In Vivo Assessment of Cogniion Related Acivated Seven Transmembrane

In vivoIn assessment of cogniion related acivated sev

verdediging van het proefschrit In vivo assessment of cogniion related cogniion related acivated acivated seven transmembrane receptors seven transmembrane receptors Hans J.C. Buiter Hans J.C. Buiter

Na aloop van de promoie

voor de recepie ter plaatse en transmembrane receptors

H.J.C. Buiter H.J.C. Buiter

In vivo assessment of cognition related activated seven transmembrane receptors

Hans J.C. Buiter

The research described in this thesis was performed at the department of Radiology & Nuclear Medicine at the VU University Medical Center Amsterdam, the Netherlands and was inancially supported by:

Coops Foundation

Financial support for the printing of this thesis was kindly provided by:

Coops Foundation VU University Amsterdam Stichting Apotheek der Haarlemse Ziekenhuizen

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission from the author, or, when appropriate, from the publishers of the peer-reviewed publications.

VRIJE UNIVERSITEIT

In vivo assessment of cognition related activated seven transmembrane receptors

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad Doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magniicus prof.dr. F.A. van der Duyn Schouten, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de Faculteit der Geneeskunde op dinsdag 1 juli 2014 om 11.45 uur in de aula van de universiteit, De Boelelaan 1105

door

Hans Johan Christiaan Buiter

geboren te Delfzijl

promotoren: prof.dr. J.E.M.F. Leysen prof.dr. A.A. Lammertsma copromotor: prof.dr. A.D. Windhorst

“Those who dream by night in the dusty recesses of their minds wake up in the day to ind it was vanity, but the dreamers of the day are dangerous men, for they may act their dreams with open eyes, to make it possible. This I did.”

Col. T.E. Lawrence (1888 –1935), Seven Pillars of Wisdom.

Voor Francis, Ellen en Inge

InHoud

Chapter 1 Introduction 9 1.1 General 11 1.2 Seven transmembrane domain receptors 11 1.3 Positron emission tomography 16 1.4 Aim 23 1.5 Outline 23

Chapter 2 overview of agonist PET ligands for 7-transmembrane re ceptors 25 2.1 Dopamine receptors 27 2.2 5-Hydroxytryptamine receptors 37 2.3 Muscarinic acetylcholine receptors 45 2.4 Opioid receptors 48 2.5 Cannabinoid receptors 52 2.6 Melatonin receptors 53 2.7 Concluding remarks 54

Chapter 3 Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S) 57 3.1 Introduction 59 3.2 Material and Methods 61 3.3 Results 64 3.4 Discussion 71 3.5 Conclusion 74

Reproducible analysis of rat brain PET studies using an additional [18F]naF scan and an MR based RoI template 77 4.1 Introduction 79 4.2 Methods 80 4.3 Results 85 4.4 Discussion 91 4.5 Conclusions 93

Chapter 5 [11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors 95 5.1 Introduction 97 5.2 Materials and methods 98 5.3 Results 102 5.4 Discussion 108 5.5 Conclusions 111

Chapter 6 Radiosynthesis and preclinical evaluation of 11 [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor 113 6.1 Introduction 115 6.2 Methods and materials 116 6.3 Results 120 6.4 Discussion 128 6.5 Conclusion 130

Chapter 7 Summary, general discussion and future perspectives 133 7.1 Summary 135 7.2 General discussion 138 7.3 Future perspectives 139

Chapter 8 141 nederlandse samenvatting: In vivo bepaling van aan cognitie gerelateerde geactiveerde zeven transmembraan domein receptoren 8.1 introductie 143 8.2 Resultaten en conclusies van het onderzoek 144 8.3 Algemene discussie 147 8.4 Toekomstperspectieven 148

References 151 List of publications 169 Dankwoord 171 Curriculum vitae 174 Fotoverantwoording 175

1 Introduction

Introduction 1.1 GEnERAL 1 Positron emission tomography (PET) is the method of choice for imaging and assessing molecular targets in the living body (in vivo). Most PET tracers are ligands that are labelled with a short living positron emitting radionuclide and that selectively bind to a particular receptor. Very interesting targets for PET studies are seven transmembrane domain receptors (7-TMRs), previously known as G-protein coupled receptors (GPCRs). To date, 802 7-TMRs are known and/or predicted based on an analysis of the human genome and they form the largest receptor class (for review see [1-3]). At present, 7-TMRs are the most successful drug targets and 7-TMR ligands are marketed for treatment of a wide variety of disorders including psychiatric disorders (schizophrenia, bipolar disorder, depression), neurological disorders (Parkinson’s disease and other movement disorders), pain, migraine, cardiovascular disorders, pulmonary disorders, inlammation, allergy and gastrointestinal impairments [4]. The irst successful 7-TMR agonist PET ligand was the opioid agonist [11C]carfentanil [5]. So far, however, most PET studies have concentrated on 7-TMR antagonist ligands, possibly because many therapeutic agents are 7-TMR antagonists. Nevertheless, the potential of 7-TMR agonists as therapeutic agents is increasingly recognized and these can best be evaluated using agonist PET ligands. In addition, agonist PET ligands can provide unique information, as they can distinguish between the active state of a 7-TMR, to which they bind with high ainity, and the inactive state, for which they have low ainity, whilst antagonists have the same ainity for all states of a 7-TMR.

This chapter provides background on general features of 7-TMRs, principles of PET, choice between agonist and antagonist PET ligands, general requirements for PET ligand development, PET signal quantiication, and 7-TMR targets for treatment of cognitive dysfunction.

1.2 SEVEn TRAnSMEMBRAnE doMAIn RECEPToRS

1.2.1 7-TMRs: basic structure and activation mechanisms A 7-TMR is a membrane protein monomer with 7 hydrophobic domains that form 7 transmembrane helices. 7-TMRs have an extracellular N-terminus and an intracellular C-terminus and its helices are connected through intra- and extra-cellular loops (Figure 1). This structure is characteristic for the entire receptor family. 7-TMRs have diverse natural ligands, as diferent as photons, organic odorants, Ca2+-ions, small chemical neurotransmitters, nucleotides, nucleosides, peptides, lipids and proteins. They have been classiied into 5 families according to their structure and ligand binding criteria: 1)

Chapter 1 rhodopsin family (also designated as family A), 2) secretin family (family B), 3) glutamate family (family C), 4) adhesion family and 5) frizzled/taste 2 family (for review see [1-3,6])

Figure 1: General structure of 7-transmembrane domain receptors.

7-TMRs are located in the plasma membrane of the cell and can form complexes with several auxiliary proteins, the most characteristic of which is the G-protein. The latter mostly, but not always, mediates signal transduction between the 7-TMR and an efector protein that generates a second messenger, which subsequently triggers a response in the cell (for review see [7]). G-proteins are trimers, consisting of an α, β and γ subunit. The

Figure 2: Schematic representation of 7-TMR activation and deactivation, adapted from [11].

Introduction

α subunit interacts with the intracellular loops of a 7-TMR and, subsequently, hydrolyses GTP into GDP (Figure 2). The α subunits have been classiied according to the efector 1 protein and second messengers that they activate. The Gαi-protein inhibits adenylate + cyclase, Gαo couples to K -channels, Gαs stimulates adenylate cyclase and Gαq/11 activates phospholipase C (for review see [8-10]).

1.2.2 7-TMR conformational states: ‘Ternary complex models’. In the 7-TMR activation process, the receptor, a G-protein and an agonist ligand form a ternary complex which can exist in various activation states. The ainity of the agonist is determined by the activation state of the 7-TMR- G-protein-agonist complex (for review see [12]). The initial ternary complex model (Figure 3A), proposed by De Lean et al. [13], represents the complex in its most simple form, where the receptor (R) can interact with the agonist (A) or the G-protein (G) and these complexes can react further with the third reaction partner to form a ternary complex. This simple model, however, does not describe the entire process of activation of 7-TMRs. Therefore the model was developed further, resulting in the extended ternary complex model (Figure 3B), proposed by Lefkowitz [14,15]. The extension was based on the observation that 7-TMRs can have basal activity in the absence of an agonist and that mutant 7-TMRs can have enhanced agonist independent activity. The model takes into account occurrence of the naked receptor in two conformational states (Ri = inactive state, Ra = active state), occurrence of agonist-receptor complexes (ARi, ARa), a receptor-G-protein complex (RaG) and the ternary complex (ARaG), where both RaG and ARaG lead to a response (see Figure 3), Until today, the extended ternary complex model is used to describe the working mechanism of diferent types of receptor ligands, such as full agonists, partial agonists, neutral antagonists and inverse agonists. Yet, not all 7-TMR properties can be explained by this extended ternary complex model, as it does not support multiple conformational states of the receptor. Consequently, Kenakin proposed the cubic ternary complex model

Figure 3: Ternary complex models describing 7-TMR activation by complex formation between agonist (A), receptor (R) and G-protein (G). (A) Initial simple ternary complex (B) Extended ternary complex (C) Cubic ternary complex. For details see [12] (reprinted with permission).

Chapter 1

(Figure 3C), which allows for interaction of inactive receptor (Ri and ARi) with G proteins, but without resulting signal [12]. Several functional and biophysical studies have been reported that support this model of multiple conformational states of 7-TMR [16].

1.2.3 Classes of exogenous ligands that target 7-TMRs 7-TMRs have an orthosteric binding site where the natural ligand binds. This can be located either within the transmembrane domains or in the extracellular N-terminus [1]. In addition, 7-TMRs can have other ligand binding sites, distinct from the orthosteric site. These are allosteric binding sites, usually located within the transmembrane domains. Ligand binding to either orthosteric or allosteric binding sites can result in various ef- fects. A schematic representation of the various classes of pharmacological ligands for 7-TMRs is shown in Figure 4.

Figure 4: Schematic representation of classes of exogenous ligands for 7-TMRs, adapted from [17].

Ligands that bind to the orthosteric site and produce full receptor activation are called orthosteric full agonists. Ligands that can only activate the receptor partially and do not produce a maximal response are called partial agonists. Ligands that do not produce a response, but inhibit binding of agonists and thereby block receptor activation by an agonist are called (neutral) antagonists, and ligands that can block both constitutive receptor activity and receptor activation by an agonist are called inverse agonists [18]. Ligands that bind to allosteric sites can afect receptor activity in various ways. Al- losteric agonists activate the receptor through interaction at a site distinct from that of the orthosteric agonist. They can be allosteric full agonists or allosteric partial agonists. Negative allosteric modulators block receptor function, but do not necessarily interfere

Introduction with agonist binding to the orthosteric site. Positive allosteric modulators potentiate the efects of orthosteric agonists on the receptor; for a review see [17,19]. 1

1.2.4 7-TMR traicking The process of 7-TMR maturation, reaching its residence at the plasma membrane of the cell, receptor internalization into the cell, recycling back to the plasma membrane or degradation is called 7-TMR traicking (Figure 5). 7-TMRs located in plasma membranes are available for activation by agonists. Strong and persistent activation can trigger receptor internalization, as a consequence of which the receptors become localized at intracellular organelles, where they exist in an inactive state. The process of 7-TMR internalization involves multiple steps and various auxiliary proteins. The receptor is phosphorylated (P) by receptor kinases (GRKs), leading to the recruitment of β-arrestins. Subsequently β-arrestins, through their interaction with clathrin and adaptor protein-2 (AP-2), target the receptor/arrestin complexes to clathrin coated pits. A GTPase enzyme, dynamin, regulates the pinching of from the cell surface of clathrin-coated pits. When these so called clathrin-coated vesicles (CCVs) are formed, the receptor is internalized into endosomes. Next, the receptor is dephosphorylated before returning back to the cell surface by an exocytotic process, or the receptor is degraded in lysosomes [20-22].

Figure 5: General model for 7-TMR traicking: internalization and recycling of 7-TMRs. 1) Upon agonist binding: receptor phosphorylation and recruitment of β-arrestins. 2) Beta-arrestins target the receptor/ arrestin complexes to clathrin coated pits. 3) Beta-arrestins also bind tyrosine-protein kinase (c-Src). 4) Dynamin regulates pinching of from the cell surface. 5) Formation of clathrin-coated vesicles and receptor internalisation into endosomes. 6) Receptor is dephosphorylated before returning to the cell surface, 7) Receptor is degraded in lysosomes (adapted from [22]).

Chapter 1

1.3 PoSITRon EMISSIon ToMoGRAPHY

1.3.1 Basic principles PET has been recognized as a valuable technique for imaging and quantifying, amongst others, ligand-receptor interactions in vivo. Using PET ligands for 7-TMRs allows for in vivo investigations of a multitude of questions. It can provide information on receptor distribution and concentrations in brain regions and peripheral organs. Comparative studies in healthy and disease states can provide information on receptor alterations related to the disease or due to drug treatment. In addition, the technique enables early evaluation of novel drugs in man, including receptor occupancy studies to verify whether and to what extent drugs reach their molecular targets at various locations, and studies to optimize drug-dosing regimens [23]. For PET imaging radionucliedes that decay by emitting positrons are used. A positron has the same mass as an electron, but with a positive charge. In tissue, such a positron will collide with neighbouring electrons and lose its kinetic energy. Ultimately, the positron will combine with an electron. This combination of positron and electron is very unstable and will decay (annihilate) almost immediately (half life 100 ns) by emitting two gamma rays (photons), each with a ixed energy of 511 keV (corresponding with the mass of electron and positron), in opposite directions. These annihilation photons can be detected by coincidence detection (i.e. simultaneous detection by two opposing detectors) (Figure 6).

Figure 6. Schematic representation of detection principles of PET. Reprinted from [24] with permission.

For brain imaging, especially in small animals, PET systems require high sensitivity and high spatial resolution in order to acquire accurate data. Clinical whole body PET systems have a relatively low spatial resolution, limiting proper quantiication of small brain regions such as the hippocampus and striatum in laboratory animals. The ECAT

Introduction

High Resolution Research Tomograph (HRRT; CTI/Siemens, Knoxville, TN, USA) (Figure 7) is a dedicated brain PET system with high sensitivity and relatively high spatial resolu- 1 tion when compared with clinical PET systems [25]. The HRRT allows for improved quantiication and identiication of small brain nuclei [26]. The spatial resolution typically is expressed in terms of full width at half maximum (FWHM). The HHRT has a resolution of 2.5 to 3 mm FWHM, making this PET system also suitable for small animal imaging [27]. In addition the system is very sensitive and, due to the double layer of scintillation crystals, its resolution is more homogeneous.

Figure 7. HRRT exterior (left) and interior (right). The interior clearly shows the 8 HRRT detector heads.

1.3.2 Choice of PET ligand: agonists versus antagonists for 7-TMRs For PET imaging of 7-TMRs the type of ligand, i.e. antagonist or agonist, will afect outcome. Radiolabelled antagonists usually bind with equal ainity to 7-TMRs in all conformational states, including active and inactive ones. Moreover, lipophilic antagonists can enter cells and cell organelles, where they can bind to the inactive internalized 7-TMR. Hence, antagonists do not distinguish between activated and inactive 7-TMRs and between 7-TMRs on plasma membranes and intracellular organelles. This will afect the outcome of studies designed for receptor quantiication in vivo, as well as that of competitive binding studies with either endogenous ligands or exogenous non-labelled competitors, in particular agonists. In contrast, agonists usually show high ainity (indicated by a high ainity dissociation constant KH) for the activated 7-TMR-G-protein coupled receptor complex and low ainity (low ainity dissociation constant KL) for the inactive uncoupled receptor. When used as radioligand at low concentrations around or below the KH, an agonist will only bind to the activated 7-TMR. Radioligand receptor binding studies in vitro, using either cell membrane preparations or tissue sections for radioligand autoradiography, have shown that antagonist radioligands bind to a larger number of a particular 7-TMR than agonist radioligands in the same tissue preparation

Chapter 1

[28,29]. The extent of the diference in binding of a particular 7-TMR by an agonist or antagonist radioligand can difer between the region where the 7-TMR is present. Therefore agonist PET ligands are preferred for: · Gathering information on the active state of the receptor, e.g. in health versus disease or following chronic drug treatment. · Assessing receptor occupancy by endogenous ligands when their release is trig- gered. · Assessing receptor occupancy by agonist therapeutic agents.

1.3.3 Required features of PET ligands for 7-TMRs in the brain PET has the advantage that most of the short-lived positron emitting radionuclides available are isotopes of elements that occur in natural products and synthetic pharmacological agents, i.e. elements such as carbon, nitrogen, oxygen and luorine. Incor- poration of such a radionuclide (carbon-11, nitrogen-13, oxygen-15 and luorine-18) into a molecule as replacement of its stable counterpart does not alter the chemical structure or afect the biological activity of the molecule and thereby provides a unique opportunity to study biological processes in living subjects. Short-lived radionuclides are typically produced on site by a cyclotron (particle accelerator). Properties of the two most applied positron emitting radionuclides, i.e. carbon-11 ([11C]) and luorine-18 ([18F]) are presented in Table 1. Use of radioligands labelled with short lived tracers such as carbon-11 allows for multiple PET studies in the same subject on the same day.

Table 1. Main characteristics of the two most applied radionuclides in PET 11C 18F Half-life (min) 20.4 109.6 Nuclear reaction: 14N(p,α)11C 18O(p,n)18F Mode of decay: β+ (100%) β+ (97%) Maximal energy (MeV) 0.97 0.64 Tissue penetration (mm) 4 2

PET ligands for imaging 7-TMRs in the brain must meet a number of requirements. Because of the short half-life, radiosynthesis should be performed with a precursor molecule amenable for introduction of the radionuclide in a rapid way. Good chemical stability of the radioligand is desired, especially since radiolysis can be a cause of increased instability. The radioligand should be suiciently lipophilic to allow passage across the blood brain barrier, but at the same time it should not be too lipophilic, as this can result in excessive non-speciic, i.e. non-displaceable binding. The radioligand should not be a substrate for the P-glycoprotein elux transporter and/or related processes, as this would hamper or prevent uptake into the brain. A suiciently high speciic

Introduction

activity of the radioligand is required to allow for administration of a minimal mass of radioligand, thereby meeting requirements of tracer conditions. Ideally, less than 1 5% of the biological target should become occupied by the radioligand to minimize relevant signal in the brain. A detailed discussion on required features of PET ligands pharmacological responses. Radioligand metabolism should be such that radiolabelled can be found elsewhere [23,30,31]. metabolites do not interfere with the relevant signal in the brain. A detailed discussion 2.3.4PETon required signal quantification features of PET ligands can be found elsewhere [23,30,31].

In the most simple form of binding equilibrium between a ligand and a single binding site, the1.3.4 reaction is described PET signal according quantiication to the law of mass action, When, in aqueous solution,In equilibrium the most exists simple between form receptor of concentrati bindingon ( Requilibrium), concentration of betwe free en a ligand and a single binding ligand site, (F) and the concentration reaction of receptoris described bound ligand according (B), this equilibrium to the can law be of mass action, When, in aqueous describedsolution, by Eq. 1 (forequilibrium review see [32]). exists between receptor concentration (R), concentration of free ligand (F) and concentration of receptor bound ligand (B), this equilibrium can be de-

scribed by kEq.on 1 (for review see [32]). [R] + [F] [B]

Kkonoff R  F   B (1) (1)        Ligand receptor bindingK off is driven by the concentrations of receptor and ligand, and the associationLigand (kreceptoron) and dissociation binding (koff is) ratedriven constants by the of the concentratio ligand-receptor ns of receptor and ligand, and the complex. The receptor concentration usually is low (maximal total number is in the association (kon) and dissociation (kof) rate constants of the ligand-receptor complex. The range receptor of pmoles/g concentration tissue). To obtain complex usually formation, is low the (maximal ligand concentration total number is in the range of pmoles/g must betissue). adapted To according obtain to thecomplex koff/kon, i.e. formation, the equilibrium the dissociation ligand constant concentration must be adapted accord- (KD) of the ligand (Eq 2). ing to the kof/kon, i.e. the equilibrium dissociation constant (KD) of the ligand (Eq 2).

koff R  F  kD   (2) (2) kon B  It is important to realize that ideal equilibrium conditions will never meet the required It is important to realize that ideal equilibrium conditions will never meet the required betweencriteria compounds of the in solution,Law of since Mass the Action receptor is[33]. membr Theane equati bound, iton is is not only applicable for reactions criteria of the Law of Mass Action [33]. The equation is only applicable for reactions present27 between in soluble compounds form. Further, the in model solution, implies since interaction the ofreceptor the ligand is with membrane a bound, it is not pres- single entbinding in sitesoluble on the receptor.form. Further, This often maythe notmodel be the impliescase, as receptors interaction can of the ligand with a single occur binding in different site conformational on the receptor. states and ligandsThis often may bind may to not orthosteric be the and case, as receptors can occur in diferent conformational states and ligands may bind to orthosteric and allosteric allosteric sites, as discussed in par 1.1.3. Nevertheless, these simple equations form sites, as discussed in par 1.1.3. Nevertheless, these simple equations form the basis for the basis for experimental determinations of KD-values of a ligand and for estimations

experimental determinations of KD-values of a ligand and for estimations of the maximal of the maximal receptor concentration (Bmax) in a tissue (Eq 3). receptor concentration (Bmax) in a tissue (Eq 3).

B max F (3) B  (3) KD F

In PET studies often the term ‘binding potential’ is used, which is also based on in vitro

In PETradioligand studies often thebinding term 'binding [34]. potential' This is is a u sed,relatively which is simple also based concept on in to indicate the amount of vitro radioligand binding [34]. This is a relatively simple concept to indicate the amount of radioligand uptake in the brain and is determined by the ratio between 19 receptor density (Bmax) and radioligand affinity (KD) [34,35].

B max B BP   (4) KD F

In order to obtain the binding potential of a radioligand in a particular brain area, PET data need to be quantified. Therefore anatomical regions of interest (ROI) are defined directly on the PET image or, alternatively on a high-resolution anatomical MRI image which is co- registered with the PET data. Next, these ROIs are projected on all time frames of a dynamic scan, thereby generating time-activity curves (TACs) for the corresponding anatomical regions. These TACs can then be analysed using tracer kinetic models that describe uptake 28

between compounds in solution, since the receptor is membrane bound, it is not present in soluble form. Further, the model implies interaction of the ligand with a single binding site on the receptor. This often may not be the case, as receptors can occur in different conformational states and ligands may bind to orthosteric and allosteric sites, as discussed in par 1.1.3. Nevertheless, these simple equations form the basis for experimental determinations of KD-values of a ligand and for estimations of the maximal receptor concentration (Bmax) in a tissue (Eq 3).

B max F B  (3) KD F

In PET studies often the term 'binding potential' is used, which is also based on in Chapter 1 vitro radioligand binding [34]. This is a relatively simple concept to indicate the amount of radioligand uptake in the brain and is determined by the ratio between radioligand uptake in the brain and is determined by the ratio between receptor density receptor density (Bmax) and radioligand affinity (KD) [34,35]. (B ) and radioligand ainity (K ) [34,35]. max D

B max B BP   (4) (4) KD F

In order to obtain the binding potential of a radioligand in a particular brain area, PET In order to obtain the binding potential of a radioligand in a particular brain area, PET data data need to be quantiied. Therefore anatomical regions of interest (ROI) are deined need to be quantified. Therefore anatomical regions of interest (ROI) are defined directly on directly on the PET image or, alternatively on a high-resolution anatomical MRI image the PET image or, alternatively on a high-resolution anatomical MRI image which is co- which is co-registered with the PET data. Next, these ROIs are projected on all time frames registered with the PET data. Next, these ROIs are projected on all time frames of a dynamic scan, of thereby a dynamic generating scan, time-activity thereby curves generating (TACs) for the correspondingtime-activity anatomical curves (TACs) for the corresponding regions.anatomical These TACs can regions. then be analysed These using TACs tracer cankinetic thenmodels bethat describeanalysed uptake using tracer kinetic models that 28 describe uptake in and clearance from tissue using the time course of activity within arterial plasma (so called input function). In general, models used in PET are compartment models, in which the possible distribution of a tracer is divided into a limited number of discrete compartments as shown in Figure 8. The most common compartment models are a single tissue compartment model, where the tracer is taken up by tissue, but has no further interactions within the tissue (Figure 8b), and a two tissue compartment model with exchange of the tracer between the two tissue compartments, where the irst compartment represents for example free tracer in tissue and the second bound tracer (Figure 8c) [36]. The latter model best describes kinetics of a receptor ligand, but if the exchange between the two compartments is too fast, they cannot be distinguished from each other and the model reverts to a single tissue compartment model.

Figure 8: Structure of the most common compartment models used in PET with (a) zero, (b) one, and (c) two tissue compartments. Reprinted from [36] with permission.

Most PET tracer kinetic models for 7-TMRs in the brain are two tissue compartment models. These models consider concentration of radioactivity in plasma, non-displaceable radioactivity in tissue, speciically bound radioactivity in tissue and four rate constants. Models are based on the assumption that the radioligand is distributed homogenously within the diferent compartments and that measured radioactivity represents unchanged radioligand [34]. These models describe exchange of the radioligand between the diferent compartments. The cerebral signal is the sum of 1) radioligand binding to one particular receptor type, 2) free radioligand, and 3) non-speciically bound ligand.

Introduction

Theoretically these models would represent a three tissue compartment model, but in practice it is assumed that kinetics of the nonspeciic compartment are very fast, so that 1 free and nonspeciic compartments can be combined into a single non-displaceable compartment (shown in Figure 9).

Figure 9: Schematic diagram of the two tissue compartment model used for receptor studies. CP, CND, and

CS represent arterial plasma, and non-displaceable and speciic tissue concentrations of radioactivity. K1 to k4 represent rate constants characterizing transport of radioactivity between compartments and VB blood volume within the PET region. Reprinted from [36] with permission.

In this particular model (Figure 9) 4 rate constants are needed to describe kinetics of the radioligand. The model is only applicable when arterial blood sampling during PET measurements is possible. This often is not the case for PET studies using small laboratory animals. For the latter, a reference tissue model can be applied, that makes use of a reference region devoid of the receptor of interest to determine non-displaceable binding. In Figure 10, the simpliied reference tissue model (SRTM) is presented [37].

Figure 10: Schematic diagram of the simpliied reference tissue model (SRTM). Reprinted from [36] with permission.

Figure 10: Schematic diagram of the simplified reference tissue model (SRTM). Reprinted from [36] with permission.

Chapter 1 SRTM allows for the calculation of BPND i.e. the binding potential relative to the non- displaceable concentration of radioligand measured in the reference tissue. BPND is SRTM allows for the calculation of BP i.e. the binding potential relative to the non- defined as the ratio, at equilibrium, of specificallyND bound to nondisplaceable

displaceable concentration of radioligand measured in the reference tissue. BPND is radioligand in tissue, and represents the product of density of receptors available for deined as the ratio, at equilibrium, of speciically bound to nondisplaceable radioligand -1 binding (Bavail), the reciprocal of the dissociaton constant (KD ) and the free fraction in tissue, and represents the product of density of receptors available for binding (Bavail), of radioligand in the non-displaceable tissue compartment (fND-1) [35]: the reciprocal of the dissociaton constant (KD ) and the free fraction of radioligand in the

non-displaceable tissue compartment (fND) [35]:

Bavail BPND  fND (5) (5) KD

SRTM contains three parameters: R1, k2 and BPND, in which R1 = K1/K’1 is introduced to

SRTMcontrol contains forthree diference parameters: in R 1rates, k2 and of BP deliveryND, in which to R1target = K1/K' a1nd is introducedreference tissues, respectively. to controlThis PET for differencetracer kinetic in rates model of delivery is based to on targ twoet and assump referencetions: tissues, (1) the volume of distribution of the non-displaceable, i.e. free plus non-speciically bound, radioligand is similar respectively. This PET tracer kinetic model is based on two assumptions: (1) the in target and reference regions, and (2) the blood brain barrier is intact. 31 1.3.5 7-TMR targets for treatment of cognitive dysfunction Cognitive decline is a feature of various neurological and psychiatric disorders. Apart from disorders with dementia as primary feature, with Alzheimer’s disease being the most prevalent, cognitive decline can be associated with e.g. Parkinson’s disease, Hun- tington’s disease, schizophrenia and depression [38-41]. Several central neurotransmitter systems, including the cholinergic, glutamatergic, dopaminergic, noradrenergic and serotonergic system, have a role in cognitive functions. Of these, the cholinergic system has been explored most. Thogether with extracellular amyloid plaques in cortical regions and intracellular neuroibrillary tangels, degeneration of basal forebrain cholinergic neurones is are hallmarks of Alzheimer’s disease [42-44]. In spite of extensive research over the past two decades, clinical trials with potential disease modifying therapies for Al- zheimer’s disease have not been successful [43]. A symptomatic approach for improving cognitive dysfunction may remain a necessary adjuvant therapy, not only for Alzheimer’s disease, but also for other disorders with cognitive decline. Stimulating central cholinergic neurotransmission is one possibility. A current treatment used in Alzheimer’s disease consists of reduction of acetylcholine breakdown by acetylcholine esterase inhibitors. Unfortunately, however, at present success of this treatment is limited. Direct stimulation of cholinergic receptors that mediate cognitive functions may provide symptom alleviation over a longer course. Acetylcholine signalling is mediated by ionotropic nicotinic receptors and metabotropic muscarinic receptors. The latter are 7-TMRs and of the ive identiied muscarinic acetylcholine receptor subtypes, the M1 receptor is the most prevalent in the brain and has a role in memory processes [45]. Muscarinic M1 agonists are being explored as potential treatment for memory symptoms in Alzheimer’s disease and schizophrenia. Other potential 7-TMR targets for treatment of cognitive dysfunction

Introduction are e.g. subtypes of serotonergic receptors such as the 5-HT , 5-HT and 5-HT receptors, 1A 4 6 which have a role in modulation of cholinergic neurotransmission, and the dopamine D1 1 receptor, stimulation of which improves attention [46,47]. Further, allosteric modulators of metabotropic glutamate receptors are being investigated [48]. To elucidate the function of these 7-TMRs and the eicacy of their ligands in vivo, a non-invasive molecular imaging technique, such as PET, could be useful.

1.4 AIM

At the onset of the studies described in this thesis, development of PET ligands for 7-TMRs was focussed primarily on receptor antagonists. The aim of this thesis was to develop PET ligands for 7-TMRs with agonistic activity, focussing on agonist ligands that can stimulate cholin\ergic activity directly by acting on the M1 muscarinic acetylcholine receptor

(M1ACh-R) or indirectly by acting on the 5-HT4 serotonergic receptor (5-HT4-R). The goal was to select agonist ligands for these receptors that were amenable for labelling with carbon-11, to synthesize these radioligands and to investigate their potential application for assessing receptors in laboratory animals using both ex vivo techniques and PET.

1.5 ouTLInE

Chapter 2 provides a comprehensive overview of all 7-TMR agonist PET ligands that have been labelled with carbon-11 or luorine-18 and that have been evaluated for imaging 7-TMRs in the brain using PET. Chapter 3 describes radiosynthesis and biological evaluation of [11C]AF150(S) as an agonist ligand for the M1 muscarinic acetylcholine receptor. In Chapter 4, a new method is developed and validated that provides a reproducible and tracer independent ROI analysis method of rat brain PET data using an additional [18F]NaF scan together with an MR based ROI template. In Chapter 5 the in vivo evaluation of [11C]AF150(S) in rats is investigated. In these studies brain uptake and binding of [11C]AF150(S) was studied using PET in order to explore its suitability as an agonist PET ligand for M1ACh-R. In addition, in vivo speciicity of binding and sensitivity to changes in extracellular levels of acetylcholine were examined by treatment with various pharmacological agents. Chapter 6 describes radiosynthesis and preclinical evaluation of [11C]prucalopride as a potential agonist PET ligand for the 5-HT4 serotonergic receptor. Finally, in Chapter 7, a summary of chapters 2 to 6 is provided together with a general discussion and future perspectives.

2 overview of agonist PET ligands for 7-transmembrane receptors

Hans J.C. Buiter, Albert D. Windhorst, Adriaan A. Lammertsma and Josée E. Leysen

Manuscript in preparation

Chapter 2

Overview of agonist PET ligands for 7-transmembrane receptors

2.1 doPAMInE RECEPToRS

In the brain, dopaminergic neurotransmission occurs via three pathways projecting to conined forebrain regions: the nigro-striatal, meso-limbic and meso-cortical pathways. In addition, there are tubero-infundibular projections to the pituitary. Dopamine medi- 2 ates in multiple central functions comprising motor functions, feelings of pleasure and reward, interest and motivation, attention and mood [49-52]. Five distinct subtypes of dopamine receptors, all belonging to the class of 7-TMRs, have been identiied. They were classiied into two classes, based on their main signalling pathway. The D1-like receptors, consisting of the subtypes D1 and D5, couple to the Gαs protein and stimulate cAMP production. The D2-like receptors, comprising the subtypes D2, D3 and D4, couple + to Gαi/o proteins and inhibit cAMP production and/or open K -channels [53,54]. In the brain, the density of D1 and D2 receptors is a factor 100 higher than that of D3, D4 and D5 receptors. High densities of both D1 and D2 receptors are found in the caudate nucleus, putamen and nucleus accumbens. They have distinct cellular localisations with D1 receptors occurring on neurones of the direct pathway of the basal ganglia circuit and high concentration are observed in cortical areas, and D2 receptors occurring both on neurones of the indirect pathway and on striatal inter-neurones. D1 and D2 receptors are located in synaptic and extra-synaptic parts of neuronal cells. D3 receptors occur exclusively in the brain with high abundance in the Islands of Calleja and lower levels in the nucleus accumbens, cortical regions and cerebellum [55-57]. D4 and D5 receptors have been detected in cortical areas.

All antipsychotic drugs developed to date have D2 and D3 antagonistic properties, some are relatively selective (certain benzamides, e.g. raclopride and sulpiride and the butyrophenone, haloperidol), but most antipsychotics have a rather broad receptor proile [58]. Nevertheless, based on clinical studies with a wide variety of compounds, it is believed that D2 antagonism is responsible for mitigation of positive symptoms of schizophrenia. High D2 receptor occupancy/blockade, however, may lead to extrapyramidal symptoms and prolactin elevation. D2 agonists are used for treating symptoms of Parkinson’s disease, restless leg syndrome and suppression of lactation [59,60]. Although no truly selective D3 agonists or antagonists are available, a growing body of evidence suggests that dopamine D3 receptors are involved in mechanisms of drug dependence and abuse [57]. Clinical studies with D1 antagonists revealed aggravation of symptoms of schizophrenia, hypofrontality and worsening of cognitive functions. In contrast, studies with D1 agonists in monkeys demonstrated improvement of attention and cognitive functions. Unfortunately, existing D1 agonists penetrate the brain poorly and sufer from pharmacokinetic draw-backs, so that clinical studies in man have not been successful.

Dopamine D2/3 receptors have been at the forefront of agonist PET ligand research, and this has resulted in a variety of potential agonist radioligands (see Finnema et al. [61] for

Chapter 2

a detailed review). So far, compounds acting on D4 and D5 receptors have not led to useful therapeutic drugs. In the following paragraphs an overview of dopamine receptor radioligands that were evaluated in vivo is provided.

2.1.1 dopamine d1 selective agonist PET ligands R- and S-[11C]n-Methyl-nnC 01-0259 11 The D1 receptor partial agonists (R)- and (S)-[ C]N-Methyl-NNC 01-0259 were labelled with carbon-11 and their characteristics were assessed in a preliminary PET study in non-human primates (Table 1). The (S)-enantiomer appeared to be superior over the (R)- enantiomer. Pre-treatment with the D1 receptor selective antagonist (R)-(+)-SCH23390 reduced uptake in striatum, expressed as the speciic binding ratio (SBR), pointing to 11 possible selective D1 receptor labelling. In vivo uptake of (S)-[ C]N-Methyl-NNC 01- 0259 was found to be insensitive to an amphetamine induced increase in extracellular dopamine levels [62,63]. The radioligand appeared to be a substrate for cathechol-O- methyltransferase yielding lipohilic radiolabelled metabolites, which showed ainity for 11 cerebral D1 receptors [64]. The selectivity of (S)-[ C]N-methyl-NNC 01-0259 for D1 receptors and its sensitivity to an increase in endogenous dopamine levels was investigated further in a recent study [65]. In non-human primates, striatal non-displaceable binding 11 potential (BPND) [35] of (S)-[ C]N-methyl-NNC 01-0259, corrected for radiolabelled metabolites, could be reduced by the dopamine D1 receptor antagonist (R)-(+)-SCH 23390.

Treatment with the 5-HT2A selective antagonist MDL-100907, however, also showed an approximately 30% decrease in binding in the neocortex, pointing to potential binding to 5-HT2A receptors. Administration of D-amphetamine to enhance extracellular dopamine concentrations did not afect its binding. Based on the study results authors concluded that the ligand showed speciic binding to dopamine D1 receptors in striatum, possibly representing binding to the high-ainity state of the receptors. In the same study, (S)-[11C]N-Methyl-NNC 01-0259 uptake in the brain also appeared to be insensitive to increased endogenous dopamine concentrations. It was argued that the lack of sensitivity to changes in extracellular dopamine could be due to either the fact that partial dopamine D1 receptor agonists have lower intrinsic activity than dopamine, or the small number of dopamine D1 receptors that would be present in the high ainity state. Alternative explanations, such as extra-synaptic localisation of D1 receptors, were not considered.

Overview of agonist PET ligands for 7-transmembrane receptors = ND 2 = (striatum-cerebellum)/cerebellum, BP = (striatum-cerebellum)/cerebellum, cer cer 1.5-3.5 Demo [62,63] 0.6-2.6 Demo [62] 0.4-1.6 n.i. [64] 0.2-1.8 Demo [65] cer cer cer ND ND BP BP SBR SBR Non-human primate Non-human primate primate primate Ref. PET species Speciic binding Selectivity Ref. (cLogP eLogD) (cLogP : inhibitions equilibrium constant. Speciic binding ratio expressed as: SBR expressed Speciic binding ratio : inhibitions equilibrium constant. i 3 CH 11 N H O C]N-methyl-NNC 01-0259 Cl receptor selective agonist PET ligands receptor 1 11 D )-[ S C]N-methyl-NNC 01-0259 220 nM [65] C = 4.05 Cb Non-human C]N-methyl-NNC 01-0259 4.9 nM [65] C = 4.05 Cb Non-human / 11 11 R ( HO )-[ )-[ R S HO Table 1. Table Ligand Ki Ref. Lipophilicity non-displaceable [35]. binding potential ( C = calculated Log P, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands) ; demo : selectivity demonstrated by (pre)treatment with the Netherlands) ; demo : selectivity (pre)treatment Groningen, by demonstrated 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation Cb : own P, Log C = calculated K n.i.: selectivityselective inhibitors; not investigated. (

Chapter 2

Table 2. Apomorphine derived D2/3 agonist PET ligands R1

OH HO

N H R2 Ligand R1 R2 Ki Ref. Lipophilicity Ref. PET species Speciic Selectivity Ref. cLogP eLogD binding

11 [ C]MNPA -O*CH3 -C3H7 0.17 nM [80] c= 3.69 [61] Human BPND 0.8 n.i. [69]

Non-human BPND 1.0 n.i. [81] primate

Non-human BRcer 2 n.i. [66] primate

Non-human BPND 2.2 Demo [67] primate

Non-human BPND 1.3 n.i. [70] primate

Non-human BPND 0.8-1.0 Demo [82] primate

Rat BPND 0.9 n.i. [68]

Rat BPND 0.8 n.i. [81] 11 [ C]NPA -H -*CH2C2H5 0.8 nM [83] e7.4=2.54 [84] Human BPND 0.8-0.9 n.i. [85]

Non-human BRcer 2.9 Demo [72] primate

Non-human BPND 1.2-1.3 n.i. [86] primate

Pig BPND 1.2-1.3 n.i. [87]

Cat BPND 1.5 n.i. [88] 11 [ C]-2-OH-NPA -OH -*CH2C2H5 0.053 nM [80] c= 0.77 [61] Non-human non speciic n.i [66] primate

18 [ F]MCL 524 -C2H4*F -C3H7 3.7 nM [89] c= 3.77 Cb Non-human BPND 1.99 n.i. [78] primate

Non-human BPND 2.2 Demo [79] primate

18 [ F]MCL 534 -C2H4*F -CH3 1.19 nM [89] c=2.71 Cb Non-human BPND 0.20 n.i. [78] primate

18 [ F]MCL 538 -C2H4*F -C2H5 0.83 nM [89] c=3.23 Cb Non-human BPND 0.34 n.i. [78] primate * = labelled position, carbon-11 or luorine-18; C = calculated Log P; Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental LogD at indicated pH; demo = selectivity demonstrated by (pre) treatment with selective inhibitors; ; n.i.: selectivity not investigated; Ki: inhibitions equilibrium constant. Speciic binding is expressed as: BPND = non-displaceable binding potential or BRcer = Binding ratio striatum/cerebellum.

Overview of agonist PET ligands for 7-transmembrane receptors

2.1.2 dopamine d2/3 selective agonist PET ligands Radiolabelled apomorphines Various radiolabelled apomorphine derivatives that have been investigated using PET are presented in Table 2. The dopamine agonist, (R)-2-[11C]methoxy-N-n-propyl- nor-apomorphine ([11C]MNPA) was irst radiosynthesized and evaluated in non-human 2 primates in 1992. The radioligand readily passed the blood brain barrier and marked uptake of radioactivity in striatum was demonstrated with a ratio over cerebellum of 2 [66]. Fourteen years later, the characteristics of [11C]MNPA uptake in the brain were investigated in non-human primates, using baseline PET and blocking studies with raclopride, a selective D2/3 receptor antagonist. Results showed high uptake in dopamine

D2 receptor rich brain areas with a maximal striatum to cerebellum ratio of 2.3 at 78 min post injection. Raclopride predosing resulted in a reduction of brain uptake, reducing the striatum to cerebellum ratio to 1.26 [67]. In a study in rats, [11C]MNPA was used to assess occupancy of central dopamine D2/3 receptors by endogenous dopamine [68].

In that study, baseline BPND increased twofold after dopamine depletion by reserpine and α-methyl-para-tyrosine pre-treatment. D2/3 receptor occupancy by endogenous dopamine was estimated to be about 55% [68]. Furthermore, striatal binding was displaceable with raclopride but not with BP 897, a partial dopamine D3 receptor agonist. A study with [11C]MNPA in 10 healthy human volunteers showed that its regional distribution was in agreement with previous PET studies in non-human primates and with human PET studies using D2 receptor antagonist radioligands [69]. The efect of amphetamine on [11C]MNPA binding was evaluated in non-human 11 primates, comparing results directly with those of the D2/3 receptor antagonist [ C]raclopride [70]. This study demonstrated a larger amphetamine induced reduction of 46% in [11C]MNPA binding than the 23% reduction in [11C]raclopride binding. This efect is in contrast with results observed for the D1 PET ligands, where no efect between antagonist and angonsit binding could be observed. This can be explained by the proportion of the receptor that is present in the high ainity state. For D2 receptors, most of the receptors occur in the high ainity state, whereas for D1 receptors the receptor is mostly present in the low ainity state [71]. In non-human primates, PET Imaging using carbon-11 labelled N-n-propyl-nor-apomorphine ([11C]NPA) showed a high striatum/cerebellum binding ratio (BR) of 2.8 [72]. A comparative study of [11C]NPA and [11C]raclopride in non-human primates provided in vivo KD values of 0.16 and 1.59 nM in striatum, respectively. Corresponding Bmax values were 21.6 and 27.3 nM, respectively. Based on these Bmax values it seems that, in non- human primates, 79% of the dopamine D2 receptors in vivo are in the high ainity state, [73]. In addition, the sensitivity of [11C]NPA to amphetamine induced changes in endogenous dopamine levels was investigated in non-human primate PET studies, again using

Chapter 2 a direct comparison with [11C]raclopride studies in the same animal [74]. The efect of amphetamine was larger for [11C]NPA than for [11C]raclopride, suggesting that 70% of the D2 receptors would be present in the high ainity state [74,75]. These indings were conirmed in human volunteers [76]. The amphetamine induced increase in synaptic 11 dopamine in human striatum resulted in a 49 to 90% larger decrease in BPND of [ C]NPA 11 compared with that of [ C]raclopride. The carbon-11 labelled 2-OH analogue of N-n-propyl-nor-apomorphine, (R)-[propyl- 11C]2-OH-NPA, was evaluated in non-human primates, but this radioligand ligand did not appear to pass the blood brain barrier, suggesting that it may not be suitable for CNS imaging [66]. Initial radiolabelling of apomorphine-like structures with luorine-18 did not result in PET ligands that were successful with respect to brain uptake and speciic binding, assessed using ex vivo biodistribution experiments [77]. Recently, three new luorine-18 radiolabelled ligands were synthesized and evaluated in non-human primates in vivo. Stephanov et al. [78] showed that [18F]MCL 524, [18F]MCL 534 and [18F]MCL 538, all passed the blood brain barrier, yielding BPND values of 1.99, 0.20 and 0.34, respectively. 18 Additional [ F]MCL 524 studies in non-human primates resulted in a striatal BPND value of 2.2, which could be reduced by 43 and 87% following treatment with D-amphetamine and raclopride, respectively. It was concluded that [18F]MCL 524 has very promising characteristics to study dopamine release in human subjects [79].

Radiolabelled aminotetralins Radiolabelled aminotetralins, as evaluated in PET studies, are listed in Table 3. In non-human primates, carbon-11 labelled (±)-2-(N,N-dipropyl)amino-5-hydroxytetralin ([11C]-5-OH-DPAT) showed selective binding in striatum in vivo with a striatum over cerebellum binding ratio of 2 [90]. Speciicity of in vivo binding, however, was not reported. (R,S)-2-(N-propyl-N-5′-luoropentyl)amino-5-hydroxytetralin (5-OH-FPPAT) has also been labelled with luorine-18. In non-human primates, [18F]-5-OH-FPPAT showed selective uptake in the striatum with a striatum over cerebellum ratio of approximately 2 [91], similar to that seen for [11C]-5-OH-DPAT. Aminotetralin analogues with sub-nanomolar ainity for the D2/3 receptors, (±)-2-(N-phenethyl-N-propyl)amino-5-hydroxytetralin (PPHT) and (±)-2-(N-cyclohexylethyl-N-propyl)amino-5-hydroxytetralin (ZYY339), have been radiolabelled with carbon-11 and evaluated in non-human primates [92]. Both [11C]PPHT and [11C]ZYY339 showed increased uptake in striatum and thalamus, but the binding ratios over cerebellum of 1.5 for [11C]PPHT and 1.3 for [11C]ZYY339 were lower than for other radiolabelled aminotetralins. It was suggested that this could be due to the relatively slow washout of nonspeciic signal from the cerebellum, which was ascribed to the increased lipophilicity compared with 5-OH-DPAT (experimental Log distribution between 1-octanol and phosphate bufer at indicated pH (eLogD) of 3 vs. 1.4)

Overview of agonist PET ligands for 7-transmembrane receptors

Table 3. Aminotetralin derived D2/3 agonist PET ligands.

R1 N R2 2

O H Ligand R1 R2 7-TMR Ki Ref. Lipophilicity Ref. PET Speciic Selectivity Ref. cLogP eLogD species binding

11 [ C]-5-OH-DPAT -C3H7 -*CH2C2H5 D2/3 [2.5 nM] [93] e7.4 = 1.37 [93] Non- BRcer 2 n.i. [90] human primate

18 [ F]-5-OH-FPPAT -C3H7 -C5H10*F D2/3 [6.95 nM] [91] e7.4 = 1.6 [91] Non- BRcer 2 n.i [91] human primate

11 [ C]-PPHT -*CH2C2H5 -C2H4C6H5 D2/3 [0.65 nM] [93] e7.4 = 3.03 [93] Non- BRcer 1.5 n.i [92] human primate

11 [ C]-ZYY-339 -*CH2C2H5 -C2H4C6H11 D2/3 [0.01 nM] [93] e7.4 = 3.09 [93] Non- BRcer 1.5 n.i [92] human primate * = labelled position, carbon-11 or luorine-18; C = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental LogD at indicated pH; demo = selectivity demonstrated by (pre) treatment with selective inhibitors; ; n.i.: selectivity not investigated. Ki: inhibitions equilibrium constant, igures between brackets are IC50 –values. Speciic binding is expressed as: BRcer = Binding ratio striatum/cerebellum.

[92]. An additional confounding factor could have been the fact that both radioligands consisted of mixtures of enantiomers. A possible contribution of binding to D3 receptors in the cerebellum by these high ainity radioligands was not considered by the authors.

Chapter 2

11 Table 4. The D2/3 agonist PET ligand, [ C]-(+)PHNO.

11CH HO N 2 O

[11C]-(+)PHNO Ligand Ki Ref. Lipophilicity Ref. PET species Speciic Selectivity Ref. (cLogP eLogD) binding

11 [ C]-(+)PHNO D2: 0.24 nM [105] e7.4=2.14 [105] Human BPND 3.0 – 4.17 Demo [94]

D3: 0.60 nM Human BPND 2.1-3.6 n.i. [95]

Non-human BPND 2.1-3.9 n.i. [88] primate

Cat BPND 4 Demo [88] e = experimental LogD at indicated pH; demo = selectivity demonstrated by (pre)treatment with selective inhibitors; n.i.: selectivity not investigated; Ki: inhibitions equilibrium constants. Speciic binding is expressed as: BPND = non-displaceable binding potential [35].

[11C]-(+)PHno (+)-4-propyl-3,4,4a,5,6,10b-hexahydro-2H-naphtho[1,2-b][1,4]oxazin-9-ol ((+)PHNO) is an aminotetralin in which the amine function, as present in dopamine, is located in the oxazin ring structure. The compound showed selective D2/3 agonist properties and labelling with carbon-11 yielded a promising PET ligand, which has been investigated extensively [84,88,94,95] (Table 4). Using PET studies in cats, [11C]-(+)PHNO binding was studied after pre-treatment with the D1 selective antagonist, SCH23390, the D2/3 selective antagonists raclopride and haloperidol, and the D3 selective antagonist SB-277011. 11 Results showed speciic and selective [ C]-(+)PHNO binding to D2 receptors. Moreover, [11C]-(+)PHNO was more sensitive to the dopamine releasing efect of amphetamine than the antagonist ligand [11C]raclopride, with reductions in binding of 83 and 56%, respectively [88]. In human volunteers, uptake of [11C]-(+)PHNO was seen in caudate, putamen, and globus pallidus. In addition, uptake was seen in substantia nigra/ventral tegmental area. Pre-treatment with haloperidol reduced binding potential in D2 receptor rich regions. Using repeat studies in the same subjects, [11C]-(+)PHNO showed higher uptake in the globus pallidus than [11C]raclopride. In contrast to [11C]-(+)PHNO, [11C] raclopride did not show uptake in the substantia nigra/ventral tegmental area [94]. [11C]-(+)PHNO binding in extra striatal regions, in particular in the globus pallidus, seems to involve a large fraction of D3 receptor binding, Evidence for this was found in the 59% reduction of [11C]-(+)PHNO uptake in the globus pallidus following pre- treatment with the partial D3 agonist BP897, which was signiicantly larger than the corresponding 29% reduction of [11C]raclopride uptake in the same area [96]. A human

Overview of agonist PET ligands for 7-transmembrane receptors

PET study with [11C]-(+)PHNO and amphetamine challenge, compared with placebo, showed signiicant reduction in BPND in caudate, putamen and ventral striatum, but not in globus pallidus [97]. In addition, a within subject study between [11C]-(+)PHNO and [11C]raclopride in human subjects was reported, the comparison between brain uptake, distribution, and binding characteristics of the two D radioligands showed that [11C]-(+) 2 2 PHNO had a preferential distribution in the globus pallidus and in the ventral striatum, while [11C]raclopride showed preferential uptake in the dorsal striatum, providing two 11 distinct views for the D2/3 system of the human brain [98]. Recently, [ C]-(+)PHNO PET imaging was used in patients with schizophrenia and in healthy controls (age and sex matched) to study diferences in D2 and/or D3 receptor binding, yet no diferences were observed [99]. Studies with [11C]-(+)PHNO and [11C]raclopride in both patients on antipsychotics and healthy human volunteers showed that in antipsychotic treated patients, for both tracers, higher occupancies were observed in dorsal striatum [100]. In 11 another study, uptake of [ C]-(+)PHNO in globus pallidus could be blocked with the D3 receptor preferring agonist pramipexole. As an explanation for this phenomena it was argued that antipsychotics would have blocked both high- and low-ainity states of the

D2 receptors in the brain, whilst antipsychotic treatment most likely would not block 11 11 the [ C]-(+)PHNO signal in D3 receptor rich regions [101]. Further studies with [ C]-(+)

PHNO in humans performed to assess its D3 binding contribution are reported in [102] and [103]. Another study with both [11C]-(+)PHNO in non-human primates together with an in vitro study using a tritiated analogue in D2- and D3 knock-out mice compared with wild- 11 3 type mice demonstrated a predominantly D3 related component of [ C/ H](+)-PHNO binding in extrastriatal and midbrain regions. Changing the labelling position in the (+)-PHNO molecule to [3-11C]-(+)-PHNO resulted in comparable results as [11C]-(+)PHNO [104], indicating that the labelling position does not inluence the in vivo characteristics of the radioligand. Furthermore, PET studies with [3-11C]-(+)PHNO in rats demonstrated speciic uptake of radioactivity in the cerebellum [57], a region that contains D3 receptors, but is devoid of D2 receptors [55]

2.1.3 d3 receptor partial PET agonists ligands N-[4-[4-(2-methoxyphenyl)piperazin-1-yl]butyl]benzo[b]thiophene-2-carboxamide

(FAUC346), a partial agonist with high selectivity for D3 over D2 receptors, was labelled with carbon-11 (Table 5). [11C]FAUC346 demonstrated speciic uptake in rat brain, but in a baboon PET study it failed to show speciic uptake after competition with the D3 partial agonist BP 897 [106]. Fluorine-18 labelled analogues were also radiosynthesized, but preliminary evaluation in rat brain using ex vivo autoradiography and biodistribution did not warrant further PET experiments [107].

Chapter 2

Table 5. D3 partial agonist PET ligands.

O N O N N O N O N N 11 CH3 H S H 18F S

[11C]FAUC346 [18F]5

Ligand 7-TMR target Ki R Ref. Lipophilicity Ref. PET species Speciic binding Selectivity Ref. (cLogP eLogD)

11 [ C]FAUC346 D3 D3 0.23nM [109] e7.4= 2.36 [109] Non-human Non speciic n.i. [106]

D2 52 nM primate

18 [ F]5 D3 D3 0.17 nM [108] c= 4.67 Cb Non-human DVR-1 0.13-0.23 n.i. [108] primate C = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental LogD at indicated pH; demo = selectivity demonstrated by (pre)treatment with selective inhibitors; n.i.: selectivity not investigated. Ki: inhibitions equilibrium constant. Speciic binding is expressed as: DVR-1 = binding potential non-displacable derived from volumes of distribution measured with arterial plasma concentrations of radioligand [35].

A series of FAUC346 related compounds was radiolabelled and evaluated in non- human primates using small animal PET studies. The agonist PET ligand [18F]N-[4- [4-(2-(2-luoroethoxy)phenyl)piperazine-1-yl]butyl]-4-(3-thienyl)benzamide ([18F]5) demonstrated variable uptake in regions known to be rich in D3 receptors (Table 5). Pre-treatment with lorazepam, a GABA-A receptor modulator which decreases activity of dopaminergic neurons, resulted in signiicantly increased binding potential (derived from volumes of distribution measured with arterial plasma concentrations of this radioligand [35] (DVR-1)), in caudate, putamen and thalamus [108]. This inding suggests relatively tight binding of dopamine to D3 receptors under normal conditions, making D3 receptors not readily available for PET ligands in vivo, which is in line with observations made in ex vivo autoradiography studies [6].

2.1.4 General discussion In various species, [11C]-(+)PHNO showed the most favourable characteristics of all dopamine selective agonist PET ligands available to date. These characteristics include good lipophilicity, excellent receptor ainity and good brain uptake. Although initially 11 [ C]-(+)PHNO was developed as a D2/3 agonist PET ligand, it should now be considered as a D3 preferring agonist PET ligand. Its potential to bind to extrastriatal D3 receptors makes it a useful tool to study these receptors in vivo. Although the cerebellum is used as reference tissue for many (dopaminergic) ligands, it is questionable whether this is valid for D3 ligands, as the cerebellum is not devoid of D3 receptors.

Overview of agonist PET ligands for 7-transmembrane receptors

11 11 Both the dopamine D2/3 agonist ligands [ C]NPA and [ C]MNPA allowed for an in vivo evaluation of the high ainity state of D2/3 receptors using PET. Both ligands were found to be more sensitive to endogenous ligand displacement than the commonly used antagonist [11C]raclopride. The recently developed luorine-18 labelled apomorphine analogue [18F]MCL 524 holds promise as an agonist PET ligand, given its favourable 2 initial results. Overall, PET studies using dopamine agonist radioligands have provided diferential information in particular with regard to the activity state of the dopamine receptor subtypes. Additional information compared with antagonist radioligands included: (1) percentage of receptors in the activated state, (2) activated receptors accessible under normal conditions and after dopamine depletion or inhibition of dopaminergic signalling, (3) percentage of receptors that become occupied by endogenous dopamine following amphetamine challenge. Attempts have already been made to study diferences in activated state of D2/3 receptors in patients with schizophrenia as compared with normal volunteers, and before and after prolonged treatment with dopamine receptor blocking antipsychotics. In the future, this type of studies may shed further light on the role of D2/3 receptors in various diseases and their activity state following various treatments.

2.2 5-Hydroxytryptamine receptors In mammals, central serotonergic neuronal projections originate in the raphe nuclei and project widely to the forebrain, hindbrain and spinal cord [47,110-112]. There are two types of serotonergic neurones in the brain, thin D-ibres that do not form synaptic con- tacts and act through volume transmission, and thick M ibres that do form synapses. Serotonergic neurones display tonic activity and no burst iring. Fourteen diferent

5-HT receptors, divided into seven subclasses, have been identiied. The 5-HT1A/1B/1D/1E/1F,

5-HT2A/2B/2C, 5-HT4, 5-ht5A/5B, 5-HT6 and 5-HT7 receptors are 7-TMRs, whilst the 5-HT3 receptor belongs to the class of ligand-gated ion channels [110]. All but one of the 5-HT receptor subtypes occur in the brain, showing distinct distributions.

All 5-HT1 receptor subtypes are coupled to a subtype of G proteins type that inhibit ad- + 2+ enylyl cyclase (Gi/o) and induce opening of K channels and closing of Ca channels, and stimulate phospholipase Cb and mitogen-activated protein kinase [113,114]. The highest density of 5-HT1A receptors is found in the hippocampus, followed by septum, some of the amygdaloidal nuclei, cortical layers and raphe nuclei. 5-HT1A receptors have a role in the regulation of serotonergic neuronal activity and are involved in mood, anxiety, emotion and cognitive functions. 5HT1A agonism is a component of certain anxiolytic, antidepressant and antipsychotic drugs. Selective ligands for the 5HT1A receptor have been investigated for the treatment of various psychiatric diseases, but no selective compounds made it to the market.

Chapter 2

High densities of 5-HT1B receptors can be found in the globus pallidus and substantia nigra pars reticulate, followed by caudate, putamen, nucleus accumbens, periaque- ductal gray and hippocampal formation [115,116]. 5-HT1B receptors are located presynaptically, as auto-receptors on serotonergic neurones and as hetero-receptors on various neurones. 5-HT1B receptors regulate release of 5-HT and other neurotransmitters, and have been a target for developing antidepressant drugs. In peripheral tissue, they occur on blood vessels, substance P and calcitonin gene related peptide containing neurons from the trigeminal ganglia, where they have a role in migraine. In fact, 5-HT1B agonists are marketed as anti-migraine agents. Presence of 5-HT1D receptors in the brain is limited, as shown by in situ hybridisation studies [115,116]. The available ligands lack sybptype selectivy and bind to 5-HT1B and 5-HT1D receptors, and in vitro studies indicate that the 5-HT1B receptor is predominant over the 5-HT1D receptor, thus the 5HT1D receptor ligand will be hardly detectable in the brain [115,116]. 5HT1D mRNA is abundantly found in calcitonin gene-related peptide containing trigeminal neurones, but this does not result in. 5-HT1E receptors are highly expressed in the human frontal cortex and hippocampus, suggesting a possible role in cognitive functions, but only a few selective ligands are available [117]. The 5-HT1F receptor has been identiied, but little is known about its distribution and function.

The 5-HT2 receptor class comprises the 5-HT2A, 5-HT2B and 5-HT2C receptor subtypes.

They couple to the Gq/11-protein with activation of phospholipase C, leading to formation of inositol 1,4,5-trisphosphate and diacylglycerol, further triggering elevation of cytosolic Ca2+, activation of protein kinase C and downstream intracellular processes. In addition, activation of phospholipase A2 and release of arachidonic acid have been described [118,119]. 5-HT2A receptors are abundant in the telencephalon, but are scarcely found in midbrain and hindbrain. They occur in high density in the frontal cortex and further throughout the cortex, especially in neocortical laminae, claustrum, nucleus accumbens, olfactory tubercle, caudate and putamen. 5-HT2A receptors have been localized in synaptic regions, but also outside synapses, e.g. on dendrites of pyramidal neurons [120]. 5-HT2A receptors occur on blood platelets, peripheral smooth muscle and on blood vessels. 5-HT2B receptors have not been detected in the brain, they are present in peripheral tissues e.g. in the stomach fundus, lungs and myocardium. 5-HT2C receptors are abundantly present in the telecephalon, midbrain, hindbrain, cerebellum and spinal cord. The highest density occurs in the choroid plexus.

Blocking 5-HT2A receptors is part of the mechanism of action of antipsychotic and antidepressant drugs. 5-HT2A agonists are hallucinogenic agents. They are responsible for the action of e.g. LSD and hallucinogens in mushrooms. Blocking 5-HT2C receptors is associated with various antipsychotics and tricyclic antidepressants, and, as such, explains side efects such as increased food intake leading to excessive weight gain.

Overview of agonist PET ligands for 7-transmembrane receptors

5-HT2C agonists are applied for treatment of obesity [121] (for a review of the 5-HT2 class of receptors and ligands, see Leysen 2004 [118]).

5-HT4 receptors occur in both brain and peripheral tissues. They are present in olfactory tubercles, nucleus accumbens, amygdala, corpus striatum, globus pallidus, substantia nigra, hippocampus and septum [29]. Apart from mediating cognitive functions, 2 several other functions of central 5-HT4 receptors have been postulated [110,122]. In the periphery, they are present in intestines and heart. 5-HT4 agonists are on the market for treatment of constipation and have been studied for treatment of cognitive impairment.

5-ht5a/5b receptors were discovered by gene cloning. They appeared to be coupled primarily to Gi/o proteins, leading to inhibition of adenylyl cyclase [123,124], yet no selective ligands have been reported and their pharmacological properties remain to be elucidated.

5-HT6 receptors seem to occur exclusively in the brain. High levels of 5-HT6 receptor mRNA were found in striatum, nucleus accumbens and olfactory tubercles, and limbic and forebrain regions, including hippocampus and cortex [47,125]. 5-HT6 receptors are

Gs coupled and stimulate cAMP formation. They are thought to play a role in cognitive functions and feeding behaviour [125]. 5-HT6 antagonism is a component of some antipsychotic drugs. Several 5-HT6 antagonists have been in development for potential treatment of schizophrenia and cognitive dysfunction. In addition, they have been investigated for possible treatment of anxiety. However, no selective 5-HT6 ligand reached the market [47,125].

5-HT7 receptors were also identiied by gene cloning [126]. Radioligand autoradiography and mRNA in situ hybridization in rat brain sections demonstrated that 5-HT7 receptors are located in supericial layers of neocortex, lateral septum, hypothalamus, some nuclei of dorsal and midline thalamus, and parts of the hippocampal formation

[127]. 5-HT7 receptors couple to Gs proteins and stimulate adenylyl cyclase. They have a role in the regulation of circadian rhythm and thermoregulation, and are thought to be involved in mood disorders [128]. 5-HT7 receptors occur in various peripheral tissues.

5-HT7 antagonism is associated with some antipsychotic drugs. Selective 5-HT7 antagonists have been investigated for the treatment of depression. Finally, 5-HT7 agonists were reported to reduce nerve injury [129].

In summary, in line with the multiple central roles of serotonin, 5-HT receptor subtypes are involved in various central functions including mood, cognition, anxiety, aggression, feeding and sleep. Moreover, in most of these functions several of the 5-HT receptor subtypes play a role. Serotonergic dysfunction has been implicated in psychiatric disorders including depression, anxiety, schizophrenia, and in neurological disorders such as Alzheimer’s disease, and epilepsy. For an extensive overview of 5-HT radiolabelled antagonists and agonists for PET and SPECT see Patterson et al. [130].

Chapter 2

cer 3 [135] [133] [139] [137] CH O N 11 N n.i. Demo Demo n.i. C]CUMI-101 11 O N N N C]MMP / [ 2.9 Demo [131] 6.25 12-31 9.70-32.62 11 cer cer F F O BR - BP BP 3 CH 11 O Non-human primate Non-human primate Non-human primate N N Ref. PET species Speciic binding Selectivity Ref. O N N N (cLogP eLogD) (cLogP C]MMT [ O 11 O C 11 3 H N = binding potential free in plasma [35] for various brain regions. various brain in plasma [35] for free = binding potential F N O agonists PET ligands (part 1). 1A N N N 5-HT C]MPT [ 11 O [ C]MPT 1.36 nM [131] c= 2.2 Cb Non-human primate BR C]MMTC]MMP/CUMI-101 0.15 nM 1.1 nM [133] [132] c= 1.1 c= 1.57 Cb Cb Human Non-human primate Non speciic n.i. [132] 11 11 11 C = calculated Log P; , Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental LogD at indicated pH; demo = indicated at the Netherlands); LogD e = experimental Groningen, 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation , Cb : own P; Log C = calculated as: BR Ki: Speciic binding is expressed inhibitions equilibrium constant. n.i.: selectivity with selective inhibitors; not investigated. selectivity (pre)treatment by demonstrated Table 6. Table Ligand Ki Ref. Lipophilicity = hippocampus/cerebellum, BP = hippocampus/cerebellum, [ [ [

Overview of agonist PET ligands for 7-transmembrane receptors

2.2.1 5-HT1A selective agonist PET ligands

Several 5-HT1A agonists were radiolabelled and evaluated in vivo, but with limited success so far. 5-HT1A agonist radioligands are presented in Tables 6 and 7. Three piperazine- butyl-triazine-diones, showing very high 5-HT receptor ainity, were labelled with 1A carbon-11 (table 6). [O-methyl-C-11]2-[4-[4-(7-methoxynaphthalen-1-yl)piperazin-yl] 2 butyl]-4-methy1-2H-[1,2,4]triazine-3,5-dione ([11C]MPT) was studied in non-human primates and showed speciic uptake in brain regions rich in 5-HT1A receptors. The signal could be blocked by pre-treatment with the 5-HT1A antagonist WAY100635 and the 5-HT1A agonist (±)-8-OH-DPAT [131]. A major drawback of this ligand, however, was the rather slow washout from the reference tissue (cerebellum), compromising proper quantiication of binding potential. The 3-methoxyphenyl analogue of MPT, [O-methyl-C-11]2-[4- [4-(3-methoxyphenyl)piperazin-1-yl]-butyl]-4-methyl-2H-[1,2,4]-triazine-3,5-dione ([11C] MMT), did cross the blood brain barrier, but sufered from low speciic binding and fast clearance, making if unsuitable for clinical studies [132]. The 2-methoxyphenyl analogue of MPT, [O-methyl-C-11]2-(4-(4-(2-methoxyphenyl) piperazin-1-yl)butyl)-4-methyl-1,2,4- triazine-3,5(2H,4H)dione ([11C]MMP), also designated as [11C]CUMI-101, showed good and speciic uptake in 5-HT1A receptor rich brain regions in non-human primates [133]. In addition, in these regions uptake was reduced by pre-treatment with WAY-100635 and (±)-8-OH-DPAT. A PET study in human volunteers conirmed rapid brain uptake and variable washout from diferent brain regions relected the known distribution of 5-HT1A receptors [134,135]. In these studies binding was expresses as the binding potential free in plasma (BPF) which is deined as the ratio at equilibrium of the concentration of speciically bound radioligand in tissue to the concentration of free radioligand in tissue [35]. In a recent, preliminary study in two non-human primates, [11C]CUMI-101 showed sensitivity to in vivo displacement of binding by endogenous extracellular serotonin, elicited by pre-treatment with either the serotonin re-uptake blocker citalopram or the serotonin releasing agent fenluramine [136,137]. Using these pharmacological challenges, receptor occupancy by endogenous serotonin was estimated to be 15 to 30%. It was concluded, however, that it is unlikely that changes in intrasynaptic 5-HT levels can be detected under physiological conditions. It should be noted that similar studies with the antagonist ligands [11C]WAY100635 and [18F]MPPF were not successful in demonstrating endogenous ligand competition following treatment with amphetamine or fenluramine [138].

Further radiolabelled 5-HT1A selective agonists are presented in Table 7. In non-human primates, [O-Methyl-11C]4-[3-[4-(2-methoxyphenyl)-piperazin-1-yl]propoxy]-4-aza- tricyclo-[5.2.1.02,6]dec-8-ene-3,5-dione ([11C]4), showed rapid uptake in the brain [140].

However, rapid washout and lack of retention in 5-HT1A receptor rich regions renders this ligand not useful as a 5-HT1A receptor agonist ligand for clinical studies (Table 7).

Chapter 2

F 18 NH O F]S14506 18 N 3 CH N 11 1.2-1.5 Demo [143] O 1.1-1.6 n.i. [140] C]S14506 / [ cer cer cer 11 [ SUVr 1.5 SUVr n.i. [142] SUVr 1.35SUVr n.i. [142] BR F 3 18 CH Cl 11 O primate primate primate O N N OH F Ref. PET species Speciic binding Selectivity Ref. O N H N C]thiochromane 11 N S [ N = striatum/cerebellum. F]F15599 cer = 2.8 [140] Non-human 18 [ 7.4 O (cLogP eLogD) (cLogP C 11 3 N H O N O N O 4 C] 11 [ 5 nM [141] c= 4.40 Cb cat non speciic n.i. [141] 0.021 nM [140] e agonists PET ligands (part 2). 1A 5-HT 4 F]F15599 2.24 nM [145] c= 1.17 Cb cats BR F]S14506 0.15 nM [144] c=4.68 Cb Non-human C] C]S14506 0.15 nM [144] c=4.68 Cb Non-human C] 18 18 11 11 11 Table 7. Table [ [ [ [ = standard uptake value ratio ratio (striatum/cerebellum) or BR (striatum/cerebellum) ratio ratio uptake value = standard C = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands) ; e = experimental LogD at indicated pH; demo = indicated at the Netherlands) LogD ; e = experimental Groningen, 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation Cb : own P;, Log C = calculated as: SUVr Ki: Speciic binding is expressed inhibitions equilibrium constant. n.i.: selectivity with selective inhibitors; not investigated. selectivity (pre)treatment by demonstrated Ligand Ki Ref. Lipophilicity thiochromane [

Overview of agonist PET ligands for 7-transmembrane receptors

ND 2 NH 2 C]Cimbi-88 11 O O C 11 I 3 H 3 CH O 11 0.46 Demo [147] 0.17 n.i. [148] 0.49 n.i. [148] 0.43 n.i. [148] 0.17 n.i. [148] 0.45 n.i. [148] 0.32 n.i. [148] 0.60 n.i. [148] 0.32 n.i. [148] 0.82 Demo [148] NH ND ND ND ND ND ND ND ND ND ND Speciic binding Selectivity Ref. O species Ref. PET O C]Cimbi-31 [ Br 11 O (cLogP eLogD) (cLogP H N agonist PET ligands. 2A 2.2 nM C=4.45 Cb Pig BP 1.5±0.35 nm c=4.45 Cb Pig BP 1.01±0.17 nM c=4.24 Cb Pig BP 2.89±1.05 nM c=4.09 Cb Pig BP 0.35±0.05 nM c=4.63 Cb Pig BP O O C 3 3 3 3 11 I 3 3 C]Cimbi-29 [ H 11 -O*CH 3 R 3 O -I -O*CH -I -CH -Br -O*CH -Cl -O*CH -I -F 12.5±3.1 nM c=4.67 Cb Pig BP -I -OH 1.12±0.08 nM c=3.81 Cb Pig BP -CF 3 3 3 H N 3 3 3 3 O O 1[4-halo-2,5 alkylamine 5-HT dimethoxyphenyl] derived 1 R 2 C]Cimbi-R [ C]Cimbi-5 -CH C]Cimbi-5-2 -O*CH C]Cimbi-36 -CH C]Cimbi-82 -CH C]Cimbi-21 -O*CH C]Cimbi-88 - - - 47.2±16.3 nM c=2.36 Cb pig BP C]Cimbi-27 -O*CH C]Cimbi-138 -CH C]Cimbi-31 - - - 0.16±0.04 nM c=4.44 Cb Pig BP C]Cimbi-29 - - - 1.36±0.37 nM c=4.95 Cb Pig BP 11 11 11 11 11 11 11 11 11 11 11 R [ [ [ * = labelled position, carbon-11 or luorine-18; C = calculated Log P; Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands; demo = Groningen, 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation * = labelled position, carbon-11 Cb : own or luorine-18; P; Log C = calculated as: BP Ki: Speciic binding is expressed inhibitions equilibrium constant. n.i.: selectivity with selective inhibitors; not investigated. selectivity (pre)treatment by demonstrated [ [ [ [ [ [ = cortical non-displaceable binding potential [35]. Table 8. Table [ Ligand R1 R2 R3 Ki Lipophilicity [

Chapter 2

8[[3-[4-(2-[11C]methoxyphenyl)piperazin-1-yl]-2-hydroxypropyl]oxy]thiochromane ([11C] thiochromane) suffered a similar fate. This radioligand was evaluated in one awake cat and failed to show an in vivo distribution that was consistent with known in vitro data

[141]. Based on in vivo images, it appeared to bind to D2 receptors in striatum, although this was not further investigated using inhibition with selective compounds.

Another 5-HT1A agonist, (1-[2-(4-luorobenzoylamino)ethyl]-4(7-methoxy-naphthyl) piperazine) or S14506, has been labelled with both carbon-11 ([11C]S14506) and luorine-18 ([18F]S14506). Both radioligands were evaluated in rat and non-human primate, but did not show 5-HT1A receptor binding, as pre-treatment with WAY-100635 did not cause reduction of uptake, not even in case of P-glycoprotein inhibition using cyclospo- rine A [142]. 18 The labelled 5-HT1A agonist [ F]F15599 was evaluated in vivo in cats. It readily entered the brain and although only few radiolabelled metabolites were observed in the brain, in vivo uptake in brain regions did not match the known in vitro distribution of 5-HT1A receptors [143].

2.2.2 5-HT2A selective agonist PET ligands Derivatives of 1-(2,5-dimethoxyphenyl)-alkylamines, some of which are known as hallucinogenic, potent 5-HT2A agonists [146], have been radiolabelled with carbon-11 [102, 103]. These radioligands are presented in Table 8. 2-(4-iodo-2,5-dimethoxyphenyl)-N- 11 11 (2-[ C-OCH3]methoxybenzyl)ethanamine ([ C]Cimbi-5) was evaluated in pigs and its speciic uptake pattern in brain was in accordance with the 5-HT2A receptor distribution known from in vitro studies. Uptake in cortical areas could be reduced to cerebellar levels by pre-treatment with the 5-HT2A antagonist ketanserin. The measured BPND value of 0.46 in the cortex was comparable with the BPND value of 0.4 obtained with the 5-HT2A antagonist, [18F]altanserin. Moreover, the regional distribution of speciic [11C]Cimbi-5 binding in pig brain was comparable with measures antagonist radioligand distributions in pigs and humans [147]. More radiolabelled dimethoxyphenyl alkylamine derivatives were evaluated in pig studies, both without and with ketanserin pre-treatment to investigate binding selectivity [148]. Of all radiolabelled analogues tested, [11C]Cimbi36 showed the most promising results with the largest cortical BPND value of 0.60 and a reduction of 60% following ketanserin pre-treatment [148]. In non-human primates, the regional uptake pattern of [11C]Cimbi-36 appeared to be comparable with that of the 11 5-HT2A receptor antagonist [ C]MDL 100907 [149]. In addition, a challenge with (±)-fenluramine suggested that [11C]Cimbi-36 receptor binding may be sensitive to changes in extracellular levels of endogenous serotonin.

Overview of agonist PET ligands for 7-transmembrane receptors

2.2.3 General discussion

Some of the 5-HT1A selective agonist radioligands showed favourable characteristics in vivo. [11C]CUMI-101 appeared to be the most promising with respect to both brain uptake and selectivity. This ligand was successfully studied in humans and, in non-human primates, showed in vivo sensitivity to changes in extracellular levels of endogenous 2 serotonin. The latter inding, however, was obtained under non-physiological conditions and it was deemed unlikely that physiological changes in extracellular serotonin would be detectable in vivo in human PET studies. A major pitfall of all 5-HT1A agonist PET ligand studies to date is the use of cerebellum as reference tissue, as it is known that the cerebellum is not devoid of 5-HT1A receptors. 11 The successful 5-HT2A agonist PET ligand, [ C]Cimbi-36, was discovered by screening an extended series of radiolabelled [2,5 dimethoxyphenyl]alkylamines. [11C]Cimbi-36 appeared to be the most promising, showing good brain uptake and selective binding. The potential sensitivity to changes in extracellular levels of endogenous serotonin war- rants further evaluation in humans. The 5-HT2A agonist PET ligand could provide a more physiological relevant measure of the 5-HT2A receptor active state compared to antagonist PET ligands. Exploration of the active state of the 5-HT2A receptor in normal versus disease states would be very interesting for future explorations of the pathophysiology and therapeutics in 5-HT2A receptor related disorders. Further, investigation of the regulation of 5-HT2A receptors in the active state following treatment with drugs possessing

5-HT2A antagonist properties or following drugs of abuse with 5-HT2A agonist activity would be amongst the possibilities to prospect pharmacologically-induced changes in endogenous serotonin levels in the living brain of animals and humans.

In spite of existing selective agonists for 5-HT1B, 5-HT4, 5-HT6 and 5-HT7 receptor subtypes and their therapeutic potential for CNS disorders, no suitable PET agonist radioligands for these receptor subtypes have been reported.

2.3 MuSCARInIC ACETYLCHoLInE RECEPToRS

Acetylcholine is a core neurotransmitter that controls vital functions in both central (including cognitive and motor functions, arousal and regulation of the sleep/waking cycle) and peripheral (e.g. heart rate, intestinal motility, glandular secretion) nervous systems [150]. Cholinergic neurotransmission is processed via nicotinic and muscarinic acetylcholine (MACh) receptors. The former are ligand gated ion channels and the latter metabotropic 7-TMRs. The MACh receptor family comprises ive characterized subtypes

(M1-M5) [151]. M1, M3 and M5 receptors preferentially couple to Gq/11 G-proteins, whilst M2 and M4 receptors activate Gi/Go type proteins [152]. In the CNS M1ACh receptors are the most prevalent followed by M4ACh and M5ACh receptor subtypes. M1ACh

Chapter 2 [160] [161] [162] –values; –values; 50 n.i. n.i. n.i. = binding potential = binding potential ND : 0.10-0.59 n.i. [165] ND DVs 563-730 DVs 15.7 - 26.5 DVs n.i. non-human primate Non-human primate Ref. PET species Speciic binding Selectivity Ref. 2.4 [163] Non-human primate 13-22 DVs n.i. [163] =2.9=0.83 [163] Non-human primate [165] 131-177 DVs Non-human primate BP n.i. [163] =2.9 [163] Non-human primate 17-25 DVs n.i. [163] 7.4 7.4 7.4= 7.4 (cLogP eLogD) eLogD) (cLogP N M1 2.3 μM M2 2.4 [164] e 3 C]Cl-979 N 11 CH O -*CH3 M2 112 nM M1 119 [159] e -*CH3-CH3 [7 nM] [2 nM] [158] [158] c=4.4-*CH3 c=3.35 M2 24 nM M1 51 [159] Cb Cb Non-human primate e Non-human primate str ~ 1.8 ctx SUVr: ~ 1.5 Demo str ~2.0, ctx SUVr: ~ 1.6 Demo [158] [158] 11 3 3 *CH *F -CH3F M1 7.4 nM M2 2.2nM [159] -*CH3 M2 38 nM M1 59 c=2.41 [159]CF e Cb Human 13 6 6 6 7 4 H H H H H H 6 3 3 3 3 2 1 R -- -SC N M1/M2ACh agonist PETM1/M2ACh ligands S N N P-TZTP -SC 3 2 C]Xanomeline/TZTP’s [ C]P-TZTP -SC F]F C]CI-979 C]Xanomeline -OC C] F]FP-TZTP -SC C]FP-TZTP -SC 11 11 11 11 11 11 18 11 R [ [ [ [ [ [ [ * = labelled position, carbon-11 or luorine-18; C = calculated Log P; Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental Groningen, 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation * = labelled position, carbon-11 Cb : own or luorine-18; P; Log C = calculated IC are brackets between Ki: with selective igures inhibitors; inhibitions equilibrium constant; (pre)treatment pH; demo = selectivity by indicated at demonstrated LogD (Milameline) ButylthioTZTP Ligand R1 R2 Ki Ref. Lipophilicity Table 9. Table [ n.i.: selectivity not investigated. Speciic binding is expressed as: SUVr = standard uptake value ratio of (brain region)/cerebellum for striatum or cortex, BP for region)/cerebellum of (brain ratio uptake value = standard as: SUVr Speciic binding is expressed n.i.: selectivity not investigated. non-displaceable [35] for thalamus, striatum, occipital ctx, temporal ctx and cerebellum using a white matter region as reference; or DVs = distribution volumes (ml/ml) [35] in cortical = distribution volumes or DVs as reference; region matter using a white ctx, ctx striatum, occipital temporal and cerebellum thalamus, non-displaceable [35] for thalamus , basal ganglia and cerebellum. regions,

Overview of agonist PET ligands for 7-transmembrane receptors receptors are located primarily on postsynaptic nerve terminals in forebrain structures, including basal ganglia, hippocampus and cortical areas [153] and M4ACh receptors are autoreceptors that are present in the same regions [154]. M2ACh and M3ACh receptors are distributed throughout the body including the CNS. In the CNS, M2ACh receptors are located primarily in presynaptic nerve terminals in cortical areas and cerebellum 2 [154]. Muscarinic receptors (M1, M2 and M4ACh) in forebrain structures, including basal ganglia, hippocampus and cortical areas, play a role in motor control and cognition and dysfunction may mediate psychosis. They are assumed to be implicated in pathologies such as schizophrenia, depression, Parkinson’s disease and Alzheimer’s disease [155,156]. A number of PET ligands for the MACh receptor have been developed, but only a few were subtype selective and even less were agonists [157]. An overview of agonist PET ligands for the MACh receptor are that have been evaluated in vivo is listed in Table 9.

2.3.1 M1ACh receptor agonist PET ligands Two M1ACh receptor agonists have been labelled and evaluated as potential PET ligands, namely [11C]xanomeline and [11C]butylThio-TZTP. These radioligands showed good brain penetration with speciic retention in M1ACh receptor rich brain regions in both non-human primate and human PET studies. Unfortunately, because of high non M1ACh speciic binding due to ainity of these ligand for M4ACh receptors and σ-sites, and poor pharmacokinetics, these radioligands were considered to be not suitable for further evaluation [158]. Pre-treatment studies with selective M1 agonists/antagonists for assessing binding selectivity have not been reported.

2.3.2 M2/M1ACh receptor agonist PET ligands Butylthio-TZTP analogues showed good ainity for M2ACh receptors and also ainity for M1Ach receptors. Several derivatives were radiolabelled and evaluated in vivo in non- human primates. The luoroethyl analogue [18F]FE-TZTP displayed high uptake in vivo, which was inhibited only weakly by co-injection of unlabelled P-TZTP. The luoropropyl analogue [18F]FP-TZTP showed similar brain uptake, and inhibition of uptake was apparent upon co-injection with unlabelled P-TZTP or the muscarinic M2/M3 antagonist L-687,306. Moreover, uptake of [18F]FP-TZTP in the brain showed an in vivo distribution, which was in agreement with the known distribution of M2ACh receptor [162]. It should be noted that the latter inding is rather inconclusive, as the M2ACh receptor is distributed almost homogeneously throughout the brain. In a PET study in healthy human volunteers, [18F]FP-TZTP showed signiicantly higher volumes of distribution throughout much of the cerebellum, cortex and sub-cortical areas in older compared with younger subjects. Findings were explained tentatively by a possible lower concentration of extracellular acetylcholine in brains of some older subjects [160]. A further study in older subjects who were carriers and non-carriers of

Chapter 2

APOE-ε4, an allele that is associated with increased susceptibility to Alzheimer’s diseases, showed larger volumes of distribution of [18F]FP-TZTP in APOE-ε4 subjects. The APOE- ε4 allele is expected to have an adverse efect on the central cholinergic system with aging, possibly leading to lower acetylcholine in the synapse [166]. Carbon-11 11 11 11 labelled TZTP analogues, [ C]FP-TZTP, [ C]P-TZTP and [ C]F3P-TZTP were evaluated in non-human primates. Results indicate that the volume of distribution (DV; the volume of distribution is the ratio of the concentration of radioligand in a region of tissue to that in plasma [35]). of [11C]FP-TZTP and [11C]P-TZTP were comparable with that of [18F] 11 FP-TZTP, whilst the volume of distribution of [ C]F3P-TZTP was higher. Uptake of these carbon-11 labelled analogues of TZTP was signiicantly reduced following co-injection of unlabelled compounds [163].

2.3.3 non selective MACh receptor agonist PET ligand [11C]CI-979 (milameline) was synthesized and evaluated in a PET study in non-human primates. This study revealed extensive and rapid uptake of radioactivity in the basal ganglia and temporal occipital cortices. In cortical regions, this uptake could be de- creased in a dose dependent manner following infusion of unlabelled CI-979 [165]. No further studies with this particular radioligand were reported.

2.3.4 General discussion No truly selective M1ACh receptor agonist PET ligands have been developed, simply because fully selective agonists with nM ainity are not available yet. The agonist PET ligands developed for the M2ACh receptor, i.e. the labelled butylthio-TZTP analogues, have in common that they have limited muscarinic receptor subtype selectivity. Since they have almost equal ainity for M2ACh and M1ACh receptors, interpretation of imaging results is complicated. In addition, M2ACh receptors are homogeneously distributed throughout the brain and thus no suitable reference region is available. At present, no suitable orthosteric agonists are available for M3ACh, M4ACh or M5Ach receptors, whereas, recently positive allosteric modulators for M4ACh and M1Ach receptors have been discovered [167].

2.4 oPIoId RECEPToRS

The opioid receptor family consists of four receptor subtypes, classiied and renamed as OP1-4, (previous names between brackets below) of which three subtypes have been well characterized and OP4 remaining an orphan receptor. The OP1 (δ), OP2 (κ) and OP3

(μ) receptors are Gi/o coupled receptors that are widely distributed throughout the brain, the spinal cord, and peripheral sensory and autonomic nerves [168,169]. The OP1 recep-

Overview of agonist PET ligands for 7-transmembrane receptors tor distribution in the CNS is more restricted than those of the other opioid receptors.

The highest OP1 receptor densities are found in olfactory bulb, neocortex, caudate putamen and nucleus accumbens. Brain regions as thalamus, hypothalamus and brainstem have a more moderate to poor density of OP receptors [169]. Relative high OP receptor 1 2 densities are observed in nucleus accumbens, claustrum, dorsal endopiriform nucleus 2 and interpeduncular nucleus. OP3 receptors are distributed throughout the brain and highest densities are reported in caudate putamen, followed by neocortex, thalamus, nucleus accumbens, hippocampus and amygdala [169]. OP1 receptors play a role in analgesia, motor integration and gastro-intestinal motility. OP2 receptors are thought to be implicated in the regulation of several functions in CNS such as nociception, feeding, neuro-endocrine secretions, olfaction, cognitive function and mood driven behaviour [169]. OP3 receptors have a prominent role in nociception and also play a role in cardiovascular and respiratory functions, intestinal transit, feeding, learning and memory. In addition, OP3 receptors are the target for morphine-like drugs of abuse.

Quite remarkably, for opioid receptors and in particular OP3 receptors, numerous high ainity agonists, belonging to various structural classes, have been discovered [170]. In contrast, very few antagonists are known. Compounds with morphine-like structure usually are not selective for the OP receptor subtypes. Certain structural series, e.g. fentanyl derivatives, show greater OP3 receptor selectivity. Also more selective OP1 and

OP2 agonists have been developed. For reviews on PET ligands for the opioid system see [171] and [172]. An overview of available agonist PET ligands evaluated in vivo is presented in Table 10.

2.4.1 oP1 receptor selective agonist PET ligands

The OP1 selective agonist (+)-4-[(αR)-α-((2S,5R)-4-Allyl-2,5-di methyl-1-piperazinyl)-3- methoxybenzyl]-N,N-diethylbenzamide has been labelled with carbon-11 ([11C]SNC80) and evaluated in non-human primate PET studies [176]. [11C]SNC80 showed low brain permeability and a rather uniform distribution. It was suggested that the low brain uptake, despite a favourable lipophilicity for passive difusion (cLogP 4.93), could be due to [11C]SNC80 being a substrate for the P-glycoprotein elux transporter, although this was not studies using pre-treatment of the animals with a P-glycoprotein inhibitor.

2.4.2 oP2 receptor selective agonist PET ligands 11 The OP2 selective agonist salvinorin A has been labelled with carbon-11. [ C]salvinorin A, evaluated in non-human primates, showed extremely rapid brain uptake and fast clearance. High uptake was observed in cerebellum and notable concentrations were found in visual cortex. The regional distribution of [11C]salvinorin A was not reduced after pre-treatment with the OP3/OP2/OP1 antagonist naloxone [174], rendering this ligand not useful as an agonist PET ligand for the OP2 receptor.

Chapter 2 [5] –values. Speciic –values. 50 demo [175] OH R F]FE-PEO 18 O 1.33-3.52 O C]PEO/[ ND N : 0.8 - 6.5 n.i. [177] T 11 BP n.q. n.q. HO N 3 = binding potential non-displaceable [35] or BRcer = striatum/cerebellum. non-displaceable = striatum/cerebellum. = binding potential [35] or BRcer CH ND 11 C]Carfentanil [ Non-human primate / human Non-human primate O O N 11 O Ref. PET species Speciic binding Selectivity Ref. 3 CH 11 O O O =3,28 Cb Non-human primate substrate Pgp n.i. [176] 7.4 (cLogP eLogD) (cLogP O H H O [177] c=3.63 Cb rats V [5] c=2.26 Cb rats 2.5, FCtx 2.5 Str SUVr: demo [5] O O C]Salvinorin A [ 11 0.4 nM 0.12 nM 0.18 nM 1.6 nM O 2 3 1 2/3 [1.31 nM] [176] e OP OP 4.3-10 nM [173] c=1.59 Cb Non-human primate Non speciic n.i. [174] 0.051 nM [175] e= 3.89 [170] Humans (6) OP OP 3 CH 11 O 1 1/2/3 1/2/3 2 3 = brainregion/cerebellum, for striatum (str) or frontal cortex (FCtx), BP (str) or frontal striatum for = brainregion/cerebellum, OP

3 N N H Opioid agonist PET ligands. O N C]SCN80 [ 11 [ = Region-of-interest distribution volume based on 1 tissue compartment distribution= Region-of-interest volume model. C]SCN80 - OP C]PEO -*CH F]FE-PEO -C2H4*F OP C]Salvinorin A - OP C]Carfentanil - OP T 11 11 18 11 11 V Table 10. Table Ligand R 7-TMR Ki Ref. Lipophilicity [ binding is expressed as: SUVr binding is expressed [ [ C = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); e = experimental LogD at indicated pH; demo = selectivity indicated at the Netherlands); LogD e = experimental Groningen, 13.0 (PerkinElmer, Ultra using ChemBioDraw calculation Cb : own P;, Log C = calculated IC are brackets between Ki: igures inhibitions equilibrium constant; n.i.: selectivity with selective inhibitors not investigated. (pre)treatment by demonstrated [ [

Overview of agonist PET ligands for 7-transmembrane receptors

2.4.3 oP3 receptor selective agonist PET ligands

The very irst agonist that was radiolabelled for PET imaging of a 7-TMR is the OP3 selective agonist carfentanil. Carfentanil is an extremely potent opioid agonist and narcotic analgesic (more than 7000 times more potent than morphine) and binds predominantly to OP receptors. Radiolabelled carfentanil ([11C]carfentanil) was evaluated in non-human 2 3 primates and humans. Its brain distribution showed high concentrations of radioactivity in basal ganglia and thalamus, intermediate concentrations in frontal and parietal cerebral cortex and low concentrations in cerebellum and occipital cortex. The in vivo 11 distribution of [ C]carfentanil corresponds with the known regional distribution of OP3 receptors measured in vitro. The heterogeneous brain uptake could be blocked by pre- treatment with naloxone [178]. In vivo sensitivity for endogenous ligand competition was demonstrated in a study in healthy human volunteers, who received subcutaneous capsaicin administration to elicit pain. A pain related decrease in brain OP3 binding was observed in contralateral thalamus consistent with possible competitive binding between [11C]carfentanil and acutely released endogenous opioid peptides [179]. This latter inding is rather remarkable in view of carfentanil’s extremely high ainity for and slow dissociation rate from OP3 receptors.

2.4.4 oP2/3 receptor agonist PET ligands (20R)-4,5-a-Epoxy-17-methyl-3-hydroxy-6-[11C]methoxy-a,17-dimethyl-a-(2- 11 phenylethyl)-6,14-ethenomorphinan-7-methanol ([ C]PEO) is a labelled mixed OP2 and

OP3 agonist that was evaluated in rats using PET studies in rats. It accumulated in brain regions known to be rich in OP3 receptors. Blocking studies with naloxone demonstrated that binding of [11C]PEO could be reduced in striatum, thalamus and frontal cortex by more than 95%, demonstrating probable selectivity for the OP3 receptor in rat brain [5,180]. The luorine-18 analogue of PEO, (20R)-4,5-a-epoxy-6-(2-18F-luoroethoxy)-3- hydroxy-a,17-dimethyl-a-(2-phenyleth-1-yl)-6,14-ethenomorphinan-7-methanol ([18F]- FE-PEO) was recently evaluated in rodents [177]. [18F]-FE-PEO showed high distribution volume values in thalamus, parts of the striatum and frontal cortex, with lowest uptake in cerebellum. It should be noted that no pre-treatment studies were performed with

OP2 or OP/3 selective ligands. In addition, more studies are necessary to determine suit- 18 ability and selectivity of [ F]-FE-PEO as an agonist PET ligand for the OP2/3 receptor.

2.4.5 General discussion To date, agonist PET ligands for opioid receptors have shown limited opioid receptor subtype selectivity. Only [11C]Carfentanil was successfully used in humans for quantitati- ication of the OP3 receptor. The success of this very irst agonist PET ligand probably has hampered investigations of other OP3 agonist PET ligands, in spite of the availability of numerous OP3 agonists with varying properties in terms of receptor ainity, dissociation

Chapter 2 rate and duration of pharmacological action [170]. Comparative PET studies between opioid agonist and antagonist PET ligands have not been reported, as in the early days, PET radioligands were evaluated based on their therapeutic application rather than on their antagonist or agonist characteristics. Further PET studies with opioid agonist radioligands in humans to investigate opioid receptor activated status in conditions of chronic pain or opiate dependence are warranted.

2.5 CAnnABInoId RECEPToRS

The cannabinoid (CB) receptor family comprises the CB1 and CB2 7-TMRs, which both are coupled to Gi/o proteins that can inhibit adenylyl cyclase activity or induce opening + of K -channels and activate mitogen activated protein kinase [181]. The CB1 subtype is distributed throughout the brain and high densities are found in cerebral cortex, hippocampus, basal ganglia, and cerebellum. Lower receptor densities are observed in hypothalamus and spinal cord. The CB2 receptor subtype is known to have sparse occurrence in the brain, but it can be present in both central and peripheral immune cells as a result of upregulation after neuroinlammation.

The CB1 receptor is involved in appetite regulation, emesis, muscle spasms, craving, pain, anxiety and depression, whilst the CB2 subtype is thought to be implicated in im-

Table 11. Cannabinoid Agonist PET ligand.

OH H

H 18F O [18F](-)5’18F-∆8 THC

Ligand Ki Ref. Lipophilicity Ref. PET Speciic Selectivity Ref. (cLogP eLogD) species binding [18F](-)5’18F-∆8 THC 47.6 nM [181] c=6.51 Cb Non- Non n.i. [183] human speciic primate c = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); n.i.: selectivity not investigated. Ki: inhibitions equilibrium constant; igures between brackets are IC50 –values.

Overview of agonist PET ligands for 7-transmembrane receptors

mune system regulation and pain [181]. The prominent presynaptic localization of CB1 receptors and their ability to inhibit Ca2+-channels and activate K+-channels suggests that they may modulate neurotransmission and afect neuronal excitability [182]. The CB1 agonist fuorine-18 labelled therahydrocannabinol, [18F](-)5’18F-∆8THC, has been evaluated in mice using ex vivo biodistribution studies and in non-human primates 2 using PET (Table 11). It was shown that its uptake was similar in striatum, thalamus and cerebellum and that it was cleared rapidly from the brain. It was concluded that this uptake pattern appeared to be nonspeciic [183]. No other CB1 selective agonist PET ligands have been reported thus far.

2.6 MELATonIn RECEPToRS

The melatonin (MT, 5-methoxy-N-acetyltryptamine) receptor family consists of the MT1 and MT2 7-TMRs. Both subtypes are distributed throughout the brain. In human brain they are located in cerebellum, cerebral cortex, thalamus, hippocampus, and suprachi- asmatic nucleus. Melatonin plays a key role in the regulation of the circadian system, and it circulates nightly to provide timing cues for melatonin receptors to inluence daily and seasonal rhythms of the sleep/wake cycle and behaviour [184].

Table 12. Melatonin agonist PET ligand.

O HN 11 CH3 O

N H [11C]Melatonin

Ligand Ki Ref. Lipophilicity Ref. PET Speciic Selectivity Ref. (cLogP eLogD) species binding [11C]Melatonin 0.59 nM [187] c=1.03 Cb Human Non n.i. [185] speciic c = calculated Log P;, Cb : own calculation using ChemBioDraw Ultra 13.0 (PerkinElmer, Groningen, the Netherlands); n.i.: selectivity not investigated. Ki: inhibitions equilibrium constant.

Chapter 2

[11C]melatonin was assessed using PET in a sigle human volunteer. Maximum uptake of radioactivity in the brain was observed at 8.5 min post intravenous injection, but no speciic binding could be observed [185] (Table 12). Therefore, it was considered not suitable for clinical imaging. An alternative compounds for imaging of MT receptors is ramelteon. The MT selective ligand ramelteon is marketed for sleep induction under the name Roserem™ and its radiosynthesis has been reported [186].

2.7 ConCLudInG REMARKS

Agonist PET ligands for dopamine D2/3, 5-HT1A, 5-HT2A and OP3 receptors have successfully been developed and used. Studies, in particular those using D2/3 agonist PET ligands, clearly have shown that agonist PET ligands can provide diferential information with regard to the receptor population being imaged (i.e. active state rather than total pool). In addition, sensitivity to extracellular levels of endogenous neurotransmitter was shown. Further studies of receptor regulation in disease states and following drug treatment could provide important insight in the role of various receptor systems in the course of a disease and guide the development of novel drugs for therapeutic intervention. Given the current view that agonist PET ligands can be applied to image the active state of the receptor, it is remarkable that so few 7-TMR agonist PET ligands have been explored. D2/3, 5-HT1A and OP3 receptors belong to the Gi/o protein coupled 7-TMRs, whilst

5-HT2A receptors are Gq/11 protein coupled. For the former type of 7-TMRs more high ainity agonists are available, indicating that this type of receptor is more suitable for agonist development, yet agonists for other Gi/o coupled receptors that are available, were not explored as potential PET ligands. A gap in the development of agonist PET ligands lies in the ield of Gs coupled, adenylate cyclase stimulating 7-TMRs, to which certain 5-HT receptor subtypes and also adrenergic receptor subtypes belong. Nevertheless, for several of these receptors, high ainity agonists, with potential for central therapeutic application, have been reported. Concerning cholinergic 7-TMRs, the scarcity of agonist PET ligands is probably related to the limited availability of compounds with suitable properties. Other classes of 7-TMRs for major neurotransmitter systems that are not discussed in this review are glutamatergic and GABAergic metabotropic receptors. For the former, drug development has focussed on allosteric receptor modulators as the orthosteric ligands are not subtype selective. Allosteric receptor modulators do not directly act on the 7-TMR agonist binding site, therfore we choose not to include them in the present review. For the GABAergic metabotropic receptors, only few ligands are available, and no agonist radioligands were reported yet.

Overview of agonist PET ligands for 7-transmembrane receptors

From this review, criteria can be corroborated that are required for developing successful agonist PET ligands are similar to those for antagonist PET ligands. Criteria include: (1) a ligand must have suitable physicochemical properties, in particular moderate lipophylicity to permit passive difusion across the blood brain barrier and avoid high non-speciic binding, (2) a ligand must have at least nM receptor binding ainity to fulil 2 the tracer principle, (3) pharmacokinetic properties of a ligand must be suitable for characterisation within the time frame of a PET scan, (4) the ligand should not be a substrate for elux transporters like P-glycoprotein pump transporters, and (5) metabolism of the ligand should be limited and resulting radiolabelled metabolites should not cross the blood brain barrier. Although it is diicult to fulil all criteria at the same time, examples presented in this review indicate that suitable ligands do exist. The main limitation of antagonist PET ligands is their inability to distinguish the inactivated (low-ainity) from the activated (high-ainity) states of 7-TM receptors. Agonist PET ligands are able to distinguish between the activated and inactivated receptor, as their binding is preferentially to the activated receptor. Next to this, only agonist PET ligands are thought to be fully sensitive to the efects of endogenous ligand. These speciic advantages of agonist over antagonist PET ligands, should be a stimulus for further developments.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

Hans J.C. Buiter, Josée E. Leysen, Robert C. Schuit, Abraham Fisher, Adriaan A. Lammertsma and Albert D. Windhorst

Published in: Journal of Labelled Compounds and Radiopharmaceuticals 55 (2012), pp. 264 - 273

Chapter 3

ABSTRACT

Purpose: M1 muscarinic acetylcholine receptor (M1ACh-R) agonists are potential therapeutics for cognitive disorders. The aim of this study was to develop an orthosteric agonist PET ligand for the M1ACh-R, which would provide a tool to explore the active G-protein coupled receptor. To that extend, the radiosynthesis, receptor binding in vitro and brain uptake of [11C]AF150(S), a selective M1ACh-R agonist, were investigated.

Procedures: [11C]AF150(S) was synthesized, its binding properties were investigated using autoradiography on rat brain sections and brain uptake was determined in biodistribution studies ex vivo in rats. Metabolites in brain and blood were measured using HPLC.

Results: Yield of the [11C]AF150(S) radiosynthesis was 69±9% with radiochemical purity

>99%. The logDoct.7.4 was 0.05. Autoradiography showed binding in M1ACh-R rich brain areas and selective inhibition by muscarinic agents, in conditions promoting agonist binding. Biodistribution studies revealed rapid and high brain uptake, to levels exceeding 5 times the blood level. In M1ACh-R rich areas speciic uptake versus cerebellum was apparent, remaining constant over 60 min, in spite of rapid decline of radioactivity from the brain and rapid peripheral metabolism. Brain uptake of [11C]AF150(S) may involve facilitated transport.

Conclusion: [11C]AF150(S) was successfully synthesized, showed M1ACh-R agonist binding properties and high brain uptake. It provides an interesting lead for the development of an M1ACh-R agonist PET ligand, meant to explore the active receptor state.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

3.1 InTRoduCTIon

PET ligands for labelling receptors are an increasingly applied tool in translational medicine, allowing studying receptors in vivo in a non-invasive way, including in humans. They are applied to study receptor distribution and levels in normal and disease state and to investigate occupancy of receptors by drugs in relation to drug action. There is also great interest in applying receptor PET ligands to explore receptor binding by the endogenous natural agonist. Most investigated PET ligands target G-protein coupled 3 receptors and the majority of successful ligands are receptor antagonists. However, in case of GPCRs, the receptor protein can occur in various states [14,188,189]. In its active form the receptor is coupled to a G-protein, by which it shows high ainity for the endogenous ligand and for agonistic agents. The uncoupled receptor binds endogenous and exogenous agonists with substantial lower ainity. Antagonists however bind with equally high ainity to the G-protein coupled and uncoupled receptor and hence do not distinguish between the active and inactive receptor. GPCRs occur on the plasma membrane of cells, where they exert their function, but they also occur on intracellular organelles, as a consequence of receptor internalization following persistent activation or due to de-novo receptor synthesis. Lipophilic antagonist PET ligand, usually, can rapidly enter cells and will neither distinguish between externalized and internalized receptors. Recent work with dopamine D2 receptor PET ligands [68,73], have shown that agonist PET ligands label a lower number of D2 receptors than antagonists; the binding of the former is probably restricted to the active G-protein coupled receptor. It also appeared that the receptor labelling by the agonist PET ligands is more sensitive to inhibition by non-labelled agonists and by endogenous dopamine [70,73,74]. There is an increasing interest in the development of drugs that act as receptor agonists, for example to ind new therapeutics for cognitive disorders. One of the many challenges, is the assessment of the appropriate dose that causes suicient receptor stimulation, while avoiding overstimulation and receptor down regulation. Agonist PET ligands, much more so than antagonist PET ligands, can provide a useful tool to explore the functional receptor. We have been particularly interested in the M1ACh-R, the one of ive G-protein coupled acetylcholine receptor subtypes, that predominantly occurs in the brain [155].

The M1ACh-R is an excitatory receptor that couples to the Gαq/11 protein. It is located primarily in postsynaptic terminals in forebrain structures, including basal ganglia, cortical areas and hippocampus [153]. M1ACh-R play a role in motor control and cognition and are thought to be implicated in pathologies such as schizophrenia, Parkinson’s disease and Alzheimer’s disease. Studies with M1 agonists in laboratory animals have revealed improvement in cognitive performance and early clinical studies in patients with

Chapter 3

Alzheimer’s disease and schizophrenia with muscarinic agonists pointed to possible amelioration of cognitive functions and attenuation of psychotic behaviour [155,190]. MACh-R antagonists that have been radiolabelled with carbon-11 are e.g. [11C]scopolamine, [11C]NMPB, [N-11C-methyl]-benztropine, [11C]QNB and [11C]3NMPB [191-195]. In addition to labelling the total MACh-R receptor pool, these antagonists are non-selective with regard to MACh-R subtypes. The M1ACh-R agonists that have been radiolabelled and evaluated, such as [11C]xanomeline, [11C]butylThio-TZTP and [18F]luoroxanomeline, are not fully selective for MACh-R subtypes and also bind to σ-sites [158,196].

In our search for suitable selective M1ACh-R agonists a prioritization of agents reported in the literature was made, based on pharmacological properties and performed clinical research. Cevimeline (AF102B), a rigid analogue of ACh, was selected as an interesting lead; as it is marketed for human medicinal use and is likely to act as an orthosteric agonist, however, radiolabelling AF102B is not straightforward. AF150(S) (2,8-Dimethyl- 1-thia-3,8-diaza-spiro[4.5]dec-2-ene; Figure 1), together with AF267B, belongs to a series of novel selective M1ACh-R agonists, related to AF102B [197]. Their pharmacological and potential therapeutic properties have been investigated extensively in animal models of neurological diseases, where improvement in cognitive functions, reduction of amyloid plaques and decrease in TAU phosphorylation were demonstrated [198-202].

AF150(S) has moderate ainity (Kd,H 200 nM) for M1ACh-R in rat cerebral cortex and shows functional selectivity for M1ACh-R in transfected cells [203]. Its chemical structure is amenable to rapid introduction of carbon-11. The present study reports the radiosynthesis of [11C]AF150(S), characterisation of its binding to M1ACh-R in vitro and investigation of its biodistribution and metabolism ex vivo in rats.

11 11 Figure 1. Radiosynthesis of [ C]AF150(S): [ C]CH3I, CH3CN, 5 min, 60 °C.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

3.2 MATERIAL And METHodS

3.2.1 Material AF150(S) and AF400 (Figure 1) were synthesized at the Israel Institute for Biological Re- 11 search (Ness-Ziona, Israel). [ C]CO2 was provided by BV Cyclotron VU (Amsterdam, The Netherlands). Xanomeline was obtained from Metina AB (Lund, Sweden). Pirenzepine, atropine and haloperidol were purchased from Sigma-Aldrich (Zwijndrecht, the Neth- erlands). Chemicals were from Merck KGaA (Darmstadt, Germany) or Sigma-Aldrich. 3 They were of analytical grade and were used without further puriication. Disposable materials were obtained from VWR International B.V. (Amsterdam, the Netherlands). HPLC solvents (gradient grade) were purchased from Mallinckrodt-Baker B.V. (Deventer, the Netherlands). Devices used for automated radiosynthesis were home made [204]. The semi-preparative HPLC system consisted of a Jasco (Ishikawa-cho, Japan) PU-1587 HPLC pump, a six-way VICI injector (VA EPC6W; VICI AG International, Schenkon, Swit- zerland) with a 5 mL loop, a Phenomenex (Torrance, CA, USA) Gemini 5µm C18 100 Å 250x10 mm HPLC column, a Jasco UV1575 UV detector and a custom made radioactivity detector. Radioactivity was measured using a Veenstra (Joure, the Netherlands) VDC-405 dose calibrator. The analytical HPLC system consisted of a Jasco PU-1580 HPLC pump, a Rheodyne 7724I injector (IDEX Health & Science, Wertheim-Mondfeld, Germany) with a 20 μL loop, a Phenomenex Luna 5μm C5 100 Å 150x4.6 mm HPLC column, a Jasco UV-2075 Plus UV detector and a NaI radioactivity detector (Raytest, Straubenhardt, Ger- many). Male Wistar rats (210-310 g) were obtained from Harlan Netherlands B.V. (Horst, the Netherlands). All animal experiments were in compliance with Dutch law and approved by the University Animal Ethics Committee. For carbon-11 radioactivity measurements a Wallac 1282 Compugamma CS (LKB Wal- lac, Turku, Finland) was used. Brain sections were obtained using a CM3050 S cryostat from Leica Microsystems Nederland B.V. (Rijswijk, the Netherlands). Phosphor storage screens were read with a StormTM 860 Phosphorimager (GE Healthcare Europe GmbH, Diegem, Belgium). For the semi-preparative HPLC system Jasco ChromPass software (version 1.8.6.1) was used and for the analytical HPLC system Raytest GINA Star software (version 2.14). GraphPad Prism (version 4.00) was from GraphPad Software (San Diego, CA, USA), Im- ageQuant (version 5.2) from GE Healthcare Europe GmbH. Xanomeline and haloperidol were dissolved in ethanol and diluted with assay bufer; the inal concentration of ethanol was 0.01%. Pirenzepine and atropine were dissolved and diluted in assay bufer. [11C]AF150(S) was obtained in a solution of saline containing

7.09 mM NaH2PO4 and 20% ethanol. This solution was diluted with assay bufer for in vitro experiments and with saline for in vivo application.

Chapter 3

3.2.2 Automated radiosynthesis and HPLC

11 Methyl iodide, labelled with carbon-11, was prepared from [ C]CO2 as described previ- 11 11 ously [205-209]. [ C]AF150(S) was synthesized by methylation of AF400 with [ C]CH3I in a solution of 1.0 mg AF400 (as free base) in 500 μL dry acetonitrile. The reaction mixture was heated for 5 min at 60 °C and subsequently quenched with 2 mL water containing 0.2% diisopropylamine. The complete reaction mixture was transferred to a remotely controlled HPLC injection system and subjected to semi-preparative HPLC puriication with acetonitrile/water/diisopropylamine 20/80/0.2 (v/v/v) as the eluent at a low rate of 4 mL/min. The fraction containing [11C]AF150(S) was collected and diluted with 120 mL sterile water for injection. The product was isolated by solid-phase extraction using Waters tC18 Plus Sep-Pak® SPE cartridge, pre-conditioned by rinsing with 10 mL of ethanol and 10 mL of sterile water. Following passage of the [11C]AF150(S) containing solution, the cartridge was washed with 20 mL sterile water and the product was eluted with 1 mL ethanol followed by 5 mL saline containing 7.09 mM NaH2PO4. The eluate was iltered through a sterile Millex GV 0.22 μm membrane ilter using helium overpressure. The purity of the product was analyzed using the analytical HPLC system eluted with acetonitrile/water/diisopropylamine 30/70/0.2 (v/v/v) at a low rate of 1 mL/min. The speciic activity was determined by HPLC using calibration curves based on the ultraviolet signal.

11 3.2.3 determination Logdoct, 7.4 of [ C]AF150(S) The distribution of [11C]AF150(S) between 1-octanol and 0.2 M phosphate bufer (pH = 7.4) was measured in triplicate at room temperature as described previously [210], with slight modiication. Briely, 1 mL of a 20 MBq/mL solution of [11C]AF150(S) in 0.2 M phosphate bufer (pH = 7.4) was vigorously mixed with 1 mL 1-octanol for 1 min at room temperature using a vortex. After a settling period of 30 min, ive aliquots of 100 μL were taken from both layers, carefully avoiding cross contamination between the phases. Five aliquots of 100 μL of the 20 MBq/mL solution of [11C]AF150(S) in 0.2 M phosphate bufer (pH = 7.4) were taken as reference for determining recovery. All aliquots were counted 10 for radioactivity. The LogDoct,7.4 value was calculated according to LogDoct,7.4 = Log(Aoct/

Abufer), where Abufer and Aoct represent average radioactivity concentrations of the 5 bufer and 5 1-octanol samples, respectively.

3.2.4 Rat brain sections Rat brain sections were obtained by adapting a method described previously [211]. Briely, rats were decapitated and brains were removed quickly, snap frozen at -35 °C in isopentane and stored at -20 °C (at least for 24 h). Coronal sections of 20 μm thickness were thaw-mounted on Menzel Superfrost® plus slides (Gerhard Menzel, Glasbearbei- tungswerk GmbH & Co. KG, Braunschweig, Germany). Sections were collected from the

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S) beginning of the striatum until the end of the medulla oblongata, corresponding to plates 13 to 130 from the Paxinos and Watson atlas of the rat brain [212].

3.2.5 [11C]AF150(S) autoradiography Assay conditions for incubation of [11C]AF150(S) on rat brain sections were optimised by varying radioligand concentration (0.02 to 0.5 μM, corresponding to 1 to 25 MBq/ mL), incubation time (5 to 30 min), presence of MnCl (1 mM), and washing procedure. 2 Optimal assay conditions were as follows. Brain sections were thawed and incubated 3 in a container with freshly prepared assay bufer (50 mM Tris-HCl, 1 mM MnCl2, pH 7.4) for 15 min at room temperature. Slides were removed from the container, liquid was shaken of, and brain sections were covered individually with 900 µL assay bufer and pre-incubated for 30 min at room temperature. The solution was removed and replaced by 900 μL assay bufer containing 0.4 μM [11C]AF150(S) and incubated for 10 min at room temperature, to determine baseline binding. Reactions were terminated by shaking of the excess of liquid immediately followed by dipping in ice-cold assay bufer and two times in ice-cold deionised water. Slides were dried gently under a stream of cold air and exposed to a phosphor storage screen. Non-speciic/non-selective binding was deined on adjacent sections with assay bufer containing 0.4 μM [11C]AF150(S) and 100 nM atropine or 100 nM haloperidol. Baseline binding and non-speciic/non-selective binding were determined in quadruplicate. Competition binding experiments (in triplicate) were performed by adding compounds (concentration 0.1 to 1000 nM) to 0.4 μM [11C]AF150(S) in assay bufer. Image data were analysed using ImageQuant software, after identiication of anatomical regions using the rat brain atlas by Paxinos and Watson [212]. Brain sections containing striatum and hippocampus were used to determine [11C]AF150(S) binding to M1ACh-R, and brain sections containing cerebellum and medulla oblongata to determine [11C] AF150(S) non-speciic/non-selective binding, respectively. Statistical analyses were performed with the two-tailed unpaired Student’s t-test using GraphPad Prism software, and diferences were considered signiicant at a level of p < 0.05.

3.2.6 Biodistribution The biodistribution of [11C]AF150(S) was measured in rats using a slight modiication of the method described by Airaksinen et al. [213]. Under 2% isolurane anaesthesia, 5 groups of 4 rats received an intravenous injection (tail vein) of [11C]AF150(S) (54 ± 4 MBq at the start of the experiment). Under anaesthesia, rats were sacriiced by cervical dislocation at 5, 10, 15, 30 and 60 min post injection. Whole blood was collected by cardiac puncture, and heart, lungs, liver, kidneys and brain were excised. The following brain regions were dissected: olfactory bulb, hippocampus, striatum, frontal and posterior cortex (including occipital cortex), thalamic region, cerebellum, and the rest of the

Chapter 3 brain. All organs and brain regions were weighed and counted for radioactivity using an automatic gamma counter. The %ID/g for every organ or brain region was calculated as percentage of total radioactivity injected divided by the weight of the tissue.

3.2.7 In vivo metabolism In vivo stability of [11C]AF150(S) in plasma and brain was determined in rats under 2% isolurane anaesthesia. Two groups of 5 rats received intravenous injections of [11C] AF150(S) via the tail vein. At 10 and 45 min post injection, whole blood was collected by cardiac puncture and the brain was excised. Whole blood was centrifuged for 5 min at 4000 rpm, and a 1 mL plasma aliquot was taken, acidiied using 10 μL 5 M HCl and diluted with 2 mL of water. The excised brain was washed in ice-cold saline and subsequently one hemisphere was homogenized in 12 mL ice-cold saline and centrifuged for 5 min at 4000 rpm. Plasma and brain samples were passed over a Waters tC18 Plus ® Sep-Pak SPE cartridge (Waters Chromatography B.V, Etten-Leur, the Netherlands) (1 g, preconditioned with 10 mL ethanol followed by 10 mL water) washed with 5 mL water and eluted with 2 mL acidic methanol. All fractions, including the Sep-Pak® SPE cartridge, were counted for radioactivity using an automatic gamma counter. The methanol fractions were analysed using the semi-preparative HPLC method (see above). The identity of [11C]AF150(S) was conirmed by co-injection with reference AF150(S). Percentage of metabolite and parent compound fractions were obtained by dividing the corresponding HPLC radioactivity peaks by the total amount of radioactivity within all fractions and Sep-Pak® SPE cartridge.

3.3 RESuLTS

3.3.1 Radio- and automated synthesis

11 11 [ C]AF150(S) was obtained by methylation of AF400 using [ C]CH3I (Figure 1). First radiosynthesis attempts with DMF as solvent (150 μL) and tetrabutylammonium hydrox- ide as base at either room temperature or 50 °C, and reaction for 2 min, resulted in 1.6 11 and 3.3% incorporation of [ C]CH3, respectively. Increasing the reaction temperature to 80 °C resulted in the formation of side products rather than the desired product. Omitting the base from the reaction mixture and heating at 110 °C for 5 min resulted in 11 99% incorporation of [ C]CH3. Since DMF disturbed HPLC puriication, this solvent was replaced by acetonitrile. Using acetonitrile and heating at 60°C for 5 min also resulted in 11 quantitative incorporation of [ C]CH3.

11 For the automated radiosynthesis, trapping eiciency of [ C]CH3I in 150 μL acetoni- 11 trile at room temperature appeared to be insuicient, as only 60 – 70 % of [ C]CH3I was

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

Figure 2. Representative HPLC chromatograms of semi-preparative HPLC of reaction mixture (a UV chromatogram; b radiochromatogram; [11C]AF150(S) elutes at approximately 12 min) and analytical HPLC of inal radiotracer product (c UV chromatogram; d radiochromatogram), spiked with non radioactive AF150(S) (e UV chromatogram; [11C]AF150(S) elutes at approximately 4.7 min).

Chapter 3 trapped. Increasing the amount of acetonitrile to 500 μL and pre-cooling the reaction 11 11 vessel to -25 °C before [ C]CH3I distillation yielded the optimal trapping eiciency of [ C] 11 CH3I of 92 ± 6 %. The formation of [ C]AF150(S) was conirmed by analytical HPLC using co-injection of reference AF150(S). Representative chromatograms of both semi-preparative and analytical HPLC puriication are shown in Figure 2. Formulated [11C]AF150(S) was obtained with a decay corrected overall yield of 69 ± 12 % (n =16), corresponding to 3200 – 7000 MBq at the end of the synthesis time of 50 ± 5 min. Radiochemical purity was >99% and speciic activity at the end of synthesis was 23 – 118 GBq/μmol.

11 3.3.2 Logdoct 7.4 of [ C]AF150(S) 11 The measured LogDoct,7.4 value of [ C]AF150(S) was 0.050 ± 0.004 (n = 3). The recovery of radioactivity in the summed 1-octanol and phosphate bufer fractions was 98.9 ± 1.6%.

3.3.3 [11C]AF150(S) autoradiography [11C]AF150(S) binding was investigated on rat brain sections and analyzed using autoradiography. Variation of radioligand concentration (0.02 – 0.5 μM [11C]AF150(S) corresponding to 1 – 25 MBq/mL) revealed a concentration dependent increase in labelling.

At concentrations below the Kd,H of 200 nM labelling remained weak, at concentrations exceeding the Kd,H-value labelling became more pronounced. Representative autoradiograms of coronal brain sections at the level of the hippocampus are shown in Figure 3;

Figure 3. Autoradiograms of [11C]AF150(S) after incubation at various concentrations on coronal rat brain sections at the level of the hippocampus.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S) at the higher concentrations of [11C]AF150(S), clear labelling of cortex and hippocampus is apparent.

Using 0.4 µM [11C]AF150(S), efects of the presence of Mn2+ ions on agonist binding to M1ACh-R were assessed, quantitative data are presented in Table 1. In the absence of Mn2+, labelling was observed in various brain areas, but 100 nM atropine, a muscarinic antagonist [214], only caused signiicant inhibition of [11C]AF150(S) binding in the frontal cortex. In the presence of 1 mM Mn2+, total labelling with [11C]AF150(S), measured as 3 pixel intensity, was reduced but clear inhibition by atropine was observed in brain areas known to be enriched in M1ACh-R. Signiicant inhibition by atropine, was observed in striatum (50 %), hippocampus (26 %) and cortical areas (30 – 39 %). No signiicant inhibition was observed in M1ACh-R poor areas, i.e. medulla oblongata and cerebellum. 11 Possible non-selective binding of 0.4 μM [ C]AF150(S) to σ1 binding sites was examined using 100 nM haloperidol [215,216], a D2 antagonist that also has high-ainity for σ1 binding sites. Haloperidol did not cause any statistically signiicant decreases in [11C] AF150(S) binding.

Table 1. Quantitative data a of binding of [11C]AF150(S) (0.4 μM, room temperature) to rat brain sections at various levels in the absence and presence of Mn2+ ions and in the absence and presence of atropine

(inhibition of MACh-R labelling) or haloperidol (inhibition of σ1 binding site and dopamine D2 receptor labelling). Brain Region Total binding Atropine Haloperidol Total binding Atropine Haloperidol (100 nM) (100 nM) (100 nM) (100 nM) no Mn2+ no Mn2+ no Mn2+ 1mM Mn2+ 1mM Mn2+ 1mM Mn2+ Striatum 583.9±296.7 520.9±247.1 554.0±234.3 406.3±197.2 201.1±125.1*** 401.9±145.4 Hippocampus 231.7±66.2 248.2±46.5 244.4±61.5 173.1±42.1 128.8±43.5*** 167.8±54.6 Frontal cortex 343.7±58.4 273.7±40.7** 356.2±54.8 238.3±39.7 145.3±43.0*** 235.5±35.5 Posterior 235.0±73.8 221.7±60.0 235.5±71.8 160.2±31.7 111.2±32.2*** 142.4±37.8 cortex Cerebellum 343.7±76.1 317.5±32.6 304.21±41.1 213.3±53.2 185.4±36.2 174.3±59.6 Medulla 446.1±93.2 410.8±75.9 395.1±80.3 242.5±43.6 218.7±60.8 210.7±31.3 oblongata a Binding of [11C]AF150(S) relects average intensity of all pixels in a ROI. Data are presented as mean ± SD. Students t-test unpaired (n = 4): signiicance of diference between total binding and binding in the presence of atropine or haloperidol **p<0.01 ***p<0.001

Chapter 3

Figure 4. Inhibition by xanomeline of 0.4 μM [11C]AF150(S) binding to coronal rat brain sections in Tris bufer with 1mM Mn2+ at room temperature. Baseline binding is set to 100%. Inhibition of binding by various concentrations of xanomeline is expressed as % of baseline binding and represented as mean ± SD (n = 3). Two-tailed unpaired Student’s t-test; *p<0.05, **p<0.01, ***p<0.001, compared with percentage baseline binding.

Competition binding assays on rat brain sections, using 0.4 μM [11C]AF150(S) and assay bufer containing 1 mM Mn2+, were performed with the M4/M1ACh-R agonist xanomeline [217,218] and the M1ACh-R selective antagonist pirenzepine [219]. Analysis was performed at the level of M1ACh-R rich regions. Both compounds showed signiicant inhibition of 0.4 μM [11C]AF150(S) binding in striatum, hippocampus, frontal and posterior cortex, at almost all concentrations investigated. The degree of inhibition was not concentration dependent in the investigated, relatively high concentration range of the compounds. Data are shown in Figures 4 and 5.

3.3.4 Biodistribution The biodistribution of radioactivity, following IV injection of [11C]AF150(S) was determined ex vivo in brain regions, peripheral organs and blood at 5, 10, 15, 30 and 60 min. Findings are presented in Table 2.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

Figure 5. Inhibition by pirenzepine of 0.4 μM [11C]AF150(S) binding to coronal rat brain sections in Tris bufer with 1mM Mn2+ at room temperature. Baseline binding is set to 100%. Inhibition of binding by various concentrations of pirenzepine is expressed as % of baseline binding and represented as mean ± SD (n = 3). Two-tailed unpaired Student’s t-test; *p<0.05, **p<0.01, ***p<0.001, compared with percentage baseline binding.

Five min post IV injection of [11C]AF150(S), high uptake of radioactivity in brain areas was observed. Radioactivity levels in olfactory bulb, hippocampus, striatum and cortical areas were considerably higher (30 to 80%) than in thalamus, cerebellum and the rest of the brain. In the former brain areas, levels were in the range of liver and lung levels, and 3 times higher than in the heart. Remarkably, the concentration of radioactivity in these brain areas was up to 5 times higher than in blood. The radioactivity concentration in blood remained constant over 30 min and only slightly declined at 60 min. Radioactivity levels in brain areas were declined by 20 to 40% at 10 min, by over 60% at 30 min and by over 80% at 60 min. At peak value, the brain radioactivity concentration amounted to 10.6 nM (calculated using molecular weight of the parent compound). The ratio of brain region over cerebellum, calculated at ive time points up to 60 min remained approximately constant and ranged from 1.33 to 2.02 in olfactory bulb, from 1.51 to 1.77 in hippocampus, from 1.20 to 1.78 in frontal cortex, from 1.25 to 1.51 in posterior cortex, and from 1.27 to 1.39 in striatum. In the thalamic region and the rest of the brain, values did not exceed the level of the cerebellum.

Chapter 3

Table 2. Biodistribution of [11C]AF150(S)a in rats at various time points following IV injection. %ID/g (mean ± SD, n = 4) Organ/Region 5 min 10 min 15 min 30 min 60 min Olfactory bulb 1.05 ± 0.37 0.69 ± 0.06 0.46 ± 0.01 0.43 ± 0.04 0.22 ± 0.03 Hippocampus 1.03 ± 0.06 0.83 ± 0.07 0.56 ± 0.03 0.42 ± 0.08 0.17 ± 0.02 Frontal cortex 0.99 ± 0.05 0.62 ± 0.06 0.52 ± 0.11 0.42 ± 0.04 0.14 ± 0.02 Striatum 0.91 ± 0.11 0.65 ± 0.05 0.44 ± 0.03 0.33 ± 0.04 0.14 ± 0.01 Posterior cortex 0.86 ± 0.05 0.64 ± 0.05 0.50 ± 0.02 0.32 ± 0.03 0.15 ± 0.02 Thalamic region 0.69 ± 0.06 0.49 ± 0.03 0.34 ± 0.03 0.24 ± 0.03 0.10 ± 0.01 Cerebellum 0.66 ± 0.03 0.52 ± 0.03 0.33 ± 0.01 0.23 ± 0.01 0.11 ± 0.01 Rest of the brain 0.50 ± 0.03 0.68 ± 0.06 0.34 ± 0.08 0.17 ± 0.01 0.12 ± 0.01 Kidney 2.24 ± 0.12 2.65 ± 0.31 1.52 ± 0.21 1.26 ± 0.05 0.99 ± 0.20 Liver 1.07 ± 0.08 1.14 ± 0.14 0.93 ± 0.14 0.87 ± 0.09 0.62 ± 0.10 Lungs 0.86 ± 0.11 0.80 ± 0.05 0.66 ± 0.01 0.50 ± 0.06 0.29 ± 0.05 Heart 0.35 ± 0.02 0.37 ± 0.01 0.27 ± 0.02 0.21 ± 0.01 0.15 ± 0.01 Blood 0.21 ± 0.01 0.30 ± 0.01 0.20 ± 0.02 0.19 ± 0.01 0.14 ± 0.01 Brain region/cerebellum ratio Olfactory bulb 1.58 ± 0.49 1.33 ± 0.07 1.42 ± 0.01 1.83 ± 0.13 2.02 ± 0.25 Hippocampus 1.57 ± 0.10 1.60 ± 0.07 1.68 ± 0.11 1.77 ± 0.25 1.51 ± 0.09 Frontal cortex 1.51 ± 0.12 1.20 ± 0.03 1.58 ± 0.34 1.78 ± 0.15 1.25 ± 0.05 Striatum 1.38 ± 0.17 1.27 ± 0.06 1.34 ± 0.07 1.39 ± 0.14 1.27 ± 0.06 Posterior cortex 1.31 ± 0.07 1.25 ± 0.06 1.51 ± 0.01 1.38 ± 0.12 1.37 ± 0.08 Thalamic region 1.05 ± 0.08 0.95 ± 0.03 1.03 ± 0.08 1.02 ± 0.11 0.91 ± 0.06 Rest brain 0.76 ± 0.03 1.32 ± 0.07 1.02 ± 0.24 0.71 ± 0.03 1.08 ± 0.03 a Total injected amount 54 ± 4.2 MBq

Table 3. Radiolabelled parent compound and metabolites of [11C]AF150(S) at 10 and 45 min post I.V. injection in rats.a 10 min 45 min Plasma Brain Plasma Brain Metabolites in water extract 10.8 ± 4.8 8.2 ± 3.7 19.5 ± 6.1 20.8 ± 6.9 Metabolites in methanol extract 49.0 ± 3.6 8.7 ± 3.6 68.2 ± 10.5 23.9 ± 9.4 Parent compound in methanol extractb 40.2 ± 2.6 83.1 ± 2.8 12.3 ± 7.3 55.3 ± 11.8 a Percentage of total amount of radioactivity, data are presented as mean ± SD (n = 5). b Parent in methanol extract, identiied on HPLC

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

3.3.5 In vivo metabolic stability [11C]AF150(S) metabolites were measured ex vivo in plasma and brain, 10 and 45 min post IV injection. In plasma 40.2 ± 2.6 and 12.3 ± 7.3% of parent compound was detected at 10 and 45 min post injection, respectively. In brain, this was 83.1 ± 2.8 and 55.3 ± 11.8%, respectively. An overview of metabolites in water and methanol extracts is given in Table 3. 3 3.4 dISCuSSIon

In search of a potential M1ACh-R agonist PET ligand, AF150(S) was selected for radiolabelling and investigation of its biological properties because of its relatively good M1ACh-R selectivity, full agonist activity, interesting central pharmacological properties and amenability for labelling with carbon-11. From a structural point of view, AF150(S) belongs to a series of rigid analogues of ACh, therefore it can be expected to bind to the orthosteric site of the M1ACh-R and to directly compete with the natural ligand, ACh. Preliminary studies on agonist-induced Ca2+ response of engineered M1 - M5ACh-R, expressed in CHO cells, showed full agonist activity of AF150(S) at M1ACh-R and partial agonism at M3ACh-R, with a three times lower potency. No efect on M2ACh-R, M4ACh-R or M5ACh-R was observed in the concentration range of, 1.10-8 to 1.10-4 M. Its weak ainity for M3ACh-R is not to be expected to interfere with M1ACh-R labelling in the brain, as M3ACh-R are found primarily in the peripheral nervous system.

3.4.1 Radio- and automated synthesis Radiochemistry conditions for [11C]methylation of the precursor did not require presence of a base in the reaction mixture. Apparently, the basicity of AF400 is strong enough, so that a suicient proportion of the precursor is deprotonated allowing methylation to occur. Addition of base resulted in formation of side-products at higher tempera- 11 tures, which are likely due to reaction of DMF with [ C]CH3I or breakdown products of [11C]AF150(S). Because DMF interfered with semi-preparative HPLC puriication it was replaced by acetonitrile as solvent. The automated, optimized radiosynthesis yielded a formulated product with a yield of 69 ± 12% (decay corrected) at the end of synthesis, good speciic activity and high radiochemical purity, within a total synthesis time of 50 ± 5 min.

3.4.2 [11C]AF150(S) autoradiography Preliminary experiments showed that assay conditions are critical for [11C]AF150(S) binding. Addition of Mn2+ to the assay bufer proved to be required to promote [11C] AF150(S) speciic binding in M1ACh-R rich brain areas. This is in accordance with previ-

Chapter 3 ous studies by Potter et al. [220], which demonstrated that addition of Mn2+ is essential for observing high-ainity agonist binding to M1ACh-R in vitro. The transition metal ion is thought to promote G-protein receptor coupling. The concentration of 0.4 μM [11C]

AF150(S), required for optimal binding, corresponds to approximately 2 times the Kd,H (= 200 nM) of AF150(S) for the high-ainity state of rat M1ACh-R in the cerebral cortex. Hence, 0.4 μM would be adequate for labelling up to 70 % of M1ACh-R in the high- ainity state. Determination of non-speciic binding using 0.4 μM [11C]AF150(S) was performed with 100 nM atropine. This concentration of atropine was approximately 100 times its

Ki value (0.82 nM) for M1ACh-R [221], i.e. suicient to achieve maximal displacement of

M1ACh-R binding and low enough not to afect σ1 binding sites (Ki value > 10 μM).[222] The concentration of haloperidol of 100 nM was chosen in order to avoid an efect on

M1ACh-R binding (Ki value of 2140 ± 572 nM)[215], whilst having a maximal efect on σ1 11 binding sites (Ki value of 0.900 ± 0.067 nM) [222]. Inhibition of [ C]AF150(S) binding by atropine was highest in striatum and frontal cortex and was somewhat less in posterior cortex and hippocampus This is in accordance with the distribution of M1ACh-R in rat brain [151,153,223]. No signiicant inhibition of [11C]AF150(S) binding by haloperidol was found in medulla oblongata and striatum, areas rich in σ1 binding sites and in 11 dopamine D2 receptors, respectively. This indicates that [ C]AF150(S) does not label σ1 binding sites nor dopamine D2 receptors [215,224,225]. Xanomeline inhibited [11C]AF150(S) binding in frontal and posterior cortex, striatum and hippocampus pointing to agonist-agonist competition at the M1ACh-R. Xanomeline [190,226-228] also has ainity for M4 receptors, but AF150(S) does not. The inhibition by xanomeline showed a tendency of being concentration dependent. Due to the variable reduction at some concentrations, no curve could be itted and a pIC50 value could not be derived. Pirenzepine inhibited [11C]AF150(S) binding in frontal cortex, hippocampus, striatum and posterior cortex. The observed inhibition was similar at all concentrations used (20-25%). As pirenzepine has a Kd of 2.7 nM for M1ACh-R, substantial inhibition of speciic binding could be reached at the lowest applied concentration of 1 nM, and further inhibition at higher concentrations could not be distinguished. Using autoradiography on rat brain sections in vitro, speciic M1ACh-R binding of [11C] AF150(S) was demonstrated at sub µmolar concentration. The fact that conditions promoting M1ACh-R – G-protein coupling were required, suggests that it concerns labelling of the active, G-protein coupled state of the receptor.

3.4.3 Logdoct 7.4 and brain uptake LogD was used as parameter for lipophilicity rather than LogP, as LogP relates to lipophilicity for the charge-neutral forms of the radioligand, whilst LogD takes into account

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S) the sum of unionized and ionized forms of the radioligand at the physiological pH of 7.4 [229].

PET ligands with moderate lipophilicity indicated by LogDoct,7.4 values in the approxi- mate range 2.0 – 3.5 show optimal passive brain entry. Exceptionally, some useful radiotracers with lower or higher LogDoct,7.4 values also enter the brain, but for unclear reasons. Passive entry into the brain irst requires partitioning of the radiotracer from plasma into the lipid bilayer of the cells that form the blood brain barrier [31]. This is unlikely to happen for a compound with LogD of 0.05, as was experimentally measured for 3 oct,7.4 AF150(S) in this study. However, Fisher et al., reported that AF150(S) showed a preference for brain over plasma [201] and, in the present study, rapid uptake of [11C]AF150(S) in the brain was observed, leading to levels in M1ACh-R rich areas that exceeded up to 5-times the blood level. This can impossibly be achieved by passive difusion and likely involves facilitated transport. Similar observations were made for rapid uptake of nicotine in the brain to levels more than double of plasma levels. Also for nicotine facilitated transport was postulated, although the lipophilicity of nicotine (log D = 0.45) is higher than that of AF150(S) [230,231]. In this respect, the structural resemblance between AF150(S) and nicotine is interesting to note; both are relatively small molecules (MW 183.29 and

162.24 g/mol), the N-CH3-piperidine of AF150(S) is isosteric to the N-CH3-pyrrolidine of nicotine and for both, the saturated ring is substituted with an unsaturated heterocyclic amine.

3.4.4 Biodistribution and in vivo metabolism [11C]AF150(S) uptake in brain regions peaked at 5 min post injection and radioactivity levels in brain declined rapidly over time. Parent compound, measured by HPLC analysis, was found to persist in brain longer than in blood, in accordance with a fast peripheral and a slow brain metabolism. The highest radioactivity levels were observed in M1ACh-R rich regions, i.e. olfactory bulb, hippocampus, frontal and posterior cortex, and striatum. In these regions, apparent speciic uptake, given by the ratio over cerebellum, was observed at all time points up to 60 min. This high uptake in the brain, exceeding up to 5 times the blood levels, and the subsequent rapid decrease of radioactivity levels in the brain could possibly be explained by the rapid clearance of parent [11C]AF150(S) from blood, in combination with the postulated facilitated transport mechanism. Facilitated transport may be a ligand concentration dependent process [232], and only the initial blood concentrations of parent [11C]AF150(S) may be suicient to drive the transport. It was indeed found that at 10 and 45 min, only 40 and hardly 10 % of parent [11C]AF150(S) compound remained in the blood. The rapid decline in total brain radioactivity, in contrast to the constant level of radioactivity in blood, suggests that the peripherally formed radioactive metabolites are not taken up in the brain.

Chapter 3

The increased ratios of brain region over cerebellum in M1ACh-R rich areas remaining constant over the measured time frame of 60 min could possibly be due to retention of parent [11C]AF150(S) by M1ACh-R binding. However, since the estimated [11C]AF150(S) concentration in the brain was maximally 10.6 nM, which is 20-times below the reported

Kd,H of AF150(S), only a minor portion of the activated G-protein coupled receptor can have been labelled.

3.5 ConCLuSIon

Some of the characteristics of [11C]AF150(S) may not be in line with the projected properties for a good PET ligand, as derived from PET studies with high ainity, lipohilic, mainly antagonist PET ligands for receptors. The hydrophylicity, rapid metabolism, radioactive metabolites, and relatively low M1ACh-R binding ainity found for [11C]AF150(S), may be a challenge for PET studies and may require development of novel methods and models. Nevertheless, the ability to pass the blood brain barrier possibly in a facilitated way, higher brain versus blood levels and selective labelling of M1ACh-R on rat brain sections, suggest that [11C]AF150(S) is an interesting lead compound in search for an agonist PET ligand for the M1ACh-R, meant to explore the active state of the receptor.

Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)

4 Reproducible analysis of rat brain PET studies using an additional [18F]naF scan and an MR based RoI template

Hans J.C. Buiter, Floris H.P. van Velden, Josée E. Leysen, Abraham Fisher, Albert D. Windhorst, Adriaan A. Lammertsma and Marc C. Huisman

Published in: International Journal of Molecular Imaging Volume 2012 (2012) doi:10.1155/2012/580717

Chapter 4

ABSTRACT

Background: An important step in the analysis of positron emission tomography (PET) studies of the brain is the deinition of regions of interest (ROI). Image co-registration, ROI analysis and quantiication of brain PET data in small animals can be observer dependent. The purpose of the present study was to investigate the feasibility of ROI analysis based on a standard MR template and an additional [18F]NaF scan.

Methods: A procedure for co-registration of MR template and [18F]NaF images was developed and implemented. The precision of ROI analysis was determined using a simulation study. In addition, [18F]NaF scans of 10 Wistar rats were co-registered with a standard MR template by 3 observers and derived transformation matrices were applied to corresponding [11C]AF150(S) images. Uptake measures, calculated as injected dose per gram (%ID/g), were derived for several brain regions delineated using the MR template. Overall agreement between the 3 observers was assessed by interclass correlation coeicients (ICC) of uptake data. In addition, [11C]AF150(S) ROI data (%ID/g) were compared with ex vivo data (%ID/g)

Results: For all brain regions, ICC analysis showed excellent agreement between observers. Reproducibility, estimated by calculation of standard deviation of the between-observer diferences, was demonstrated by an average of 17% expressed as coeicient of variation. Uptake of [11C]AF150(S) derived from ROI analysis of PET data closely matched ex vivo biodistribution data.

Conclusions: The proposed method provides a reproducible and tracer independent method for ROI analysis of rat brain PET data.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

4.1 InTRoduCTIon

An important step in the analysis of positron emission tomography (PET) data is the deinition of regions of interest (ROI). For brain studies in man and large mammals, like primates and pigs, it becomes more common practice to use magnetic resonance (MR) based templates [233-236]. For brain studies in small animals, e.g. rats, several methods to deine ROI have been proposed, e.g. spatial normalization to an MR brain atlas [237], pre-deined PET templates [238,239], direct ROI deinition on PET images [240], ROI deinition using a co-registered segmented rat brain atlas (based on autoradiography) [241], and probabilistic atlases based on PET data (e.g. [18F]FDG, [18F]FECT and [11C]raclopride) for voxel-based functional mapping [242]. 4 In case of small animals, automated image co-registration of PET data with MR templates is diicult, because of the relatively large diference in spatial resolution between MR and PET. In this study, small animal imaging was performed on a high resolution PET scanner (i.e. ECAT High Resolution Research Tomograph, HRRT) with a spatial resolution of 2.5 to 3 mm and an MR scanner with a spatial resolution of 0.1 mm, resulting in an MR to PET diference of at least 25 in one direction, and 253 in all directions. Furthermore, MR – PET co-registration techniques require a certain level of morphological correspondence between both images, i.e. there should be suicient mutual information in both types of images to allow automatic co-registration. Consequently, characterization of new PET tracers with highly speciic uptake patterns can hamper direct co-registration of an MR template to a PET image, either manually or automatically. A possible way to obtain anatomical information is to perform an additional [18F]NaF scan, which can be acquired immediately following an experimental PET study and which can then be used for co-registration to the MR template. As experiments are performed under anaesthesia and head ixation device, an implicit registration between both PET data sets can be assumed. [18F]NaF scans provide anatomical (skull) information, as uptake of [18F]NaF in bone is high with much lower uptake in brain tissue [243-245]. The outline of the skull in an [18F]NaF scan provides an image that, indirectly, is also available from an MR scan, where the skull can be visualized by inverting the image colour scale and setting this scale to maximum intensity. This provides a two-step method for MR based ROI analysis of PET images in rat brains. The purpose of the present study was to investigate the reproducibility of MR based ROI analysis using a standard MR template, co-registered with an additionally acquired [18F]NaF scan. This method was tested using [11C]AF150(S) in the rat brain [246]. [11C]AF150(S) is a functionally selective agonist radioligand with moderate ainity for the muscarinic acetylcholine receptor M1 (M1ACh-R) in vitro and is currently evaluated as a potential PET ligand for central nerve system imaging.

Chapter 4

4.2 METHodS

4.2.1 Animals PET experiments were performed using 10 male Wistar rats (274 ± 24 g; Harlan Nether- lands B.V., Horst, The Netherlands), who were kept in conditioned housing under a regular light/dark cycle (12/12 h), allowing food and water ad libitum. All animal experiments were performed in compliance with Dutch laws and were approved by the University Animal Ethics Committee (VU University Amsterdam, the Netherlands).

4.2.2 Tracer synthesis Fluorine-18 was made via (p, n) on Oxygen-18 enriched water on a IBA cyclotrpn 18/9. [18F]NaF for injection was prepared by passing the [18F]luoride solution from the target, over a Waters quaternary methyl ammonium (QMA) light anion exchange cartridge (Waters Chromatography B.V., Etten-Leur, The Netherlands), followed by elution from the QMA cartridge with 1 mL 1.4% (w/v) of NaHCO3 in sterile H2O. This solution was collected in 10 mL of saline containing 7.09 mM NaH2PO4. Subsequently, the whole mixture was passed over a sterile Millex GV 0.22 μm ilter (Millipore B.V., Amsterdam, The Netherlands), yielding a sterile and pyrogen free solution with radiochemical purity >99.9% and a pH of 7. Quality control of the [18F]NaF solution was performed using an analytical high- performance liquid chromatography (HPLC) system consisting of a Jasco PU-2080 HPLC pump (Jasco, Ishikawa-cho, Japan), a Rheodyne 7724I injector (IDEX Health & Science, Wertheim-Mondfeld, Germany) with a 20 μL loop, a Bio-Rad Aminex Fermentation Monitoring Column 150x7.8 mm (Bio-Rad, Hercules, CA, USA) with a Jasco UV-2075 Plus UV detector (Jasco, Ishikawa-cho, Japan) and a NaI radioactivity detector (Raytest, Straubenhardt, Germany). The purity of the product was determined using this analytical HPLC system eluted with water containing 0.001 M sulphuric acid at a low rate of 0.8 mL·min-1. [11C]AF150(S) was prepared as described previously [246]. Briely, [11C]AF150(S) was 11 11 obtained via methylation of AF400 with [ C]CH3I. An incorporation yield of 90% of [ C]

CH3 was achieved in a 5-minute reaction time at 60°C in CH3CN. The product was puriied by HPLC, recovered by solid phase extraction and subsequently passed over a sterile Millex GV 0.22 μm ilter, yielding an isotonic, sterile and pyrogen free solution. The inal solution of [11C]AF150(S) was obtained with a decay corrected overall yield of 69 ± 12% (n=26), corresponding to 3200 – 7000 MBq at the end of the synthesis (i.e. 50 ± 5 min). Radiochemical purity was >99% and speciic activity at the end of synthesis was 23 – 118 GBq∙μmol-1.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

4.2.3 PET data acquisition PET measurements were performed using an ECAT HRRT (CTI/Siemens, Knoxville, TN, USA). The ECAT HRRT is a dedicated human brain PET scanner, with design features that enable high spatial resolution combined with high sensitivity, making it also suitable for small animal imaging. This scanner has a transaxial ield of view of 312 mm and an axial ield of view of 250 mm. The spatial resolution ranges from 2.3 to 3.2 mm full width at half maximum (FWHM) in transaxial direction and from 2.5 to 3.4 mm FWHM in axial direction [27]. First a transmission scan was performed using a 740 MBq 2D fan-collimated 137Cs rotat- ing point source. Following this transmission scan, a 3D emission acquisition of 45 min was initiated immediately after an intravenous tail vein injection of 9.9 ± 0.6 MBq [11C] 4 AF150(S). After the [11C]AF150(S) scan, an intravenous dose of 14.2 ± 3.5 MBq [18F]NaF was administered, followed 30 min later by a 3D emission of 30 min. Acquired data were stored in 64-bit list mode format and, for [11C]AF150(S), subsequently histogrammed into 21 time frames (7 x 10, 1 x 20, 2 x 30, 2 x 60, 2 x 150 and 7 x 300 s). [18F]NaF data were rebinned in a single frame of 30 min. Data were reconstructed using 3D ordered subsets weighted least squares (3D-OSWLS) [247]) using 7 iterations and 16 subsets. All data were normalized and corrected for attenuation, randoms, scatter, decay and dead-time. All images were reconstructed into a matrix of 256 x 256 x 207 voxels with a voxel size of 1.218 x 1.218 x 1.218 mm3.

4.2.4 MR template MR data were provided by the Image Sciences Institute of the University Medical Center Utrecht and consisted of a typical Wistar rat brain image, obtained by averaging single MR images of 8 Wistar rat brains (mean weight 305 g) acquired on a 4.7 Tesla MR system (Varian, Palo Alto, CA, USA) using a 3D Fast Spin Echo sequence (TR/TE = 1500/10 ms, 128 x 128 x 128 voxels, 32 x 40 x 32 mm3). Final image dimensions were 144 x 112 x 152 voxels with voxel dimensions of 0.125 x 0.125 x 0.156 mm3. Voxel dimensions of the MR image were scaled to 0.1 x 0.1 x 0.13 mm3, as brain size did not correspond with that in the present study. As an index of brain size, the distance between left and right dorsal lateral sides of the tympanic bulla was used. This distance was 14.4 mm in the original average MR image and 11.4 ± 0.1 mm for rats used in the present study (as was calculated from the [18F]NaF images). The MR template was generated by deining ROI on this average MR image using AMIDE (version 0.9.1; http://amide.sourceforge.net; [248]). ROI were deined from the beginning of the striatum until the end of the medulla oblongata, corresponding to plates 13 to 130 from the Paxinos and Watson atlas of the rat brain [212].

Chapter 4

4.2.5 MR template co-registration Using AMIDE, the [18F]NaF image of each individual animal was co-registered manually with the MR template image by three diferent observers according to a predeined protocol. An overview of the steps involved in this manual image co-registration is depicted in Figure 1. Briely, the centre of an [18F]NaF image was moved until it roughly overlaid the MR template. Next, rotational settings were adjusted until the [18F]NaF image paralleled the MR template and, inally, the centre was adjusted again until the image itted the MR template. MR template and [18F]NaF image its were considered successful when (a) the medial surface of the cranial case was aligned and the hemispheres were distributed ipsilaterally in a coronal cross-section, (b) tympanic bullae and frontal bones of the anterior part of the cranial cavity were aligned in a transverse cross-section, and

Figure 1. Overview of the various steps to manually co-register [18F]NaF and [11C]AF150(s) images with the MR template.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

Figure 2. Example of an [18F]NaF image co-registered with MR image and the MR template. (I) [18F]NaF image, co-registered with (II) MR image and (III) MR template for (a) coronal, (b) transverse and (c) sagittal cross-sections. Arrows indicate: (1) medial surface cranial case, (2) ipsilateral line distributing hemispheres, (3) frontal bones, (4) tympanic bulla, (5) surface of cranial case, and (6) articulation of parietal and occipital bones.

(c) the surface of the cranial case and the articulation of parietal and occipital bones were aligned in a sagittal cross-section, as illustrated in Figure 2. Finally, dynamic [11C] AF150(S) data were co-registered by directly applying the (rotational) settings obtained from the corresponding [18F]NaF co-registration.

4.2.6 RoI analysis Validation of ROI analysis software was performed on simulated data using AMIDE. ROI of the MR template were projected onto two simulated PET images that consisted of voxels with a theoretical content of either 0 or 1 kBq, arranged in a 3D checkerboard pattern. One image had voxel dimensions equal to the MR image, whereas the other

whereas the other had voxel dimensions equal to the PET image. Since there is a

considerable difference in spatial resolution with a factor of 1390 difference in voxel Chapter 4 volumes, the purpose of this simulation was to check whether ROI volumes and hadmean voxel ROI dimensions activity equal were tosimilar the PET for image. both images. Since there is a considerable diference in spatial resolution with a factor of 1390 diference in voxel volumes, the purpose of this simulationFor ROI was analysis to check of experimentalwhether ROI volumes data, the and MR mean templat ROI eactivity was projectedwere similar onto for the co- both images. registered [11C]AF150(S) images, using the corresponding [18F]NaF image. The For ROI analysis of experimental data, the MR template was projected onto the co- registeredpercentage [11C]AF150(S) injected images, dose perusing gram the corresponding tissue (%ID/g) [18 F]NaF was calculimage.ated The percent- for each brain age injected dose per gram tissue (%ID/g) was calculated for each brain region using the followingregion usingequation: the following equation: regional radioactivity concentration (kBq/ mL) %ID / g   100% injected dose (kBq)  density braintissue g /( mL

in whichwhich aa mean mean brain brain tissue tissue density density of 1.045 of 1.045 g/mL g/mL was usedwas [249].used [249].

4.2.7 Evaluation of image co-registration Inter6.2.7Evaluation observer variability ofof the image image co-registration co-registration method was investigated by comparing results of the three independent observers for all brain regions speciied. TheInter reproducibility observer variabilityof image co-registration of the image was co-registratio assessed by nconsidering method wasthe variability investigated by within subjects not attributable to possible systematic diferences between observers. comparing results of the three independent observers for all brain regions specified. It was assumed that the within-subject variability was the same for all subjects when thereThe were reproducibility no statistically of signiicant image diferences, co-registration and it was was estimated assessed by calculating by considering the the standard deviation of the between-observer diferences recorded for each subject. Ex- variability within subjects not attributable to possible systematic differences between pressed as a coeicient of variation (COV), this standard deviation provides a dimension- lessobservers. measure of It reproducibility. was assumed Reliability that the between within-subject the 3 observers variabil wasity quantiied was the sameusing for all the intra class correlation coeicient for absolute agreement (ICCA) and consistency subjects when there were no statistically significant differences, and it was estimated (ICCC). Validity of the image co-registration method was veriied comparing PET derived %ID/g data with previously obtained measurements of %ID/g in ex vivo biodistribution by calculating the standard deviation of the between-observer differences recorded studies. Briely, 5 groups of 4 of rats received an intravenous injection of [11C]AF150(S) andfor were each sacriiced subject. at 5, Expressed 10, 15, 30 or as 60 a min coefficient post injection. of variation Brains were (COV), excised this and standard regions dissected. All brain regions were weighed and counted for radioactivity using an automaticdeviation gamma provides counter a dimensionless (Wallac 1282 Compugamma measure of CS, reproducibi LKB Wallac,lity. Turku, Reliability Finland) between andthe the 3 %ID/g observers was calculated was quantified as percentage using of thetotal intraradioactivity class correlinjected,ation divided coefficient by for the weight of the tissue.

absolute agreement (ICCA) and consistency (ICCC). Validity of the image co- 4.2.8 Statistics registration method was verified comparing PET derived %ID/g data with previously All statistical analyses were performed using SPSS Statistics (version 17.0; SPSS Inc., Chicago,135 IL, USA). The Shapiro-Wilk test was used to conirm that all data were normally distributed. A two-tailed paired Student’s t-test was used for ROI analysis of the checkerboard simulation study. The correlation between ROI results (ROI volume and mean ROI activity concentration) of MR and PET checkerboard images was assessed using squared

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

Pearson’s correlation. A two-way ANOVA with Bonferroni post test was used for comparison of %ID/g measures obtained by the various observers who delineated ROI for [11C]AF150(S) PET data, and for the comparison of PET and ex vivo derived %ID/g values. The inter-observer reliability for [11C]AF150(s) PET data (mean ROI %ID/g) was computed by intraclass correlation coeicient using a two-way mixed efects model for absolute agreement (ICCA) and consistency (ICCC). ICC scores range form 0 to 1, representing a level of agreement: ≤ 0.40 poor to fair, 0.41 – 0.60 moderate, 0.61 – 0.80 substantial, 0.81 – 1.00 almost perfect [250]. Conidence intervals were also calculated for each ICC score. Data are reported as mean ± SD and diferences were considered signiicant at a level of p < 0.05. 4

4.3 RESuLTS

4.3.1 MR template Table 1 provides an overview of the ROI deined on the MR image.

Table 1. Regions and volumes of ROI deined on the MR template. Brain region Volume (mm3) Left Striatum 16.4 Right Striatum 15.1 Hippocampus 26.4 Frontal cortex (incl. part parietal cortex) 46.1 Posterior cortex (incl. occipital cortex) 49.6 Medulla oblongata (part) 15.7 Cerebellum 86.6

4.3.2 RoI analysis Both ROI volumes and mean ROI activity concentrations in left and right striatum, posterior cortex and medulla oblongata were slightly lower for the PET than for the MR resolution checkerboard images. The opposite was true for hippocampus, frontal cortex and cerebellum, as shown in Table 2. More importantly, both ROI volumes and mean ROI activities derived from MR and PET checkerboard images were not signiicantly diferent from each other. Correlation analysis of mean ROI activity concentrations in MR and PET resolution checkerboard images yielded a slope of 0.982, an intercept of 0.173 and an R2 of 0.998 (Figure 3).

Chapter 4

Table 2. Volume, mean activity concentration and total activity in various ROIs projected onto the checkerboard images with MR and PET resolution. MR image PET image Mean voxel Mean Mean voxel Mean Volume Volume Activity ROI Activity ROI Brain region (mm3) (mm3) Concentrationa Activity Concentrationb Activity (kBq.mL-1) (Bq) (kBq.mL-1) (Bq) Left Striatum 16.4 0.50 8.19 15.3 0.52 7.95 Right Striatum 15.1 0.50 7.56 14.7 0.50 7.30 Hippocampus 26.4 0.50 13.22 26.8 0.50 13.38 Frontal cortex 46.1 0.50 23.03 48.6 0.50 24.13 Posterior cortex 49.6 0.50 24.80 48.9 0.49 23.96 Medulla 15.7 0.50 7.85 15.8 0.49 7.75 oblongata Cerebellum 86.7 0.50 43.35 86.0 0.49 42.14 a 3 1 MR voxel = 0.0013 mm . b 3 1 PET voxel = 1.81 mm .

Figure 3. Pearson’s correlation plot of mean ROI activity for all brain regions in PET vs. MR-checkerboard.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

4.3.3 Evaluation of image co-registration Figure 4 shows a typical example of a coregistered MR image and template with PET images. Quantitative reproducibility of image co-registration was assessed using [11C] AF150(S) data by comparing %ID/g data between observers at ive diferent time points (Table 3). The Shapiro-Wilk test conirmed that these data were consistent with the nor-

Figure 4. Overview of co-registered MR image and template with PET images. (I) MR image, (II) MR 18 11 template, (III) [ F]NaF image and (IV) [ C]AF150(S) image(0 - 45 min) for (a) coronal, (b) transverse and (c) sagittal cross-sections.

Chapter 4 Region StriatumLeft Right Striatum Hippocampus cortex Frontal cortex Posterior Medulla oblongata Cerebellum Mean COV 5 min 1.06±0.16 1.04±0.17 0.98±0.16 0.95±0.13 0.92±0.14 0.67±0.13 0.80±0.14 16.12 10 min 0.70±0.11 0.71±0.11 0.67±0.10 0.64±0.11 0.64±0.10 0.46±0.08 0.53±0.08 16.08 15 min 0.57±0.09 0.57±0.10 0.54±0.09 0.52±0.08 0.52±0.08 0.38±0.06 0.44±0.07 15.96 30 min 0.33±0.05 0.33±0.05 0.32±0.05 0.30±0.05 0.31±0.05 0.25±0.04 0.27±0.05 15.81 45 min 0.23±0.04 0.23±0.05 0.23±0.04 0.21±0.04 0.22±0.04 0.18±0.04 0.20±0.04 18.49 5 min 1.05±0.17 1.05±0.18 0.99±0.16 0.99±0.17 0.95±0.17 0.66±0.12 0.80±0.14 17.26 10 min 0.70±0.12 0.72±0.12 0.68±0.11 0.67±0.13 0.65±0.12 0.47±0.08 0.53±0.09 17.19 15 min 0.57±0.10 0.58±0.10 0.55±0.09 0.54±0.10 0.53±0.09 0.39±0.06 0.44±0.08 17.10 30 min 0.33±0.05 0.33±0.05 0.33±0.05 0.31±0.05 0.31±0.05 0.24±0.04 0.28±0.05 16.67 45 min 0.23±0.04 0.24±0.04 0.23±0.04 0.22±0.04 0.22±0.04 0.18±0.03 0.20±0.04 18.08 5 min 1.06±0.16 1.04±0.17 0.98±0.15 0.93±0.14 0.89±0.14 0.71±0.14 0.80±0.14 16.44 10 min 0.70±0.12 0.71±0.12 0.67±0.11 0.64±0.11 0.61±0.10 0.48±0.07 0.53±0.09 16.52 15 min 0.56±0.09 0.57±0.10 0.54±0.09 0.51±0.08 0.50±0.08 0.40±0.06 0.44±0.07 16.51 30 min 0.32±0.06 0.33±0.05 0.32±0.05 0.30±0.05 0.29±0.05 0.25±0.04 0.28±0.05 16.70 45 min 0.22±0.04 0.23±0.04 0.23±0.04 0.21±0.04 0.21±0.04 0.19±0.04 0.20±0.04 18.86 C]AF150(S) uptake (%ID/g) derived from various brain regions by three observers following image co-registration. observers three following by regions C]AF150(S) various brain from uptake (%ID/g) derived 11 [ Table 3. Table Number of subjects (n = 10) (n = 10) mean ± SD, represent Data Observer 1 Observer 2 Observer 3

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template 95% CL 95% CL C C 4 95% CL ICC 95% CL ICC C]AF150(S) uptake (% ID/g). brain 11 A A Observer 1 vs. 3 Observer 1 vs. Overall 95% CL ICC 95% CL ICC C C 95% CL ICC 95% CL ICC A A ICC 0.956 (0.826,0.989) 0.962 (0.856,0.990) 0.927 (0.102,0.987) 0.978 (0.913,0.994) 0.856 (0.062,0.971) 0.939 (0.776,0.985) 0.911 (0.605,0.979) 0.959 (0.885,0.989) 0.966 (0.879,0.991) 0.967 (0.873,0.992) 0.904 (-0.022,0.984) 0.985 (0.940,0.996) 0.895 (0.306,0.977) 0.945 (0.794,0.986) 0.919 (0.608,0.981) 0.966 (0.904,0.991) Observer 2 vs. 3 Observer 2 vs. ICC Intraclass correlation coeicients and corresponding 95% conidence intervals for regional [ regional intervals for 95% conidence and corresponding coeicients Intraclass correlation Table 4. Table Number of subjects (n = 10) ROI 2 Observer 1 vs. Left StriatumLeft 0.998 (0.991,0.999) 0.997 (0.990,0.999) 0.997 (0.988,0.999) 0.997 (0.988,0.999) Left StriatumLeft 0.997 (0.988,0.999) 0.997 (0.989,0.999) 0.997 (0.992,0.999) 0.997 (0.992,0.999) Right Striatum 0.995 (0.978,0.999) 0.995 (0.981,0.999) 0.998 (0.991,0.999) 0.998 (0.992,0.999) Right Striatum 0.998 (0.991,0.999) 0.997 (0.990,0.999) 0.997 (0.990,0.999) 0.997 (0.991,0.999) Hippocampus 0.997 (0.989,0.999) 0.997 (0.989,0.999) 0.992 (0.769,0.999) 0.997 (0.989,0.999) Hippocampus 0.987 (0.632,0.998) 0.996 (0.983,0.999) 0.992 (0.947,0.998) 0.997 (0.991,0.999) Frontal cortex Frontal 0.930 (0.491,0.985) 0.961 (0.853,0.990) 0.985 (0.886,0.997) 0.991 (0.963,0.998) Frontal cortex Frontal 0.907 (0.080,0.983) 0.968 (0.878,0.992) 0.937 (0.682,0.985) 0.972 (0.922,0.992) Posterior Posterior cortex Posterior Posterior cortex Medulla oblongata Medulla oblongata Cerebellum 0.993 (0.972,0.998) 0.992 (0.969,0.998) 0.970 (0.499,0.994) 0.988 (0.954,0.999) Cerebellum 0.957 (0.663,0.991) 0.977 (0.910,0.994) 0.974 (0.881,0.994) 0.986 (0.960,0.996) Level of agreement: ≤ 0.40 poor to fair, 0.41 – 0.60 moderate, 0.61 – 0.80 substantial, 0.81 – 1.00 almost perfect. 0.61 – 0.80 substantial, 0.41 – 0.60 moderate, fair, ≤ 0.40 poor to of agreement: Level

Chapter 4

Figure 5. Uptake (%ID/g) of [11C]AF150(S) in various brain regions obtained from PET and ex vivo biodistribution experiments. PET studies, dark gray (n = 10, average of 3 observers) and ex vivo biodistribution experiments, light gray (n =4 per time point). Data are presented as mean ± SD. Level of signiicance was evaluated using a two-way ANOVA with Bonferroni post test; * p<0.05.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template mal distribution (W statistic ≥ 0.936, p ≥ 0.511). The comparison of %ID/g data revealed that for each time point none of the observations were signiicantly diferent from each other (p>0.722). In addition, COV values were not signiicantly diferent between the 3 observers (p>0.467).

Inter-observer reliability of the image co-registration method was assessed by calculating ICCA and ICCC of mean uptake in each region, as derived by the 3 observers. These results are summarized in Table 4 and, for all regions, show almost perfect agreement

(ICCA> 0.856) between observers. In addition, for all regions, there was almost perfect consistency (ICC >0.939) between observers. C 4 Observed regional brain uptake of [11C]AF150(S) was compared between PET and ex vivo studies. The Shapiro-Wilk test (W statistic ≥ 0.885, p ≥ 0.361) conirmed that these data were consistent with the normal distribution. A graphical overview for the various time points is given in Figure 5, indicating no signiicant diferences between both data sets at any of the time points, except for cerebellum at 5 min and hippocampus at 10 min.

4.4 dISCuSSIon

4.4.1 data acquisition The proposed generic method for ROI analysis of PET data acquired in rats is based on the acquisition of an additional [18F]NaF scan. In the present study, the additional acquisition time was 30 min, starting 30 min after injection of ~14 MBq of [18F]NaF. This time frame, however, can be adapted according to experimental design in future studies.

4.4.2 RoI analysis In order to determine efects of diferent spatial resolutions (1390 MR voxels correspond with one PET voxel) on ROI analyses, calculation of average ROI values and volumes was validated using a simulated checkerboard scan. In general, less than 7% diference in volumes and less than 5% in mean ROI activity values were observed between PET images with PET and MR resolution (table 2) and none of these diferences were signiicant. In addition, mean ROI activity values derived from both images showed near perfect correlation (r2 > 0.998; slope = 0.982). Therefore, applying MR based ROI onto PET images is reproducible with respect to both ROI volumes and mean ROI activity values.

Chapter 4

4.4.3 Evaluation of the image co-registration paradigm All three observers successfully co-registered all ten [18F]NaF images with the MR brain template, following the procedure as outlined in igure 1. Derived uptake of [11C] AF150(S) was not signiicantly diferent between observers for any brain region and any time point. Following image co-registration of [11C]AF150(S) scans, inter-observer agreement (ICCA and ICCC) of uptake was excellent for all brain regions. This implies that the image co-registration procedure is very reliable. Moreover, uptake values were similar to those obtained from an ex vivo study. The minor diferences at 10 min for hippocampus and at 5 min for cerebellum might be ascribed to individual variability in combination with the relatively small groups of animals. Interestingly, both PET and ex vivo biodistribution results indicate that [11C]AF150(S) uptake is highest in striatum, followed by hippocampus, then cortex and then cerebellum, closely following the distribution of M1ACh-R in rat brain [153]. Clearly, some user dependent uncertainty may remain in the co-registration of MR template and [18F]NaF images. Based on the good inter-observer correspondence in the manual procedure, semi-automatic or fully automatic image co-registration methods could be investigated for easier and faster ROI analysis of small animal PET studies. Although several methods are available for PET co-registration in small animals [237- 239,241,242], reproducibility was not assessed for all those methods. In two previous small animal studies reproducibility of brain uptake has been investigated [240,251] using direct ROI deinition on PET images or pre-deined PET templates for [11C]raclopride scans. Reproducibility of [11C]raclopride uptake was found to be between 10% and 22%. These values are similar to the reproducibility of [11C]AF150(S) uptake observed in the present study (average COV of 17%), suggesting that there is no additional uncertainty due to application of the presently described two-step method of ROI analysis. The advantage of the present method is that it allows exact localization of the brain, independent of radioligand and PET scanner used, and without the need to perform an individual MR scan for each rat. In the near future, multi-modality imaging may provide better means for analysis of small animal PET studies. The presently available combined PET/CT small animal scan- ners do have certain notable shortcomings, i.e. the inability to perform simultaneous data acquisition and lesser contrast among soft tissues compared to MR. Nevertheless, the CT component of PET/CT images may be used to co-registrate these images with MR. Ultimately, the emerging role of hybrid PET/MR small animal imaging may provide better means for tracer independent data analysis of rat brain studies, and should be further investigated.

Reproducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template

4.5 ConCLuSIonS

Reproducible analysis of PET data is possible using co-registration of an MR based ROI template with an additional [18F]NaF scan. This method can serve as a reliable and reproducible two-step method for tracer independent data analysis of rat brain studies, as demonstrated for [11C]AF150(S) data.

5 [11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

Hans J.C. Buiter, Albert D. Windhorst, Marc C. Huisman, Maqsood Yaqub, Dirk L. Knol, Abraham Fisher, Adriaan A. Lammertsma and Josée E. Leysen

Published in: European Journal of Nuclear Medicine and Molecular Imaging Research 3 (2013) doi:10.1186/2191-219X-3-19

Chapter 5

ABSTRACT:

Background: The M1 muscarinic acetylcholine receptor (M1ACh-R) is a G-protein coupled receptor than can occur in interconvertible coupled and uncoupled states. It is enriched in basal ganglia, hippocampus, olfactory bulb and cortical areas, and plays a role in motor and cognitive functions. Muscarinic M1 agonists are potential therapeutic agents for cognitive disorders. The aim of this study was to evaluate [11C]AF150(S) as a putative M1ACh-R agonist PET ligand, which, owing to its agonist properties, could provide a tool to explore the active G-protein coupled receptor.

Methods: Regional kinetics of [11C]AF150(S) in rat brain were measured using a High Resolution Research Tomograph (HRRT), both under baseline conditions and following pre-treatment with various compounds or co-administration of non-radioactive AF150(S). Data were analysed by calculating standard uptake values and by applying the simpliied reference tissue model (SRTM)

Results and discussion: [11C]AF150(S) was rapidly taken up in the brain, followed by a rapid clearance from all brain regions. Analysis of PET data using SRTM revealed a binding potential (BPND) of 0.25 for striatum, 0.20 for hippocampus, 0.16 for frontal and 0.15 for posterior cortical area, all, regions rich in M1ACh-R. BPND values were signiicantly reduced following pre-treatment with M1ACh-R antagonists. BPND values were not af- fected by pre-treatment with a M3ACh-R antagonist. Moreover BPND was signiicantly reduced after pre-treatment with haloperidol, a dopamine D2 receptor blocker that causes an increase in extracellular acetylcholine (ACh). The latter may compete with [11C]AF150(S) for binding to the M1ACh-R, further pharmacological agents were applied to investigate this possibility. Upon injection of the highest dose (49.1 nmol.kg-1) of [11C]AF150(S) diluted with non-radioactive AF150(S), brain concentration of AF150(S) reached 100 nmol.l-1 at peak level. At this concentration no sign of saturation in binding to M1ACh-R was observed.

Conclusions: The agonist PET ligand, [11C]AF150(S) was rapidly taken up in brain and showed an apparent speciic M1ACh-R related signal in brain areas that are rich in M1ACh-R. Moreover binding of the agonist PET ligand [11C]AF150(S), appears to be sensitive to changes in extracellular ACh levels. Further studies are needed to evaluate the full potential of [11C]AF150(S) for imaging the active pool of M1ACh-R in vivo.

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

5.1 InTRoduCTIon

Acetylcholine (ACh) is a core neurotransmitter that controls vital functions via the peripheral nervous system (e.g. heart rate, intestinal motility, glandular secretion) and the central nervous system (CNS) (e.g. cognitive and motor functions) [150]. Cholinergic neurotransmission is processed via nicotine and muscarinic acetylcholine receptors (MACh-R). The former are ligand gated ion channels, the latter metabotropic G-protein coupled receptors (GPCRs). The MACh-R family comprises ive characterized subtypes (M1-M5) [151]. M1ACh-R is the most prevalent MACh-R subtype in the CNS and acts as an excitatory receptor that is coupled to the Gαq/11 protein [252]. It is located primarily in postsynaptic nerve terminals in forebrain regions including basal ganglia, hippocampus and cortical areas, the cerebellum is essentially devoid of M1ACh-R [153]. M1ACh-R in the forebrain regions play a role in motor control and cognition and are thought to be 5 implicated in pathologies such as schizophrenia, Parkinson’s disease and Alzheimer’s disease [155,156]. Studies using M1ACh-R agonists in laboratory animals have shown improvement in cognitive functions [190]. Clinical studies in patients with Alzheimer’s disease and schizophrenia revealed attenuation of psychotic behaviour and showed improvement of cognition [253]. A non-invasive molecular imaging technique such as positron emission tomography (PET) could be useful to elucidate the function of M1ACh-R in vivo. In the past, several compounds with ainity for M1ACh-R have been labelled with carbon-11, e.g. [11C]scopolamine, [11C]NMPB, [N-11C-methyl]-benztropine and [11C]QNB [191-194]. These PET ligands demonstrated good brain uptake and allowed mapping of MACh-R in primate and human brain. However, these ligands are not M1ACh-R selective and they show slow ligand-receptor dissociation kinetics, by which ligand-receptor binding equilibrium cannot be reached in vivo [254,255]. In addition, the above mentioned radioligands are antagonists that bind to the total pool of GPCRs. Consequently, they cannot distinguish between activated GPCRs and uncoupled inactive receptors. In contrast, agonists usually show high ainity for G-protein-coupled receptors and low ainity for uncoupled receptors. Therefore, at low concentrations, they bind to the activated receptors only. The occurrence of interconvertible ainity states of GPCRS has been amply documented [256] and was also demonstrated for M1ACh-R in in vitro experiments [220,257]. Studies with agonist PET ligands for the dopamine-D2 receptor, such as [11C](+)-PHNO, revealed that the agonist PET ligand labelled in vivo a smaller receptor pool than an antagonist PET ligand and moreover, agonist PET ligands appeared more sensitive to displacement by released endogenous dopamine [88]. This can be taken as an indication that the agonist PET ligand preferentially labels the active G-protein-coupled receptor pool, which is the target for the endogenous neurotransmitter.

Chapter 5

Besides for the dopamine D2 [91,258-261] and the µ-opiate receptor [5], few agonist PET ligands have been investigated for other GPCRs. The tested muscarinic agonists, [11C] xanomeline and [11C]butylThio-TZTP, appeared not selective for the M1ACh-R subtype, and show high binding ainity for σ-sites, leading to substantial non-selective binding, in addition, they sufer from poor pharmacokinetics [158]. In the search for a suitable selective M1ACh-R agonist PET ligand, AF150(S) was selected based on its pharmacological properties. AF150(S) has moderate ainity with an apparent equilibrium dissociation constant for high ainity binding sites, Kd,H = 200 nM for M1ACh-R in rat cerebral cortex and shows functional selectivity for M1ACh-R in transfected cells [203]. AF150(S) belongs to a series of selective M1ACh-R agonists, related to Cevimeline (AF102B) [197]. Cevimeline, a rigid analogue of ACh, is marketed for human medicinal use and is likely to act as an orthosteric agonist. The pharmacological and potential therapeutic properties of AF150(S) have been investigated extensively in animal models of neurological diseases, where improvement in cognitive functions, reduction of amyloid plaques and decrease in tau phosphorylation were demonstrated [198,200,201]. AF150(S) and its congener, AF267B have been considered as potential dual symptomatic and disease modifying treatments for Alzheimer’s disease [262]. Such important possible therapeutic application, further supports the interest for investigating [11C]AF150(S) as a potential agonist M1ACh-R PET ligand. Recently, [11C]AF150(S) has been synthesized successfully. Its binding properties and speciicity for M1ACh-R were evaluated by in vitro autoradiography; high uptake in rat brain, possibly by facilitated transport mechanisms, was demonstrated in studies ex vivo [246]. In the present study, brain uptake and binding of [11C]AF150(S) were studied in rats using PET in order to explore its suitability as an agonist PET ligand for M1ACh-R. In vivo speciicity of binding and sensitivity to changes in extracellular levels of ACh were examined by treatment with various pharmacological agents.

5.2 MATERIALS And METHodS

5.2.1 Materials Xanomeline was obtained from Metina AB (Lund, Sweden), pirenzepine, trihexyphenidyl and haloperidol were purchased from Sigma-Aldrich (Zwijndrecht, the Netherlands), darifenacin was obtained from Sequoia Research Products Ltd. (Pangbourne, United Kingdom), AF-DX 384 was purchased from Tocris Bioscience (Bristol, United Kingdom) and rivastigmine was purchased from AvaChem Scientiic LLC (San Antonio, U.S.A). Both AF400, the precursor for radiolabelling, and reference AF150(S) were provided by the Israel Institute for Biological Research, Ness-Ziona, Israel.

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

Compounds were dissolved in saline for in vivo application, except for xanomeline and haloperidol which were dissolved in ethanol and further diluted with saline (inal concentration of ethanol <10%). AF-DX 384 was dissolved in DMSO and diluted with saline (inal concentration of DMSO < 5%). Cold (non-radiolabelled) AF150(S) was dissolved in saline to obtain a 10 mmol∙L-1 stock solution, from which further dilutions were made. Concentrations of AF150(S) in all solutions were veriied by HPLC.

5.2.2 Animals Experiments were performed with male Wistar rats (304 ± 43 g; Harlan Netherlands B.V. Horst, the Netherlands). Rats were kept in conditioned housing under a regular light/ dark cycle (12/12h) and allowed food and water ad libitum. All animal experiments were in compliance with Dutch law and approved by the VU University Animal Ethics Com- mittee. 5

5.2.3 Radiochemistry [11C]AF150(S) was synthesized as previously described by methylation of the desmethyl precursor (AF400) using [11C]methyl iodide [246]. The inal product had a radiochemical purity >99% and a speciic activity (SA) of 23 – 118 GBq.μmol-1, at the end of synthesis. [18F]NaF was prepared as described previously [263]. Both radiotracers were formulated in an isotonic, sterile and pyrogen free solution for IV injection.

5.2.4 PET studies

5.2.4.1 PET scanner PET measurements were performed using an ECAT High Resolution Research Tomograph (HRRT) (CTI/Siemens, Knoxville, TN, USA). The ECAT HRRT is a dedicated human brain PET scanner, with design features that enable high spatial resolution combined with high sensitivity, making it also suitable for small animal imaging. This scanner has a ield of view of 312 and 250 mm in transaxial and axial directions, respectively. The spatial resolution ranges from 2.3 to 3.2 mm full width at half maximum (FWHM) in transaxial direction and from 2.5 to 3.4 mm FWHM in axial direction, depending on the distance from the centre as described previously [27].

5.2.4.2 Scan protocol Anaesthesia was induced and maintained by constant insulation of 1.5 - 2% isolurane in pure oxygen. Animals were placed in a ixation device with tooth bar to secure a ixed and immobile horizontal position of the head during scanning. Body temperature was kept constant at 37 °C with a heating-pad coupled to a thermostat, which was connected

Chapter 5 to a rectal thermometer. A cannula was inserted into the vena femoralis for radiotracer injection. Transmission measurements of 6 min duration were performed using a 740 MBq 2D fan-collimated 137Cs moving point source. An intravenous (IV) bolus of 15.5 ± 4.2 MBq of [11C]AF150(S) was injected, at the irst injection for baseline scanning, the speciic activity (SA) was 71 ± 16 GBq·μmol-1, at the second injection, following pre-treatment, SA was 9 ± 2 GBq·μmol-1. A 3D emission scan was acquired, starting immediately prior to the IV injection and lasting for 45 min. At the end of the [11C]AF150(S) scans, [18F]NaF (15.2 ± 4.5 MBq) was injected and, 30 min later, an emission scan was acquired for 30 minutes.

5.2.4.3 Image reconstruction Acquired PET data were stored in 64 bit list mode format and, for [11C]AF150(S), were subsequently histogrammed into 21 frames with progressive duration (7 x 10, 1 x 20, 2 x 30, 2 x 60, 2 x 150 and 7 x 300 s). The [18F]NaF data were histogrammed into a single frame of 30 min. Data were reconstructed using 3D ordered subsets weighted least squares (3D-OSWLS) [247] using 7 iterations and 16 subsets. All data were normalized and corrected for attenuation, randoms, scatter, decay and dead-time. All images were reconstructed into a matrix of 256 x 256 x 207 voxels with a voxel size of 1.218 x 1.218 x 1.218 mm3.

5.2.4.4 Pre-treatment with muscarinic agents A total of twenty rats was used to examine the speciicity of binding of [11C]AF150(S) in brain. Rats were randomized in 5 groups of 4 rats and for each rat, two consecutive [11C]AF150(S) scans were acquired with a 30 min interval. The irst scan was performed under baseline conditions, providing information on regional distribution and kinetics of [11C]AF150(S) in the brain. The second, scan was performed following pre-treatment with either the M4/M1ACh-R agonist xanomeline (5 or 30 mg.kg-1 subcutaneous (SC)) [218], the M1ACh-R antagonists pirenzepine (30 mg.kg-1 SC) [219] or trihexyphenidyl (3 mg.kg-1 SC) [222], or the M3ACh-R antagonist darifenacin (3 mg.kg-1 IV) [264]. Xanomel- ine, pirenzepine and trihexyphenidyl were administered 30 min and darifenacin 15 min prior to [11C]AF150(S) injection.

5.2.4.5 Pre-treatment with agents afecting extracellular ACh-levels

Twelve rats were used to examine indirect efects of haloperidol (a dopamine D2 antagonist and σ-site ligand) [265] and AF-DX 384 (an M2/M4ACh-R antagonist) [266], the latter with and without the acetylcholine esterase (AChE) inhibitor rivastigmine [267], on [11C] AF150(S) binding. Rats were randomized in 3 groups of 4 rats. In each animal, two consecutive scans were performed with a 45 min interval. The irst scan was performed under baseline conditions and the second after pre-treatment with haloperidol (1 mg.kg-1

100

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

SC), the combination of AF-DX 384 (5 mg.kg-1 intraperitoneal (IP)) and rivastigmine (2.5 mg.kg-1 SC), or AF-DX 384 (5 mg.kg-1 IP). Haloperidol and AF-DX 384 were administered 30 min, and rivastigmine 45 min prior to the [11C]AF150(S) injection.

5.2.4.6 Co-injection of [11C]AF150(S) with cold AF150(S) Twelve rats were used to investigate the efect of co-injection of cold AF150(S) on uptake of total amount of AF150(S) in the brain. 1, 5 and 15 nmol respectively, of cold AF150(S) was added to the dose of [11C]AF150(S) in the syringe, right before intravenous injection. Rats were randomized in 3 groups of 4 animals, each animal underwent two consecutive scans with a 30 min interval. First, each rat received an injection of undiluted [11C] AF150(S), to register a baseline scan, before the second scan rats received an injection of [11C]AF150(S) diluted with cold AF150(S), 5.2.5 data Analysis 5

5.2.5.1 Region of interest analysis [18F]NaF scans were co-registered with a standard MR rat brain template to delineate regions of interest (ROIs), this method was previously described and validated with [11C[AF150(S) in rats [268]. The following ROIs were used: left striatum (15 mm3), right striatum (15 mm3), hippocampus (left and right together, 27 mm3), frontal plus parietal cortical area (49 mm3), posterior plus occipital cortical area (49 mm3), and cerebellum (86 mm3). ROIs were transferred onto dynamic PET images in order to calculate regional radioactivity concentrations (kBq.mL-1) and to generate regional time activity curves (TACs). TACs were also normalised for generation of standardized uptake values (SUV) [269].

5.2.5.2 Binding potential The simpliied reference tissue model (SRTM) [37], with cerebellum as reference tissue, was used to calculate non-displaceable binding potential (BPND) as an outcome measure of speciic binding.

5.2.6 STATISTICAL AnALYSIS

Diferences in regional BPND values at baseline were assessed using one-way ANOVA with Bonferroni correction and diferences in BPND values before and after treatment with various agents or co-administration of non-radioactive AF150(S).were tested using a general Linear Mixed Model (SPSS Statistics version 17.0; SPSS Inc., Chicago, IL, USA).

101

Chapter 5

All results are expressed as the mean ± SD and values of p < 0.05 were considered to be statistically signiicant.

5.3 RESuLTS

5.3.1 Baseline uptake [11C]AF150(S) uptake in brain regions under baseline conditions was examined in a total of 44 rats. Average time-activity curves for 5 brain regions are shown in Figure 1A. Following intravenous injection of [11C]AF150(S), radioactivity in all brain regions rapidly increased and reached its maximum within the irst minute after injection. Up- take in striatum, hippocampus and the two cortical areas was higher than in cerebellum. Clearance of radioactivity from the brain was relatively fast, especially in cerebellum. The maximum brain concentration measured was, on average, 155 ± 26 kBq.mL-1 corresponding to 2.9 ± 0.5 nM [11C]AF150(S). SUV ratios (SUVr) of brain region over cerebellum as function of time are shown in Fig- ure 1B. SUVr values were maximal at 5.8 min, ranging from 1.19 ± 0.07 for the posterior cortical area to 1.35 ± 0.05 for striatum.

[11C]AF150(S) speciic binding in M1ACh-R rich brain areas was determined using SRTM with cerebellum as reference tissue. BPND values are presented in Table 1. BPND under baseline conditions ranged from 0.25 to 0.15, with a regional rank order: striatum > hippocampus > frontal cortical area ≈ posterior cortical area. Findings were quite consistent across the 5 groups of 4 rats. Representative images of baseline [11C]AF150(S) uptake and corresponding parametric BPND are shown in Figure 2.

5.3.2 Pre-treatment with muscarinic agents

In Table 1 the BPND values in ive brain regions following pre-treatment with xanomeline, pirenzepine, trihexyphenidyl and darifenacin are shown. Following pre-treatment with -1 11 xanomeline at 5 mg∙kg , [ C]AF150(S) BPND signiicantly increased by 20 and 19% in left striatum and hippocampus, respectively. Xanomeline at a higher dose (30 mg∙kg-1) did not result in signiicant changes in BPND in any of the brain areas investigated. This high dose caused severe side efects, such as strongly increased heart rate, salivation and irregular breathing. Pre-treatment with pirenzepine and trihexyphenidyl caused signiicant reductions in BPND. With pirenzepine reductions of 37, 32 and 16% were seen in frontal cortical area, posterior cortical area and left striatum, respectively. Trihexyphe- nidyl caused reductions in frontal cortical area, posterior cortical area and hippocampus of 53, 56 and 13%, respectively. Pre-treatment with darifenacin had no measurable efect on [11C]AF150(S) binding in any brain region.

102

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

Figure 1: Standard Uptake Values of [11C]AF150(S) in various brain regions obtained from PET experiments. (A) Regional brain concentration, expressed as SUV, as function of time following IV injection of [11C]AF150(S) and (B) relative to cerebellar concentration (SUVr). Each data point represents the mean of 44 animals. Vertical lines indicate the SD.

103

Chapter 5 #2 #3 #1 #4 #5 Mean (n = 20) IV) -1 Darifenacin (3 mg∙kg all regions (except for left striatum), for (except all regions #3 SC) -1 Trihexyphenidyl (3 mg∙kg SC) -1 all regions (except for right striatum), striatum), right for (except all regions Pirenzepine (30 mg∙kg #2 SC) -1 all region (except for frontal cortical frontal area). for (except all region Xanomeline (30 mg∙kg #5 compared to all other brain regions, regions, all other brain to compared #1 SC) -1 Xanomeline (5 mg∙kg C]AF150(S) before and after treatment with various muscarinic with various agents. and after treatment C]AF150(S) before 11 of [ ND BP all regions (except for posterior cortical area) or, cortical posterior or, for area) (except all regions Table 1. Table Right StriatumHippocampus cortical areaFrontal 0.24±0.04 cortical areaPosterior 0.12±0.07 0.27±0.01 0.20±0.03 and SC = subcutaneous. IV = intravenous mean ± SD (n = 4 per group), represent Data 0.12±0.06 0.10±0.04 0.23±0.08 0.24±0.01* with baseline: *p<0.05, **p<0.01, ***p<0.001. compared diference Signiicant 0.10±0.04 0.17±0.06 0.19±0.04 0.25±0.06 with p<0.001: diferent, Signiicantly 0.15±0.06 0.14±0.04 0.19±0.06 0.25±0.06 0.13±0.04 0.16±0.05 0.20±0.03 0.23±0.03 0.13±0.03 0.10±0.04** 0.18±0.02 0.22±0.01 0.11±0.03 0.09±0.03** 0.11±0.05 0.19±0.01 0.22±0.01 0.05±0.03* 0.05±0.03* 0.16±0.01** 0.26±0.05 0.22±0.02 0.24±0.02 0.23±0.01 0.26±0.03 0.20±0.06 0.22±0.04 0.18±0.05 0.24±0.05 0.16±0.06 0.20±0.03 0.15±0.06 #4 Brain regionBrain Baseline Pre-treated Baseline Pre-treated Baseline Pre-treated Baseline Pre-treated Baseline Pre-treated Baseline Left StriatumLeft 0.25±0.05 0.30±0.04* 0.22±0.07 0.22±0.05 0.25±0.04 0.21±0.05* 0.24±0.02 0.21±0.01 0.29±0.03 0.28±0.04 0.25±0.05

104

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

Figure 2: Example of an [11C]AF150(S) PET image of the brain. Summed (10 – 45 min after injection) baseline [11C]AF150(S) uptake image together with corresponding parametric non-displaceable binding potential (BPND) image. Corresponding MR slices and fused images are shown for anatomical reference.

5.3.3 Pre-treatment with agents afecting extracellular ACh levels

Table 2 provides BPND values in ive brain areas following pre-treatment with haloperidol, AF-DX 384 and AF-DX 384 plus rivastigmine. Haloperidol pre-treatment resulted in signiicant reductions in BPND in right striatum and hippocampus of 27 and 15%, respectively. The animals showed severe rigidity, following the haloperidol treatment. Pre- treatment with AF-DX 384, however, did not show any signiicant changes in BPND. The addition of rivastigmine to AF-DX 384 pre-treatment resulted in signiicant increases in 11 BPND of [ C]AF150(S) in right striatum, hippocampus and frontal cortical area of 22, 12

105

Chapter 5

11 Table 2. BPND of [ C]AF150(S) without and with pre-treatment with agents that increase extracellular levels of ACh. Haloperidol AF-DX 384 AF-DX 384 & Rivastigmine (1 mg.kg-1 SC) (5 mg.kg-1 IP) (5 mg.kg-1 IP and 2.5 mg.kg-1 SC) Brain region Baseline Pre-treated Baseline Pre-treated Baseline Pre-treated Left Striatum 0.23±0.04 0.18±0.01 0.27±0.06 0.32±0.04 0.25±0.05 0.27±0.03 Right Striatum 0.28±0.05 0.21±0.03* 0.30±0.04 0.33±0.05 0.23±0.05 0.29±0.03* Hippocampus 0.19±0.02 0.16±0.01* 0.26±0.04 0.26±0.02 0.22±0.02 0.25±0.01* Frontal cortical 0.10±0.06 0.06±0.04 0.24±0.02 0.24±0.06 0.17±0.03 0.21±0.02** area Posterior 0.07±0.07 0.06±0.04 0.21±0.03 0.22±0.05 0.19±0.04 0.19±0.04 cortical area Data represent mean ± SD (n = 4 per group), SC = subcutaneous and IP = intraperitoneal. Signiicant diference compared with baseline: *p<0.05, **p<0.01.

Figure 3: Efect of co-injection of cold AF150(S) on regional brain uptake of AF150(S) in various brain regions. Concentrations of cold AF150(S) in those brain areas were calculated based on measured radioactivity concentrations of [11C]AF150(S) at 5.8 min and the speciic activity (SA) at time of injection. Solid lines are the non-linear its of the data using GraphPad Prism (version 4.00, San Diego, CA, USA). Possible metabolism of AF150(S) was not taken into account.

106

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors and 24%, respectively. These treatments did not have an apparent efect on the animal’s behaviour or body posture.

5.3.4 Co-injection with cold AF150(S) In Table 3 doses of co-injected cold AF150(S) and resulting SA of injected [11C]AF150(S) are shown. The aim of the experiment was to explore whether an AF150(S) concentration occupancy curve could be achieved. Efects of co-injection of cold AF150(S) on regional brain uptake of total AF150(S) (radioactive + cold calculated according to the SA of the injected sample), measured at 5.8 min after injection, are shown in Figure 3. The increase in brain concentration of total AF150(S) was larger than based on a linear relationship between injected dose and brain concentration. Possible metabolism of

AF150(S) was not taken into account. Calculated BPND values under baseline conditions and after co-administration of cold AF150(S) were not signiicantly diferent, see Table 4. Hence a saturation curve of speciic binding of AF150(S) could not be constructed. 5

Table 3. Overview of injected doses of [11C]AF150(S) and AF150(S). Speciic Activity of with Cold AF150(S) [11C]AF150(S) AF150(S) N cold diluted [11C]AF150(S) co-injected Injected radioactivity Total dose at injection per rat (MBq) (nmol.kg-1) (GBq.μmol-1) 0 12 16.9 ± 0.8 82.9 ± 8.4 0.7 ± 0.1 1 nmol 4 17.8 ± 0.3 6.4 ± 0.0 10.9 ± 0.3 5 nmol 4 15.7 ± 2.1 2.4 ± 0.3 21.7 ± 2.2 15 nmol 4 18.2 ± 1.1 1.1 ± 0.1 49.1 ± 1.2

11 Table 4. BPND of [ C]AF150(S) without and with co-injection of cold AF150(S).

+ 1 nmol of cold AF150(S) + 5 nmol of cold AF150(S) + 15 nmol of cold AF150(S) Brain region Baseline Co-injected Baseline Co-injected Baseline Co-injected Left Striatum 0.23±0.02 0.21±0.02 0.26±0.06 0.25±0.09 0.28±0.08 0.31±0.11 Right Striatum 0.24±0.01 0.23±0.02 0.26±0.06 0.26±0.09 0.30±0.07 0.32±0.11 Hippocampus 0.20±0.01 0.18±0.03 0.23±0.04 0.23±0.07 0.24±0.04 0.26±0.08 Frontal cortical 0.11±0.04 0.08±0.03 0.16±0.06 0.14±0.09 0.17±0.09 0.18±0.12 area Posterior cortical 0.08±0.05 0.06±0.04 0.16±0.05 0.13±0.07 0.14±0.07 0.13±0.09 area Statistical analysis: signiicance of diference between baseline and co-injected p ≥ 0.12

107

Chapter 5

5.4 dISCuSSIon

5.4.1 [11C]AF150(S) brain uptake under baseline conditions Under baseline conditions, brain kinetics of [11C]AF150(S) were very rapid with peak values being reached within 1 minute after injection, followed by a rapid decline. The calculated maximal brain concentration of [11C]AF150(S) of 2.9 nM remained ~70 times below the Kd.H (200 nM), of AF150(S) for M1ACh-R, under baseline conditions. In consecutive scans (baseline and following pre-treatment), the same synthesis batch of [11C] AF150(S) was used. As a consequence, the SA was about 8 times lower at the second injection, and to keep the injected amount of radioactivity constant, a larger dose of AF150(S) was injected. SUVr data showed increased retention in M1ACh-R rich areas. The rank order of uptake, striatum ≥ hippocampus > frontal cortical area > posterior cortical area >> cerebellum, corresponds with the rank order of M1ACh-R density measured in vitro in rat brain [153]. These indings also conirm previous biodistribution data with [11C]AF150(S) uptake measured ex vivo [246]. Note, that the parametric image under baseline conditions 11 clearly shows [ C]AF150(S) uptake in the hippocampus (Figure 2). BPND values could successfully be derived by applying SRTM referring to cerebellum, a region which is essentially devoid of M1ACh-R [153]. This particular reference model best itted the [11C] AF150(S) PET data. Furthermore a reference tissue model was chosen for data analysis, as no arterial input function could readily be obtained from rats.

5.4.2 Pre-treatment with muscarinic agents Pre-treatment conditions, i.e. time of injection and dose of the applied compounds, were chosen according to their maximal pharmacological activity in rats. Pre-treatment with the M1ACh-R selective antagonists pirenzepine and trihexyphenidyl caused a decrease 11 in BPND of [ C]AF150(S) in frontal and posterior cortical areas. In addition, pirenzepine caused a signiicant reduction in BPND in left striatum, and trihexyphenidyl in hippocampus, indicative of a reduction in speciic binding to M1ACh-R in these regions. The M4/M1ACh-R agonist xanomeline, selected to investigate agonist-agonist com- -1 petition, at a dose of 5 mg∙kg caused an increase in BPND in striatum and hippocampus, whereas a decrease would have been expected if [11C]AF150(S) and xanomeline are competing for the same binding site. Since GPCRs are dynamic structures that undergo conigurational changes upon agonist binding, several explanations are possible for this apparent anomalous inding. Xanomeline is reported to induce M1ACh-R activation via both orthosteric and ectopic binding sites [270,271]. Xanomeline induced activation of the ectopic site could cause increased M1ACh-R G-protein coupling by which more activated orthosteric sites become available for [11C]AF150(S) binding, leading to higher 11 BPND of [ C]AF150(S). An alternative explanation could be the efect of xanomeline on

108

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors

M4ACh-R. M4ACh-R is an autoreceptor, predominantly present in striatum and hippocampus [272]. Agonist stimulation of M4ACh-R leads to a decrease in acetylcholine release [273], potentially resulting in reduced competition between endogenous extracellular acetylcholine and [11C]AF150(S), which in turn may result in increased binding of [11C]AF150(S). However, a higher dose of xanomeline (30 mg∙kg-1) that caused severe peripheral and cardiac efects, and glandular secretion, did not show any signiicant 11 efect on BPND of [ C]AF150(S) in M1ACh-R rich areas. Therefore the observed efects of xanomeline seem to be dose dependent and pharmacokinetic efects may play a role. In vitro experiments, have demonstrated partial agonist/antagonist efects of AF150(S) on M3ACh-R (Fisher et al., unpublished results). M3ACh-R are present in glands and can also be found on smooth muscles of blood vessels and at low density in the brain [274]. Darifenacin, an M3 antagonist with central activity, was selected to check for possible labelling of M3ACh-R by [11C]AF150(S). Pre-treatment with darifenacin, however, did not signiicantly reduce BP of [11C]AF150(S) in any brain region. Hence, [11C]AF150(S) 5 ND appears not to show measurable binding to M3ACh-R in the brain in vivo.

5.4.3 Pre-treatment with agents afecting extracellular ACh-levels Pre-treatment conditions, i.e. time of injection and dosing, were chosen to give maximal ACh release at 5.8 min post intravenous injection of [11C]AF150(S), in order to have maximal competition. Haloperidol pre-treatment causes blockade of D2 receptors, which in turn leads to reduced inhibition of the cholinergic system, in particular in striatum and hippocampus. This results in increased extracellular levels of acetylcholine [275]. In- crease in ACh release following haloperidol treatment, is demonstrated in in vivo microdialysis studies [276]. Further evidence for increased cholinergic activity is apparent in behavioural studies, where haloperidol causes severe catalepsy, as a result of cholinergic over-activity in the striatum [277]. Also in this study, at the applied dose, haloperidol caused immobility and severe rigidity in the animals. The signiicant reductions in BPND of [11C]AF150(S) in right striatum (-27%) and hippocampus (-15%) following haloperidol treatment can likely be ascribed to increased extracellular ACh levels, which may have competed with [11C]AF150(S) for M1ACh-R binding in these particular brain areas. AF-DX 384, is an M2/M4 selective muscarinic antagonist that acts on M2 autoreceptors located presynaptically on cholinergic nerve terminals. Blockade of the M2 autoreceptor will cause increased release of acetylcholine [278], which could cause competition between ACh and [11C]AF150(S) for the M1ACh-R, located postsynaptically. In microdialysis studies in rats, AF-DX 384 given IP at a dose of 5 mg·kg-1, resulted in a long-lasting signiicant increase in acetylcholine release, 30 minutes after administration, in both cortex and hippocampus [266]. In those studies breakdown of extracellular ACh was prevented by adding the AChE inhibitor, physostigmine, in the microdialysis perfusate. In the present study AChE inhibition as protection against ACh breakdown, was attempted by

109

Chapter 5 using co-treatment with rivastigmine [267]. AF-DX 384 and rivastigmine, at the present dose, did not cause animal immobility or catalepsy and such efect on behaviour by AF- DX 384 treatment, has neither been reported in the literature. Therefore, the efect-size of M2ACh-R blockade on acetylcholine release probably is much less pronounced than that of haloperidol. Pre-treatment with only AF-DX 384 also did not have an efect on

BPND in M1ACh-R rich brain areas. Pre-treatment with AF-DX 384 combined with rivastigmine, surprisingly, resulted in 11 signiicant increases in BPND of [ C]AF150(S) in M1ACh-R rich brain areas, e.g. right striatum, hippocampus and frontal cortical area. A tentative explanation for these observed increases in BPND could be agonist mediated changes in M1ACh-R ainity/availability. 11 This has been proposed for the dopamine D2 receptor radioligand [ C]raclopride which demonstrated increased BP after pre-treatment with L-dopa [279]. Unfortunately, experimental evidence to explain such phenomena is hard to obtain. Nevertheless, [11C] AF150 binding appears to be sensitive to changes in extracellular ACh levels, in particular when behavioural efects are apparent. Ideally, and in analogy with the dopamine D2 agonist PET ligand studies, the sensitivity of [11C]AF150(S) to changes in extracellular ACh should be assessed in direct comparison with an M1ACh-R antagonist PET ligand to investigate diference in sensitivity between them. Indeed for the dopamine D2 agonist PET ligands [11C]MNPA, [11C]NPA and [11C](+)-PHNO a larger reduction in binding, by increased endogenous dopamine levels following amphetamine pre-treatment, was found as compared to the antagonist PET ligand [11C]raclopride [75,76,88,280]. This can be taken as indication that the agonist PET ligands preferentially label the G-protein- coupled receptors that are the target for the endogenous neurotransmitter. The present observations, warrant further studies, preferentially in larger animals, to investigate and further characterize the sensitivity of speciic [11C]AF150(S) binding in brain regions to alterations in extracellular ACh levels.

5.4.4 Co-injection of [11C]AF150(S) with cold AF150(S) Increasing amounts of co-injected cold AF150(S) resulted in a more than linear increase in corresponding brain concentrations of total AF150(S) in M1ACh-R rich brain areas (Figure 3), whereas in M1ACh-R poor brain areas, such as cerebellum, this efect was less pronounced. After co-administration of the highest dose of cold AF150(S), BPND values were slightly increased compared to baseline conditions, yet due to variability, these values were statistically not signiicantly diferent (Table 4). The more than linear increase in brain concentrations of [11C]AF150(S) in M1ACh-R rich brain areas is probably not due to difusion, but a result of facilitated transport of AF150(S) in the brain, that was hypothesized from analyses of results in our previous study of [11C]AF150(S) uptake into the brain ex vivo [246]. The hypothesis was mainly based on (1) the measured low 11 lipophilicity of [ C]AF150(S), logDpH 7.4 = 0.05, which hampers difusion through lipid cell

110

[11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors membranes, (2) the 8x higher brain as compared to plasma levels of [11C]AF150(S) and (3) the structural similarity between AF150(S) and nicotine, for which facilitated transport into the brain has been demonstrated. The highest administered molar dose of non-radiolabelled AF150(S) (~50 nmol·kg-1) resulted in a brain concentration of ~100 nM AF150(S) in vivo, a value that still lays 50

% under the in vitro Kd,H of AF150(S) (200 nM). At the highest applied dose no sign of saturation in binding to M1ACh-R was observed, indeed, the latter is only expected to occur at a concentration of 4x Kd,H. Therefore, saturation should be investigated using a substantial higher dose of non-radiolabelled AF150(S).

5.4.5 Further evaluation of [11C]AF150(S) as a potential agonist PET ligand for M1ACh-R The indings in this preliminary PET study with [11C]AF150(S) in rat are indicative of the potential of [11C]AF150(S) for speciic labelling of M1ACh-R. However, PET studies in rat 5 sufer from several limitations such as the diiculty of getting an arterial input function, the small brain regions and the resolution of the PET camera. Therefore, [11C]AF150(S) should be further explored in PET studies in primates and humans. The chances of success and merit of such studies are supported by the following. M1ACh-R density in primate and human brain is about 50 % higher than in rat brain, according to in vitro autoradiography studies [257]. No species diferences in the binding ainity of compounds for M1ACh-R between varies species including rat, primate and human have become apparent [281], in this respect, it was found that the potency of AF150(S) to stimulate human M1ACh-R expressed in cells fully matches its binding ainity for M1ACh-R in rat brain homogenates (personal observations, manuscript in preparation). AF150(S) shows a beneicial safety proile and close congeners of AF150(S) have been in clinical studies and are being investigated as potential treatment (symptomatic and disease modiication) for Alzheimer’s disease [262].

5.5 ConCLuSIonS

The agonist radioligand [11C]AF150(S) was rapidly taken up in brain and showed fast kinetics with a signiicant M1ACh-R related signal in brain areas that are rich in M1ACh-R. Moreover, binding of the agonist PET ligand [11C]AF150(S), appears to be sensitive to changes in extracellular ACh levels. Further preclinical and clinical studies are needed to evaluate the full potential of [11C]AF150(S) for imaging the active pool of M1ACh-R in vivo.

111

6 Radiosynthesis and preclinical evaluation of [11C]prucalopride as a potential agonist

PET ligand for the 5-HT4 receptor

Hans J.C. Buiter, Albert D. Windhorst, Marc C. Huisman, Joris H. De Maeyer, Jan A.J. Schuurkes, Adriaan A. Lammertsma and Josée E. Leysen

Published in: European Journal of Nuclear Medicine and Molecular Imaging Research 3 (2013) doi:10.1186/2191-219X-3-24

Chapter 6

ABSTRACT

Background: Serotonin 5-HT4 receptor (5-HT4-R) agonists are potential therapeutic agents for enterokinetic and cognitive disorders and are marketed for treatment of constipation. The aim of this study was to develop an agonist positron emission tomography

(PET) ligand in order to label the active G-protein coupled 5-HT4-R in peripheral and central tissues. For this purpose prucalopride, a high ainity, selective 5-HT4-R agonist, was selected.

Methods: [11C]prucalopride was synthesized from [11C]methyl trilate and desmethyl prucalopride, and its LogDoct,pH7.4 was determined. Three distinct studies were performed with [11C]prucalopride, administered IV, in male rats. (1) The biodistribution of radioactivity was measured ex vivo. (2) Kinetics of radioactivity levels in brain regions and peripheral organs was assessed in vivo under baseline conditions and following pre- treatment with tariquidar, a P-glycoprotein elux pump inhibitor. (3) In vivo stability of [11C]prucalopride was checked ex vivo in plasma and brain extracts using HPLC.

Results & discussion: [11C]prucalopride was synthesized in optimised conditions with a yield of 21 ± 4% (decay corrected) and a radiochemical purity (>99%), its LogDoct,pH7.4 was 0.87. Ex vivo biodistribution studies with [11C]prucalopride in rats showed very low levels of radioactivity in brain (maximal 0.13 %ID·g-1) and 10 times higher levels in certain peripheral tissues. The PET studies conirmed very low brain levels of radioactivity under baseline conditions, however it was 3 times increased after pre-treatment with tariquidar. [11C]Prucalopride was found to be very rapidly metabolized in 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 pump in the gut.

Conclusion: [11C]Prucalopride demonstrated low radioactivity levels in rat brain, a combination of reasons may include rapid metabolism in the rat in particular, low passive difusion, potential P-glycoprotein substrate. Further investigation of [11C]prucalopride for imaging the active state of 5-HT4-R, in humans, is worthwhile in view of therapeutic applications of 5-HT4 agonists for treatment of GI motility disorders.

114

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor

6.1 InTRoduCTIon

In mammalian species, serotonin or 5-hydroxytryptamine (5-HT), acts as a neurotransmitter and paracrine agent, that mediates a wide variety of functions, including cognitive and emotional processes, regulation of sleep and food intake and cardiovascular and gastrointestinal mechanisms. Serotonergic neurons originate in the raphé nuclei of the brain stem and project widely to forebrain, hind brain and spinal cord. 5-HT is also synthesized in enterochromain cells in the gut. The latter contain 90% of the 5-HT in the body, from where it is released in the blood to exert a paracrine actions. To date 14 diferent 5-HT receptors, classiied into 7 subclasses, have been identiied The 5-HT1A/1B/1D/1E/1F,

5-HT2A/2B/2C, 5-HT4, 5-ht5A/5B, 5-HT6 and 5-HT7 receptors are G-protein coupled receptors

(GPCR), the 5-HT3 receptor belongs to the class of ligand gated ion channels [282].

The 5-HT4 receptor (5-HT4-R) is of interest for its role in the central nervous system

(CNS), and in peripheral tissues. For the latter, 5-HT4-R agonists are applied therapeuti- cally to treat laxative-resistant constipation [283], and in the CNS, 5-HT4-R agonists have been shown to improve memory, and cognition in animal models [284,285]. In the human brain 5-HT -Rs have been localized in the basal ganglia, the hippocampal 6 4 formation, and the cortical mantle [29]. A non-invasive molecular imaging technique such as positron emission tomography 11 (PET) could be useful to explore the function of 5-HT4-R in vivo. To date, only [ C]

SB207145, a high ainity 5-HT4-R antagonist [286,287], has been evaluated in large mammals and man. In general, successful PET ligands for GPCRs are antagonists, nevertheless, antagonists have an important disadvantage. GPCRs are known to occur in interconvertible active G-protein coupled states and in inactive uncoupled or desensitized states. Antagonists show equal ainity for these diferent GPCR states and consequently do not distinguish between the active and the inactive receptor. In contrast, agonists have high-ainity for the active G-protein coupled receptors and low ainity for inactive uncoupled receptors. Therefore, agonist PET ligands will preferentially label the activated receptors [288].

Prucalopride is a potent, selective 5-HT4-R agonist [289], currently marketed for human use for the treatment of laxative-resistant constipation [283]. Studies with [3H] prucalopride demonstrated favourable radioligand binding properties in vitro [29]. 3 Ainity was high (KD = 2 nM) and using autoradiography in vitro [ H]prucalopride clearly labelled 5-HT4-R in striatum, hippocampus, frontal cortex and substantia nigra in human brain hemispheres cryosections [29]. Comparison of 5-HT4-R densities measured with the agonist [3H]prucalopride and with the antagonist [3H]R116712 in various brain regions revealed that Bmax values measured with the former, represented 16 to 54% of the Bmax values of the latter [29]. This indicates that the active G-protein coupled 5-HT4- R state may vary between various brain regions and potentially also between peripheral

115

Chapter 6

tissues. This illustrates the importance of developing an agonist 5-HT4-R PET ligand that would allow investigating the active 5-HT4-R state in vivo. The aim of the present study was to label prucalopride with carbon-11, optimize its radiosynthesis and investigate [11C]prucalopride in biodistribution ex vivo and in vivo studies in rats.

6.2 METHodS And MATERIALS

8.2.1 Chemicals Prucalopride succinate (4-amino-5-chloro-N-[1-(3-methoxypropyl)piperidin-4-yl]-2,3- dihydro-1-benzofuran-7-carboxamide monobutanedioate) and desmethyl prucalopride (4-amino-5-chloro-N-[1-(3-hydroxypropyl)piperidin-4-yl]-2,3-dihydro-1-benzofuran-7- carboxamide) were provided by Shire-Movetis N.V. (Turnhout, Belgium). Tariquidar, a P-glycoprotein drug elux pump inhibitor [290-293], was obtained from Haupt Pharma Wüling GmbH (Gronau, Germany). All other reagents were from Merck (Schiphol-Rijk, the Netherlands) or Sigma-Aldrich (Zwijndrecht, the Netherlands). They were of analytical grade and used without further puriication. High-performance liquid chromatography (HPLC) solvents (gradient grade) were purchased from Mallinckrodt-Baker B.V. (Deventer, the Netherlands).

6.2.2 HPLC The semi-preparative HPLC system consisted of a Jasco (Ishikawa-cho, Japan) PU-1587 HPLC pump, a six-way VICI injector (VA EPC6W, VICI AG International, Schenkon, Swit- zerland) with a 5 mL loop, a Jasco UV1575 UV detector, a custom made radioactivity detector and a Phenomenex (Torrance CA, USA) Synergi 10μm Hydro-RP 80 C18 250x10 mm HPLC column and an eluent of methanol/0.1 M Phosphate bufer (pH 4.3) 34/66 (v/v) was used with a low of 6 mL·min-1. Radioactivity was measured using a Veenstra (Joure, the Netherlands) VDC-405 dose calibrator. The analytical HPLC system consisted of a Jasco PU-1580 HPLC pump, a Rheodyne 7724I injector (IDEX Health & Science, Wertheim-Mondfeld, Germany) with a 20 μL loop, a Jasco UV-2075 Plus UV detector, a NaI radioactivity detector (Raytest, Straubenhardt, Germany) and a Phenomenex Gemini 5 µm C18 150x4.6 mm column and an eluent of methanol/0.1 M Phosphate bufer (pH 4.3) 30/70 (v/v) was used with at a low of 1 mL·min-1. The peak of parent prucalopride was identiied by co-injection of a sample of cold prucalopride.

6.2.3 Preparation of [11C]prucalopride The automated radiolabelling [204] of [11C]prucalopride was performed by addition of [11C]methyltrilate [294] to a solution of 350 μL acetonitrile containing 1.0 mg (2.7 μmol)

116

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor des-methyl prucalopride and 1.9 μL (4.3 μmol) tetrabutylammoniumhydroxide (60% solution in water) at -25 °C. After distillation and trapping [11C]methyltrilate, the temperature was raised to 85 °C for 5 min, after which the reaction mixture was quenched with 0.4 mL of phosphate bufer (0.1 M, pH 4.3) and diluted with 1.5 mL of HPLC eluent. This mixture was subjected to semi-preparative HPLC puriication. The fraction containing [11C]prucalopride was collected and diluted with 60 mL of sterile water. The product was trapped by solid-phase extraction using a pre-conditioned (10 mL of ethanol followed by 10 mL of sterile water) Waters tC18 Plus Sep Pak® cartridge (Waters Chromatography B.V., Etten-Leur, the Netherlands), which subsequently was washed with 20 mL sterile water. Next, the product was eluted from the Sep Pak cartridge with 1 mL ethanol followed by 9 mL saline containing 7.09 mM NaH2PO4 and iltered through a sterile Millex GV 0.22 μm membrane ilter (Millipore B.V., Amsterdam, the Netherlands), using helium overpressure. Devices used for the automated radiosyntheses were home- made [204]. The purity of the product was analysed using the analytical HPLC system and speciic activity (SA) was determined by HPLC based on a, calibration curve (performed in triplicate with 4 concentrations of parent prucalopride in the range 1.10-4 – 1.10-7 mol.L-1) and measurement of the ultraviolet (UV) signal. 6

11 6.2.4 determination of the Logdoct,pH7.4 value of [ C]prucalopride The distribution of [11C]prucalopride between 1-octanol and 0.2 M phosphate bufer (pH 7.4) was measured in triplicate at room temperature by adapting a method previously described [210]. Briely, 1 mL of a 10 MBq·mL-1 solution of [11C]prucalopride in 0.2 M phosphate bufer (pH = 7.4) was mixed vigorously with 1 mL 1-octanol for 1 min at room temperature, using a Vortex. After centrifugation for 5 min at 4000 rpm and a settling period of 30 min, ive aliquots of 100 μL were taken from both layers, carefully avoiding cross contamination between the phases. Five aliquots of 100 μL of the 10 MBq·mL-1 solution of [11C]prucalopride in 0.2 M phosphate bufer (pH = 7.4) were taken as reference for determining recovery. All aliquots were counted for radioactivity using an automated gamma counter, Wallac 1282 Compugamma CS (LKB Wallac, Turku, Finland). 10 The LogDoct,pH7.4 value was calculated according to: LogDoct,pH7.4 = Log(Aoct/Abufer), with

Abufer as the average radioactivity of 5 bufer samples and Aoct as the average radioactivity of 5 1-octanol samples.

6.2.5 drug solutions [11C]prucalopride and [18F]NaF, prepared as described previously [263], were diluted with saline to prepare an isotonic, sterile and pyrogen free solution, for IV injection. Radio- chemical purity was >99%.

117

Chapter 6

A tariquidar solution in 20% ethanol was diluted with 5% glucose in saline to a concentration of 7.5 mg/mL for IV injection. The inal concentration of ethanol was 10% for the PET studies.

6.2.6 Animals Experiments were performed using male Wistar rats (265 ± 15 g; Harlan Netherlands B.V. Horst, the Netherlands). Rats were kept in conditioned housing under a regular light/ dark cycle (12/12 h) and allowed food and water ad libitum. All animal experiments were in compliance with Dutch law and approved by the VU University Animal Ethics Com- mittee.

6.2.7 [11C]prucalopride biodistribution The biodistribution of radioactivity following IV injection of [11C]prucalopride was measured in male Wistar rats using a modiication of the method described by Airaksinen et al. [213]. Under 2% isolurane anaesthesia, 4 groups of 4 rats received an intravenous injection (tail vein) each of [11C]prucalopride (53 ± 2 MBq at the start of the experiment). Rats (4 per time point) were sacriiced, under anaesthesia, by cervical dislocation at 5, 15, 30 and 60 min post injection. Whole blood was collected by cardiac puncture, and heart, lungs, liver, colon, kidneys and brain were excised. The following brain regions were dissected: olfactory bulb, hippocampus, striatum, frontal and posterior cortex (including occipital cortex), thalamic region, medulla oblongata, cerebellum and the rest of the brain. All organs and brain regions were weighed and counted for radioactivity using an automated gamma counter. Values were corrected for decay and the %ID·g-1 for every tissue was calculated as percentage of the total injected radioactivity dose, divided by the weight of the tissue.

6.2.8 In vivo stability of [11C]prucalopride Wistar rats (2 groups of 3) received an intravenous injection of [11C]prucalopride (100 ± 40 MBq at the start of the experiment) via the tail vein, under 2% isolurane anaesthesia. At 5 and 30 min post injection, whole blood was collected by cardiac puncture and the brain was excised. Whole blood was centrifuged for 5 min at 4000 rpm, and a 1 mL plasma aliquot was taken and acidiied using 10 μL 5 M HCl and diluted with 2 mL of water. The excised brain was washed in ice-cold saline and subsequently one hemisphere was homogenized in 12 mL ice-cold saline and centrifuged for 5 min at 4000 rpm. Plasma and brain extract were passed over a Waters tC18 Plus Sep Pak® cartridge (1 g, pre-conditioned with 10 mL ethanol followed by 10 mL water), washed with 5 mL water and eluted with 2 mL acidic methanol. All fractions, including the Sep Pak® cartridge, were counted for radioactivity using an automatic gamma counter. The

118

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor methanol fraction was analysed using the semi-preparative HPLC method for the presence of parent [11C]prucalopride.

6.2.9 [11C]prucalopride PET studies

6.2.9.1 PET system PET measurements were performed using an HRRT (CTI/Siemens, Knoxville, TN, USA) scanner. The HRRT is a dedicated human brain PET scanner, with design features that enable high spatial resolution combined with high sensitivity, making it also suitable for small animal imaging. This scanner has a transaxial ield of view of 312 mm and an axial ield of view of 250 mm. The spatial resolution ranges from 2.3 to 3.2 mm full width at half maximum (FWHM) in transaxial direction and from 2.5 to 3.4 mm FWHM in axial direction [27].

6.2.9.2 Rat treatment and PET measurements Anaesthesia was induced and maintained by constant insulation of 1.5 - 2% isolurane in pure oxygen into the nose. Rats were placed in a ixation device with tooth bar to 6 secure a ixed and immobile horizontal position of the head during scanning. Body temperature was kept constant at 37 °C with a heating-pad coupled to a thermostat which was connected to a rectal thermometer. A cannula was inserted into the vena femoralis for radiotracer injection. Transmission measurements of 6 min duration were performed using a 740 MBq 2D fan-collimated 137Cs moving point source. A bolus of 10.6 ± 2.6 MBq of [11C]prucalopride was injected IV. A 3D emission acquisition of 45 min duration was started immediately prior to the IV injection. Four rats underwent two consecutive PET scans with a 30 min interval. The irst PET scan was performed under baseline conditions, the second PET scan was performed following pre-treatment with tariquidar, 15 mg·kg-1 IV given as slow bolus, 30 min prior to the [11C] prucalopride injection. At the end of the [11C]prucalopride PET scans, [18F]NaF (~15 MBq) was injected and after 30 min a 30 min emission scan was acquired.

6.2.9.3 Image acquisition and reconstruction Acquired [11C]prucalopride data were stored in 64-bit list mode format and were subsequently histogrammed into 21 time frames of 7 x 10, 1 x 20, 2 x 30, 2 x 60, 2 x 150, 7 x 300 s duration. [18F]NaF data were stored as a single 30 min frame. All data were normalized and corrected for attenuation, randoms, scatter, decay and dead-time. Data were reconstructed using 3D ordered subsets weighted least squares (3D-OSWLS) [247] using 7 iterations and 16 subsets. The image matrix was 256 x 256 x 207 voxels with a voxel size of 1.218 x 1.218 x 1.218 mm3.

119

Chapter 6

6.2.9.4 PET data analysis

Region of interest analysis [18F]NaF scans of the brain were co-registered with a standard MR template to delineate regions of interest (ROIs) as previously described [268]. The following ROIs were used: striatum (30 mm3), hippocampus (27 mm3), frontal plus parietal cortical area (49 mm3), posterior plus occipital cortical area (49 mm3), and cerebellum (86 mm3). Peripheral ROIs of heart, liver, small intestine (jejunum and ileum), colon and the caecum were manually drawn directly into the dynamic PET images. ROIs were transferred onto the dynamic PET images, and regional time activity curves (TACs in kBq·mL-1) were generated [269].

Binding potential quantiication The simpliied reference tissue model (SRTM) [37], with cerebellum as reference tissue, was used to calculate non-displaceable binding potential (BPND) as outcome measure of speciic binding in the analyzed brain regions [35].

Statistical analysis

Diferences in BPND values between baseline and post tariquidar scans were tested using a general Linear Mixed Model in SPSS Statistics (version 17.0; SPSS Inc., Chicago, IL, USA). All results are expressed as mean ± standard deviation (SD) and values of p < 0.05 were considered to be statistically signiicant.

6.3 RESuLTS

6.3.1 Radiolabelling and automated synthesis of [11C]prucalopride Optimized radiosynthesis of [11C]prucalopride (Figure 1) followed by preparative puriication on HPLC, isolation by solid phase extraction and formulation, yielded 1.1 to 1.3 GBq of formulated [11C]prucalopride. Total preparation time was 45 ± 5 min, yield 21 - 25%, radiochemical purity >99% and speciic activity (SA) 52 ± 19 GBq·μmol-1. Radiochemical yields for several reaction conditions and parameters investigated are presented in Table 1. These yields are decay corrected and either determined by analysing samples taken from the reaction mixture (analytical yields) or from activities at the start of synthesis and in the formulated product.

The methylation reaction occurred with signiicant formation of side-product. In addition, there was a signiicant amount of unreacted [11C]methyl trilate (Figure 2). Formation of [11C]prucalopride was conirmed by analytical HPLC using co-injection of

120

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor reference prucalopride. Representative chromatograms of both semi-preparative- and analytical HPLC puriication are shown in Figure 2.

Figure 1. Chemical structures of desmethyl prucalopride and [11C]prucalopride. Optimized reaction 11 conditions and reagents for the radiosynthesis: [ C]methyl trilate, CH3CN, TBAOH, 5 min, 85 °C. 6

Table 1. Overview of parameters investigated to optimize radiosynthesis of [11C]prucalopride. Amount desmethyl Base Amount Temp Methylation Yield Yield N prucalopride Base (°C) agent (%) (μmol) (μmol)

11 3.0 NaOH 25.0 25 [ C]CH3I 0 Analytical 1 11 3.0 NaOH 25.0 85 [ C]CH3I 1 Analytical 1 11 3.0 NaOH 5.0 25 [ C]CH3I 6 Analytical 1 11 3.0 NaOH 5.0 85 [ C]CH3I 7 Analytical 1 11 2.2 TBAOH 5.1 25 [ C]CH3I 11±1 Analytical 2 11 2.2 TBAOH 5.1 85 [ C]CH3I 12±1 Analytical 2 3.0 TBAOH 6.9 85 [11C]methyl 12±0 Analytical 2 trilate 2.2 TBAOH 4.4 85 [11C]methyl 16±0 Prep. 2 trilate 3.5 TBAOH 5.3 85 [11C]methyl 31±2 Prep. 2 trilate 2.2 TBAOH 3.5 85 [11C]methyl 32±1 Prep. 2 trilate 2.7 TBAOH 4.3 85 [11C]methyl 34±3 Prep. 14 trilate

For every reaction CH3CN was used as solvent, the reaction volume was 350 μL and the reaction time was 5 min.

121

Chapter 6

Figure 2. Representative HPLC chromatograms of the puriication of the reaction mixture, and formulated [11C]prucalopride. A: Chromatograms of the semi-preparative HPLC puriication of the reaction mixture. A1: UV vs. time; A2: radioactivity vs. time. Retention times were 8 min for the precursor and 11 min for [11C]prucalopride. B: Analytical HPLC chromatograms of formulated [11C]prucalopride vs. time. B1: UV vs. time; B2: radioactivity vs. time; B3: UV spiked with cold prucalopride. Retention times were 8.5, 8.8 and 8.5 min for UV, radioactivity signal of [11C]prucalopride, and UV spiked with cold prucalopride, respectively. UV and radioactivity detection were connected in series, giving a time delay of 0.3 min.

122

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor

11 6.3.2 Logdoct pH7.4 value of [ C]prucalopride 11 The measured LogDoct,pH7.4 value of [ C]prucalopride was 0.870 ± 0.004 (n = 3). The recovery of radioactivity in the sum of 1-octanol and phosphate bufer fractions was 94.1 ± 1.7%.

6.3.3 [11C]prucalopride biodistribution studies The biodistribution of radioactivity after IV injection of [11C]prucalopride in rats was determined ex vivo in brain regions, peripheral organs and blood at 5, 15, 30 and 60 min. Results are presented in Table 2. At 5 min post IV injection of [11C]prucalopride, low levels of radioactivity in brain were observed, the highest %ID·g-1 value being 0.13 in the olfactory bulb. Levels of radioactivity in olfactory bulb, striatum, and hippocampus were higher than in cortical areas, thalamus, medulla oblongata, cerebellum and the rest of the brain. Over time, radioactivity in brain areas luctuated, over the sixty minutes, a decrease of 60 – 70% was observed. In peripheral tissues, radioactivity levels at 5 min were highest in kidney followed by liver and lung, the level was lower in colon and very low in heart and blood. Radioactivity in kidney and lung was reduced to 1/3 at 15 min. Disappearance of radioactivity with time was slower for other tissues and in particular 6 in the liver.

Table 2. Distribution of radioactivity in several brain regions and peripheral organs in rats, measured ex vivo at various time points following IV injection of 53 ± 2 MBq (SA 42 ± 8 GBq·μmol-1) of [11C] prucalopride*. %ID/g (mean ± SD, n = 4 per time point) Organ/Region 5 min 15 min 30 min 60 min Olfactory bulb 0.13 ± 0.04 0.05 ± 0.03 0.10 ± 0.05 0.04 ± 0.01 Striatum 0.07 ± 0.01 0.05 ± 0.02 0.08 ± 0.05 0.04 ± 0.01 Hippocampus 0.09 ± 0.02 0.05 ± 0.02 0.05 ± 0.01 0.03 ± 0.01 Medulla oblongata 0.06 ± 0.00 0.03 ± 0.01 0.05 ± 0.00 0.03 ± 0.00 Frontal cortex 0.06 ± 0.00 0.03 ± 0.02 0.03 ± 0.00 0.02 ± 0.00 Thalamic region 0.05 ± 0.00 0.03 ± 0.01 0.03 ± 0.01 0.02 ± 0.01 Posterior cortex 0.06 ± 0.01 0.04 ± 0.02 0.03 ± 0.01 0.03 ± 0.01 Rest of the brain 0.06 ± 0.00 0.03 ± 0.02 0.05 ± 0.00 0.03 ± 0.00 Cerebellum 0.08 ± 0.01 0.04 ± 0.02 0.04 ± 0.00 0.02 ±0.00 Liver 1.60 ± 0.58 1.32 ± 0.59 1.53 ± 0.19 0.83 ± 0.05 Kidney 2.57 ± 0.24 1.06 ± 0.49 0.79 ± 0.10 0.43 ± 0.04 Lungs 1.55 ± 0.61 0.51 ± 0.23 0.39 ± 0.13 0.24 ± 0.04 Colon 1.01 ± 0.29 0.54 ± 0.26 0.32 ± 0.07 0.22 ± 0.03 Heart 0.39 ±0.04 0.21 ± 0.10 0.19 ± 0.01 0.10 ± 0.01 Blood 0.16 ± 0.01 0.08 ± 0.04 0.10 ± 0.01 0.06 ± 0.00 * Tissue samples were not lushed with saline prior to the radioactivity measurement. Experiments were performed on two separate days, day 1: 5 and 15 min, day 2: 30 and 60 min

123

Chapter 6

6.3.4 In vivo stability of [11C]prucalopride In vivo stability of [11C]prucalopride was measured at 5 and 30 min post IV injection of [11C]prucalopride, by analysing the presence of parent compound in plasma and brain methanol extracts using HPLC. The complete procedure was performed with a radioactivity recovery of >90%. Most of the radioactivity was recovered in the water fraction of the Seppak, which would be polar radiolabelled metabolites of [11C]prucalopride, however, these products were not identiied. The methanol extract of the Seppak column, expected to contain parent [11C]prucalopride contained no parent compound for both plasma and brain extracts taken at 5 and 30 min post injection.

6.3.5 [11C]prucalopride PET studies in rats Following IV injection of [11C]prucalopride at baseline conditions, all rats showed low cerebral levels of radioactivity. Radioactivity levels were highest at 30 s, and, for analyzed brain regions, corresponded to a SUV of about 0.6, which was declined to ≤ 0.3 at 40 min (Figure 3).

Figure 4. Summed (0 – 20 min after injection) [11C]prucalopride images. Baseline (left), baseline co- registered with MR template (middle) and following tariquidar (right) (15 mg·kg-1 IV) predosing. Arrows indicate (a) eye, (b) Harderian gland and (c) brain.

11 Table 3. BPND for [ C]prucalopride in rat brain regions versus cerebellum, at baseline and following pre- treatment with tariquidar (n=4 for both conditions). Tariquidar Brain region Baseline Pre-treatment Striatum 0.00±0.08 0.07±0.09** Hippocampus -0.05±0.05 0.07±0.05* Frontal cortex 0.04±0.07 0.03±0.04 Posterior cortex 0.00±0.05 0.03±0.04 Signiicant diference compared with baseline: *p<0.05, **p<0.01,

124

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor

In all animals, tariquidar pre-treatment resulted in higher cerebral concentrations of radioactivity following the IV injection of [11C]prucalopride than at baseline. In fact, brain radioactivity concentrations were approximately three-fold higher in all brain regions with a peak SUV of 1.3 ± 0.2, which subsequently declined to SUVs ≤ 1 within 20 min. (Figure 3). Representative PET images are shown in Figure 4.

Figure 3. Time-activity curves in various brain regions, measured using PET, following IV injection of 15±4 MBq [11C]prucalopride (SA 65±5GBq.μmol-1). Open triangles (with dashed line) represent SUVs at baseline, solid triangles (with solid line) represent SUVs after tariquidar pre-treatment (15 mg.kg-1 IV). In both cases n=4.

125

Chapter 6

The BPND values for striatum, hippocampus, frontal cortex, posterior cortex and medulla oblongata, using cerebellum as reference tissue are, shown in Table 3. At baseline, BPND values were essentially zero. Following tariquidar pre-treatment, however, positive BPND values were obtained for striatum and hippocampus.

Representative whole body PET images following IV injection of [11C]prucalopride at baseline and after pre-treatment with tariquidar (15 mg·kg-1 IV) are shown in Figure 5. High levels of radioactivity over time (summed image 0 - 45 min) wereseen in bladder, liver, jejunum, colon and caecum (the beginning of the colon). Time activity curves based on SUV values in selected tissues are shown in igure 6. The hearth showed high SUV values (maximal SUVs of 10.4 ± 1.3) only within the irst minute after IV injection of [11C]prucalopride, where after SUV values were <1. Tariquidar pre-treatment did not af- fect the time activity curve of SUV values in the heart. In liver a maximal SUV of 4.6 ± 0.8 was observed at 8.75 min post injection which slowly declined over time. Pre-treatment with tariquidar aforded lower maximal SUVs of 3.6±0.4 which declined to 2.8±0.3 at the

Figure 5. Summed (0 – 45 min after injection) whole body PET images following IV injection of [11C] prucalopride in rat. Baseline (left) and after pre-treatment (right) with tariquidar (15 mg·kg-1 IV); Arrows indicate: a) brain region, b) lungs, c) spine, d) stomach, e) liver, f) jejunum, g) caecum and h) bladder.

126

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor

Figure 6. Time activity curves in selected peripheral tissues following IV injection of 15±4 MBq [11C] prucalopride (SA 65 ± 5 GBq.μmol-1). Open triangles (with dashed line) represents SUVs at baseline, solid triangles (with solid line) represents SUVs after tariquidar pre-treatment (15 mg·kg-1 IV). Data represent mean ± SD (n = 4).

end of the scan. Radioactivity levels in small intestine were initially higher for baseline compared to post tariquidar, after 12.5 min SUVs were ~3.3 under both conditions and remained constant over time. For the colon, levels of radioactivity under baseline were maximal at 17.5 min with SUVs of 3.1±1.6 which declined to SUVs of 2.5±0.4 at the end of the scan. Post tariquidar colon SUVs were ~1.8 over time. The ROI placed on the caecum showed that SUVs, at baseline conditions increased to maximal SUVs of 9.6 ± 6.2 at 27.5 min post injection. Post tariquidar measurements aforded SUVs of ~3.5, which remained constant over time.

127

Chapter 6

6.4 dISCuSSIon

6.4.1 Radiolabelling and automated synthesis

11 11 11 Radiolabelling of [ C]prucalopride was performed using both [ C]CH3I and [ C]methyl 11 trilate. Using NaOH as base, [ C]CH3 incorporation was low and labelled side products (80-90%) were formed primarily. Using TBAOH as base, resulted in less side products 11 and incorporation of [ C]CH3 of 7 – 11%, possibly due to reduced deprotonation of the secondary amine group of the precursor. Using [11C]methyl trilate and still lower 11 amounts of TBAOH further increased [ C]CH3 incorporation. Optimal conditions were achieved with 1.5 to 1.6 molar equivalents of base, resulting in an incorporation yield of 11 30 – 36% of [ C]CH3. Using these optimal conditions, automated radiosynthesis resulted in a formulated product with a total yield of 21 ± 4% (decay corrected) at the end of an overall preparation time of 45 ± 5 min, with good speciic activity and high radiochemical purity.

11 6.4.2 Logdoct, pH7.4 value of [ C]prucalopride LogD was used as parameter for lipophilicity rather than LogP, as LogP relates to lipophilicity for the charge-neutral form of the radioligand only, whilst LogD takes into account the sum of unionized and ionized forms of prucalopride, that contains a primary, secondary and tertiary amine function, at a physiological pH of 7.4 [229].

The measured LogDoct,pH7.4 value of 0.87, corresponding to a distribution ratio of 7.76 between 1-octanol and phosphate bufer of pH 7.4, points to a relatively low lipophilic nature of prucalopride. Pike (2009) reported that PET ligands with moderate lipophilicity indicated by logDoct,pH7.4 values in the range of 2.0 – 3.5 showed optimal passive brain entrance. Exceptionally, some useful radiotracers with lower or higher logDoct,pH7.4 values also entered the brain, but mostly for unclear reasons [31]. The relatively low lipophilicity of prucalopride may hamper its passive difusion into the brain.

6.4.3 [11C]prucalopride stability in vivo In this in vivo stability study in rats, parent [11C]prucalopride was not detectable in blood and brain extracts, at 5 or 30 min following IV injection, whereas diferent radiolabelled products were detected at both time points. This conirms, indings of previous studies, showing that in male rats prucalopride is extensively metabolised, reportedly through hydroxylation and/or O-demethylation. Such extensive metabolism was not seen in other species (Shire-Movetis, unpublished results; partly reported in [295]). In this study, hydroxylation of [11C]prucalopride would yield a radiolabelled metabolite which is highly hydrophilic (ClogP 0.007) and is thus expected not to pass the blood-brain-barrier readily. O-demethylation of [11C]prucalopride by CYP1A2 and other isoenzymes in the 11 liver [296-298] would yield unlabelled hydroxylated prucalopride and [ C]CH3OH and

128

11 Radiosynthesis and preclinical evaluation of [ C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor

11 possibly [ C]CH2O. Formation of the latter was demonstrated in diferent enzymatic demethylation reactions and it was found that CH2O formed in tissue, readily and spontaneously forms condensation products e.g. with arylethylamines, such as catecholamines 11 11 and indoleamines [299,300]. [ C]CH3OH and condensation products of [ C]CH2O may have contributed to nonspeciic levels of radioactivity in brain and peripheral tissues.

6.4.4 [11C]prucalopride, a potential PET ligand? Rats, commonly used and readily available species for initial evaluation of potential PET ligands, were also used for the present ex vivo biodistribution and in vivo PET studies. Unfortunately, evaluation of [11C]prucalopride as a potential PET ligand in the rat was hampered by rapid metabolism, which in the male rat is much faster than in other species [295]. Nevertheless, some interesting indings were made. Information was obtained that can explain the low radioactivity levels in the brain of [11C]prucalopride in this study. In rats, brain levels of radioactivity following IV injection of [11C]prucalopride was extremely low and under baseline conditions probably represented minute to no parent [11C]prucalopride, due to the fast metabolism. Surprisingly, following pre-treatment with tariquidar levels of radioactivity in all brain areas was increased three-fold within 6 seconds after IV injection of [11C]prucalopride (Figure 3). Tariquidar is an inhibitor of the P-glycoprotein ABC transporter. This transporter is located in capillary endothelial cells of the blood-brain-barrier, where its function is to translocate xenobiotics out of the brain, and in the intestinal epithelium, where it translocates toxic metabolites and xenobiotics from cells and blood into the intestinal lumen [301,302]. Tariquidar also is an inhibitor of CYP1A2, an enzyme involved in metabolism of prucalopride. Therefore, increased levels of radioactivity in the brain following tariquidar pre-treatment could be a result of slower metabolism and/or higher blood concentrations of unchanged [11C]prucalopride. On the other hand, it may also indicate that prucalopride is a P-glycoprotein substrate and that its removal from the brain is reduced by inhibition of the pump. The extremely fast appearance of the efect could suggest that the radioactivity that appears in the brain within seconds after injection, represents for the major part parent[11C]prucalopride. Hence, low levels of radioactivity in rat brain under baseline conditions could be due to (1) rapid metabolism, in particular in male rats, (2) limited passive difusion owing to its low lipophilicity and (3) the possibility of being a P-glycoprotein substrate. In a pilot PET study in one pig, parent compound was detected in blood with limited to no uptake in brain (unpublished observation), supporting the low brain uptake of [11C]prucalopride . Radioactivity levels in peripheral tissues following IV injections of [11C]prucalopride were substantially higher than in the brain, as shown in both the ex vivo biodistribution and in the in vivo PET study. PET images in the gut deserve particular attention in view of the therapeutic application of prucalopride in GI motility disorders. In this rat PET study, the caecum could clearly be delineated, although no analysis was done on the

129

Chapter 6 identity of the radioactivity in the gut. SUV values in the caecum and colon as a whole, were reduced after tariquidar pre-treatment as compared to baseline. This is in line with the role of the P-glycoprotein pump in the gut [303]. If prucalopride is a P-glycoprotein substrate, its transport from blood into the intestinal lumen should indeed be reduced by inhibiting P-glycoprotein.

5-HT4-R have been localized in the colonic mucosa and circular muscles [304].[ As prucalopride is used to treat constipation, the ability to investigate the active state of

5-HT4-R in the colon and the intestine in general, in vivo, would be highly interesting.

It could provide information on the active 5-HT4-R for example in cases of reduced gastric motility and it would allow monitoring possible 5-HT4-R desensitisation during 11 treatment with 5-HT4-R agonists. Further evaluation of [ C]prucalopride as a potential agonist PET ligand for 5-HT4-R in humans, where prucalopride is slowly metabolized

[283], seems worthwhile. In human studies, the labelling of peripheral 5-HT4-R could be fully explored and the uptake into the brain could be further checked.

6.5 ConCLuSIon

[11C]prucalopride was successfully synthesized. However, because of extremely fast metabolism, the male rat appeared not an appropriate species to assess the value of [11C]prucalopride as PET ligand. Because of low lipophilicity and the possibility of it being a P-glycoprotein substrate, [11C]prucalopride may not be suitable for in vivo imaging 11 of central 5-HT4-R. However, further investigation of [ C]prucalopride for imaging the active state of 5-HT4-R, in particular in human, is worthwhile in view of therapeutic applications of 5-HT4 agonists for treatment of GI motility disorders.

130

7 Summary, general discussion and future perspectives

Summary, general discussion and future perspectives

7.1 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 diferential 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 speciically 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 luorine-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, pharma- 7 cological and pharmacokinetic properties of corresponding agonist PET ligands. Spe- ciic 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 bufer 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. Biodistribu- tion studies revealed rapid and high brain uptake to levels exceeding ive times the level seen in blood. In M1ACh-R rich areas, speciic 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

135

Chapter 7 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 deining 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 coeicients (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 speciicity of uptake associated with M1ACh-R 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 deined using the MR template method and by applying the simpliied 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-speciic 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 signiicant reduction in BPND. BPND was not afected by pre-treatment with an M3ACh-R antagonist. Moreover, BPND was signiicantly 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 M1ACh-R saturation, [11C]AF150(S) was co-injected with the unlabelled substance. At the highest dose

136

Summary, general discussion and future perspectives

(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 speciic M1ACh-R-related signal, which could be inhibited by M1ACh-R 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 M1ACh-R binding ainity, speciic 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 ainity 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 trilate 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 elux transporter, and (3) in vivo stability of [11C]prucalopride was checked ex vivo in plasma and brain extracts 7 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 conirmed 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 difusion and/or being a substrate for P-glycoprotein.

137

Chapter 7

7.2 GEnERAL dISCuSSIon

The studies described in this thesis provide an insight in agonist PET ligand development and biological evaluation of two PET ligands with agonistic activity for 7-TMRs. 7-TMRs have been a main target for PET in the last decades, but although one of the irst 11 successful PET ligands was the opioid OP3 receptor agonist [ C]carfentanil [5], primarily antagonist PET ligands have been developed and evaluated. Recently, interest in the development and study of agonist PET ligands has been boosted by work performed on dopamine (D2/D3) [91,258-260,305] and serotonin (5-HT1A) receptors [133]. These particular agonist PET ligands were successfully applied to determine in vivo receptor occupancy and to assess sensitivity to changes in endogenous ligand concentrations, which therefore allow for the study of neurotransmitter release. Unfortunately, agonists that meet the criteria, put forward for successful PET ligands, appear to be rare. These criteria include high ainity for the target receptor (nM to pM), speciic binding to the target receptor, optimal lipophilic properties to allow passive difusion into the brain, not being a substrate for P-glycoprotein, and metabolic stability. Remarkably, in the overview of agonists for 7-TMRs (chapter 2), it would appear that the G-protein which couples to the 7-TMR, i.e. Gi/o, Gs, or Gq/11, might be a factor in the chance for successful discovery of high ainity agonist ligands. Indeed, it appeared that for Gi/o-coupled receptors, e.g. the OP3, D2/D3 and 5-HT1A receptors, more high ainity agonist ligands are available. Agonists for these receptors could chemically be optimized for good physicochemical properties, chemical and metabolic stability and good brain penetration to yield successful agonist PET ligands (chapter 2). Nevertheless, agonists for 7-TMRs that couple to Gq/11 or Gs proteins may provide interesting therapeutic agents for diseases with high medical need. For example, agonists for the Gq/11 coupled M1ACh-

R or the Gs coupled 5-HT4 or 5-HT6-R may improve cognitive impairment and agonists for the Gq/11 coupled 5-HT2C-R may help in the control of feeding behaviour. The fact that existing agonists for aforementioned receptors do not meet all imposed criteria for successful PET ligands should be a guide for the development of better ligands for these receptors. The agonist PET ligand [11C]AF150(S) (Chapters 3 & 5) does not comply with the “standard” criteria of a high ainity for the target receptor (nM to pM) and optimal lipophilicity to allow passive difusion into the brain. Yet, [11C]AF150(S) did show good brain uptake despite this poor lipophilicity. Moreover, it apparently showed speciic binding to M1ACh-R, despite the fact that the ainity of AF150(S) for the receptor was not favourable (Kd = 200 nM) and uptake of [11C]AF150(S) in speciic brain areas appeared to be sensitive to competition with endogenous ACh. 11 The investigated 5-HT4-R agonist [ C]prucalopride (Chapter 6) met certain “standard” criteria, such as moderate lipophilicity and nanomolar ainity for the target receptor. Still, this ligand proved to be unsuccessful for PET imaging, at least in male rats, where

138

Summary, general discussion and future perspectives

[11C]prucalopride showed poor brain uptake. It turned out that the ligand is a substrate for P-glycoprotein. In spite of the many challenges, PET with agonist ligands may provide an essential tool for future research that will aid in exploring the role and functioning of 7-TMRs in the human body and the development of new therapeutics.

7.3 FuTuRE PERSPECTIVES

The chances for developing a successful agonist PET ligand for a 7-TMR could potentially be increased by a more systematic approach with investigation of series of agonist PET ligands instead of one single candidate. The evaluation of a series of substituted 11C- phenethylamines, as agonists PET ligand candidates for the 5-HT2A receptor, constitutes a good example and resulting data are highly promising [148]. Using in vivo evaluation of a series of ligands, a new candidate (Cimbi-36) was discovered that showed improved in vivo binding potential compared with the initial candidate ligand (Cimbi-5). In addition, the same study provided insight in the delicate balance between receptor ainity and ligand lipophilicity. Unfortunately, evaluation of a series of agonist PET ligands is very time consuming and requires a substantial efort. Research and development of agonist PET ligands could be aided by the development of bio-mathematical models, aimed 7 at predicting in vivo performance of radioligands from in silico and in vitro data. Such a model has been developed for antagonist PET ligands [306]. In that particular study the authors correlated modelling data with data obtained using in vivo PET measurements with promising results. An alternative approach that may prove to be valuable in understanding in vivo 7-TMR characteristics of Gs, Gi/o and Gq coupled 7-TMRs, could be the development of PET ligands that do not bind to the orthosteric site, as most agonists do. Eforts could be directed towards the development of positive allosteric modulators as PET ligands for 7-TMRs. For instance, recently there has been a breakthrough in the development of a PET ligand for the metabotropic glutamate subtype 5 receptors (mGluR-5). Allosteric modulators of mGluR-5, in contrast to non-selective orthosteric mGlu-R agonists, have proven to be successful. They show better subtype selectivity, due to the fact that they bind in the cell transmembrane domain of the receptor, an area which is more speciic towards the receptor subtype compared with the orthosteric site [307]. Positive allosteric ligands deserve further investigation and in vivo evaluation. They may provide diferential information for those 7-TMRs for which no suitable orthosteric agonist PET ligands are available.

139

8 nederlandse samenvatting: In vivo bepaling van aan cognitie gerelateerde geactiveerde zeven transmembraan domein receptoren

Nederlandse samenvatting

8.1 InTRoduCTIE

Ons lichaam bestaat uit ongeveer 100.000 miljard cellen. Vrijwel elke cel bestaat uit een celkern met daar omheen een celmembraan met daarin receptoren. Deze receptoren stellen de cel instaat om te kunnen binden met moleculen in de extracellulaire ruimte. Er zijn veel verschillende type receptoren, die allemaal een andere lichaamseigen stof (ligand) herkennen, en in reactie daarop een speciiek signaal doorgeven dat leidt tot een biologisch efect. De grootste groep receptoren zijn de zeven transmembraan domein receptoren (7-TMRs), deze zijn ook wel bekend als G-eiwit gekoppelde receptoren (GPCR). Deze receptoren worden zo genoemd omdat ze binnen de cel het signaal doorgeven via speciale eiwitten: de zogenoemde G-eiwitten. Er zijn meer dan 800 verschillende 7-TMRs in het menselijk lichaam bekend die allemaal net iets anders zijn en een ander biologisch efect teweeg brengen als ze geactiveerd worden. 7-TMRs zijn vrijwel bij elk biologisch proces in ons lichaam betrokken, en vormen daardoor een interessant doel- wit voor geneesmiddelontwikkeling. Door middel van organisch-chemische synthese zijn liganden zo te maken dat ze net als het natuurlijke ligand in de receptor passen en zo de activiteit en dus het biologische efect van een dergelijke receptor beïnvloeden. Er zijn verschillende type synthetische liganden bekend die de receptor kunnen activeren (agonist) of blokkeren (antagonist). Ongeveer 40% van al onze geneesmiddelen grijpt aan op een 7-TMR. Hierbij valt te denken aan bètablokkers, anti-parkinsonmiddelen, antipsychotica, anticholinergica, antidepressiva en antihistaminica. Om de functie en de rol van de 7-TMRs in de mens te kunnen bestuderen kan er gebruik gemaakt worden van Positron Emissie Tomograie (PET). PET is een medische beeldvormende techniek 8 die het mogelijk maakt fysiologische en metabole processen in het lichaam zichtbaar te maken. PET maakt hiervoor gebruik van diverse radioactief gelabelde liganden, ook wel PET liganden genoemd. Voor 7-TMRs in het centraal zenuwstelsel (CZS) zijn tot op heden enkele succesvolle PET liganden ontwikkeld, welke voornamelijk antagonisten zijn. Dit komt waarschijnlijk door de gunstige farmacologische, fysisch-chemische en farmacokinetische eigenschappen van de antagonisten. Voor de ontwikkeling van agonisten als PET liganden voor het CZS is minder aandacht. Waarschijnlijk omdat er minder agonisten beschikbaar zijn als geneesmiddel voor aandoeningen van het CZS. Toch bieden juist agonist PET liganden de mogelijkheid om diferentiële informatie en inzicht te verkrijgen in receptorfunctie en receptorregulatie welke niet te verkrijgen is met antagonist PET liganden. Dit komt doordat antagonist PET liganden binden aan het totale aantal van actieve en inactieve 7-TMRs die beschikbaar zijn. Terwijl, de agonist PET liganden speciiek kunnen binden aan de functionele en geactiveerde 7-TMR. Het doel van de in dit proefschrift beschreven studies was het ontwikkelen van PET liganden met agonistische activiteit voor 7-TMRs die betrokken zijn bij cognitie. Als

143

Chapter 8 onderzoeksgebied zijn hiervoor de muscarine M1 acetylcholine receptor (M1ACh-R ) en de serotonine 5-HT4 receptor subtype (5-HT4-R ) geselecteerd.

8.2 RESuLTATEn En ConCLuSIES VAn HET ondERZoEK

Hoofdstuk 2 geeft een uitgebreid overzicht van agonist PET liganden die voor de 7-TMRs ontwikkeld zijn die: (1) gelabeld zijn met kortlevende radionucliden zoals koolstof-11 of luor-18 en (2) in vivo geëvalueerd zijn als PET ligand voor 7-TMRs in de hersenen. Voor elke 7-TMR is er een samenvatting opgenomen van de rol van de betrefende receptor. Verder is er een beschouwing gegeven van de chemische, farmacologische en farmacokinetische eigenschappen van de betrefende agonist PET liganden. Ook wordt er speciieke informatie besproken met betrekking tot de receptorfuncties welke verkregen zijn met behulp van agonist PET liganden. Tot slot wordt er ingegaan op de voordelen van agonist PET liganden ten opzichte van antagonist PET liganden.

In hoofdstuk 3 wordt de ontwikkeling van een orthosterische agonist PET ligand voor de M1ACh-R beschreven. Dit begint met de radiosynthese van de M1ACh-R selectieve agonist, [11C]AF150(S). Verder zijn de receptor bindingseigenschappen van deze radioligand onderzocht met behulp van autoradiograie experimenten op hersencoupes van de rat. Ook is de opname van de radioligand in de hersenen na intraveneuze toediening bepaald in ratten met behulp van biodistributie studies. Het metabolisme van de [11C] AF150(S) in de hersenen en bloed is gemeten met behulp van vloeistofchromatograie (HPLC). Verder is de partitiecoëiciënt van de radioligand tussen 1-octanol en fosfaatbufer pH 7,4 (LogDoct) bepaald. De radiosynthese opbrengst (gecorrigeerd voor radioactiefverval) van [11C]AF150(S) was ongeveer 70%, met een radiochemische zuiverheid >99%. De verdeling van [11C]AF150(S) tussen 1-octanol en fosfaatbufer pH 7,4 was 0,05. Autoradiograie studies toonden binding aan in M1ACh-R rijke hersengebieden in vitro, welke onder invloed van competitieve muscarine acetylcholine receptorliganden ger- educeerd kon worden. De biodistributie studies lieten een snelle en hoge opname in de hersenen zien met een meer dan vijf keer zo hoge concentratie als de bloedconcentratie. In de M1ACh-R rijke hersengebieden was speciieke opname tot 60 minuten na injectie van [11C]AF150(S) waarneembaar, in vergelijking tot het cerebellum (hersengebied wat vrijwel geen M1ACh-R bevat). Dit ondanks een snelle daling van radioactiviteit in de hersenen en een snel perifeer metabolisme. Ook is het redelijk ongebruikelijk dat de opname van [11C]AF150(S) in de hersenen relatief snel en hoog is, zeker gezien het hydroiele karakter van de ligand. Dit duidt mogelijk op de betrokkenheid van actief transport. Concluderend, [11C]AF150(S) is met succes gesynthetiseerd en liet agonist-

144

Nederlandse samenvatting speciieke bindingseigenschappen zien voor de M1ACh-R en een hoge opname in de hersenen.

In hoofdstuk 4 is een nieuwe methode beschreven voor een reproduceerbare analyse van PET data bij hersenstudies in ratten. Deze methode maakt gebruik van een standaard magnetische resonantie (MRI) template met daarin gedeineerde regions of interest (ROI) en een additionele [18F]NaF scan. De laatste maakt een duidelijke afbakening van de schedel mogelijk, zodat de MRI gecoregistreerd kan worden met de PET scan. Een standaard procedure is ontwikkeld om eenduidig deze wijze van coregistratie van de MRI template met de [18F]NaF scans te kunnen uitvoeren. Ter validatie van de methode zijn er tien [18F]NaF scans van ratten gecoregistreerd met de standaard MRI template, elk door 3 waarnemers. De hierbij verkregen transformatie matrices werden toegepast op de bijbehorende [11C]AF150(S) PET scans. Zo kon met de in de MRI template gedeinieerde hersenen gebieden de opname van radioactiviteit, uitgedrukt als het percentage geïnjecteerde dosis per gram (% ID.g-1), bepaald worden voor de [11C]AF150(S) PET scans. Deze waarden zijn vervolgens vergeleken met resultaten uit biodistributie studies van [11C]AF150(S). Ook zijn er interclass correlatiecoëiciënten (ICC) berekend voor de coregistratie door de 3 verschillende waarnemers. De ICC analyse toonde voor alle bepaalde hersengebieden een uitstekende correlatie aan tussen de waarnemers. Daarnaast was de opname van [11C]AF150(S) in de hersengebieden, afgeleid uit de PET data, niet verschillend van de biodistributie studies data. In conclusie kan gesteld worden dat deze nieuwe methode, een reproduceerbare en PET ligand onafhankelijke methode vormt voor de analyse van PET data van rattenhersenen. 8

In hoofdstuk 5 is de kinetiek en speciiciteit van de opname van [11C]AF150(S) in de hersenen van de rat beschreven. Zo zijn PET studies uitgevoerd onder basale condities en na voorbehandeling met verschillende farmacologisch actieve liganden of co- injectie met niet-radioactief AF150(S). Na intraveneuze toediening werd [11C]AF150(S) snel opgenomen in de hersenen, gevolgd door snelle klaring uit alle hersengebieden. Verdere analyse van de PET data met behulp van het simpliied reference tissue model (SRTM), waarbij gebruik gemaakt is van het cerebellum als referentiegebied, leverde een bindings potentiaal (BPND ; een maat voor speciieke opname) op van 0,25 voor het striatum, 0,20 voor de hippocampus, 0,16 voor frontale cortex en 0,15 voor posterior cortex. Deze hersengebieden bevatten een relatief hoge dichtheid in M1ACh-R. De PET studies waarbij de ratten werden voorbehandeld met selectieve M1ACh-R antagonisten resulteerden in een signiicante vermindering van BPND waarden. BPND waarden werden niet beïnvloed door voorbehandeling met een selectieve M3ACh-R antagonist. Verder werden de BPND waarden signiicant verminderd na voorbehandeling met haloperidol, een dopamine D2 receptor blokker welke voor een toename in extracellulair acetylcho-

145

Chapter 8

line (ACh) zorgt. Deze vermindering in BPND waarden na haloperidol voorbehandeling kan mogelijk ontstaan zijn door de extracellulair verhoogde concentratie ACh welke competitie is aangegaan met [11C]AF150(S) voor binding aan de M1ACh-R. Om deze mogelijkheid verder te onderzoeken zijn de efecten van voorbehandeling met andere farmacologisch actieve liganden die de extracellulaire niveaus van ACh kunnen ver- hogen bestudeerd. Hiervoor zijn AF-DX 384, een M2/M4ACh-R receptor antagonist en rivastigmine, een acetylcholine-esterase remmer gebruikt. Ook is de mogelijke saturatie van binding van [11C]AF150(S) aan M1ACh-R bestudeerd. Hiervoor is [11C]AF150(S) ge- coïnjecteerd met het niet-radioactief AF150(S). Bij de hoogste toegediende dosis van niet-radioactief AF150(S) (49,1 nmol.kg-1) werd in de hersenen een concentratie van 100 nmol.l-1 bereikt. Bij deze concentratie werden geen tekenen van saturatie van binding aan de M1ACh-R waargenomen. Deze studie toonde aan dat de agonist PET ligand [11C] AF150(S) snel opgenomen wordt in de hersenen en er een potentieel M1ACh-R gerelateerde signaal waarneembaar is in M1ACh-R rijke hersengebieden, welke geremd kon worden door voorbehandeling met M1ACh-R antagonisten. Daarnaast lijkt de mate van in vivo binding van de agonist PET ligand [11C]AF150(S) mogelijk gevoelig voor veranderingen in extracellulaire ACh concentraties. Waarschijnlijk door de relatief lage binding- sainiteit van AF150(S) voor de M1ACh-R, trad er geen saturatie van speciieke binding op. Dit is mogelijk veroorzaakt door een te beperkt gekozen concentratiegebied.

Beschreven in hoofdstuk 6 is de ontwikkeling van een agonist PET ligand voor de serotonine 5-HT4-R. Prucalopride is agonist met een hoge ainiteit voor de 5-HT4-R en wordt toegepast als een geneesmiddel voor behandeling van constipatie, en zou naar verluidt de hersenen kunnen binnendringen. [11C]prucalopride werd gesynthetiseerd uit [11C]methyltrilaat en desmethyl prucalopride precursor. Ook is de partitiecoëiciënt tussen 1-octanol en fosfaatbufer pH 7,4 (LogDoct) bepaald. Er zijn drie verschillende studies met intraveneuze toediening van [11C]prucalopride uitgevoerd in mannelijke ratten: (1) een biodistributie studie van radioactiviteit, (2) een PET studie naar de kinetiek van radioactiviteit in de hersenen en perifere organen, onder basale condities en na voorbehandeling met tariquidar, een remmer van het P-glycoproteïne, en (3) een in vivo studie naar de stabiliteit van [11C]prucalopride, bepaald met ex vivo gemeten bloedplasma en hersenextracten met behulp van HPLC. [11C]Prucalopride kon onder geopti- maliseerde omstandigheden gesynthetiseerd worden met een opbrengst van 21 ± 4 % (voor radioactiefverval gecorrigeerd) en een radiochemische zuiverheid van >99%. De partitiecoëiciënt tussen 1-octanol en fosfaatbufer pH 7,4 was 0.87. De biodistributie studie bij mannelijke ratten liet zeer lage niveau’s van radioactiviteit in de hersenen zien (maximaal 0,13% ID.g-1) en een tienmaal hogere concentratie werd waargenomen in bepaalde perifere weefsels. De PET studie bevestigde de zeer lage hersenconcentraties van radioactiviteit onder basale condities. Na voorbehandeling van de ratten met tariq-

146

Nederlandse samenvatting uidar nam de hersenconcentratie met een factor drie toe. Verder bleek [11C]Prucalopride zeer snel gemetaboliseerd te worden in mannelijke ratten en was niet meer aantoon- baar in bloedplasma en hersenextracten 5 minuten na intraveneuze toediening. Analyse van de PET data liet wel een speciieke opname van radioactiviteit in perifere weefsels zien. In het bijzonder was er een duidelijke PET signaal waarneembaar in een deel van de dikke darm (caecum), welke verminderd kon worden door voorbehandeling met tariquidar. Deze geobserveerde vermindering is in overeenstemming met de rol van het P-glycoproteïne in het maag-darmstelsel. Concluderend, [11C]prucalopride liet een lage concentratie radioactiviteit zien in de hersenen van de mannelijke rat. Dit wordt mogelijk veroorzaakt door een snel metabolismen, een lage mate van passieve difusie en / of het zijn van een substraat voor het P-glycoproteïne.

8.3 ALGEMEnE dISCuSSIE

De in dit proefschrift beschreven studies geven een inzicht in de ontwikkeling van agonist PET liganden voor 7-TMRs en de preklinische evaluatie van twee agonist PET liganden, [11C]AF150(S) en [11C]prucalopride. 7-TMRs vormen een van de belangrijkste moleculaire aangrijpingspunten voor PET in de laatste decennia. Ondanks het feit dat een van de eerste succesvolle PET liganden een agonist is, [11C]Carfentanil, selectief voor de opioïde OP3 receptor, zijn er tot nu toe voornamelijk antagonist PET liganden ontwikkeld en geëvalueerd. De recente toename in de ontwikkeling van agonist PET liganden is voornamelijk gestimuleerd door de resultaten van de PET studies met ago- 8 nist PET liganden voor dopamine (D2/D3) en serotonine (5-HT1A) receptoren. Met deze agonist PET liganden kon met succes in vivo de graad van receptorbezetting bepaald worden. Daarnaast bleek de de mate van in vivo binding van deze PET liganden gevoelig te zijn voor veranderingen in endogene ligandconcentraties. Deze gevoeligheid maakt het mogelijk om met deze PET liganden neurotransmitter afgifte in vivo te bestuderen. Helaas zijn agonisten, die aan de ideale eigenschappen voor een succesvolle PET ligand voldoen een schaars goed. De ideale eigenschappen waar een agonist idealiter aan zou moeten voldoen zijn: een hoge ainiteit voor de te onderzoeken receptor (nM to pM), receptor speciieke en selectieve binding, qua fysisch-chemische eigenschappen een optimale lipoiliciteit voor snelle passieve difusie over de bloed-hersenbarrière, geen substaat voor P-glycoproteïne, en minimaal metabolisme van de PET ligand. Uit het in hoofdstuk 2 beschreven overzicht van ontwikkelde agonist PET liganden voor 7-TMRs, kan afgeleid worden dat juist het type G-eiwit dat koppelt aan de 7-TMR

(Gi/o, Gs, of Gq/11) een kritische een factor kan zijn in de ontdekking van een succesvolle agonist PET ligand met een hoge receptor ainiteit. Zo lijkt het dat er met name voor de

Gi/o gekoppelde receptoren, zoals de opioïde OP3, dopamine D2/D3 en serotonine 5-HT1A

147

Chapter 8 receptoren, er meer agonisten met hoge receptor ainiteit beschikbaar zijn. Toch zijn agonisten die aangrijpen op 7-TMRs die Gq/11 of Gs gekoppeld zijn mogelijk ook interessant en kunnen mogelijk als geneesmiddel ingezet worden. Zo kunnen bijvoorbeeld agonisten voor de M1ACh-R (Gq/11 gekoppeld) of serotonine 5-HT4 of 5-HT6 receptor (beide GS gekoppeld) mogelijk verlichting geven van cognitieve stoornissen. Agonisten voor de serotonine 5-HT2C receptor (Gq/11 gekoppeld) kunnen mogelijk bijdragen aan een verbeterde controle over eetgedrag.

Voor de hierboven beschreven type receptoren voldoen de beschikbare agonist liganden qua eigenschappen niet altijd aan de ideale criteria om zo een succesvolle PET ligand te kunnen ontwikkelen. De agonist PET ligand [11C]AF150(S) (beschreven in hoofdstuk- ken 3 & 5) beschikt niet over de “ideale eigenschappen” zoals een hoge ainiteit voor de doelreceptor (nM tot pM) en beschikt ook niet over een optimale lipoiliciteit voor gegarandeerde passieve difusie in de hersenen. Ondanks zijn matige lipoiele eigenschappen liet [11C]AF150(S) toch een hoge opname zien in de hersenen. Daarnaast was mogelijk speciieke binding voor de M1ACh-R waarneembaar, ondanks het feit dat de ainiteit van AF150(S) voor de receptor niet gunstig is (Kd = 200 nM). Verder was het opmerkelijk dat de binding van [11C]AF150(S), in speciieke hersengebieden, mogelijk gevoelig is voor competitie met de endogene neurotransmitter acetylcholine. 11 De serotonine 5-HT4 receptor agonist [ C]prucalopride (beschreven in hoofdstuk 6) voldeed voor een groot gedeelte wel aan de “ideale eigenschappen”, zoals een redelijke lipoiliciteit en nM ainiteit voor de doelreceptor. Toch bleek deze radioligand niet ges- chikt als agonist PET ligand. In studies met mannelijke ratten liet [11C]prucalopride een zeer matige hersenopname zien. Uit verder onderzoek blijkt de ligand een substraat te zijn voor het P-glycoproteïne. Ondanks vele uitdagingen, is en blijft PET onderzoek met agonist PET liganden essentieel voor wetenschappelijk onderzoek naar de rol en functie van 7-TMRs in het menselijke lichaam evenals voor de ontwikkeling van nieuwe geneesmiddelen.

8.4 ToEKoMSTPERSPECTIEVEn

De kans om met succes een agonist PET ligand voor een 7-TMR te ontwikkelen zou verder kunnen worden vergroot door een meer systematische aanpak. Door het ontwikkelen van een reeks structuur analoge agonist PET liganden in plaats van de focus op één enkele kandidaat ligand kunnen de kansen vergroot worden op het vinden van een succesvolle PET ligand en wordt er meer fundamenteel inzicht verkregen. Een goed voorbeeld van een dergelijke aanpak, is de evaluatie van een reeks gesubstitueerde en met koolstof-11 gelabelde fenethylaminen die als mogelijke agonist PET liganden

148

Nederlandse samenvatting

voor de serotonine 5-HT2 receptor zouden kunnen dienen. Uit de in vivo evaluatie van deze reeks van PET liganden werd een nieuwe kandidaat (Cimbi-36) ontdekt die een nog betere in vivo bindings potentiaal had dan de initeel geselecteerde ligand (Cimbi-5). Daarnaast geeft de studie een goed inzicht in de ijne balans tussen receptor ainiteit en lipoiliciteit en daaruit voortvloeiende mate van hersenopname van de PET liganden. Het is voor de hand liggend dat de evaluatie van reeksen agonist PET liganden veel inspanning en middelen zal vergen in vergelijk met een enkele kandidaat agonist PET ligand. De ontwikkeling van agonist PET liganden zou mogelijk verder ondersteund kunnen worden door ontwikkeling van bio-mathematische modellen. Deze modellen kunnen mogelijk de in vivo karakteristieken van radioliganden voorspellen op basis van in silico en in vivo data van eerder ontwikkelde agonist PET liganden. Voor antagonist PET liganden is er een dergelijk bio-mathematisch model reeds beschikbaar. Een alternatieve benadering, die mogelijk ook waardevol kan blijken te zijn is de ontwikkeling van PET liganden die niet binden aan de orthosterische bindingplaats, zoals de meeste agonisten doen. Deze liganden binden in het transmembrane domein en worden ook wel positieve allosterische modulatoren genoemd. Dit type ligand kan mogelijk ook ontwikkeld worden als PET liganden voor 7-TMRs. Ter illustratie, onlangs is er een doorbraak geweest in de ontwikkeling van een PET ligand voor de metabotrope glutamaatreceptor (mGluR). In tegenstelling tot de beschikbare orthostere mGluR agonisten, bleken juist de allosterische modulatoren succesvolle PET liganden te zijn voor de mGluR5. Dit kan mogelijk verklaard worden doordat de agonisten niet receptor subtype selectief bleken te zijn, terwijl de allosterische liganden een betere receptor subtype selectiviteit lieten zien. Dit 8 komt omdat ze in het cellulaire transmembrane domein binden en voor de mGluR deze regio speciieker is per receptor subtype dan de othosterische bindingsplaats. Daarom verdienen positieve allosterische liganden nader onderzoek en in vivo evaluatie, omdat zij mogelijk een uitkomst kunnen bieden voor die 7-TMRs waarvoor nog geen geschikte orthostere agonist PET ligand ontwikkeld is.

149

References

REFEREnCES

1. Bockaert J, Pin JP: Molecular tinkering of G protein-coupled receptors: an evolutionary success. EMBO J 1999, 18: 1723-1729. 2. Fredriksson R, Lagerström MC, Lundin LG, Schiöth HB: The G-Protein-Coupled Receptors in the Human Genome Form Five Main Families. Phylogenetic Analysis, Paralogon Groups, and Finger- prints. Mol Pharmacol 2003, 63: 1256-1272. 3. Fredriksson R, Schiöth HB: The Repertoire of G-Protein-Coupled Receptors in Fully Sequenced Genomes. Mol Pharmacol 2005, 67: 1414-1425. 4. Flower DR: Modelling G-protein-coupled receptors for drug design. BBA-Rev Biomembranes 1999, 1422: 207-234. 5. Frost JJ, Wagner HNJ, Dannals RF, Ravert HT, Links JM, Wilson AA et al.: Imaging Opiate Receptors in the Human Brain by Positron Tomography. J Comp Ass Tomo 1985, 9: 231-236. 6. Rosenbaum DM, Rasmussen SGF, Kobilka BK: The structure and function of G-protein-coupled receptors. Nature 2009, 459: 356-363. 7. Vögler O, Barceló JM, Ribas C, Escribá PV: Membrane interactions of G proteins and other related proteins. BBA-Rev Biomembranes 2008, 1778: 1640-1652. 8. Rodbell M: The complex regulation of receptor-coupled G-proteins. Adv Enzyme Regul 1997, 37: 427-435. 9. Oldham WM, Hamm HE: How do Receptors Activate G Proteins? In Advances in Protein Chemistry Mechanisms and Pathways of Heterotrimeric G Protein Signaling. Edited by Stephen RS. Academic Press; 2007: 67-93. 10. Oldham WM, Hamm HE: Heterotrimeric G protein activation by G-protein-coupled receptors. Nat Rev Mol Cell Biol 2008, 9: 60-71. 11. Repapetilto. Cartoon depicting the GPCR-G-protein activation/deactivation cycle. In: cycle.jpg G, editor.Adobe Illustrator CS3. 2010. 12. Kenakin T. Principles: Receptor theory in pharmacology. Trends Pharmacol Sci 2004, 25: 186-192. 13. De Lean A, Stadel JM, Lefkowitz RJ: A ternary complex model explains the agonist-speciic binding properties of the adenylate cyclase-coupled beta-adrenergic receptor. J Biol Chem 1980, 255: 7108-7117. 14. Lefkowitz RJ, Cotecchia S, Samama P, Costa T: Constitutive activity of receptors coupled to gua- nine nucleotide regulatory proteins. Trends Pharmacol Sci 1993, 14: 303-307. 15. Lefkowitz RJ: G Proteins in Medicine. N Engl J Med 1995, 332: 186-187. 16. Kobilka BK, Deupi X: Conformational complexity of G-protein-coupled receptors. Trends Pharma- col Sci 2007, 28: 397-406. 17. Kenakin T: Inverse, protean, and ligand-selective agonism: matters of receptor conformation. FASEB J 2001, 15: 598-611. 18. Barker EL, Westphal RS, Schmidt D, Sanders-Bush E: Constitutively active 5-hydroxytryptamine2C receptors reveal novel inverse agonist activity of receptor ligands. J Biol Chem 1994, 269: 11687- 11690. 19. Schwartz TW, Holst B: Ago-Allosteric Modulation and Other Types of Allostery in Dimeric 7TM Receptors. J Recep Sig Transd 2006, 26: 107-128. 20. Ferguson SSG: Evolving Concepts in G Protein-Coupled Receptor Endocytosis: The Role in Recep- tor Desensitization and Signaling. Pharmacol Rev 2001, 53: 1-24.

151

References

21. Drake MT, Shenoy SK, Lefkowitz RJ: Traicking of G Protein-Coupled Receptors. Circ Res 2006, 99: 570-582. 22. Delom F, Fessart D: Role of Phosphorylation in the Control of Clathrin-Mediated Internalization of GPCR. Int J Cell Biol 2011. 23. Gee AD: Neuropharmacology and drug development. Br Med Bull 2003, 65: 169-177. 24. Verel I, Visser GWM, van Dongen GA: The Promise of Immuno-PET in Radioimmunotherapy. J Nucl Med 2005, 46: 164S-171S. 25. van Velden FHP, Kloet RW, van Berckel BNM, Buijs FL, Luurtsema G, Lammertsma AA et al.: HRRT Versus HR+ Human Brain PET Studies: An Interscanner Test-Retest Study. J Nucl Med 2009, 50: 693-702. 26. Leroy C, Comtat C, Trébossen R, Syrota A, Martinot JL, Ribeiro MJ: Assessment of 11C-PE2I Binding to the Neuronal Dopamine Transporter in Humans with the High-Spatial-Resolution PET Scanner HRRT. J Nucl Med 2007, 48: 538-546. 27. De Jong HWAM, van Velden FHP, Kloet RW, Buijs FL, Boellaard R, Lammertsma AA: Performance evaluation of the ECAT HRRT: an LSO-LYSO double layer high resolution, high sensitivity scanner. Phys Med Biol 2007, 52: 1505-1526. 28. Battaglia G, Titeler M: [3H]N-propylapomorphine and [3H]spiperone binding in brain indicate two states of the D2-dopamine receptor. Eur J Pharmacol 1982, 81: 493-498. 29. Bonaventure P, Hall H, Gommeren W, Cras P, Langlois X, Jurzak M et al.: Mapping of serotonin 5-HT4 receptor mRNA and ligand binding sites in the post-mortem human brain. Synapse 2000, 36: 35-46. 30. Halldin C, Gulyas B, Farde L: PET studies with carbon-11 radioligands in neuropsychopharmaco- logical drug development. Curr Pharm Des 2001, 7: 1907-1929. 31. Pike VW: PET radiotracers: crossing the blood-brain barrier and surviving metabolism. Trends Pharmacol Sci 2009, 30: 431-440. 32. Hulme EC, Trevethick MA: Ligand binding assays at equilibrium: validation and interpretation. Br J Pharmacol 2010, 161: 1219-1237. 33. Leysen J, Langlois X, Heylen L, Lammertsma A: Receptors: Binding Assays. In Encyclopedia of Psychopharmacology. Edited by Stolerman I. Springer Berlin Heidelberg; 2010: 1127-1134. 34. Mintun MA, Raichle ME, Kilbourn MR, Wooten GF, Welch MJ: A quantitative model for the in vivo assessment of drug binding sites with positron emission tomography. Ann Neurol 1984, 15: 217- 227. 35. Innis RB, Cunningham VJ, Delforge J, Fujita M, Gjedde A, Gunn RN et al.: Consensus nomenclature for in vivo imaging of reversibly binding radioligands. J Cereb Blood Flow Metab 2007, 27: 1533- 1539. 36. Lammertsma A: Compartmental Modeling in Emission Tomography. In Handbook of Particle Detection and Imaging. Edited by Grupen C, Buvat In. Springer Berlin Heidelberg; 2012: 1065-1081. 37. Lammertsma AA, Hume SP: Simpliied Reference Tissue Model for PET Receptor Studies. NeuroIm- age 1996, 4: 153-158. 38. Terry AV, Buccafusco JJ: The cholinergic hypothesis of age and Alzheimer’s disease-related cognitive deicits: Recent challenges and their implications for novel drug development. J Pharmacol Exp Ther 2003, 306: 821-827. 39. Browne SE, Beal MF: Oxidative damage in Huntington’s disease pathogenesis. Antiox Redox Sign 2006, 8: 2061-2073. 40. Phukan J, Pender NP, Hardiman O: Cognitive impairment in amyotrophic lateral sclerosis. Lancet Neurol 2007, 6: 994-1003.

152

References

41. Buccafusco JJ: Emerging cognitive enhancing drugs. Exp Opin Emerg Drugs 2009, 14: 577-589. 42. Mesulam MM: Cholinergic circuitry of the human nucleus basalis and its fate in Alzheimer’s disease. J Comp Neurol 2013, 521: 4124-4144. 43. Selkoe DJ: SnapShot: Pathobiology of Alzheimer’s Disease. Cell 2013, 154: 468. 44. Teipel S, Heinsen H, Amaro E, Grinberg LT, Krause B, Grothe M. Cholinergic basal forebrain atrophy predicts amyloid burden in Alzheimer’s disease. Neurobiol Aging 2014, 35[3]: 482-491. 45. Bubser M, Byun N, Wood M, Jones C: Muscarinic Receptor Pharmacology and Circuitry for the Modulation of Cognition. In Muscarinic Receptors. Edited by Fryer AD, Christopoulos A, Nathanson NM. Springer Berlin Heidelberg; 2012: 121-166. 46. Terry J, Buccafusco JJ, Wilson C: Cognitive dysfunction in neuropsychiatric disorders: Selected serotonin receptor subtypes as therapeutic targets. Behav Brain Res 2008, 195: 30-38. 47. King MV, Marsden CA, Fone KCF: A role for the 5-HT1A, 5-HT4 and 5-HT6 receptors in learning and memory. Trends Pharmacol Sci 2008, 29: 482-492. 48. Buchanan RW, Freedman R, Javitt DC, bi-Dargham A, Lieberman JA: Recent Advances in the Development of Novel Pharmacological Agents for the Treatment of Cognitive Impairments in Schizophrenia. Schizophr Bull 2007, 33: 1120-1130. 49. Carlsson A: The occurrence, distribution and physiological role of catecholamines in the nervous system. Pharmacol Rev 1959, 11: 490-493. 50. Hornykiewicz O: Dopamine (3-hydroxytyramine) and brain function. Pharmacol Rev 1966, 18: 925-964. 51. Beninger RJ: The role of dopamine in locomotor activity and learning. Brain Res Rev 1983, 6: 173- 196. 52. Jaber M, Robinson SW, Missale C, Caron MG: Dopamine receptors and brain function. Neurophar- macology 1997, 35: 1503-1519. 53. Kim A.Neve, Jeremy K.Seamans, Heather Trantham-Davidson: Dopamine Receptor Signaling. J Recep Sig Transd 2004, 24: 165-205. 54. Girault J: The neurobiology of dopamine signaling. Arch Neurol 2004, 61: 641-644. 55. Schotte A, Janssen PFM, Bonaventure P, Leysen JE: Endogenous dopamine limits the binding of antipsychotic drugs to D-3 receptors in the rat brain: A quantitative autoradiographic study. Histochem J 1996, 28: 791-799. 56. Kiss B, Ferenc H, Amrita B: In Vitro and In Vivo Comparison of [H-3](+)-PHNO and [H-3]Raclopride Binding to Rat Striatum and Lobes 9 and 10 of the Cerebellum: A Method to Distinguish Dopa- mine D-3 From D-2 Receptor Sites. Synapse 2011, 65: 467-478. 57. Heidbreder CA, Gardner EL, Xi ZX, Thanos PK, Mugnaini M, Hagan JJ et al.: The role of central dopamine D3 receptors in drug addiction: a review of pharmacological evidence. Brain Res Rev 2005, 49: 77-105. 58. Leysen JE: Atypical Antipsychotics. In Dopamine in the CNS II. Edited by Chiara G. Springer Berlin Heidelberg; 2002: 473-490. 59. Beaulieu JM, Gainetdinov RR: The Physiology, Signaling, and Pharmacology of Dopamine Recep- tors. Pharmacol Rev 2011, 63: 182-217. 60. Rye DB: Parkinson’s disease and RLS: the dopaminergic bridge. Sleep Med 2004, 5: 317-328. 61. Finnema SJ, Bang-Andersen B, Wikström HV, Halldin C: Current State of Agonist Radioligands for

Imaging of Brain Dopamine D2/D3 Receptors In Vivo with Positron Emission Tomography. Curr Top Med Chem 2010, 10: 1477-1498.

153

References

62. Finnema S, Bang-Andersen B, Farde L, Jorgensen M, Gulyás B, Foged C et al.: PET evaluation of the partial dopamine D1 receptor agonist radioligands R- and S-[11C]N-methyl-NNC 01-0259. NeuroImage 2006, 31: T114. 63. Finnema SJ, Bang-Andersen B, Farde L, Jorgensen M, Gulyás B, Foged C et al.: Low efect of amphetamine on the binding of the partial dopamine D-1 receptor agonist radioligand [C-11] N-methyl-NNC 01-0259. Eur J Nucl Med Mol Imaging 2006, 33: S211. 64. Finnema SJ, Bang-Andersen B, Jørgensen M, Gulyas B, Farde L, Wikström HV et al.: COMT inhibition prevents the formation of lipophilic radiometabolites of catechols, - An example with (S)-[C-11] N-methyl-NNC 01-0259. NeuroImage 2008, 41: T91. 65. Finnema SJ, Bang-Andersen B, Jørgensen M, Christofersen CT, Gulyás B, Wikström HV et al.: The dopamine D1 receptor agonist (S)-[11C]N-methyl-NNC 01-0259 is not sensitive to changes in dopamine concentration - a positron emission tomography examination in the monkey brain. Synapse 2013, 67: 586-595. 66. Halldin C, Swahn CG, Neumeyer JL, Hall H, Gao Y, Karlsson P et al.: Preparation of two potent and selective dopamine D-2 receptor agonists: (R)-[propyl-11C]-2-OH-NPA and (R)-[methyl-11C]-2- OCH3-NPA. J Label Compd Radiopharm 35, s265-s266. 1992. 67. Finnema SJ, Seneca N, Farde L, Shchukin E, Sovago J, Gulyas B et al.: A preliminary PET evaluation of the new dopamine D-2 receptor agonist [C-11]MNPA in cynomolgus monkey. Nucl Med Biol 2005, 32: 353-360. 68. Seneca N, Zoghbi SS, Skinbjerg M, Liow JS, Hong J, Sibley DR et al.: Occupancy of dopamine D-2/3 receptors in rat brain by endogenous dopamine measured with the agonist positron emission tomography radioligand [C-11]MNPA. Synapse 2008, 62: 756-763. 69. Otsuka T, Ito H, Halldin C, Takahashi H, Takano H, Arakawa R et al.: Quantitative PET Analysis of the Dopamine D(2) Receptor Agonist Radioligand (11)C-(R)-2-CH(3)O-N-n-Propylnorapomorphine in the Human Brain. J Nucl Med 2009, 50: 703-710. 70. Seneca N, Finnema SJ, Farde L, Gulyas B, Wikström HV, Halldin C et al.: Efect of amphetamine on dopamine D2 receptor binding in nonhuman primate brain: A comparison of the agonist radioligand [(11C)]MNPA and antagonist [C-11]raclopride. Synapse 2006, 59: 260-269. 71. Richield EK, Penney JB, Young AB: Anatomical and ainity state comparisons between dopamine D1 and D2 receptors in the rat central nervous system. Neuroscience 1989, 30: 767-777. 72. Hwang DR, Kegeles LS, Laruelle M: (-)-N-[C-11]propyl-norapomorphine: A positron-labeled dopamine agonist for PET imaging of D-2 receptors. Nucl Med Biol 2000, 27: 533-539. 73. Narendran R, Hwang DR, Slifstein M, Hwang Y, Huang Y, Ekelund J et al.: Measurement of the Proportion of D2 Receptors Conigured in State of High Ainity for Agonists in Vivo: A Positron Emission Tomography Study Using [11C]N-Propyl-norapomorphine and [11C]Raclopride in Baboons. J Pharmacol Exp Ther 2005, 315: 80-90. 74. Narendran R, Hwang DR, Slifstein M, Talbot PS, Erritzoe D, Huang YY et al.: In vivo vulnerability to competition by endogenous dopamine: Comparison of the D-2 receptor agonist radiotracer (-)-N-[C-11]propyl-norapomorphine ([I-11]NPA) with the D-2 receptor antagonist radiotracer [C-11]-raclopride. Synapse 2004, 52: 188-208. 75. Hwang DR, Narendran R, Laruelle M: Positron-labeled dopamine agonists for probing the high ainity states of dopamine subtype 2 receptors. Biocon Chem 2005, 16: 27-31. 76. Narendran R, Mason NS, Laymon CM, Lopresti BJ, Velasquez ND, May MA et al.: A Comparative Evaluation of the Dopamine D(2/3) Agonist Radiotracer [(11)C](-)-N-Propyl-norapomorphine and Antagonist [(11)C]Raclopride to Measure Amphetamine-Induced Dopamine Release in the Human Striatum. J Pharmacol Exp Ther 2010, 333: 533-539.

154

References

77. Zijlstra S, Visser GM, Korf J, Vaalburg W: Synthesis and in vivo distribution in the rat of several luorine-18 labeled N-luoroalkylaporphines. Appl Radiat Isotopes 1993, 44: 651-658. 78. Stepanov V, Finnema SJ, Nakao R, Sromek AW, Zhang TZ, Neumeyer JL et al.: Radiolabeling and monkey PET evaluation of the dopamine D(2) and D(3) receptor agonist radioligand [(18)F]MCL- 524. J Labelled Comp Rad 2011, 54: S9. 79. Finnema SJ, Stepanov V, Nakao R, Amini N, Sromek AW, Neumeyer JL et al.: Quantiication and Whole-Body Distribution of a Novel Dopamine D-2/D-3 Receptor Agonist, [F-18]MCL-524, in Monkeys: A Prediction for Application in Human Subjects. Eur J Nucl Med Mol Imaging 2012, 39: S181-S182. 80. Gao Y, Ram VJ, Campbell A, Kula NS, Baldessarini RJ, Neumeyer JL: Synthesis and structural requirements of N-substituted norapomorphines for ainity and activity at dopamine D-1, D-2, and agonist receptor sites in rat brain. J Med Chem 1990, 33: 39-44. 81. Tokunaga M, Seneca N, Shin RM, Maeda J, Obayashi S, Okauchi T et al.: Neuroimaging and Physi- ological Evidence for Involvement of Glutamatergic Transmission in Regulation of the Striatal Dopaminergic System. J Neurosci 2009, 29: 1887-1896. 82. Seneca N, Skinbjerg M, Zoghbi SS, Liow JS, Gladding RL, Hong J et al.: Kinetic brain analysis and whole-body imaging in monkey of [C-11]MNPA: A dopamine agonist radioligand. Synapse 2008, 62: 700-709. 83. Palner M, McCormick P, Gillings N, Begtrup M, Wilson AA, Knudsen GM: Radiosynthesis and ex vivo evaluation of (R)-(-)-2-chloro-N-[1-(11)C-propyl]n-propylnorapomorphine. Nucl Med Biol 2010, 37: 35-40. 84. Wilson AA, McCormick P, Kapur S, Willeit M, Garcia A, Hussey D et al.: Radiosynthesis and evaluation of [11C]-(+)-4-propyl-3,4,4a,5,6,10b-hexahydro-2H-naphtho[1,2-b][1,4]oxazin -9-ol as a potential radiotracer for in vivo Imaging of the dopamine D2 high-ainity state with positron emission tomography. J Med Chem 2005, 48: 4153-4160. 85. Narendran R, Frankle WG, Mason NS, Laymon CM, Lopresti BJ, Price JC et al.: Positron Emission Tomography Imaging of D(2/3) Agonist Binding in Healthy Human Subjects With the Radiotracer [(11)C]-N-Propyl-Norapomorphine: Preliminary Evaluation and Reproducibility Studies. Synapse 2009, 63: 574-584. 86. Hwang DR, Narendran R, Huang YY, Slifstein M, Talbot PS, Sudo Y et al.: Quantitative analysis of (-)-N-C-11-propyl-norapomorphine in vivo binding in nonhuman primates. J Nucl Med 2004, 45: 338-346. 87. Cumming P, Gillings NM, Jensen SB, Bjarkam C, Gjedde A: Kinetics of the uptake and distribution of the dopamine D-2,D-3 agonist (R)-N-[1-C-11]n-propylnorapomorphine in brain of healthy and MPTP-treated Gottingen miniature pigs. Nucl Med Biol 2003, 30: 547-553. 88. Ginovart N, Galineau L, Willeit M, Mizrahi R, Bloomield PM, Seeman P et al.: Binding characteristics and sensitivity to endogenous dopamine of [C-11]-(+)-PHNO, a new agonist radiotracer for imaging the high-ainity state of D-2 receptors in vivo using positron emission tomography. J Neurochem 2006, 97: 1089-1103. 89. Sromek AW, Si YG, Zhang TZ, George SR, Seeman P, Neumeyer JL: Synthesis and Evaluation of Fluorinated Aporphines: Potential Positron Emission Tomography Ligands for D(2) Receptors. ACS Med Chem Lett 2011, 2: 189-194. 90. Mukherjee J, Narayanan TK, Christian BT, Shi BZ, Dunigan KA, Mantil J: In vitro and in vivo evaluation of the binding of the dopamine D2 receptor agonist C-11-(R,S)-5-hydroxy-2-(di-n- propylamino)tetralin in rodents and nonhuman primate. Synapse 2000, 37: 64-70.

155

References

91. Shi B, Narayanan TK, Christian BT, Chattopadhyay S, Mukherjee J: Synthesis and biological evaluation of the binding of dopamine D2/D3 receptor agonist, (R,S)-5-hydroxy-2-(N-propyl-N-(5’-18F- luoropentyl)aminotetralin (18F-5-OH-FPPAT) in rodents and nonhuman primates. Nucl Med Biol 2004, 31: 303-311. 92. Mukherjee J, Narayanan TK, Christian BT, Shi BZ, Yang ZY: Binding characteristics of high-ainity dopamine D2/D3 receptor agonists, C-11-PPHT and C-11-ZYY-339 in rodents and imaging in non- human primates by PET. Synapse 2004, 54: 83-91. 93. Shi B, Narayanan TK, Yang ZY, Christian BT, Mukherjee J: Radiosynthesis and in vitro evaluation of 2-(N-alkyl-N-1’-11C-propyl)amino-5-hydroxytetralin analogs as high ainity agonists for dopamine D-2 receptors. Nucl Med Biol 1999, 26: 725-735. 94. Willeit M, Ginovart N, Kapur S, Houle S, Hussey D, Seeman P et al.: High-ainity states of human brain dopamine D2/3 receptors imaged by the agonist [C-11]-(+)-PHNO. Biol Psychiat 2006, 59: 389-394. 95. Ginovart N, Willeit M, Rusjan P, Graf A, Bloomield PM, Houle S et al.: Positron emission tomography quantiication of [C-11]-(+)-PHNO binding in the human brain. J Cereb Blood Flow Metab 2007, 27: 857-871. 96. Narendran R, Slifstein M, Guillin O, Hwang YY, Hwang DR, Scher E et al.: Dopamine (D-2/3) receptor agonist positron emission tomography radiotracer [C-11)-(+)-PHNO is a D-3 receptor preferring agonist in vivo. Synapse 2006, 60: 485-495. 97. Willeit M, Ginovart N, Graf A, Rusjan P, Vitcu I, Houle S et al.: First human evidence of d-amphetamine induced displacement of a D(2/3) agonist radioligand: A [(11)C]-(+)-PHNO positron emission tomography study. Neuropsychopharmacol 2008, 33: 279-289. 98. Graf-Guerrero A, Willeit M, Ginovart N, Mamo D, Mizrahi R, Rusjan P et al.: Brain region binding of the D-2/3 agonist [C-11]-(+)-PHNO and the D-2/3 antagonist [C-11]raclopride in healthy humans. Hum Brain Map 2008, 29: 400-410. 99. Graf-Guerrero A, Mizrahi R, Agid O, Marcon H, Barsoum P, Rusjan P et al.: The Dopamine D(2) Receptors in High-Ainity State and D(3) Receptors in Schizophrenia: A Clinical [(11)C]-(+)-PHNO PET Study. Neuropsychopharmacol 2009, 34: 1078-1086. 100. Graf-Guerrero A, Mamo D, Shammi CM, Mizrahi R, Marcon H, Barsoum P et al.: The Efect of Anti- psychotics on the High-Ainity State of D(2) and D(3) Receptors A Positron Emission Tomography Study With [(11)C]-(+)-PHNO. Arch Gen Psychiat 2009, 66: 606. 101. Rabiner EA, Slifstein M, Nobrega J, Plisson C, Huiban M, Raymond R et al.: In vivo quantiication of regional dopamine-D3 receptor binding potential of (+)-PHNO: Studies in non-human primates and transgenic mice. Synapse 2009, 63: 782-793. 102. Searle G, Beaver JD, Comley RA, Bani M, Tziortzi A, Slifstein M et al.: Imaging Dopamine D(3) Receptors in the Human Brain with Positron Emission Tomography, [(11)C]PHNO, and a Selective D(3) Receptor Antagonist. Biol Psychiat 2010, 68: 392-399. 103. Graf-Guerrero A, Redden L, bi-Saab W, Katz DA, Houle S, Barsoum P et al.: Blockade of [(11)C] (+)-PHNO binding in human subjects by the dopamine D(3) receptor antagonist ABT-925. Int J Neuropsychopharmacol 2010, 13: 273-287. 104. Egerton A, Hirani E, Ahmad R, Turton DR, Brickute D, Rosso L et al.: Further Evaluation of the Carbon11-Labeled D(2/3) Agonist PET Radiotracer PHNO: Reproducibility in Tracer Characteris- tics and Characterization of Extrastriatal Binding. Synapse 2010, 64: 301-312. 105. Seeman P, Ko F, Willeit M, McCormick P, Ginovart N: Antiparkinson concentrations of pramipexole and PHNO occupy dopamine D2high and D3high receptors. Synapse 2005, 58: 122-128.

156

References

106. Kuhnast B, Valette H, Besret L, Demphel S, Coulon C, Ottaviani M et al.: Synthesis and radiolabeling of N-[4-[4-(2-[11C]methoxyphenyl)piperazin-1-yl]butyl]benzo[b]thiophene-2-carboxamide - a potential radiotracer for D3 receptor imaging with PET. Nucl Med Biol 2006, 33: 785-795. 107. Hocke C, Maschauer S, Hubner H, Lober S, Utz W, Kuwert T et al.: A Series of (18)F-Labelled Pyridinylphenyl Amides as Subtype-Selective Radioligands for the Dopamine D3 Receptor. ChemMedChem 2010, 5: 941-948. 108. Mach RH, Tu ZD, Xu JB, Li SH, Jones LA, Taylor M et al.: Endogenous Dopamine (DA) Competes With the Binding of a Radiolabeled D(3) Receptor Partial Agonist In Vivo: A Positron Emission Tomography Study. Synapse 2011, 65: 724-732. 109. Hocke C, Prante O, Salama I, Hubner H, Lober S, Kuwert T et al.: (18)F-labeled FAUC 346 and BP 897 derivatives as subtype-selective potential PET radioligands for the dopamine D3 receptor. ChemMedChem 2008, 3: 788-793. 110. Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ et al.: International Union of Pharmacology classiication of receptors for 5-hydroxytryptamine (Serotonin). Pharmacol Rev 1994, 46: 157-203. 111. Kitson SL. 5-Hydroxytryptamine (5-HT) Receptor Ligands. Curr Pharm Des 2007, 13[25]: 2621- 2637. 112. Nichols DE, Nichols CD: Serotonin Receptors. Chem Rev 2008, 108: 1614-1641. 113. Lin SL, Setya S, Johnson-Farley NN, Cowen DS: Diferential coupling of 5-HT1 receptors to G proteins of the Gi family. Br J Pharmacol 2002, 136: 1072-1078. 114. Valdizán EM, Castro E, Pazos A: Agonist-dependent modulation of G-protein coupling and trans-

duction of 5-HT1A receptors in rat dorsal raphe nucleus. Int J Neuropsychopharmacol 2010, 13: 835-843. 115. Bonaventure P, Schotte A, Cras P, Leysen JE: Autoradiographic mapping of 5-HT1B and 5-HT1D receptors in human brain using [H-3]alniditan, a new radioligand. Receptor Channel 1997, 5: 225- 230. 116. Bonaventure P, Voorn P, Luyten WHML, Jurzak M, Schotte A, Leysen JE: Detailed mapping of serotonin 5-HT1B and 5-HT1D receptor messenger RNA and ligand binding sites in guinea-pig brain and trigeminal ganglion: clues for function. Neuroscience 1997, 82: 469-484. 117. Klein MT, Dukat M, Glennon RA, Teitler M: Towards selective drug development for the human 5-HT1E receptor: a comparison of 5-HT1E and 5-HT1F receptor structure-ainity relationships. J Pharmacol Exp Ther 2011, 337: 860-867. 118. Leysen JE: 5-HT2 receptors. Curr Drug Targets CNS Neurol Disord 2004, 3: 11-26. 119. Bonhaus DW, Bach C, DeSouza A, Salazar FHR, Matsuoka BD, Zuppan P et al.: The pharmacology and distribution of human 5-hydroxytryptamine2B (5-HT2b) receptor gene products: comparison with 5-HT2a and 5-HT2c receptors. Br J Pharmacol 1995, 115: 622-628. 120. Jakab RL, Goldman-Rakic PS: 5-Hydroxytryptamine2A serotonin receptors in the primate cerebral cortex: Possible site of action of hallucinogenic and antipsychotic drugs in pyramidal cell apical dendrites. PNAS 1998, 95: 735-740. 121. Smith SR, Prosser WA, Donahue DJ, Morgan ME, Anderson CM, Shanahan WR et al.: Lorcaserin (APD356), a Selective 5-HT2C Agonist, Reduces Body Weight in Obese Men and Women. Obesity 2009, 17: 494-503. 122. Bockaert J, Claeysen S, Compan V, Dumuis A: 5-HT4 Receptors. Curr Drug Targets CNS Neurol Disord 2004, 3[1]: 39-51.

157

References

123. Francken BJB, Jurzak M, Vanhauwe JFM, Luyten WHML, Leysen JE: The human 5-ht5A receptor couples to Gi/Go proteins and inhibits adenylate cyclase in HEK 293 cells. Eur J Pharmacol 1998, 361: 299-309. 124. Nelson DL: 5-HT5 Receptors. Curr Drug Targets CNS Neurol Disord 2004 3[1]: 53-58. 125. Holenz J, Pauwels PJ, Diaz JL, Merce R, Codony X, Buschmann H: Medicinal chemistry strategies to 5-HT6 receptor ligands as potential cognitive enhancers and antiobesity agents. Drug Discov Today 2006, 11: 283-299. 126. Ruat M, Traifort E, Leurs R, Tardivel-Lacombe J, Diaz J, Arrang JM et al.: Molecular cloning, characterization, and localization of a high-ainity serotonin receptor (5-HT7) activating cAMP formation. PNAS 1993, 90: 8547-8551. 127. Gustafson EL, Durkin MM, Bard JA, Zgombick J, Branchek TA: A receptor autoradiographic and in situ hybridization analysis of the distribution of the 5-ht7 receptor in rat brain. Br J Pharmacol 1996, 117: 657-666. 128. Bonaventure P, Dugovic C, Kramer M, De Boer P, Singh J, Wilson S et al.: Translational Evaluation of JNJ-18038683, a 5-Hydroxytryptamine Type 7 Receptor Antagonist, on Rapid Eye Movement Sleep and in Major Depressive Disorder. J Pharmacol Exp Ther 2012, 342: 429-440. 129. Brenchat A, Nadal X, Romero L, Ovalle S, Muro An, Sanchez-Arroyos R et al.: Pharmacological activation of 5-HT7 receptors reduces nerve injury-induced mechanical and thermal hypersensitivity. Pain 2010, 149[3]: 483-494. 130. Paterson LM, Kornum BR, Nutt DJ, Pike VW, Knudsen GM: 5-HT radioligands for human brain imaging with PET and SPECT. Med Res Rev 2013, 33: 54-111. 131. Kumar JSD, Majo VJ, Hsiung SC, Millak MS, Liu KP, Tamir H et al.: Synthesis and in vivo validation of [O-methyl-C-11]2-{4-[4-(7-methoxynaphthalen-1-yl)piperazin-yl]butyl}-4-m ethy1-2H-[1,2,4] triazine-3,5-dione: A novel 5-HT1A receptor agonist positron emission tomography ligand. J Med Chem 2006, 49: 125-134. 132. Prabhakaran J, Parsey RV, Majo VJ, Hsiung SC, Milak MS, Tamir H et al.: Synthesis, in vitro and in vivo evaluation of [O-methyl-C-11] 2-{4-[4-(3-methoxyphenyl)piperazin-1-yl]-butyl}-4-methyl- 2H-[1,2,4]-tria zine-3,5-dione: A novel agonist 5-HT1A receptor PET ligand. Bioorg Med Chem Lett 2006, 16: 2101-2104. 133. Kumar JSD, Prabhakaran J, Majo VJ, Milak MS, Hsiung SC, Tamir H et al.: Synthesis and in vivo evaluation of a novel 5-HT1A receptor agonist radioligand [O-methyl-C-11]2-(4-(4-(2-methoxyphenyl) piperazin-1-yl)butyl)-4-methyl-1,2,4-triazine-3, 5(2H,4H)dione in nonhuman primates. Eur J Nucl Med Mol Imaging 2007, 34: 1050-1060. 134. Milak MS, Kumar JSD, Prabhakaran J, Majo VJ, Mann JJ, Parsey RV: [C-11]CUMI-101 an agonist 5-HT1A positron emission tomography (PET) tracer in human: Preliminary test-retest data. Neuro- Image 2008, 41: T96. 135. Milak MS, DeLorenzo C, Zanderigo F, Prabhakaran J, Kumar JSD, Majo VJ et al.: In Vivo Quantiica- tion of Human Serotonin 1A Receptor Using 11C-CUMI-101, an Agonist PET Radiotracer. J Nucl Med 2010, 51: 1892-1900. 136. Mann JJ, Milak M, Kumar JSD, Majo VJ, Severance AJ, Ogden RT et al.: Modeling serotonin 1A receptors in baboon and human subjects using the agonist tracer [11C]CUMI-101 and positron emission tomography. NeuroImage 2008, 41: T18. 137. Milak MS, Severance AJ, Prabhakaran J, Kumar JSD, Majo VJ, Ogden RT et al.: In vivo serotonin- sensitive binding of [11C]CUMI-101: a serotonin 1A receptor agonist positron emission tomography radiotracer. J Cereb Blood Flow Metab 2011, 31: 243-249.

158

References

138. Kumar JSD, Mann JJ: PET tracers for 5-HT1A receptors and uses thereof. Drug Discov Today 2007, 12: 748-756. 139. Milak MS, Severance AJ, Ogden RT, Prabhakaran J, Kumar JSD, Majo VJ et al.: Modeling consider- ations for C-11-CUMI-101, an agonist radiotracer for imaging serotonin 1A receptor in vivo with PET. J Nucl Med 2008, 49: 587-596. 140. Majo VJ, Parsey RV, Prabhakaran J, Mann JJ, Kumar JSD: Synthesis and evaluation of [O-methyl-C- 11]4-[3-[4-(2-methoxyphenyl)-piperazin-1-yl]propoxyl-4-aza-t ricyclo[5.2.1.02,6]dec-8-ene-3,5- dione as a 5-HT1A receptor agonist PET ligand. J Labelled Comp Rad 2008, 51: 132-136. 141. Zimmer L, Fournet G, Joseph B, Guillaumet G, Le Bars D: Carbon-11 labelling of 8{{3-[4-(2-[C-11] methoxyphenyl)piperazin-1-yl]2-hydroxypropyl}oxy}thioch roman, a presynaptic 5-HT1A receptor agonist, and its in vivo evaluation in anaesthetised rat and in awake cat. Nucl Med Biol 2003, 30: 541-546. 142. Lu SY, Liow JS, Ichise M, Hong J, Musachio JL, Zoghbi SS et al.: Evaluation of [O-Methyl-C-11] S14506 As An Agonist Radioligand for Brain 5-Ht1A Receptors in Rat and Monkey. J Labelled Comp Rad 2005, 48: S72. 143. Lemoine L, Verdurand M, Vacher B, Blanc E, Le Bars D, Newman-Tancredi A et al.: [18F]F15599, a novel 5-HT(1A) receptor agonist, as a radioligand for PET neuroimaging. Eur J Nucl Med Mol Imag- ing 2010, 37: 594-605. 144. Lima L, Laporte AM, Gaymard C, Spedding M, Mocaër E, Hamon M: Atypical in vitro and in vivo binding of [3H]S-14506 to brain 5-HT1A receptors. J Neural Transmis 1997, 104: 1059-1075. 145. Maurel JL, Autin JM, Funes P, Newman-Tancredi A, Colpaert F, Vacher B: High-eicacy 5-HT1A agonists for antidepressant treatment: a renewed opportunity. J Med Chem 2007, 50: 5024-5033. 146. Shannon M, Battaglia G, Glennon RA, Titeler M: 5-HT1 and 5-HT2 binding properties of derivatives of the hallucinugen 1-(2,5-dimethoxyphenyl)-2-aminopropane (2,5-DMA). Eur J Pharmacol 1984, 102: 23-29. 147. Ettrup A, Palner M, Gillings N, Santini MA, Hansen M, Kornum BR et al.: Radiosynthesis and Evalu- ation of (11)C-CIMBI-5 as a 5-HT(2A) Receptor Agonist Radioligand for PET. J Nucl Med 2010, 51: 1763-1770. 148. Ettrup A, Hansen M, Santini MA, Paine J, Gillings N, Palner M et al.: Radiosynthesis and in vivo evaluation of a series of substituted (11)C-phenethylamines as 5-HT(2A) agonist PET tracers. Eur J Nucl Med Mol Imaging 2011, 38: 681-693. 149. Finnema S, Ettrup A, Stepanov V, Nakao R, Yamamoto S, Varrone A et al.: Pilot study on receptor binding and serotonin sensitivity of [11C]CIMBI-36 in monkey brain. J Nucl Med Meeting Abstracts 2011, 52: 495. 150. Bartus RT: On neurodegenerative diseases, models, and treatment strategies: Lessons learned and lessons forgotten a generation following the cholinergic hypothesis. Exp Neurol 2000, 163: 495-529. 151. Caulield MP: Muscarinic Receptors - Characterization, coupling and function. Pharmacol Ther 1993, 58: 319-379. 152. Wess J, Eglen RM, Gautam D: Muscarinic acetylcholine receptors: mutant mice provide new insights for drug development. Nat Rev Drug Discov 2007, 6: 721-733. 153. Mash DC, Potter LT: Autoradiographic Localization of M1 and M2 Muscarine Receptors in the Rat- Brain. Neuroscience 1986, 19: 551-564. 154. Levey AI: Immunological localization of m1-m5 muscarinic acetylcholine receptors in peripheral tissues and brain. Life Sci 1993, 52: 441-448.

159

References

155. Bymaster FP, Felder C, Ahmed S, McKinzie D: Muscarinic Receptors as a Target for Drugs Treating Schizophrenia. Curr Drug Targets CNS Neurol Disord 2002, 1: 163-181. 156. Bymaster FP, Whitesitt CA, Shannon HE, DeLapp N, Ward JS, Calligaro DO et al.: Xanomeline: A selective muscarinic agonist for the treatment of Alzheimer’s disease. Drug Dev Res 1997, 40: 158- 170. 157. Buiter H, Windhorst A, Huisman M, Yaqub M, Knol D, Fisher A et al.: [11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors. EJNMMI Res 2013, 3: 19. 158. Farde L, Suhara T, Halldin C, Nyback H, Nakashima Y, Swahn CG et al.: PET study of the M(1)-agonists [C-11]xanomeline and [C-11]butylthio-TZTP in monkey and man. Dementia 1996, 7: 187-195. 159. Kiesewetter DO, Lee J, Lang L, Park SG, Paik CH, Eckelman WC: Preparation of 18F-Labeled Musca- rinic Agonist with M2 Selectivity. J Med Chem 1995, 38: 5-8. 160. Podruchny TA, Connolly C, Bokde A, Herscovitch P, Eckelman WC, Kiesewetter DO et al.: In vivo muscarinic 2 receptor imaging in cognitively normal young and older volunteers. Synapse 2003, 48: 39-44. 161. Carson RE, Kiesewetter DO, Jagoda E, Der MG, Herscovitch P, Eckelman WC: Muscarinic Choliner- gic Receptor Measurements With [18F]FP-TZTP: Control and Competition Studies. J Cereb Blood Flow Metab 1998, 18: 1130-1142. 162. Kiesewetter DO, Carson RE, Jagoda EM, Herscovitch P, Eckelman WC: In vivo muscarinic binding of 3-(alkylthio)-3-thiadiazolyl tetrahydropyridines. Synapse 1999, 31: 29-40. 163. Reid AE, Ding YS, Eckelman WC, Logan J, Alexof D, Shea C et al.: Comparison of the pharmacokinetics of diferent analogs of C-11-labeled TZTP for imaging muscarinic M2 receptors with PET. Nucl Med Biol 2008, 35: 287-298. 164. Sedman AJ, Bockbrader H, Schwarz RD: Preclinical and Phase 1 clinical characterization of CI-979/ RU35926, a novel muscarinic agonist for the treatment of Alzheimer’s disease. Life Sci 1995, 56: 877-882. 165. Hartvig P, Torstenson R, Bjurling P, Fasth KJ, Langström B, Nordberg A: Regional brain distribution and binding of the muscarinic receptor agonist CI-979 studied by positron emission tomography in the monkey. Dement Geriat Cogn 1997, 8: 259-266. 166. Cohen RM, Podruchny TA, Bokde ALW, Carson RE, Herscovitch P, Kiesewetter DO et al.: Higher in vivo muscarinic-2 receptor distribution volumes in aging subjects with an apolipoprotein E-ε4 allele. Synapse 2003, 49: 150-156. 167. Jones CK, Byun N, Bubser M: Muscarinic and Nicotinic Acetylcholine Receptor Agonists and Al- losteric Modulators for the Treatment of Schizophrenia. Neuropsychopharmacol 2012, 37: 16-42. 168. Alt A, Clark MJ, Woods JH, Traynor JR: Mu and Delta opioid receptors activate the same G proteins in human neuroblastoma SH-SY5Y cells. Br J Pharmacol 2002, 135: 217-225. 169. Dhawan BN, Cesselin F, Raghubir R, Reisine T, Bradley PB, Portoghese PS et al.: International Union of Pharmacology. XII. Classiication of opioid receptors. Pharmacol Rev 1996, 48: 567-592. 170. Leysen JE, Gommeren W, Niemegeers CJE: [3H]sufentanil, a superior ligand for μ-opiate receptors: Binding properties and regional distribution in rat brain and spinal cord. Eur J Pharmacol 1983, 87: 209-225. 171. Lever JR: PET and SPECT imaging of the opioid system: Receptors, radioligands and avenues for drug discovery and development. Cur Pharm Des 2007, 13: 33-49. 172. Henriksen G, Willoch F: Imaging of opioid receptors in the central nervous system. Brain 2008, 131: 1171-1196. 173. Roth BL, Baner K, Westkaemper R, Siebert D, Rice KC, Steinberg S et al.: Salvinorin A: A potent naturally occurring nonnitrogenous κ opioid selective agonist. PNAS 2002, 99: 11934-11939.

160

References

174. Hooker JM, Xu Y, Schifer W, Shea C, Carter P, Fowler JS: Pharmacokinetics of the potent halluci- nogen, salvinorin A in primates parallels the rapid onset and short duration of efects in humans. NeuroImage 2008, 41: 1044-1050. 175. Endres CJ, Bencherif B, Hilton J, Madar I, Frost JJ: Quantiication of brain mu-opioid receptors with [C-11]carfentanil: reference-tissue methods. Nucl Med Biol 2003, 30: 177-186. 176. Pichika R, Jewett DM, Sherman PS, Traynor JR, Husbands SM, Woods JH et al.: Synthesis and in vivo brain distribution of carbon-11-labeled delta-opioid receptor agonists. Nucl Med Biol 2010, 37: 989-996. 177. Riss PJ, Hong YT, Marton J, Caprioli D, Williamson DJ, Ferrari V et al.: Synthesis and Evaluation of 18F-FE-PEO in Rodents: An 18F-Labeled Full Agonist for Opioid Receptor Imaging. J Nucl Med 2013, 54: 299-305. 178. Frost JJ, Wagner HNJ, Dannals RF, Ravert HT, Links JM, Wilson AA et al.: Imaging Opiate Receptors in the Human Brain by Positron Tomography. J Comp Ass Tomo 1985, 9: 231-236. 179. Bencherif B, Fuchs PN, Sheth R, Dannals RF, Campbell JN, Frost JJ: Pain activation of human supraspinal opioid pathways as demonstrated by [C-11]-carfentanil and positron emission tomography (PET). Pain 2002, 99: 589-598. 180. Marton J, Schoultz BW, Hjørnevik T, Drzezga A, Yousei BH, Wester HJ et al.: Synthesis and Evalua- tion of a Full-Agonist Orvinol for PET-Imaging of Opioid Receptors: [11C]PEO. J Med Chem 2009, 52: 5586-5589. 181. Howlett AC, Barth F, Bonner TI, Cabral G, Casellas P, Devane WA et al.: International Union of Pharmacology. XXVII. Classiication of Cannabinoid Receptors. Pharmacol Rev 2002, 54: 161-202. 182. Mackie K: Cannabinoid receptors as therapeutic targets. Annu Rev Pharmacol Toxicol 2006, 46: 101-122. 183. Charalambous A, Marciniak G, Shiue CY, Dewey SL, Schlyer DJ, Wolf AP et al.: PET studies in the primate brain and biodistribution in mice using (-)-5’-18F-Δ8-THC. Pharmacol Biochem Behav 1991, 40: 503-507. 184. Dubocovich ML, Delagrange P, Krause DN, Sugden D, Cardinali DP, Olcese J: International Union of Basic and Clinical Pharmacology. LXXV. Nomenclature, Classiication, and Pharmacology of G Protein-Coupled Melatonin Receptors. Pharmacol Rev 2010, 62: 343-380. 185. Bars DL, Thivolle P, Vitte PA, Bojkowski C, Chazot G, Arendt J et al.: PET and plasma pharmacokinetic studies after bolus intravenous administration of [11C]melatonin in humans. Int J Radiat Appl Instrum Part B, Nucl Med Biol 1991, 18: 357-362. 186. Takashima-Hirano M, Tazawa S, Takahashi K, Doi H, Suzuki M: Eicient Synthesis of [11C]Ramelteon as a Positron Emission Tomography Probe for Imaging Melatonin Receptors Involved in Circadian Rhythms. Chem Pharm Bull 2011, 59: 1062-1064. 187. Chen JJ, Fiehn-Schulze B, Brough PA, Snieckus V, Firnau G: Synthesis of 2-iodo- and 2-phenyl-[11C] melatonin: potential PET tracers for melatonin binding sites. Appl Radiat Isotopes 1998, 49: 1573- 1579. 188. De Lean A, Stadel JM, Lefkowitz RJ: A ternary complex model explains the agonist-speciic binding properties of the adenylate cyclase-coupled beta-adrenergic receptor. J Biol Chem 1980, 255: 7108-7117. 189. Lef P: The two-state model of receptor activation. Trends Pharmacol Sci 1995, 16: 89-97. 190. Mirza NR, Peters D, Sparks RG: Xanomeline and the Antipsychotic Potential of Muscarinic Recep- tor Subtype Selective Agonists. CNS Drug Rev 2003, 9: 159-186. 191. Vora MM, Finn RD, Boothe TE, Liskwosky DR, Potter LT: [N-Methyl-C-11]-Scopolamine - Synthesis and Distribution in Rat-Brain. J Labelled Comp Rad 1983, 20: 1229-1236.

161

References

192. Mulholland GK, Kilbourn MR, Sherman P, Carey JE, Frey KA, Koeppe RA et al.: Synthesis, in vivo biodistribution and dosimetry of [11C]N-Methylpiperidyl benzilate ([11C]NMPB), a muscarinic acetylcholine receptor antagonist. Nucl Med Biol 1995, 22: 13-17. 193. Dewey SL, Macgregor RR, Brodie JD, Bendriem B, King PT, Volkow ND et al.: Mapping Muscarinic Receptors in Human and Baboon Brain Using [N-C-11-Methyl]-Benztropine. Synapse 1990, 5: 213- 223. 194. Varastet M, Brouillet E, Chavoix C, Prenant C, Crouzel C, Stulzaft O et al.: In vivo visualization of central muscarinic receptors using [11C]quinuclidinyl benzilate and positron emission tomography in baboons. Eur J Pharmacol 1992, 213: 275-284. 195. Takahashi K, Murakami M, Miura S, Iida H, Kanno I, Uemura K et al.: Radiosynthesis and biodistribution of [11C]3NMPB enantiomers in rats for studying muscarinic cholinergic receptors using PET. J Labelled Comp Rad 1997, 40: 613-615. 196. Kiesewetter DO, Jagoda EM, Shimoji K, Ma Y, Eckelman WC: Evaluation of [F-18]luoroxanomeline {5-{4-[(6-[F-18]luorohexyl)oxy]-1 2,5-thiadiazol-3-yl}-1-methyl-1,2,3,6-tetrahydropyridine}. Nucl Med Biol 2007, 34: 141-152. 197. Fisher A, Bar-Ner N, Karton Y. Methods and compositions for treatment of central and peripheral nervous system disorders and novel compounds useful therefore. 2013 Patent [WO 03/092580 A2]. 198. Brandeis R, Sapir M, Haif N, Abraham S, Oz N, Stein E et al.: AF150(S): A new functionally selective M1 agonist improves cognitive performance in rats. Pharmacol Biochem Behav 1995, 51: 667-674. 199. Beach TG: Muscarinic agonists as preventative therapy for Alzheimer’s disease. Curr Opin Invest Drug 2002, 3: 1633-1636. 200. Beach TG, Walker DG, Potter PE, Sue LI, Fisher A: Reduction of cerebrospinal luid amyloid β after systemic administration of M1 muscarinic agonists. Brain Res 2001, 905: 220-223. 201. Fisher A, Brandeis R, Bar-Ner R, Kliger-Spatz M, Natan N, Sonego H et al.: AF150(S) and AF267B, M1 muscarinic agonists as innovative therapies for alzheimer’s disease. J Mol Neurosci 2002, 19: 145-153. 202. Caccamo A, Oddo S, Billings LM, Green KN, Martinez-Coria H, Fisher A et al.: M1 receptors play a central role in modulating AD-like pathology in transgenic mice. Neuron 2006, 49: 671-682. 203. Fisher A, Heldman E, Gurwitz D, Haring R, Meshulam H, Brandeis R et al.: AF150(S) and AF151(S): New M1 agonists mediate m1 selective signaling, neurotrophic-like efects and restore AF64A cognitive deicits in rats. Soc Neurosci Abstr 1993, 19: 1767. 204. Windhorst AD, Linden ter T, Nooij de A: J Labelled Comp Rad 2001, 44: s1052-s1055. 205. Comar D, Cartron JC, Maziere M, Marazano C: Labeling and Metabolism of Methionine-Methyl- C-11. Eur J Nucl Med 1976, 1: 11-14. 206. Crouzel C, Langstrom B, Pike VW, Coenen HH: Recommendations for A Practical Production of [C-11] Methyl-Iodide. Appl Radiat Isotopes 1987, 38: 601-603. 207. Langström B, Lundqvist H: The preparation of 11C-methyl iodide and its use in the synthesis of 11C-methyl-methionine. Int J Appl Radiat Isotopes 1976, 27: 357-363. 208. Langström B, Sjoberg S, Bergson G, Lundqvist H, Malmborg P, Stalnacke CG et al.: [C-11]Methyl- Iodide in the Synthesis of C-11-Compounds. J Labelled Comp Rad 1981, 18: 17-19. 209. Luurtsema G, Windhorst AD, Mooijer MPJ, Herscheid JDM, Lammertsma AA, Franssen EJF: Fully automated high yield synthesis of (R)- and (S)-[C-11]verapamil for measuring P-glycoprotein function with positron emission tomography. J Labelled Comp Rad 2002, 45: 1199-1207.

162

References

210. Windhorst AD, Timmerman H, Klok RP, Menge WMPB, Leurs R, Herscheid JDM: Evaluation of [18F] VUF 5000 as a potential PET ligand for brain imaging of the histamine H3 receptor. Bioorg Med Chem 1999, 7: 1761-1767. 211. Gattu M, Pauly JR, Urbanawiz S, Buccafusco JJ: Autoradiographic comparison of muscarinic M1 and M2 binding sites in the CNS of spontaneously hypertensive and normotensive rats. Brain Res 1997, 771: 173-183. 212. Paxinos G, Watson C: The rat brain in stereotaxic coordinates, 6th edn. Amsterdam: Elsevier; 2007. 213. Airaksinen AJ, Jablonowski JA, van der MM, Barbier AJ, Klok RP, Verbeek J et al.: Radiosynthesis and biodistribution of a histamine H3 receptor antagonist 4-[3-(4-piperidin-1-yl-but-1-ynyl)-[11C] benzyl]-morpholine: evaluation of a potential PET ligand. Nucl Med Biol 2006, 33: 801-810. 214. Hulme EC, Birdsall NJM, Burger ASV, Metha P: The Binding of Antagonists to Brain Muscarinic Receptors. Mol Pharmacol 1978, 14: 737-750. 215. Lever JR, Gustafson JL, Xu R, Allmon RL, Lever SZ: sigma(1) and sigma(2) receptor binding ainity and selectivity of SA4503 and luoroethyl SA4503. Synapse 2006, 59: 350-358. 216. Tam SW, Cook L: Sigma opiates and certain antipsychotic drugs mutually inhibit (+)-[3H] SKF 10,047 and [3H]haloperidol binding in guinea pig brain membranes. PNAS 1984, 81: 5618-5621. 217. Bymaster FP, Wong DT, Mitch CH, Ward JS, Calligaro DO, Schoepp DD et al.: Neurochemical efects of the M1 muscarinic agonist xanomeline (LY246708/NNC11-0232). J Pharmacol Exp Ther 1994, 269: 282-289. 218. Shannon HE, Rasmussen K, Bymaster FP, Hart JC, Peters SC, Swedberg MDB et al.: Xanomeline, an M1/M4 preferring muscarinic cholinergic receptor agonist, produces antipsychotic-like activity in rats and mice. Schizophr Res 2000, 42: 249-259. 219. Watson M, Yamamura HI, Roeske WR: A unique regulatory proile and regional distribution of [3H]pirenzepine binding in the rat provide evidence for distinct M1 and M2 muscarinic receptor subtypes. Life Sci 1983, 32: 3001-3011. 220. Potter LT, Ferrendelli CA, Hanchett HE: 2 Ainity States of M1 Muscarine Receptors. Cell Mol Neurobiol 1988, 8: 181-191. 221. Trovero F, Brochet D, Breton P, Tambuté A, Bégos A, Bizot JC: Pharmacological Proile of CEB-1957 and Atropine toward Brain Muscarinic Receptors and Comparative Study of Their Eicacy against Sarin Poisoning. Tox Appl Pharmacol 1998, 150: 321-327. 222. Hudkins RL, Haven-Hudkins DL: M1 muscarinic antagonists interact with [sigma] recognition sites. Life Sci 1991, 49: 1229-1235. 223. Ehlert FJ, Tran LP: Regional distribution of M1, M2 and non-M1, non-M2 subtypes of muscarinic binding sites in rat brain. J Pharmacol Exp Ther 1990, 255: 1148-1157. 224. Ehlert FJ, Roeske WR, Yamamura HI: Striatal muscarinic receptors: Regulation by dopaminergic agonists. Life Sci 1981, 28: 2441-2448. 225. Snyder SH, Greenberg D, Yamumura HI: Antischizophrenic drugs: Ainity for muscarinic cholinergic receptor sites in the brain predicts extrapyramidal efects. J Psychiat Res 1974, 11: 91-95. 226. Adem A, Karlsson E: Muscarinic receptor subtype selective toxins. Life Sci 1997, 60: 1069-1076. 227. El-Bakri NK, Adem A, Suliman IA, Mulugeta E, Karlsson E, Lindgren JU et al.: Estrogen and proges- terone treatment: efects on muscarinic M4 receptor subtype in the rat brain. Brain Res 2002, 948: 131-137. 228. Levey AI, Kitt CA, Simonds WF, Price DL, Brann MR: Identiication and localization of muscarinic acetylcholine receptor proteins in brain with subtype-speciic antibodies. J Neurosci 1991, 11: 3218-3226.

163

References

229. Schou M, Sovago J, Pike VW, Gulyas B, Bogeso KP, Farde L et al.: Synthesis and positron emission tomography evaluation of three norepinephrine transporter radioligands: [C-11]desipramine, [C-11]talopram and [C-11]talsupram. Mol Imaging Biol 2006, 8: 1-8. 230. Hansch C, Bjorkroth JP, Leo A: Hydrophobicity and central nervous system agents: on the principle of minimal hydrophobicity in drug design. J Pharm Sci 1987, 76: 663-687. 231. Sershen H, Lajtha A: Cerebral uptake of nicotine and of amino acids. J Neurosci Res 1979, 4: 85-91. 232. Groenendaal D, Freijer J, de Mik D, Bouw MR, Danhof M, de Lange ECM: Population pharmacokinetic modelling of non-linear brain distribution of morphine: inluence of active saturable inlux and P-glycoprotein mediated elux. Br J Pharmacol 2007, 151: 701-712. 233. Mazziotta JC, Toga AW, Evans A, Fox P, Lancaster J: A Probabilistic Atlas of the Human Brain: Theory and Rationale for Its Development : The International Consortium for Brain Mapping (ICBM). NeuroImage 1995, 2: 89-101. 234. Black KJ, Koller JM, Snyder AZ, Perlmutter JS: Template Images for Nonhuman Primate Neuroim- aging: 2. Macaque. NeuroImage 2001, 14: 744-748. 235. Andersen F, Watanabe H, Bjarkam C, Danielsen EH, Cumming P: Pig brain stereotaxic standard space: Mapping of cerebral blood low normative values and efect of MPTP-lesioning. Brain Res Bull 2005, 66: 17-29. 236. Svarer C, Madsen K, Hasselbalch SG, Pinborg LH, Haugb+©l S, Fr+©kj+ªr VG et al.: MR-based automatic delineation of volumes of interest in human brain PET images using probability maps. NeuroImage 2005, 24: 969-979. 237. Fujita M, Zoghbi SS, Crescenzo MS, Hong J, Musachio JL, Lu JQ et al.: Quantiication of brain phosphodiesterase 4 in rat with (R)-[11C]Rolipram-PET. NeuroImage 2005, 26: 1201-1210. 238. Kuntner C, Kesner A, Bauer M, Kremslehner R, Wanek T, Mandler M et al.: Limitations of Small Animal PET Imaging with [18F]FDDNP and FDG for Quantitative Studies in a Transgenic Mouse Model of Alzheimer’s Disease. Mol Imaging Biol 2009, 11: 236-240. 239. Doorduin J, Klein HC, de Jong JR, Dierckx RA, de Vries EFJ: Evaluation of [11C]-DAA1106 for imaging and quantiication of neuroinlammation in a rat model of herpes encephalitis. Nucl Med Biol 2010, 37: 9-15. 240. Alexof DL, Vaska P, Marsteller D, Gerasimov T, Li J, Logan J et al.: Reproducibility of 11C-Raclopride Binding in the Rat Brain Measured with the MicroPET R4: Efects of Scatter Correction and Tracer Speciic Activity. J Nucl Med 2003, 44: 815-822. 241. Topping GJ, Dinelle K, Kornelsen R, McCormick S, Holden JE, Sossi V: Positron emission tomography kinetic modeling algorithms for small animal dopaminergic system imaging. Synapse 2010, 64: 200-208. 242. Casteels C, Vermaelen P, Nuyts J, Van Der Linden A, Baekelandt V, Mortelmans L et al.: Construction and Evaluation of Multitracer Small-Animal PET Probabilistic Atlases for Voxel-Based Functional Mapping of the Rat Brain. J Nucl Med 2006, 47: 1858-1866. 243. Volker JF, Hodge HC, Wilson HJ, Van Voorhis SN: The Adsorption of Fluorides by Enamel, Dentin, Bone, and Hydroxyapatite as shown by the Radioactive Isotope. J Biol Chem 1940, 134: 543-548. 244. Posner AS, Eanes ED, Harper RA, Zipkin I: X-ray difraction analysis of the efect of luoride on human bone apatite. Arch Oral Biol 1963, 8: 549-570. 245. Scarpa M, Vianello F, Rigo A, Viglino P, Bracco F, Battistin L: Uptake and life time of luoride ion in rats by 19F-NMR. J Magn Reson Imaging 1993, 11: 697-703. 246. Buiter HJC, Leysen JE, Schuit RC, Fisher A, Lammertsma AA, Windhorst AD: Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S). J Labelled Comp Rad 2012, 55: 264-273.

164

References

247. van Velden FHP, Kloet RW, van Berckel BNM, Lammertsma AA, Boellaard R: Accuracy of 3-Dimen- sional Reconstruction Algorithms for the High-Resolution Research Tomograph. J Nucl Med 2009, 50: 72-80. 248. Loening AM, Gambhir SS: AMIDE: A completely free system for medical imaging data analysis. J Nucl Med 2001, 42: 827. 249. DiResta GR, Lee J, Arbit E: Measurement of brain tissue speciic gravity using pycnometry. J Neurosci Methods 1991, 39: 245-251. 250. Landis JR, Koch GG: The Measurement of Observer Agreement for Categorical Data. Biometrics 1977, 33: 159-174. 251. Hume SP, Lammertsma AA, Myers R, Rajeswaran S, Bloomield PM, Ashworth S et al.: The potential of high-resolution positron emission tomography to monitor striatal dopaminergic function in rat models of disease. J Neurosci Methods 1996, 67: 103-112. 252. Berstein G, Blank JL, Smrcka AV, Higashijima T, Sternweis PC, Exton JH et al.: Reconstitution of agonist-stimulated phosphatidylinositol 4,5- bisphosphate hydrolysis using puriied m1 musca-

rinic receptor, Gq/11, and phospholipase C-beta 1. J Biol Chem 1992, 267: 8081-8088. 253. Bodick NC, Ofen WW, Levey AI, Cutler NR, Gauthier SG, Satlin A et al.: Efects of xanomeline, a selective muscarinic receptor agonist, on cognitive function and behavioral symptoms in Al- zheimer disease. Arch Neurol 1997, 54: 465-473. 254. Zubieta JK, Koeppe AR, Frey KA, Kilbourn MR, Mangner TJ, Foster NL et al.: Assessment of muscarinic receptor concentrations in ainity and Alzheimer disease with [C-11]NMPB and PET. Synapse 2001, 39: 275-287. 255. McPherson W: Targeting Cerebral Muscarinic Acetylcholine Receptors with Radioligands for Diagnostic Nuclear Medicine Studies. In Ion Channel Localization. edited by Lopatin A, Nichols CG. 2001: 17-38. 256. Christopoulos A, Kenakin T: G Protein-Coupled Receptor Allosterism and Complexing. Pharmacol Rev 2002, 54: 323-374. 257. Flynn DD, Mash DC: Distinct Kinetic Binding-Properties of N-[H-3]-Methylscopolamine Aford Dif- ferential Labeling and Localization of M1, M2, and M3-Muscarinic-Receptor Subtypes in Primate Brain. Synapse 1993, 14: 283-296. 258. Finnema SJ, Seneca N, Farde L, Shchukin E, Sovago J, Gulyas B et al.: A preliminary PET evaluation of the new dopamine D-2 receptor agonist [C-11]MNPA in cynomolgus monkey. Nucl Med Biol 2005, 32: 353-360. 259. Mukherjee J, Narayanan TK, Christian BT, Shi BZ, Yang ZY: Binding characteristics of high-ainity dopamine D2/D3 receptor agonists, C-11-PPHT and C-11-ZYY-339 in rodents and imaging in non- human primates by PET. Synapse 2004, 54: 83-91. 260. Wilson AA, McCormick P, Kapur S, Willeit M, Garcia A, Hussey D et al.: Radiosynthesis and evaluation of [11C]-(+)-4-propyl-3,4,4a,5,6,10b-hexahydro-2H-naphtho[1,2-b][1,4]oxazin -9-ol as a potential radiotracer for in vivo Imaging of the dopamine D2 high-ainity state with positron emission tomography. J Med Chem 2005, 48: 4153-4160. 261. Shi B, Narayanan TK, Christian BT, Chattopadhyay S, Mukherjee J: Synthesis and biological evaluation of the binding of dopamine D2/D3 receptor agonist, (R,S)-5-hydroxy-2-(N-propyl-N-(5’-(18) F-luoropentyl)aminotetralin ((18)F-5-OH-FPPAT) in rodents and nonhuman primates. Nucl Med Biol 2004, 31: 303-311. 262. Fisher A: Cholinergic modulation of amyloid precursor protein processing with emphasis on M1 muscarinic receptor: perspectives and challenges in treatment of Alzheimer’s disease. J Neuro- chem 2012, 120: 22-33.

165

References

263. Czernin J, Satyamurthy N, Schiepers C: Molecular Mechanisms of Bone 18F-NaF Deposition. J Nucl Med 2010, 51: 1826-1829. 264. Maruyama S, Tsukada H, Nishiyama S, Kakiuchi T, Fukumoto D, Oku N et al.: In Vivo Quantitative Autoradiographic Analysis of Brain Muscarinic Receptor Occupancy by Antimuscarinic Agents for Overactive Bladder Treatment. J Pharmacol Exp Ther 2008, 325: 774-781. 265. Su TP: Evidence for sigma opioid receptor: binding of [3H]SKF-10047 to etorphine-inaccessible sites in guinea-pig brain. J Pharmacol Exp Ther 1982, 223: 284-290. 266. Vannucchi MG, Scali C, Kopf SR, Pepeu G, Casamenti F: Selective muscarinic antagonists difer- entially afect in vivo acetylcholine release and memory performances of young and aged rats. Neuroscience 1997, 79: 837-846. 267. Wang RH, Bejar C, Weinstock M: Gender diferences in the efect of rivastigmine on brain cholin- esterase activity and cognitive function in rats. Neuropharmacol 2000, 39: 497-506. 268. Buiter HJC, van Velden FHP, Leysen JE, Fisher A, Windhorst AD, Lammertsma AA et al.: Repro- ducible analysis of rat brain PET studies using an additional [18F]NaF scan and an MR based ROI template. Int J Mol Imaging 2012. 269. Bergstrom M, Grahnen A, Langstrom B: Positron emission tomography microdosing: a new concept with application in tracer and early clinical drug development. Eur J Clin Pharmacol 2003, 59: 357-366. 270. Jakubik J, Michal P, Machova E, Dolezal V: Importance and Prospects for Design of Selective Muscarinic Agonists. Phys Res 2008, 57: S39-S47. 271. Machova E, El-Fakahany EE, Dolezal V: Xanomeline quasi-irreversibly bound to an ectopic site can stimulate presynaptic M2 receptors via the orthosteric binding site. J Neurochem 2005, 94: 90. 272. Oki T, Takagi Y, Inagaki S, Taketo MM, Manabe T, Matsui M et al.: Quantitative analysis of binding parameters of [3H]N-methylscopolamine in central nervous system of muscarinic acetylcholine receptor knockout mice. Mol Brain Res 2005, 133: 6-11. 273. Tzavara ET, Bymaster FP, Felder CC, Wade M, Gomeza J, Wess J et al.: Dysregulated hippocampal acetylcholine neurotransmission and impaired cognition in M2, M4 and M2//M4 muscarinic receptor knockout mice. Mol Psychiatry 0 AD, 8: 673-679. 274. Flynn DD, Reever CM, FerrariDiLeo G: Pharmacological strategies to selectively label and localize muscarinic receptor subtypes. Drug Dev Res 1997, 40: 104-116. 275. Saller CF, Czupryna MJ, Salama AI: 5-Ht2 Receptor Blockade by Ici-169,369 and Other 5-Ht2 An- tagonists Modulates the Efects of D-2 Dopamine Receptor Blockade. J Pharmacol Exp Ther 1990, 253: 1162-1170. 276. Damsma G, Boer P, Westerink BHC, Fibiger HC: Dopaminergic regulation of striatal cholinergic interneurons: an in vivo microdialysis study. N-S Arch Pharmacol 1990, 342: 523-527. 277. DeBoer P, Abercrombie ED: Physiological release of striatal acetylcholine in vivo: modulation by D1 and D2 dopamine receptor subtypes. J Pharmacol Exp Ther 1996, 277: 775-783. 278. Billard W, Binch H, Crosby G, McQuade RD: Identiication of the primary muscarinic autoreceptor subtype in rat striatum as m2 through a correlation of in vivo microdialysis and in vitro receptor binding data. J Pharmacol Exp Ther 1995, 273: 273-279. 279. Hume SP, Opacka-Jufry J, Myers R, Ahier RG, Ashworth S, Brooks DJ et al.: Efect of L-dopa and 6-hydroxydopamine lesioning on [11C]raclopride binding in rat striatum, quantiied using PET. Synapse 1995, 21: 45-53. 280. Seneca N, Finnema SJ, Farde L, Gulyas B, Wikstrom HV, Halldin C et al.: Efect of amphetamine on dopamine D2 receptor binding in nonhuman primate brain: A comparison of the agonist radioligand [11C]MNPA and antagonist [C-11]raclopride. Synapse 2006, 59: 260-269.

166

References

281. Hall JM, Caulield MP, Watson SP, Guard S: Receptor Subtypes Or Species Homologs - Relevance to Drug Discovery. Trends Pharmacol Sci 1993, 14: 376-383. 282. Hoyer D, Clarke DE, Fozard JR, Hartig PR, Martin GR, Mylecharane EJ et al.: International Union of Pharmacology classiication of receptors for 5-hydroxytryptamine (Serotonin). Pharmacol Rev 1994, 46: 157-203. 283. Tack J, van Outryve M, Beyens G, Kerstens R, Vandeplassche L: Prucalopride (Resolor) in the treatment of severe chronic constipation in patients dissatisied with laxatives. Gut 2009, 58: 357-365. 284. Galeotti N, Ghelardini C, Bartolini A: Role of 5-HT4 receptors in the mouse passive avoidance test. J Pharmacol Exp Ther 1998, 286: 1115-1121. 285. Eglen RM, Wong EHF, Dumuis A, Bockaert J: Central 5-Ht4 Receptors. Trends Pharmacol Sci 1995, 16: 391-398. 286. Kornum BR, Lind NM, Gillings N, Marner L, Andersen F, Knudsen GM: Evaluation of the novel 5-HT4 receptor PET ligand [C-11]SB207145 in the Gottingen minipig. J Cereb Blood Flow Metab 2009, 29: 186-196. 287. Marner L, Gillings N, Madsen K, Erritzoe D, Baare WFC, Svarer C et al.: Brain imaging of serotonin 4 receptors in humans with [C-11]SB207145-PET. NeuroImage 2010, 50: 855-861. 288. Strange pg: G-protein Coupled Receptors: conformations and states. Biochem Pharmacol 1999, 58: 1081-1088. 289. Briejer MR, Bosmans JP, Van Daele P, Jurzak M, Heylen L, Leysen JE et al.: The in vitro pharmacological proile of prucalopride, a novel enterokinetic compound. Eur J Pharmacol 2001, 423: 71-83. 290. Mistry P, Bootie D, Liddle C, Loi R, Templeton D: Reversal of P-glycoprotein mediated multidrug resistance in vivo by XR9576. Ann Oncol 1998, 9: 568. 291. Martin C, Berridge G, Mistry P, Higgins C, Charlton P, Callaghan R: The molecular interaction of the high ainity reversal agent XR9576 with P-glycoprotein. Br J Pharmacol 1999, 128: 403-411. 292. Fox E, Bates SE: Tariquidar (XR9576): a P-glycoprotein drug elux pump inhibitor. Expert Rev Anticancer Ther 2007, 7: 447-459. 293. Bauer F, Kuntner C, Bankstahl JP, Wanek T, Bankstahl M, Stanek J et al.: Synthesis and in vivo evaluation of [11C]tariquidar, a positron emission tomography radiotracer based on a third-generation P-glycoprotein inhibitor. Bioorg Med Chem 2010, 18: 5489-5497. 294. Jewett DM: A simple synthesis of [11C]methyl trilate. Int J Radiat Appl Instr A Appl Radiat Isotopes 1992, 43: 1383-1385. 295. CHMP assessment report for resolor: http: //www.ema.europa.eu/docs/en_GB/document_library/ EPAR_-_Public_assessment_report/human/001012/WC500053997.pdf. European Medicines Agency 2009. 296. Samen E, Thorell JO, Fredriksson A, Stone-Elander S: The tyrosine kinase inhibitor PD153035: implication of labeling position on radiometabolites formed in vitro. Nucl Med Biol 2006, 33: 1005-1011. 297. Nebbia C, Ceppa L, Dacasto M, Carletti M, Nachtmann C: Oxidative Metabolism of Monensin in Rat Liver Microsomes and Interactions with Tiamulin and Other Chemotherapeutic Agents: Evidence for the Involvement of Cytochrome P-450 3A Subfamily. Drug Metab Disp 1999, 27: 1039-1044. 298. Roy SD: Kinetics of in vitro metabolism of methoxyphenamine in rats. Xenobiotica 1990, 20: 55-70. 299. Leysen J, Laduron P: Characterization of An Enzyme Yielding Formaldehyde From-5-Methyltetra- hydrofolic Acid. FEBS Lett 1974, 47: 299-303. 300. Lauwers W, Leysen J, Verhoeven H, Laduron P, Claeys M: Identiication of Alkaloids - Condensation Products of Biogenic-Amines with Formaldehyde, Enzymatically Formed from 5-Methyltetrahy- drofolic Acid. Biomed Mass Spec 1975, 2: 15-22.

167

References

301. Eisenblatter T, Huwel S, Galla HJ: Characterisation of the brain multidrug resistance protein (BMDP/ABCG2/BCRP) expressed at the blood-brain barrier. Brain Res 2003, 971: 221-231. 302. Garner CE, Solon E, Lai CM, Lin J, Luo G, Jones K et al.: Role of P-Glycoprotein and the Intestine in the Excretion of DPC 333 [(2R)-2-{(3R)-3-Amino-3-[4-(2-methylquinolin-4-ylmethoxy)phenyl]- 2-oxopyrrolidin-1-yl}-N-hydroxy-4-methylpentanamide] in Rodents. Drug Metab Disp 2008, 36: 1102-1110. 303. Kagan L, Dreiinger T, Mager DE, Hofman A: Role of P-Glycoprotein in Region-Speciic Gastroin- testinal Absorption of Talinolol in Rats. Drug Metab Disp 2010, 38: 1560-1566. 304. Hegde SS, Eglen RM: Peripheral 5-HT4 receptors. FASEB J 1996, 10: 1398-1407. 305. Shi B, Narayanan TK, Christian BT, Chattopadhyay S, Mukherjee J: Synthesis and biological evaluation of the binding of dopamine D2/D3 receptor agonist, (R,S)-5-hydroxy-2-(N-propyl-N-(5’-(18) F-luoropentyl)aminotetralin ((18)F-5-OH-FPPAT) in rodents and nonhuman primates. Nucl Med Biol 2004, 31: 303-311. 306. Guo Q, Brady M, Gunn RN: A Biomathematical Modeling Approach to Central Nervous System Radioligand Discovery and Development. J Nucl Med 2009, 50: 1715-1723. 307. Mu LJ, Schubiger PA, Ametamey SM: Radioligands for the PET Imaging of Metabotropic Gluta- mate Receptor Subtype 5 (mGluR5). Curr Top Med Chem 2010, 10: 1558-1568.

168

List of publications

LIST oF PuBLICATIonS

Articles for this doctoral study

Buiter H.J.C., Windhorst A.D, Huisman M.C., De Maeyer J.H., Schuurkes J.A.J., Lam- mertsma A.A., Leysen J.E., Radiosynthesis and preclinical evaluation of [11C]prucalopride as a potential agonist PET ligand for the 5-HT4 receptor., European Journal of Nuclear Medicine and Molecular Imaging Research, 2013, 3:24.

Buiter H.J.C., Windhorst A.D, Huisman M.C., Yaqub M, Knol D.L., Fisher A., Lammertsma A.A., Leysen J.E., [11C]AF150(S), an agonist PET ligand for M1 muscarinic acetylcholine receptors, European Journal of Nuclear Medicine and Molecular Imaging Research, 2013, 3:19.

Buiter H.J.C., Van Velden F.P.H., Lammertsma A.A., Huisman M.C., Reproducible Analysis of Rat Brain PET Studies Using an Additional [18F]NaF Scan and an MR-Based ROI Template., International Journal of Molecular Imaging, 2012, Article ID 580717, doi:10.1155/2012/580717.

Buiter H.J.C., Leysen J.E., Fisher A., Lammertsma AA, Windhorst A.D., Radiosynthesis and biological evaluation of the M1 muscarinic acetylcholine receptor agonist ligand [11C]AF150(S)., Journal of Labelled Compounds and Radiopharmaceuticals, 2012: 55 (7); 264-273. other

Buiter H.J.C., Blaauwgeers J.L.G.M., van der Spoel J.I.M., Franssen E.J.F., Erythromycin Precipitation in Vena Femoralis: Investigation of Crystals Found in Postmortem Material of an Intensive Care Unit Patient., Therapeutic Drug Monitoring 2008: 30 (1); 125-129.

Abstracts

Buiter H.J.C., Leysen J.E., Knol D.L., Fisher A., Huisman M.C., Yaqub M., Lammertsma A.A. and Windhorst A.D., “[11C]AF150(S), an agonist PET ligand for the M1 muscarinic acetylcholine receptor”. Final program & book of abstracts 12th FIGON Dutch Medicines Days, Lunteren, The Netherlands. Oral presentation, 2010

Leysen J.E., Buiter H.J.C., Fisher A., Huisman M.C., Yaqub M., Boellaard R., Lammertsma A.A. and Windhorst A.D., “[11C]AF150(S), an agonist PET ligand for in vivo imaging of the

169

List of publications

M1 muscarinic acetylcholinereceptor”. European Neuropsychopharmacology August 2010 (Vol. 20 Supplement 3, Pages S302-S303. Poster P.1.e.026

Buiter H.J.C., Leysen J.E., Knol D.L., Fisher A., Huisman M.C., Yaqub M., Lammertsma A.A. and Windhorst A.D. “[11C]AF150(S): An agonist PET ligand for the M1 muscarinic acetylcholine receptor” J NUCL MED MEETING ABSTRACTS 2010 51: 29, Oral presentation Abstract No. 29

Buiter H.J.C., Leysen J.E., Fisher A., Huisman M.C., Knol D.L., Lammertsma A.A. and Wind- horst A.D. “[11C]AF150(S), an agonist PET ligand for in vivo imaging of the M1 muscarinic acetylcholine receptor”. Final program & book of abstracts 15th European Symposium on Radiopharmacy and Radiopharmaceuticals, Edinburgh, Schotland. Poster 2010

Buiter H.J.C., Leysen J.E., Fisher A., Huisman M.C., Knol D.L., Lammertsma A.A. and Wind- horst A.D., “[11C]AF150(S), an agonist PET ligand for in vivo imaging of the M1 muscarinic acetylcholine receptor”. Journal of Labelled Compounds and Radiopharmaceuticals, 2009, Volume 52, Issue Supplement 1, s325. Poster 229

Buiter H.J.C., Fisher A., Klok R.P., Herscheid J.D.M., Verbeek J., Lammertsma A.A., Ley- sen J.E., and Windhorst A.D., “Evaluation of a putative radiolabeled ligand for in vivo imaging of the M1 muscarinic acetylcholine receptor by PET in small animals”. World Molecular Imaging Congress (WMIC) Abstract Book, Nice, France, Sep. 10–13, 2008. Poster presentation.

Buiter H.J.C., Fisher A., Klok R.P., Herscheid J.D.M., Verbeek J., Lammertsma A.A., Leysen J.E., and Windhorst A.D., “A putative radiolabeled ligand for in vivo imaging of the M1 muscarinic acetylcholine receptor by PET”. The International Journal of Neuropsycho- pharmacology, 2008, Volume 11, Supplement S1, p 222, Poster P-05.58:

170

Dankwoord dAnKWooRd

Prof. dr. J.E.M.F. Leysen, beste Josée, dank voor het in mij gestelde vertrouwen. Het is voor mij een eer om onder jouw supervisie, als jouw laatste onderzoeker in opleiding te mogen promoveren. Als internationaal vermaard receptorfarmacoloog had je al signiicante bijdragen geleverd aan geneesmiddelenonderzoek en studies naar de rol en functioneren van G-eiwitgekoppelde receptoren. Nu heb je ook een grote bijdrage geleverd aan mijn proefschrift. Jouw begeleiding en kritische maar vaak constructieve commentaar was essentieel voor de tot standkoming van mijn promotieonderzoek en dit proeschrift.

Prof. dr. A.A. Lammertsma, beste Adriaan, hartelijk bedankt voor jouw supervisie. Dank voor de geboden kans, de goede gesprekken op de moeilijke momenten en je positieve feedback op mijn manuscripten en taalkundige correcties. Ik hoop dat we in de toekomst nogmaals kunnen samenwerken. Misschien heb je gelijk en ben ik nog niet helemaal klaar met nucleaire geneeskunde.

Prof. dr. A.D. Windorst, beste Bert bedankt. Door jouw dagelijkse begeleiding heb ik als apotheker heel wat van radiochemie kunnen leren. Dank je wel dat je me de mogeli- jkheden geboden hebt om zelf alle facetten van het PET onderzoek te mogen leren kennen en in de vingers krijgen. Of het nu radiosynthese, autoradiograie, biodistributie experimenten, PET studies of modeling was, jij hebt me de kansen geboden. Verder hoop ik dat je door mijn onderzoek nog meer van radiofarmacologische en radiofarma- cokinetische studies bent gaan houden, deze bijven essentieel, zeker na een geslaagde radiosynthese van een nieuwe ligand.

Prof. dr. N.H. Hendrikse en dr. E.J.F. Franssen, beste Harry en Eric, zonder jullie was ik hier nooit aan begonnen. Beste Harry, ons oriënterende gesprek voor een bijvak tijdens mijn studie farmacie bij de afdeling nucleaire geneeskunde heeft mij de ogen geopend en mijn interesse gewekt voor de radiofarmacie. Ook nu vorm je voor mij als professor radiofarmacologie, klinisch farmacoloog en ziekenhuisapotheker met aandachtgebied radiofarmacie voor mij een voorbeeld. Beste Eric, jij heb mij tijdens mijn ziekenhuisfarmaciestage de mogelijkheid geboden om kennis te maken met klinische farmacologie en toxicologie. Dankzij jou ben ik bij de PET groep van prof. dr. Lammerstma terrecht gekomen, waarvoor mijn grote dank. Beste Harry en Eric, jullie zijn mijn voorbeeld voor de toekomst, als toekomstig ziekenhuisapotheker-klinisch farmacoloog met hopelijk te zijner tijd aandachtsgebied radiofarmacie.

171

Dankwoord

Dr. M.C. Huisman en dr. F.P.H. van Velden, beste Marc en Floris, dank voor jullie positiviteit en enthousiasme. Dankzij onze goede samenwerking hebben we een reproduceerbare methode kunnen ontwikkelen voor PET data analyse van ratten studies. Het resultaat van deze methode heeft geleidt tot een schitterend artikel welke we al snel hebben kunnen publiceren.

De leden van de promotiecommissie, prof. dr. J. Zaagsma, prof. dr. N.H. Hendrikse, prof. dr. H. Groenewegen. prof. dr. R. boellaard, prof C. Halldin, prof. dr. P.H. Elsinga en dr. J.A.J. Schuurkes wil ik hartelijk danken voor het kritisch doornemen van het manuscript en voor het zitting nemen in de promotiecommissie. Thank you very much for your role in the reading committee of my thesis and your careful examination of my thesis.

Ing. R.C. Schuit, beste Robert, dank je wel voor je ondersteuning en de analyse van de radiometabolieten. Verder heb ik de afgelopen jaren je periodieke e-mailtjes enorm gewaardeerd. Zeker de berichten zoals: “Hoe staat het? Nog steeds plezier in het feit dat je je niet druk hoeft te maken wat je dit weekend weer gaat doen. Hoe staat het met de artikelen, licht aan het einde van de tunnel, of nog geen idee of er licht is na de tunnel.” Ik weet nu dat er schitterend licht is aan het einde van de tunnel.

Dr. C.M. Molthof, Inge de Greeuw en Mariska Verlaan, beste Carla en Inge, ik wil jullie hartelijk danken voor jullie ondersteuning bij de biodistributie en in vivo PET studies. Zonder jullie hulp en ervaring waren mijn experimenten er nooit gekomen. Beste Mariska, jij ook bedankt voor je ondersteuning tijdens mijn experimenten. Ik vond het ontzettend leuk dat ik je met de Paxinos & Watson atlas op schoot de kunst van het snijden van hersencoupes heb mogen leren.

(Oud) Collega AIO’s, Joost en Pieter dank jullie wel voor de gezellig tijd en jullie hulp en ondersteuning bij experimenten. Succes met het afronden van jullie promotie! Rana, bedankt voor de gezellige diners en dat je gedurende 4 jaren mijn kamergenoot hebt willen zijn. Lyn, thank you for sharing our room and your eforts to learn me how to be polite in English. Beste Maria dank voor je support. Het werk is eindelijk klaar, hier bij mijn radiochemische en radiofarmacologische boekje!

Alle collega’s van het radionucliden centrum wil ik bedanken voor hun ondersteuning. In het bijzonder Leo, Fred, Arjan, Peter en Roy van de instumentatiedienst bedankt voor de technische ondersteuning. De stralingsveiligheidsdienst; Jan en Marcel bedankt en Tjaard tige tank foar de moaie tiid, dank voor jullie hulp. Greet bedankt voor je ondesteuning en goede zorgen. Sectie radiochemie: Anneloes, Rolph, Dennis, Arnold, Martien en Rob dank jullie wel voor het mij bijbrengen van de ijne kneepjes van het

172

Dankwoord radiochemische vak en de productie van de [18F]NaF voor de PET studies. Alle overige medewerkers van de afdeling Nucleaire Geneeskunde & PET Research wil ik bedanken voor de prettige samenwerking.

Collega’s van de Stiching Apotheek der Haarlemse Ziekenhuizen wil ik bedanken voor hun steun. In het bijzonder drs. R.T.M. van der Hoeven en drs. S.L. Verweij, beste Ruud en Sjoerd, dankzij jullie steun en het in mij gestelde vertouwen heb ik mijn promotie kunnen afronden, naast mijn werkzaamheden als ziekenhuisapotheker in opleiding.

Mijn paranimfen R.M. Dull en H.L. Hoge. Beste René en Rien, sinds het begin van onze studietijd zijn wij beste vrienden. Dank voor jullie jaren lange vriendschap en de mooie herinneringen aan de afgelopen 16 jaar (o.a. onze excursies en onze gastronomische avonturen). Ik hoop dat we nog veel samen mogen gaan genieten. Heel erg bedankt dat jullie mijn paranimfen wilden zijn!

Familie en vrienden Mijn hele familie wil ik bedanken. In het bijzonder oom Pieter en tante Aina, ome Paul en tante Chris, Efraim en Ophira bedankt voor de voortdurende interesse en steun die jullie mij de afgelopen jaren hebben gegeven. Lieve vrienden, beste Ard, Diana, Bas en Kitty-Nora, Meggy en Patrick, René, Rien en Ingeborg ik heb de afgelopen jaren de nodige weekendjes en avonden moeten missen. Ondanks dat alles zijn jullie mij bli- jven steunen en hebben jullie altijd interesse getoond. Ik hoop dat we na 1 juli snel de agenda’s kunnen trekken.

Pa en Ma, Norbert en Anne-Marije, Pap en Mam, Peter en Ankie. Graag wil ik mijn ouders, heel erg bedanken voor alles wat zij voor mij gedaan hebben. Jullie hebben mij kansen geboden en mijn voortdurende honger naar kennis proberen te stillen. Dankzij jullie steun kon en ging ik studeren. Dank jullie wel! Mijn broer Norbert, schoonzus Anne-Marije, Pap en Mam, Peter en Ankie wil ik bedanken voor hun steun tijdens mijn promotieonderzoek, maar vooral ook voor de laaste jaren.

Francis, Ellen en Inge. Tenslotte wil ik mijn vrouw en kinderen heel erg bedanken voor alles wat zij voor mij gedaan hebben. Lieve Francis, Ellen en Inge jullie hebben mij altijd gesteund en dankzij jullie heb ik de kracht kunnen vinden om deze promotie te kunnen volbrengen. Mijn lieve grietjes, ik draag dit proefschrift aan jullie op als bewijs dat jij je man en jullie je pappa weer terug gekregen hebben. Almere, mei 2014 Hans J.C. Buiter

173

Curriculum vitae

CuRRICuLuM VITAE

Op 13 mei 1980 werd Hans J.C. Buiter in Delfzijl geboren. In 1998 behaalde hij zijn VWO- diploma aan het Dockinga College te Dokkum. In september van datzelfde jaar begon hij met de studie farmacie aan de Rijksuniversiteit Groningen. Na een afstudeeronderzoek bij de vakgroep Farmacochemie aan de faculteit Wiskunde en Natuurwetenschappen van de Rijksuniversiteit Groningen, onder leiding van prof. dr. H.V. Wikström, wat zich richte op de synthese en farmacologische evaluatie van dopamine D2 agonisten, volgde er een tweede afstudeeronderzoek bij de afdeling Klinische Neurowetenschappen aan het Karolinska Instituut in Stockholm, onder leiding van prof. C. Halldin, wat zich richte op radiosynthese van dopamine D2 agonisten en PET. Al tijdens het afstudeeronderzoek werd de interesse gewekt om promotieonderzoek te gaan doen. Daarom werd na af- ronding van zijn studie farmacie en het behalen van zijn apothekerbul gestart met een baan als onderzoeker bij de afdeling Nucleaire Geneeskunde & PET Research van het VU medisch centrum te Amsterdam. Dit proefschrift is het resultaat van het verrichte onderzoekswerk. Sinds oktober 2010 is hij werkzaam in de ziekenhuisfarmacie. Eerst als projectapotheker en sinds november 2011 als ziekenhuisapotheker in opleiding bij Stichting Apotheek der Haarlemse Ziekenhuizen. Vanaf januari 2014 is hij tevens gestart met de opleiding tot klinisch farmacoloog in het VU medisch centrum.

174

Fotoverantwoording

FoToVERAnTWooRdInG

De foto’s aan het begin van elk hoofdstuk zijn gemaakt en bewerkt door R.M. Dull.

Chapter 1 Strand van Midsland aan Zee – Terschelling Chapter 2 Strand – Bergen aan Zee Chapter 3 Strand – Bergen aan Zee Chapter 4 Wad bij Hoorn – Terschelling Chapter 5 Haven – Noordpolderzijl Chapter 6 Duinen Oosterend – Terschelling Chapter 7 Noordsvaarder – Terschelling Chapter 8 Bevroren Waddenzee – Noorpolderzijl

175

In vivoIn assessment of cogniion related acivated sev

verdediging van het proefschrit In vivo assessment of cogniion related cogniion related acivated acivated seven transmembrane receptors seven transmembrane receptors

Hans J.C. Buiter

Na aloop van de promoie

voor de recepie ter plaatse en transmembrane receptors

H.J.C. Buiter H.J.C. Buiter