Preparation and Preclinical Evaluation of 18F-Labeled Olutasidenib Derivatives for Non-Invasive Detection of Mutated Isocitrate Dehydrogenase 1 (mIDH1).

Mutations of isocitrate dehydrogenase 1 (IDH1) are key biomarkers for glioma classification, but current methods for detection of mutated IDH1 (mIDH1) require invasive tissue sampling and cannot be used for longitudinal studies. Positron emission tomography (PET) imaging with mIDH1-selective radioligands is a promising alternative approach that could enable non-invasive assessment of the IDH status. In the present work, we developed efficient protocols for the preparation of four 18F-labeled derivatives of the mIDH1-selective inhibitor olutasidenib. All four probes were characterized by cellular uptake studies with U87 glioma cells harboring a heterozygous IDH1 mutation (U87-mIDH) and the corresponding wildtype cells (U87-WT). In addition, the most promising probe was evaluated by PET imaging in healthy mice and mice bearing subcutaneous U87-mIDH and U87-WT tumors. Although all four probes inhibited mIDH1 with variable potencies, only one of them ([18F]mIDH-138) showed significantly higher in vitro uptake into U87-mIDH compared to U87-WT cells. In addition, PET imaging with [18F]mIDH-138 in mice demonstrated good in vivo stability and low non-specific uptake of the probe, but also revealed significantly higher uptake into U87-WT compared to U87-mIDH tumors. Finally, application of a two-tissue compartment model (2TCM) to the PET data indicated that preferential tracer uptake into U87-WT tumors results from higher specific binding rather than from differences in tracer perfusion. In conclusion, these results corroborate recent findings that mIDH1-selective inhibition may not directly correlate with mIDH1-selective target engagement and indicate that in vivo engagement of wildtype and mutated IDH1 may be governed by factors that are not faithfully reproduced by in vitro assays, both of which could complicate development of PET probes.

8-Bicycloalkyl-CPFPX derivatives as potent and selective tools for in vivo imaging of the A1 adenosine receptor.

Imaging of the A1 adenosine receptor (A1R) by positron emission tomography (PET) with 8-cyclopentyl-3-(3-[18F]fluoropropyl)-1-propyl-xanthine ([18F]CPFPX) has been widely used in preclinical and clinical studies. However, this radioligand suffers from rapid peripheral metabolism and subsequent accumulation of radiometabolites in the vascular compartment. In the present work, we prepared four derivatives of CPFPX by replacement of the cyclopentyl group with norbornane moieties. These derivatives were evaluated by competition binding studies, microsomal stability assays and LC-MS analysis of microsomal metabolites. In addition, the 18F-labeled isotopologue of 8-(1-norbornyl)-3-(3-fluoropropyl)-1-propylxanthine (1-NBX) as the most promising candidate was prepared by radiofluorination of the corresponding tosylate precursor and the resulting radioligand ([18F]1-NBX) was evaluated by permeability assays with Caco-2 cells and in vitro autoradiography in rat brain slices. Our results demonstrate that 1-NBX exhibits significantly improved A1R affinity and selectivity when compared to CPFPX and that it does not give rise to lipophilic metabolites expected to cross the blood-brain-barrier in microsomal assays. Furthermore, [18F]1-NBX showed a high passive permeability (Pc = 6.9 ± 2.9 × 10-5 cm/s) and in vitro autoradiography with this radioligand resulted in a distribution pattern matching A1R expression in the brain. Moreover, a low degree of non-specific binding (5%) was observed. Taken together, these findings identify [18F]1-NBX as a promising candidate for further preclinical evaluation as potential PET tracer for A1R imaging.

Radiosynthesis and In Vitro Evaluation of [11C]tozadenant as Adenosine A2A Receptor Radioligand.

Tozadenant (4-hydroxy-N-(4-methoxy-7-morpholinobenzo[d]thiazol-2-yl)-4-methylpiperidine-1-carboxamide) is a highly selective adenosine A2A receptor (A2AR) antagonist and a promising lead structure for the development of A2AR-selective positron emission tomography (PET) probes. Although several 18F-labelled tozadenant derivatives showed favorable in vitro properties, recent in vivo PET studies observed poor brain penetration and lower specific binding than anticipated from the in vitro data. While these findings might be attributable to the structural modification associated with 18F-labelling, they could also reflect inherent properties of the parent compound. However, PET studies with radioisotopologues of tozadenant to evaluate its cerebral pharmacokinetics and brain distribution are still lacking. In the present work, we applied N-Boc-O-desmethyltozadenant as a suitable precursor for the preparation of [O-methyl-11C]tozadenant ([11C]tozadenant) by O-methylation with [11C]methyl iodide followed by acidic deprotection. This approach afforded [11C]tozadenant in radiochemical yields of 18 ± 2%, with molar activities of 50-60 GBq/µmol (1300-1600 mCi/µmol) and radiochemical purities of 95 ± 3%. In addition, in vitro autoradiography in pig and rat brain slices demonstrated the expected striatal accumulation pattern and confirmed the A2AR specificity of the radioligand, making it a promising tool for in vivo PET studies on the cerebral pharmacokinetics and brain distribution of tozadenant.

[18F]ALX5406: A Brain-Penetrating Prodrug for GlyT1-Specific PET Imaging.

Selective inhibition of glycine transporter 1 (GlyT1) has emerged as a potential approach to alleviate N-methyl-d-aspartate receptor (NMDAR) hypofunction in patients with schizophrenia and cognitive decline. ALX5407 is a potent and selective inhibitor of GlyT1 derived from the metabolic intermediate sarcosine (N-methylglycine) that showed antipsychotic potential in a number of animal models. Whereas clinical application of ALX5407 is limited by adverse effects on motor performance and respiratory function, a suitably radiolabeled drug could represent a promising PET tracer for the visualization of GlyT1 in the brain. Herein, [18F]ALX5407 and the corresponding methyl ester, [18F]ALX5406, were prepared by alcohol-enhanced copper mediated radiofluorination and studied in vitro in rat brain slices and in vivo in normal rats. [18F]ALX5407 demonstrated accumulation consistent with the distribution of GlyT1 in in vitro autoradiographic studies but no brain uptake in µPET experiments in naïve rats. In contrast, the methyl ester [18F]ALX5406 rapidly entered the brain and was enzymatically transformed into [18F]ALX5407, resulting in a regional accumulation pattern consistent with GlyT1 specific binding. We conclude that [18F]ALX5406 is a promising and easily accessible PET probe for preclinical in vivo imaging of GlyT1 in the brain.

Design, synthesis and biological evaluation of Tozadenant analogues as adenosine A2A receptor ligands.

With the aim to obtain potent adenosine A2A receptor (A2AR) ligands, a series of eighteen derivatives of 4-hydroxy-N-(4-methoxy-7-morpholin-4-yl-1,3-benzo[d]thiazol-2-yl)-4-methylpiperidine-1-carboxamide (SYN-115, Tozadenant) were designed and synthesized. The target compounds were obtained by a chemical building block principle that involved reaction of the appropriate aminobenzothiazole phenyl carbamates with either commercially available or readily synthesized functionalized piperidines. Their affinity and subtype selectivity with regard to human adenosine A1-and A2A receptors were determined using radioligand binding assays. Ki values for human A2AR ranged from 2.4 to 38 nM, with more than 120-fold selectivity over A1 receptors for all evaluated compounds except 13k which had a Ki of 361 nM and 18-fold selectivity. The most potent fluorine-containing derivatives 13e, 13g and 13l exhibited Ki values of 4.9 nM, 3.6 nM and 2.8 nM for the human A2AR. Interestingly, the corresponding values for rat A2AR were found to be four to five times higher. Their binding to A2AR was further confirmed by radiolabeling with 18F and in vitro autoradiography in rat brain slices, which showed almost exclusive striatal binding and complete displacement by the A2AR antagonist ZM 241385. We conclude that these compounds represent potential candidates for the visualization of the A2A receptor and open pathways to novel therapeutic treatments of neurodegenerative disorders or cancer.

Development and Evaluation of a Versatile Receptor-Ligand Binding Assay Using Cell Membrane Preparations Embedded in an Agarose Gel Matrix and Evaluation with the Human Adenosine A1 Receptor.

Guanosine-5'-triphosphate (GTP)-binding protein-coupled receptors are the target of up to 40% of prescribed medications worldwide. To evaluate the suitability of novel receptor ligands, frequently elaborate, time-consuming, and expensive receptor-ligand interaction studies have to be carried out. This work describes the development and proof of principle of a rapid, sensitive, and reliable receptor-ligand binding assay. CHO cells were stably transfected with a construct encoding the human A1 adenosine receptor (hA1AR). For ligand binding assays, membranes from these cells were prepared and embedded in low melting point agarose. These "immobilized" samples were incubated with tritiated 8-cyclopentyl-1,3-dipropylxanthine ([3H]DPCPX), a well-established receptor antagonist. The KD and Bmax values as well as kinetic parameters (kon and koff) of receptor-ligand interaction were determined. Unspecific binding of various radiotracers to either the carrier material or the agarose gel matrix was negligible. The dissociation constant (KD) for [3H]DPCPX at the hA1AR was determined by saturation, competition binding, and kinetic experiments. These studies resulted in KD values of ∼3 nM, which is in good accordance with previously published data obtained from conventional receptor-ligand binding assays. The procedure described in this study simplifies classical binding studies to a kit-like assay. The receptors retained their binding properties even when preparations were dried completely. Transport and delivery of the material are conceivable without loss of biological activity. Therefore, other laboratories can perform binding studies without special equipment or the necessity to run a cell culture laboratory and/or to dissect tissue on their own.

Synthesis and Pharmacological Evaluation of Identified and Putative Metabolites of the A1 Adenosine Receptor Antagonist 8-Cyclopentyl-3-(3-fluoropropyl)-1-propylxanthine (CPFPX).

The A1 adenosine receptor (A1 AR) antagonist [18 F]8-cyclopentyl-3-(3-fluoropropyl)-1-propylxanthine ([18 F]CPFPX), used in imaging human brain A1 ARs by positron emission tomography (PET), is stable in the brain, but rapidly undergoes transformation into one major (3-(3-fluoropropyl)-8-(3-oxocyclopenten-1-yl)-1-propylxanthine, M1) and several minor metabolites in blood. This report describes the synthesis of putative metabolites of CPFPX as standards for the identification of those metabolites. Analysis by (radio)HPLC revealed that extracts of human liver microsomes incubated with no-carrier-added (n.c.a.)[18 F]CPFPX contain the major metabolite, M1, as well as radioactive metabolites corresponding to derivatives functionalized at the cyclopentyl moiety, but no N1-despropyl species or metabolites resulting from functionalization of the N3-fluoropropyl chain. The putative metabolites were found to displace the binding of [3 H]CPFPX to the A1 AR in pig brain cortex at Ki values between 1.9 and 380 nm and the binding of [3 H]ZM241385 to the A2A AR in pig striatum at Ki values >180 nm. One metabolite, a derivative functionalized at the ω-position of the N1-propyl chain, showed high affinity (Ki 2 nm) to and very good selectivity (>9000) for the A1 AR.

New potent A1 adenosine receptor radioligands for positron emission tomography.

8-Cyclopentyl-3-(3-[18F]fluoropropyl)-1-propylxanthine ([18F]CPFPX) is meanwhile an accepted receptor ligand to examine the A1 adenosine receptor (A1AR) in humans by positron emission tomography (PET). A major drawback of this compound is its rather fast metabolic degradation in vivo. Therefore two new xanthine derivatives, namely 8-cyclobutyl-1-cyclopropymethyl-3-(3-fluoropropyl)xanthine (CBCPM; 5) and 1-cyclopropylmethyl-3-(3-fluoropropyl)-8-(1-methylcyclobutyl)xanthine (CPMMCB; 6) were designed and synthesized as potential alternatives to CPFPX. In membrane binding studies both compounds showed nanomolar affinity for the A1AR. In vitro autoradiographic studies of [18F]5 and [18F]6, using rat brain slices, showed the expected accumulation in regions known to have a high adenosine A1 receptor expression while exhibiting the necessary low unspecific binding. However, in vitro metabolite studies using human liver microsomes revealed a comparable metabolic degradation rate for both new xanthine derivatives and CPFPX.

Autoradiographic comparison of in vitro binding characteristics of various tritiated adenosine A2A receptor ligands in rat, mouse and pig brain and first ex vivo results.

The adenosine A(2A) receptor in the basal ganglia is involved in the control of movement and plays a role in movement disorders such as Parkinsonism. Developing ligands to evaluate that receptor by noninvasive methods such as positron emission tomography has a high priority. In vitro radioligand binding guides the selection of ligands for in vivo application. This study measured the binding of the adenosine A(2A) receptor antagonist [(3)H]MSX-2 (3-(3-hydroxypropyl)-8-m-methoxystyryl)-7-methyl-1-propargylxanthine) to rat, mouse and pig brain by autoradiography. Other studies measured binding to membranes from PC12 pheochromocytoma cells. Those binding parameters were compared to those of the adenosine A(2A) receptor antagonist [(3)H]ZM241385 (4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino)ethyl)phenol), the adenosine A(2A) receptor agonist [(3)H]CGS21680 (2-[p-(2-carboxyethyl)-phenethylamino]-5'-N-ethylcarboxamidoadenosine) and the unselective adenosine receptor agonist [(3)H]NECA (5'N-ethylcarboxamido)adenosine). The potency order (K(d)) in the three species was [(3)H]ZM241385<[(3)H]MSX-2<[(3)H]NECA<[(3)H]CGS21680. The density of [(3)H]MSX-2 binding sites was greater in the striatum than in the cortex. Preliminary ex vivo experiments showed that by 10min after iv injection, [(3)H]MSX-2 and [(3)H]CGS21680 crossed the blood-brain barrier to the extent of almost 1% ID/g brain tissue, but [(3)H]NECA and [(3)H]ZM241385 to only 0.2% ID/g. The prior administration of unlabeled ZM241385 significantly lowered brain uptake of [(3)H]MSX-2. In conclusion, [(3)H]MSX-2 has a high affinity and sufficient selectivity for the adenosine A(2A) receptor. It penetrates the blood-brain barrier. Sensitivity to photoisomerization is a limitation. Further investigations assess its suitability as a ligand for imaging the brain adenosine A(2A) receptor.