the radioactive signal was negligible in MAGL knockout mouse brain slices. The in vivo brain PET images of (R)-[18F]YH134 in wild-type and MAGL knockout mice demonstrated its high specificity and selectivity in mouse brain. A Logan plot with plasma input function was applied to estimate the distribution volume (V T) of (R)-[18F]YH134. V T was significantly reduced by a brain-penetrant MAGL inhibitor but was unchanged by a peripherally restricted MAGL inhibitor. The MAGL target occupancy in the periphery was estimated using (R)-[18F]YH134 PET imaging data from the brain. Conclusion: (R)-[18F]YH134 is a highly specific and selective PET tracer with favorable kinetic properties for imaging MAGL in rodent brain. Our results showed that blocking of the peripheral target influences brain uptake but not the V T of (R)-[18F]YH134. (R)-[18F]YH134 can be used for estimating the dose of MAGL inhibitor at half-maximal peripheral target occupancy.

Radiosynthesis and preclinical evaluation of [11C]SNX-ab as an Hsp90α,β isoform-selective PET probe for in vivo brain and tumour imaging

BACKGROUND

The molecular chaperone, Hsp90, is a key player in the protein quality control system that maintains homeostasis under cellular stress conditions. It is a homodimer with ATP-dependent activity, and is a prominent member of the chaperone machinery that stabilizes, matures and (re)folds an extensive list of client proteins. Hsp90 occurs as four isoforms, cytosolic Hsp90α and Hsp90β, mitochondrial TRAP1 and Grp94 present in the endoplasmic reticulum. An aberrant role of Hsp90 has been attributed to several cancers and neurodegenerative disorders. Consequently, Hsp90 has emerged as an attractive therapeutic target. However, pan-Hsp90 inhibition often leads to detrimental dose-limiting toxicities. Novel strategies for Hsp90-targeted therapy intend to avoid this by using isoform-specific Hsp90 inhibition. In this respect, the radiosynthesis of carbon-11 labeled SNX-ab was developed and [¹¹C]SNX-ab was evaluated as a Hsp90α,β isoform-selective PET probe, which could potentially allow to quantify in vivo Hsp90α,β expression.

RESULTS

[¹¹C]SNX-ab was synthesized with excellent radiochemical yields of 45% and high radiochemical purity (> 98%). In vitro autoradiography studies on tissue slices of healthy mouse brain, mouse B16.F10 melanoma and U87 glioblastoma using homologous (SNX-ab, SNX-0723) and heterologous (Onalespib and PU-H71) Hsp90 inhibitors demonstrated only limited reduction of tracer binding, indicating that the binding of [¹¹C]SNX-ab was not fully Hsp90-specific. Similarly, [¹¹C]SNX-ab binding to U87 cells was not efficiently inhibited by Hsp90 inhibitors. Ex vivo biodistribution studies in healthy mice revealed limited brain exposure of [¹¹C]SNX-ab and predominantly hepatobiliary clearance, which was confirmed by in vivo full-body dynamic µPET studies.

CONCLUSION

Our results suggest that [¹¹C]SNX-ab is not an ideal probe for in vivo visualization and quantification of Hsp90α/β expression levels in tumour and brain. Future research in the development of next-generation Hsp90 isoform-selective PET tracers is warranted to dissect the role played by each isoform towards disease pathology and support the development of subtype-specific Hsp90 therapeutics.