SUMMARY

Introduction. 7-Transmembrane receptors (7-TMRs), also known as G-Protein coupled receptors, are targets for a wide variety of therapeutic agents. Consequently, it is also interesting to  develop positron emission tomography (PET) ligands for these receptors. PET is the method of choice for non-invasive imaging of molecular targets in the living body (in vivo). To date, most successful PET receptor ligands are antagonists, likely due to their favourable pharmacological, physicochemical and pharmacokinetic properties. Nevertheless, agonist PET ligands are of potential interest, as they could provide differential information on and insight in receptor function and regulation. In contrast to antagonist PET ligands that bind to the total pool of a 7-TMR, agonist PET ligands specifically bind to the functionally active 7-TMR. The aim of the studies described in this thesis was to develop PET ligands with agonistic activity, especially for 7-TMRs involved in cognition. To this end, the M1 muscarinic acetylcholine receptor (M1ACh-R) and the 5-HT4 serotonergic receptor (5-HT4-R) were selected. 

Chapter 2 provides a comprehensive overview of agonist PET ligands for 7-TMRs that (1) have been labelled with carbon-11 or fluorine-18 and (2) have been evaluated for imaging 7-TMRs in the brain by PET in vivo. For each of 7-TMRs described, a SUMMARY is provided on its biological role, together with a critical assessment of chemical, pharmacological and pharmacokinetic properties of corresponding agonist PET ligands. Specific information on receptor function obtained with agonist ligands and advantages of agonist over antagonist ligands are discussed.

In chapter 3 the development of an orthosteric agonist PET ligand for the M1ACh-R is described, providing a potential tool to explore the active G-protein coupled receptor. The selective M1ACh-R agonist, [11C]AF150(S), was radiosynthesized, its receptor binding properties investigated using radioligand autoradiography on rat brain sections, and its brain uptake determined ex vivo using biodistribution studies in rats. Metabolites in brain and blood were measured using high-performance liquid chromatography (HPLC). The decay corrected yield of the [11C]AF150(S) radiosynthesis was around 70% with a radiochemical purity over 99%. The partitioning of [11C]AF150(S) between 1-octanol and phosphate buffer at pH 7.4 (LogDoct,pH7.4)  was 0.05. Autoradiography studies showed binding in M1ACh-R rich brain areas. In addition, selective inhibition by muscarinic agents was shown using in vitro conditions that promote agonist binding. Biodistribution studies revealed rapid and high brain uptake to levels exceeding five times the level seen in blood. In M1ACh-R rich areas, specific uptake versus cerebellum was apparent, remaining constant over 60 min in spite of rapid decline of radioactivity from the brain and rapid peripheral metabolism. The fast and high brain uptake of [11C]AF150(S), which is unusual for an hydrophilic agent,  suggest that facilitated transport may be involved. In conclusion, [11C]AF150(S) was successfully synthesized, showed M1ACh-R agonist binding properties and high brain uptake.

In Chapter 4 the feasibility of defining brain regions of interest (ROI), based on a standard magnetic resonance (MR) template of rat brain and an additional [18F]NaF scan for delineating the skull, was investigated to enable reproducible analysis of rat brain PET data. A procedure for co-registration of the MR template with [18F]NaF images was developed and the precision of ROI analysis was determined using a simulation study. Ten [18F]NaF scans of Wistar rats were co-registered with the standard MR template by 3 observers and transformation matrices obtained were applied to corresponding [11C]AF150(S) PET images. Uptake measures, expressed as percentage of injected dose per gram (%ID.g-1), were derived for several brain regions delineated using the MR template and [11C]AF150(S) ROI data from the in vivo measurements were compared with ex vivo data. Overall agreement between the 3 observers was assessed by interclass correlation coefficients (ICC) of the uptake data obtained. This analysis showed, for all brain regions, excellent agreement between observers and good reproducibility. Uptake of [11C]AF150(S) derived from ROI analysis of PET data closely matched ex vivo biodistribution data. In conclusion, this newly developed method provides a reproducible and tracer independent method for ROI analysis of rat brain PET data.

In Chapter 5 regional kinetics of [11C]AF150(S) in rat brain and specificity of uptake associated with M1AChR labelling were assessed both under baseline conditions and following pre-treatment with various pharmacological agents or co-administration of non-radioactive AF150(S). Data were analysed by calculating standard uptake values in ROIs defined using the MR template method and by applying the simplified reference tissue model (SRTM). Following IV administration, [11C]AF150(S) was rapidly taken up in the brain, followed by rapid clearance from all brain regions. In M1ACh-R rich regions, SRTM analysis using cerebellum as reference region yielded a binding potential relative to non-specific uptake (BPND) of 0.25 for striatum, 0.20 for hippocampus, 0.16 for frontal cortex and 0.15 for posterior cortex. Pre-treatment with M1ACh-R antagonists resulted in a significant reduction in BPND. BPND was not affected by pre-treatment with an M3ACh-R antagonist. Moreover, BPND was significantly reduced after pre-treatment with haloperidol, a dopamine D2 receptor blocker that causes an increase in extracellular acetylcholine (ACh). The latter may have competed with [11C]AF150(S) for binding to the M1ACh-R. To investigate this possibility, further pharmacological agents that increase extracellular levels of ACh, i.e. AF-DX 384, an M2/M4 receptor antagonist, and rivastigmine, an acetylcholine esterase inhibitor, were used. To study M1AChR saturation,  [11C]AF150(S) was co-injected with the unlabelled substance. At the highest dose (49.1 nmol.kg−1) of non-radioactive AF150(S) used, the brain concentration of AF150(S) reached 100 nmol.L−1. At this concentration, no sign of saturation in binding to M1ACh-R was observed. In conclusion, the agonist PET ligand [11C]AF150(S) was rapidly taken up in the brain and showed an apparent specific M1ACh-R-related signal, which could be inhibited by M1AChR antagonists in brain areas that are rich in M1ACh-R. Moreover, binding of the agonist PET ligand [11C]AF150(S) appeared to be sensitive to changes in extracellular ACh levels. Probably due to its relatively low M1AChR binding affinity, specific labelling was not saturable in vivo, where the concentration range that can be used is limited.

In Chapter 6 the development of an agonist PET ligand for the 5-HT4-R is described. Prucalopride, a high affinity agonist for the 5-HT4-R, marketed as an agent for treatment of constipation, but which reportedly could penetrate the brain, was selected for radiolabelling.  [11C]prucalopride was synthesized from [11C]methyl triflate and desmethyl prucalopride, and its LogDoct,pH7.4 was determined. Three distinct studies, with IV administration of [11C]prucalopride were performed in male rats: (1) the biodistribution of radioactivity was measured ex vivo, (2) in a PET study, kinetics in brain regions and peripheral organs were assessed in vivo under baseline conditions and following pre-treatment with tariquidar, an inhibitor of the P-glycoprotein efflux transporter, and (3) in vivo stability of [11C]prucalopride was checked ex vivo in plasma and brain extracts using HPLC. [11C]prucalopride was synthesized in optimized conditions with a yield of 21 ± 4% (decay corrected) and a radiochemical purity of >99%. Its LogDoct,pH7.4 was 0.87. Ex vivo biodistribution studies in male rats showed very low levels of radioactivity in brain (maximal 0.13% ID.g−1) and ten times higher levels in certain peripheral tissues. PET studies confirmed very low brain levels of radioactivity under baseline conditions, which were, however, increased by a factor of three after pre-treatment with tariquidar. [11C]Prucalopride was found to be very rapidly metabolized in male rats, with no parent compound detectable in plasma and brain extracts at 5 and 30 min following IV administration. Analysis of levels of radioactivity in peripheral tissues revealed a distinct PET signal in the caecum, which was reduced following tariquidar pre-treatment. The latter is in line with the role of the P-glycoprotein transporter in the gut. In conclusion, [11C]prucalopride showed low radioactivity levels in male rat brain. This may be due to rapid metabolism, which is especially the case in male rats, low passive diffusion and/or being a substrate for P-glycoprotein.