ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

ISSN 1651-6192ISBN 978-91-554-8843-7urn:nbn:se:uu:diva-2138632014

Digital Comprehensive Summaries of Uppsala Dissertations from the Faculty of Pharmacy 183

Development of Palladium-Promoted 11C/12C-Carbonylations and Radiosynthesis of Amyloid PET Ligands

PATRIk NoRDEmAN

Dissertation presented at Uppsala University to be publicly examined in B21, BMC,Husargatan 3, Uppsala, Friday, 21 February 2014 at 09:15 for the degree of Doctor ofPhilosophy (Faculty of Pharmacy). The examination will be conducted in English. Facultyexaminer: Professor Antony D Gee (Division of Imaging Sciences, King's College, London).

AbstractNordeman, P. 2014. Development of Palladium-Promoted 11C/12C-Carbonylations andRadiosynthesis of Amyloid PET Ligands. Digital Comprehensive Summaries of UppsalaDissertations from the Faculty of Pharmacy 183. 76 pp. Uppsala: Uppsala universitet.ISBN 978-91-554-8843-7.

In the first part of this thesis, palladium(0)-catalyzed and -mediated carbonylations arediscussed. Paper I describes a new method for the safe, efficient use of a solid carbon monoxidesource in the synthesis of primary and secondary benzamides. In total, 35 benzamides weresynthesized from aryl iodides (20 examples, 69-97% yield) and aryl bromides (15 examples,32-93% yield). Reduction-prone groups were used successfully in the reactions. In paper II,the same protocol was adopted for the palladium(0)-catalyzed synthesis of N-cyanobenzamidesfrom aryl iodides/bromides, carbon monoxide and cyanamide. In total, 22 N-cyanobenzamideswere synthesized (42-88% yield). The radiosynthesis of [11C]N-cyanobenzamides is discussedin paper III. In total, 22 compounds were synthesized from various aryl halides in 28-79% decaycorrected radiochemical yield. The protocol was then applied to the radiosynthesis of [11C]N-cyanobenzamide analogs of flufenamic acid and dazoxibene.

In the second part of this thesis, compounds of interest in relation to amyloid diseasesare discussed. Paper IV describes the solid-phase synthesis of BACE-1 enzyme inhibitorscontaining secondary and tertiary hydroxyl as the transition state isostere. In total, 22 inhibitorswere synthesized. The most potent compound (IC50= 0.19 µM) was co-crystallized at the activesite of the enzyme to reveal a new binding mode. In paper V, the evaluation of a potent BACE-1inhibitor as a potential radiotracer for use in PET is described. The radiolabeled [11C]BSI-IV wasobtained in 29±12% decay corrected radiochemical yield by a three-component palladium(0)-mediated aminocarbonylation. Its properties as a potential PET tracer were investigated in vitroby autoradiography and in vivo in rats using small animal PET-CT. A new class of amyloid-binding PET ligands is described in paper VI. Three polythiophenes were labeled with carbon-11or fluorine-18 (26-43% decay-corrected radiochemical yield). The in vitro studies showed thatthese ligands bind specifically to amyloid deposits. In vivo PET showed low uptake in the organsof interest in healthy rats and a monkey. These results suggest the labeled thiophenes derivativescould be useful as PET tracers for the study of amyloid diseases.

Keywords: Palladium, Carbonylation, Positron emission tomography, PET, Carbon-11,Fluor-18, Radiochemistry, Amyloidosis

Patrik Nordeman, Department of Medicinal Chemistry, Preclinical PET Platform, DagHammarskjöldsv 14C, Uppsala University, SE-751 83 Uppsala, Sweden.

© Patrik Nordeman 2014

ISSN 1651-6192ISBN 978-91-554-8843-7urn:nbn:se:uu:diva-213863 (http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-213863)

"A scientist in his laboratory is not a mere technician:he is also a child confronting natural phenomena that

impress him as though they were fairy tales."Marie Curie (1867-1934)

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Nordeman, P., Odell, L. R., Larhed, M. (2012) Amino-

carbonylations Employing Mo(CO)6 and a Bridged Two-Vial System: Allowing the Use of Nitro Group Substituted Aryl Iodides and Aryl Bromides. J. Org. Chem., 77(24): 11393-11398.

II Mane, R. S., Nordeman, P., Odell, L. R., Larhed, M. (2013) Palladium-catalyzed Carbonylative Synthesis of N-Cyanobenzamides from Aryl Iodides/Bromides and Cyanamide. Tetrahedron Lett. 54(51): 6912-6915.

III Nordeman, P., Antoni, G., Odell. L. R. Pd-mediated Carbonyl-ative Synthesis of 11C-N-Cyanobenzamides. Manuscript.

IV Wångsell, F., Nordeman, P., Sävmarker, J., Emanuelsson, R., Jansson, K., Lindberg, J., Rosenquist, Å., Samuelsson, B., Larhed M. (2011) Investigation of α-Phenylnorstatine and α-Benzylnorstatine as Transition State Isostere Motifs in the Search for New BACE-1 Inhibitors. Bioorg. Med. Chem. 19(1): 145-155.

V Nordeman, P., Estrada, S., Odell, L. R., Larhed, M., Antoni, G. 11C Labeling of a Potent Hydroxyethylamine BACE-1 Inhibitor and Evaluation in vitro and in vivo. Submitted.

VI Nordeman, P., Johansson, L. B. G., Bäck, M., Estrada, S., Hall, H., Westermark, G. T., Westermark, P., Nilsson, L., Ham-marström, P., Nilsson, P., Antoni G. 11C and 18F Radiolabeling of Tetra and Pentatiophenes as PET-Ligands for Misfolded Pro-tein Aggregates. Manuscript.

Reprints were made with permission from the respective publishers.

Papers Not Included in This Thesis

Datta, G. K., Nordeman, P., Dackenberg, J., Nilsson, P., Hallberg, A., Larhed, M. (2008) Enantiopure 2-aryl-2-methyl Cyclopentanones by an Asymmetric Chelation-controlled Heck Reaction Using Aryl Bromides: Increased Preparative Scope and Effect of Ring Size on Reactivity and Se-lectivity. Tetrahedron: Asymmetr. 19(9), 1120-1126. Strand, J., Nordeman, P., Honarvar, H., Altai, M., Larhed, M., Orlova, A., Tolmachev, V. Site-specific Radioiodination of HER2-targeting Affibody Molecules Using Iodophenetylmaleimide Decreases Renal Uptake of Ra-dioactivity. Manuscript.

Contents

Introduction ................................................................................................... 11 Palladium as a Catalyst in Organic Chemistry ......................................... 11 Carbonylative Palladium Chemistry ........................................................ 13

Towards Simpler Carbonylative Procedures ....................................... 16 Tracer Development in Positron Emission Tomography ......................... 18

Tracer Considerations .......................................................................... 19 Radiochemistry .................................................................................... 21

Palladium-Catalyzed Carbonylations using a Two Chamber System ........... 23 Synthesis of Benzamides (Paper I) ........................................................... 23

Background and Aim ........................................................................... 23 Results and Discussion ........................................................................ 24

Synthesis of N-Cyanobenzamides (Paper II) ............................................ 30 Background and Aim ........................................................................... 30 Results and Discussion ........................................................................ 30

Pd-mediated 11C-Carbonylative Synthesis of 11C-N-Cyanobenzamides (Paper III) ...................................................................................................... 36

Background and Aim ........................................................................... 36 Results and Discussion ........................................................................ 37

Amyloid Diseases - Enzyme Inhibitors and PET Tracers ............................. 43 Amyloid Diseases ..................................................................................... 43 Synthesis of BACE-1 Inhibitors Comprising α-Phenylnorstatine and α-Benzylnorstatine as Transition State Isosters (Paper IV) ......................... 46

Background and Aim ........................................................................... 46 Results and Discussion ........................................................................ 47

11C-Labeling of a Potent Hydroxyethylamine BACE-1 Inhibitor and Evaluation in vitro and in vivo (Paper V) ................................................. 52

Background and Aim ........................................................................... 52 Results and Discussion ........................................................................ 53

11C and 18F Radiolabeling of Tetra- and Penta-tiophenes as PET-Ligands for Amyloid Deposits (Paper VI) ............................................... 57

Background and Aim ........................................................................... 57 Results and Discussion ........................................................................ 58

Conclusions ................................................................................................... 65

Acknowledgements ....................................................................................... 67

References ..................................................................................................... 69

Abbreviations

Aβ Amyloid beta AD Alzheimer's disease IAPP Islet amyloid polypeptide AL Amyloid light chain APP Amyloid precursor protein ATTR Transthyretin BACE-1 β-Secretase BBB Blood-brain barrier BOC tert-Butyloxycarbonyl Bmax Maximum binding capacity CNS Central nervous system CO Carbon monoxide CO2 Carbon dioxide CORM CO-releasing molecule CT Computed tomography DBU 1,8-Diazabicycloundec-7-ene DIPEA N,N-Diisopropylethylamine DMAP 4-Dimethylaminopyridine DMF N,N-Dimethylformamide DPEphos Bis[(2-diphenylphosphino)phenyl] ether EOB End of bombardment EOS End of synthesis Et3N Triethylamine Fmoc 9-Fluorenylmethoxycarbonyl GC Gas chromatography HEA Hydroxyethylamine HPLC High-performance LC Kd Dissociation constant (equilibrium) LC Liquid chromatography LCO Luminescent conjugated oligothiophene LE Ligand exchange LG Leaving group mCPBA 3-Chloroperoxybenzoic acid Mesylate Methanesulfonate MeCN Acetonitrile Mesylate Methanesulfonate

Mo(CO)6 Molybdenum hexacarbonyl MS Mass spectrometry NBS N-Bromosuccinimide NIS N-Iodosuccinimide NMR Nuclear magnetic resonance NSAID Nonsteroidal anti-inflammatory drug Nonaflate Nonafluorobutanesulfonate Pd Palladium Pd(dppf)Cl2 1,1'-Bis(diphenylphosphino)ferrocene- palladium(II)dichloride Pd(PPh3)4 Tetrakis(triphenylphosphine)palladium(0) Pd(Xantphos)Cl2 Dichloro[9,9-dimethyl-4,5-

bis(diphenylphosphino)xanthene]palladium(II) PET Positron emission tomography PIB Pittsburg Compound B dppf 1,1'-Bis(diphenylphosphino)ferrocene PPh3 Triphenylphosphine PR3 Triorganyl phosphines PyBOP (Benzotriazol-1-yloxy)tripyrrolidino- phosphonium hexafluorophosphate RCP Radiochemical purity RCY Radiochemical yield ROI Region of interest SA Specific activity (specific radioactivity) Scissile bond Bond susceptible to enzymatic cleavage SPPS Solid-phase peptide synthesis SUV Standardized uptake value TBAF Tetrabutylammonium flouride TBAOH Tetrabutylammonium hydroxide TE Trapping efficiency TFA Trifluoroacetic acid THF Tetrahydrofuran Tosylate 4-Toluenesulfonate TPL Thiophene PET ligand Triflate Trifluoromethanesulfonate TS Transition state X Halide (I, Br, Cl or pseudo-halide) Xantphos 4,5-Bis(diphenylphosphino)-9,9- dimethylxanthene

11

Introduction

Palladium as a Catalyst in Organic Chemistry Palladium (Pd) is a late transition metal and an element of the platinoids (group 8-10, period 5-6). Although it was discovered over two centuries ago, it was not until the second half of the twentieth century that palladium was recognized for its properties in preparative organic synthesis. Palladium can exist in several oxidation states; Pd(0, d10, tetrahedral) and Pd(II, d8, square planar), providing two or more free coordination sites, are the most com-monly found. The low energy threshold for moving between these two states (by oxidation/reduction) is one of the main reasons for palladium becoming such a widely employed and versatile catalyst. Its relative electronegativity (2.2; Li, Mg< Pd< B, Si, Sn) also provides high chemoselectivity and excel-lent tolerance for polar functional groups.1

One of the most historically important industrial applications of palladium catalysis is the Wacker process, introduced at the end of the 1950s. In this process, ethene is oxidized to acetaldehyde by Pd(II) and water, Cu(II) is added to oxidize Pd(0) back to Pd(II), the active catalytic species, and Cu(0) is subsequently oxidized back to Cu(II) by feedstock oxygen gas, which completes the catalytic cycle.2

At the beginning of the 1970s, Heck3 and Mizoroki4 independently dis-covered what came to be known as the Mizoroki-Heck reaction. In this reac-tion, a substituted alkene is obtained through formation of a carbon-carbon bond, starting from an unsaturated halide and an alkene. The reaction proceeds by adding catalytic amounts of palladium in the presence of base to regenerate the active Pd(0) species from the Pd(II) formed in the process. In the following years and decades, several related Pd(0)-catalyzed cross-coupling reactions were discovered and these are now an integral part of modern synthetic organic chemistry (Scheme 1).

Scheme 1. Pd(0)-catalyzed reactions. R = Organic substituent.

12

They vary in the nature of the coupling partner and subsequently in the fun-damental steps of the catalytic cycle (insertion/transmetallation) and provide an enormous variety of reaction products. The reactions are named after their inventors, Kumada (Grignard reagents),5,6 Mizoroki-Heck (alkenes),3 Sono-gashira (alkynes and Cu(I) source),7,8 Negishi (organo-zinc),9,10 Stille-Migita (organo-tin),11,12 Suzuki-Miyaura (organo-boron)13 and Hiyama (organo-silanes)14. The importance of Pd(0)-catalyzed cross couplings was hig-hlighted in 2010, when the Nobel Prize in Chemistry was given to Heck, Negishi and Suzuki for their contributions to the field.15

Any molecular moiety with palladium-coordinating ability can be re-garded as a ligand.1 Ligands can be classified into three categories in organo-metallic chemistry: L, X or Z, depending on the electronic properties. L li-gands are 2-e- neutral molecules (Lewis bases), X ligands are 1-e- radicals, and Z ligands are 0-e- molecules (Lewis acids). One group of ligands that has been extensively used in palladium chemistry is the trivalent phosphines. Triorganyl phosphines (PR3) are neutral (L) ligands which donate an electron pair to the metal center, thus not altering the oxidation state. Phosphines are also able to accept back-bonding (vide infra) from Pd(0) into empty d-orbitals, which stabilizes the palladium-phosphine complex. Importantly, the phosphines also help Pd(0) to dissolve in organic solvents.

Perhaps the most important aspect of phosphines and other ligands (for example amines16) is their ability to alter the reactivity of the metal, and subsequently to favorably adjust a reaction. The properties of the metals are altered by tuning the electronic and steric properties of the phosphine li-gand.17 The electronic properties of the ligand substituents will influence the σ-donating and π-accepting character of the phosphine. The size or steric bulk of a monodentate (a single coordination site, κ1) ligand at its lowest energy could be described using cone angle (Pd-PR3) which is defined as the angle of the ligand at a distance of 228 pm from the metal centre (Figure 1).18 Phosphines with two coordination sites (κ2) can be defined as biden-tate.19 The properties of bidentate ligands (electronic and steric) may be de-scribed as a bite angle (P-Pd-P).20

The interactions between the ligand, palladium and the substrate can fur-thermore influence the regio- and stereo-selectivity.21-23,1 The steric effects are also closely linked to the dissociation of the ligand, which will determine the coordination availability of the metal.

Figure 1. Cone angle of monodentate ligands and bite angle for bidentate ligands.

13

Carbonylative Palladium Chemistry In the earlier work by Heck in the end of the 1960s, coupling reactions in-volving organo-mercury compounds, palladium salts and carbon monoxide (CO) were explored.24 In 1974, Heck demonstrated the use of catalytic amounts of Pd(0) in an alkoxy- and amino-carbonylation of aryl halides us-ing CO gas to form esters and amides.25,26 This was a new and fundamental process for the synthesis of carbonyl-containing compounds. Pd(0) can coordinate CO by two types of interactions. Firstly, an electron lone pair on the carbon σ-orbital can be donated into a vacant d-orbital of palladium. Secondly, an electron pair in a filled d-orbital can donate electron density back to an empty π*-orbital in the carbon (Figure 2). The ability of CO to act as both a Lewis base (σ-donating) and Lewis acid (π*-accepting) makes it a good ligand for d-orbital-containing elements such as palladium.27

Figure 2. CO bonding and back bonding.

Pd(0)-catalyzed carbonylative reactions are now an integral part of organic chemistry.28,29 Usually, the carbonylative system consists of a three-component mixture consisting of a substrate (aryl, vinyl, benzyl) with a ha-lide or pseudohalide (LG or X), CO, and a nucleophile (Nu) (Scheme 2).

Scheme 2. Schematic depiction of the three-component Pd-catalyzed carbonylative reaction [Substrate, CO and nucleophile (Nu)].

The carbonylation can also be internal; in this case, the substrate acts as the nucleophile to form a cyclic carbonyl derivative. Any molecule with a π bond or a free electron pair may be regarded as a nucleophile. Nucleophiles have been extensively explored in the field of palladium-catalyzed carbonyl-ative chemistry. Nucleophiles such as alcohols, organo-metallic reagents (vide supra) and amines have been used in the synthesis of esters, ketones and amides, respectively. Other nucleophiles such as hydrides and thiols can also be used to create the corresponding aldehydes30 and benzothioates31.

The properties of the leaving group (LG, X) are also important. The halo-gen triad (iodine, bromine, and chlorine) provides the most frequently em-ployed LGs, along with various sulfonate derivatives. For example, trifluo-romethanesulfonates (triflates), tosylates, mesylates32 or nonaflates33 readily

14

undergo oxidative addition and can be prepared from a phenol or ketone precursor. Arylboronic acids have also been employed as substrates in oxida-tive carbonylative reactions.34,35 Carbonylations via C-H activation have also emerged.36,37

Because ligands have such an impact on reaction outcomes, a large num-ber have been developed and are now commercially available.38 Triphenyl-phosphine (PPh3) is perhaps the most commonly used ligand in both cross-coupling reactions and carbonylations with palladium (Figure 3).39 Biden-tate ligands such as 1,1'bis(diphenylphosphino)ferrocene (dppf),40 4,5-bis(diphenylphosphino)-9,9-dimethyl-xanthene (Xantphos)41,42 and (oxydi-2,1-phenylene)bis(diphenylphosphine) (DPEphos)43 have also been frequent-ly used in palladium-catalyzed carbonylations.44

Figure 3. Selected mono- and bidentate phosphine ligands used in the present work.

The mechanism of the palladium-catalyzed carbonylation can be de-scribed in four steps, as depicted in Figure 4. Initially, a precatalyst (palla-dium source and ligands) of Pd(0) or Pd(II) is converted into an active 14-e- Pd(0) species by loss of one or more of the ligands.1,45

Figure 4. Proposed catalytic cycle for the carbonylation of an organo-halide. Oxida-tive addition (i), CO insertion (ii), nucleophilic (Nu) attack (iii) and reductive elimi-nation (iv). R = aryl, alkene, etc.; X = leaving group (LG); L = ligand.

15

Practically, this is performed by adding a palladium source [eg. Pd(II)(OAc)2 or Pd(0)(dba)2] and two or four equivalents of the mono or bidentate phos-phine ligand.46Additionally, the precatalyst may contain a pre-formed palla-dium-phosphine ligand complex such as Pd(PPh3)4.

47 The oxidative addition of Pd(0) into an aryl (or alkenyl)-X bond is the

first step of the catalytic cycle. In this step, the aryl-X bond is broken and both atoms are added to Pd(0), which is oxidized to Pd(II) giving an aryl-palladium complex. At this point, the geometry of the oxidized palladium atom is changed from tetrahedral to square planar. The LG (X) has a large influence on this step. The C-X bond energy to a large extent determines the rate of the oxidative addition, which follows the ranking C-I > C-triflate > C-Br > C-Cl. Halo-alkenes are more reactive than aryl halides.48,49 Electron-deficient aromatic groups (activated) will also favor the oxidative addition since the energy of the C-X bond will be decreased. Also, since palladium is a soft metal, a soft/soft interaction is more favorable with iodide ions. The ligands also influence this step and electron-rich phosphine ligands will in-crease the nucleophilicity of Pd(0), thus increasing the rate of this step. On the other hand, bulky ligands can negatively affect this step.50,51

After formation of the aryl-palladium species, CO insertion occurs. When a cis configuration of the CO and the aryl moiety is adopted, a 1,1-insertion (or migration) occurs, creating an acyl-palladium complex (Ar-CO-Pd(II)Ln). This process is favored by bulky, bidentate and electron-poor li-gands.52

After CO insertion, the nucleophile can attack the electrophilic Pd(II) complex and expel the carbonyl product. In the case of organo-metallic rea-gents, an initial pre-coordination and transmetallation occurs instead. For sterically hindered or less active nucleophiles, acyl-activating groups such as sodium phenoxide,53 N,N-dimethyl-4-aminoazabenzene (DMAP),54,55 or imidazole56 have been used to enhance this process (Scheme 3).

Scheme 3. Proposed mechanism using DMAP as an acylation-catalyst.

Reductive elimination is a common step in all Pd(0)-catalyzed cross-couplings. In this step, the Pd(0) catalyst is regenerated and can participate in another catalytic cycle. Bulky and bidentate ligands can promote this step.57

16

Towards Simpler Carbonylative Procedures Carbon monoxide is an important, cheap, one-carbon building block. Since there is a wide range of halide substrates and nucleophiles available, the diversification of the carbonylative reaction using CO is extensive. CO is also widely used in industry where it is the feedstock in processes such as the synthesis of acetic acid58 and methanol.59 CO is an odorless, colorless, flammable and highly toxic gas which requires special equipment and proto-cols. As a substitute for gaseous CO in small-scale applications, molecules with an inherent carbonyl source can be utilized in carbonylative reactions. In this way, CO is released from a CO-releasing molecule (CORM) by an activator or heat and is then incorporated into the carbonylation reaction (Scheme 4).

Scheme 4. CO-generation from CORMs.

The generated CO can be used as a gas [CO(g) source] or captured as a solid by a substrate and incorporated into the molecule without gaseous interme-diates (CO surrogate). Since the 1990s, methods that avoid the use of ga-seous CO are increasingly used, and new CORMs have been developed. This makes carbonylative chemistry both safer and more convenient, since the CO can be generated in situ or ex situ (Figure 5).

Figure 5. Examples of CORMs (A-C and CO-surrogates D,E).

Phenyl-60 and alkylformates61 and aldehydes (A,B) are attractive due to their low price and have been utilized for their CO-releasing capability upon heat or other activations.62 Potassium oxalate monoesters (C) have been used for the decarboxylative synthesis of aryl and alkenyl esters.63 From the carba-moylsilane (D) the corresponding gaseous-free carbonylation yielding N,N-dimethylamides could be achieved.64 The tri-chlorinated formate E can be used as a CO-surrogate through a highly activated carboxylic derivative to produce for example esters and amides from Sp2-halides.65 Other CO-surrogates such as formamides could also be used.66,67

17

The release of CO ex situ from fluorene68,69 (F) and silacarboxylic acids70 (G) has been explored extensively together with a two-chamber system (Figure 6). The technique relies upon the controlled release of CO by fluo-ride or Pd/P(t-Bu)3 which allows the CORM to be used in near stoichiome-tric amounts. These CORMs are prepared from carbon dioxide (CO2), thus providing a route for isotopic labeling using 13C and 14C.71-73 The metal com-plex molybdenum hexacarbonyl [Mo(CO)6] (I) has been used frequently as a solid source of CO in our and other groups for small scale synthesis during the past decades using both conventional and microwave heating.74-77 This CORM is relatively cheap and may release CO by ligand exchange (LE) with bases or by heat (vide infra).

Figure 6. CORMs (F, G, I), CO-surrogate (H), and gaseous CO (J) molecules. Prices are displayed as cost per mole carbon monoxide.

18

Tracer Development in Positron Emission Tomography The method of using radioactive compounds or atoms to follow biological processes (radioactive tracing) was discovered by George De Hevesy in 1911; he was awarded the Nobel Prize in 1943 for the use of isotopes as tracers in chemical processes.78 According to the International Union of Pure and Applied Chemistry, a tracer is "A foreign substance mixed with or at-tached to a given substance to enable the distribution or location of the latter to be determined subsequently".79 When a stable isotope in a tracer has been replaced by a radioactive isotope, the result is a radioactive tracer. In nuclear medicine the tracer concept can be defined as:

“A tracer is a compound or atom that can be used to follow and quantitatively study a process in a living system without to any measurable extent influence the rate or outcome of the studied process”

Positron emission tomography (PET) is the most advanced nuclear medicine modality and the technology is based on the use of tracers, compounds labeled with short-lived positron (β+) emitting radionuclides.80

The positron was first postulated by Paul Dirac in 1928 and experimental-ly observed in 1932 by Carl Anderson.81,82 Positrons are emitted from neu-tron deficient radionuclides and in the process a proton is converted into a neutron, a positron and neutrino. The development of the cyclotron in the 1930s made it possible to produce positron emitting nuclides.83 However, it was not until the seventies that PET emerged as a modality in the clinical setting.84,85 During the past thirty years, PET has undergone a tremendous development and today PET tracers are routinely produced in medical cen-ters around the world. Table 1 depicts some examples of short-lived positron emitting nuclides created from proton bombardment of stable isotopes using a low-energy cyclotron (<20 MeV).86

Table 1. Selected positron emitting nuclides and properties.

Nuclide Half-life (minutes)

β+ emission (%)

β+ Energy (MeV)

Nuclear reaction

Theoretical SA (GBq/µmol)

11C 20.3 99.8 1.0 (0.3) 14N(p,α)11C 3.41×105 13N 10.0 99.8 1.2 (0.4) 16O(p,α)13N 6.99×105 15O 2.0 99.9 1.7 (0.7) 14N(d,n)15O 3.49×106 18F 109.7 96.7 0.7 (0.2) 18O(p,n)18F 6.33×104

As the positron is emitted as a consequence of the decay, the resultant kinet-ic energy will allow the β+ particle to travel a few mm in tissue before it come to rest and interacts with an electron in an annihilation reaction in which matter (electron) and antimatter (positron) are converted into energy

19

in the form of two antiparallel 511 keV gamma rays. The gamma rays can be detected by a PET camera and an image of radioactivity concentration can then be reconstructed (Figure 7). One of the hallmarks of PET is its ability to determine tissue radioactivity quantitatively with high spatial and tempor-al resolution. Today, combined PET/computed tomography (CT) or PET/magnetic resonances imaging (MRI) systems are used to obtain both anatomical and functional information in one session.87,88 PET is a molecular imaging modality mainly used clinical applications covering neurology, oncology and cardiology and investigations could for example be:

• Studies of receptors, transporters or enzymes. • Metabolic processes, such as protein synthesis, fatty acid synthesis, glu-

cose consumption or oxygen consumption. • Physiological determinations such as blood flow or blood volume.89

Figure 7. The schematic overview of PET, a ring of detectors surrounding the sub-ject to be investigated. Courtesy of Wikimedia Commons.

Tracer Considerations In order for the earlier mentioned tracer concept to be fulfilled, the molecule must exhibit certain pharmacodynamic and pharmacokinetic properties. The following outline of these properties should be regarded as a guideline.

• The selectivity towards the target must be sufficient. • The specificity must be high enough to ensure low nonspecific binding.

20

• The affinity for the target (measured using the rate constant ratio, Kd) must be sufficiently high (low nM) to allow detection even when expres-sion of the target is low. A binding potential (Bmax/Kd) ratio of >10 can be used as a rule of thumb.90

• The kinetics (uptake and washout) should be matched to the half-life of the nuclide used; i.e. the physiological half-life of the radionuclide should match the biological half-life of the process to be studied.

• High sensitivity is required when the receptor density (measured as Bmax) is low. A high specific radioactivity (SA) level will ensure that the tracer will not saturate the target. Ideally, the receptor occupancy should only be a few per cent.91

• The tracer should have an appropriate metabolism. The radioactive me-tabolites should not interact, or accumilate non-specifically in the target organ.

• The lipophilicity of the tracer should be suitable, especially for tracers intended for use in the central nervous system (CNS).92

Additionally, there should be a viable synthetic route for radiolabeling the tracer. The possibility to obtain PET radionuclides with high specific activity (SA) is one of the important aspects of PET. The SA is defined as the amount of radioactivity related to the total mass of the substance, expressed as Bq/mol. The theoretical SA is 340 TBq/µmol for carbon-11 and 63 TBq/µmol for fluorine-18 (see Table 1). For a compound with a SA of 50 GBq/µmol this would correspond to an isotopic dilution of 6800 and 1260 for carbon-11 and fluorine-18, respectively. A high SA is needed to uphold the tracer concept (vide supra) and is of special importance in studies of receptors, transporters and other functional proteins expressed at low con-centration.93

The tracer development comprises an integrated chemistry, radiochemi-stry and preclinical evaluation process. The development of a tracer could thus proceed as follows:94

• selection of tracer candidates; • synthesis of precursors and reference compounds; • labeling of a number of potential tracer compounds; • in vitro screening using autoradiography and homogenate binding stu-

dies; • ex vivo animal studies; and • in vivo animal studies using PET.

The shorter half-life of 11C makes this nuclide suitable for research, since the same subject can be investigated several times within a short time interval; for example, a baseline study could be followed by two further studies after pharmacological challenge, all performed in the space of one day. Fluorine-

21

18 is more suitable for use in the clinical setting, since large batches of the tracer can be produced for use in multiple patients. A tracer with the poten-tial to be labeled with both 11C and 18F would be ideal, since it could cover both scenarios.

The ultimate goal in tracer development is to develop an in vivo biomark-er for the study of a biological process or a disease in human. The use of the microdosing concept95-97 (currently defined as <100 µg of radiotracer) allows first-in-man studies without the need for a costly and time consuming full toxicology study investigation.98 This means that the step from chemistry development to a radiolabeled new chemical entity for use in man can be rather short in relation to what is the normal time perspective for drug devel-opment.

Radiochemistry Because of the relatively short half-lives of most PET radionuclides, exten-sive chemical manipulations are not possible. Furthermore, since the radio-chemistry is restricted to certain precursors, the toolbox available to the ra-diochemist is rather limited. Radiochemistry development is an important part of the tracer development and new radiolabeling methods is thus of im-portance for the further development of the PET technology.99

Carbon-11 is usually produced by proton bombardment of nitrogen gas containing small amounts of a second gas, to create one of the two precur-sors: [11C]CO2 (if the second gas is O2) or [11C]methane (if the second gas is H2). In this thesis [11C]CO2 was used as the primary precursor and converted by online procedures to the secondary precursors [11C]CO and [11C]CN100 (Scheme 5).101

Scheme 5. Secondary precursors derived from the cyclotron-produced primary pre-cursor [11C]CO2.

[11C]CN and [11C]CO are versatile precursors and both can be used in metal-mediated reactions.99 The term mediated is used here, rather than cata-lyzed, since the amounts of [11C]CO produced are far less than those of the substrate and metal, indicating that the metal will only react once.102 The low concentrations of [11C]CO preclude the use of classical carbonylative me-thods.

The difficulty to quantitatively transfer and retain [11C]CO in the reaction mixture has been one of the main obstacles for using [11C]CO in radiochemi-stry. Several methods have been developed to overcome this problem. Metal complexes such as copper(I) scorpionates103,104 and borane105 have been used to increase retention of [11C]CO in solvents. Recently, xenon has been used

22

as a carrier gas to efficiently introduce [11C]CO into the reaction mixture.106 Microfluidic methods have also been used successfully.107,108 High pressure systems can concentrate the [11C]CO and this methodology was used for all carbonylations in this thesis.109

Compounds containing a 11C-carbonyl group can be obtained using for example rhodium,110,106 selenium111 or palladium (Scheme 6).

Scheme 6. Possible 11C-carbonylative reactions using rhodium, selenium or palla-dium.

Palladium-mediated 11C carbonylations have found practical use in radio-chemistry because of their broad substrate scope and tolerance for functional groups; for example, a small library of labeled compounds can be readily produced simply by altering the organohalide precursor and/or the nucleo-phile.112,113 Furthermore, the abundance of bioactive molecules containing a carbonyl group has been an incentive for the development of new methods for incorporating 11C. In principle, the carbonylative methods developed since the 1970s have been translated into radiolabeling with [11C]CO. This has allowed the labeling of a multitude of carbonyl-containing compounds, such as amides,114 ketones,102 carboxylic acids115 and esters.116,117

23

Palladium-Catalyzed Carbonylations using a Two Chamber System

Synthesis of Benzamides (Paper I) Background and Aim CO is a valuable building block in organic chemistry. The carbonylative palladium-catalyzed three-component reaction [using CO, an organo(C-sp2)-(pseudo)halide, and a nucleophile] is an efficient synthetic method for pro-ducing amides, esters and other carbonyl-containing compounds (Scheme 7). In this way, the functionality of the carbonyl group can be incorporated into the late or end stage of a multi-step reaction.

Scheme 7. Three component setup for the Pd(0)-catalyzed carbonylation of aryl halides yielding benzamides.

Mo(CO)6 is an air stable solid capable of releasing CO in situ.56,118 Carbon monoxide can be released by heating119 or by ligand exchange using a coor-dinating solvent such as acetonitrile, or a base such as 1,8-diazabicyclo-undec-7-ene (DBU).120,121 The reductive properties of Mo(CO)6 have, how-ever, precluded the use of substrates that contain, for example, a nitro group, which is reduction-prone, in one-pot carbonylative reactions. Nitro groups offer valuable functionality in that they can be converted to an amine or to nitrogen-containing heterocyclics.122,123

Moreover, the use of two metals (molybdenum and palladium) in the same reaction mixture could create a complicated reaction matrix with for-mation of unwanted by-products. To overcome this problem, a two-chamber reaction vial was used to separate the release of CO from the subsequent capture and carbonylative reaction.

The aim of this paper was to design, fabricate, and evaluate a two-chamber reaction vial as an apparatus for palladium-catalyzed carbonyl-ations. The reactions were designed to be performed at laboratory scale di-mensions, preferably at a level of 0.5-2.0 mmol. This is appropriate for late-stage incorporation of carbonyl functionality or even for multi-step (<5)

24

synthesis, dependent on the carbonyl product. The aim of the synthesis process was to evaluate the use of reductive-prone functional groups, with focus on the nitro group. Aryl iodides were investigated initially, followed by the less oxidative-addition-prone aryl- bromides and chlorides.

Results and Discussion Two-chamber Reaction Vial Various glass vials were evaluated in the development of a two-chamber reaction vial. The fusion of two Smith™ microwave vials (2-5 mL capacity; made from cylindrical borosilicate glass) allowed the two-chamber reaction vial to fit into a DrySyn™ heating block (Figure 8). The chambers were capped by two gas-tight lids intended for use with the microwave vials to provide effective retention of CO.75,76 After the reaction was completed, the vial was washed with 50% aqueous nitric acid and reused >30 times without notable wear.

Figure 8. A photograph and schematic depiction of the two-chamber vial. The total inner volume was 22 cm³. The two compartments C1 and C2 were used with each 3 mL of solvent.

Development of Aminocarbonylations Using Primary and Secondary Amines Initially, a model reaction was set up to evaluate the release/capture and incorporation of CO in a typical aminocarbonylation reaction. Pd(PPh3)4 was chosen as the catalyst as it has previously been used successfully in carbony-lations with aryl iodides. Because primary amines are suitable nucleophiles in Pd(0) carbonylative reactions, n-butylamine was used with 4-iodoanisole and triethylamine (Et3N) for this model reaction. As the solvent, 1,4-dioxane

25

was used throughout the study; a volume of 3 mL was used in each chamber to prevent spillover associated with the intense stirring and to allow suffi-cient heating on the heating block. The reaction parameter optimization study is outlined in Table 2.

The use of 0.5 mmol of 4-iodoanisole, 5% Pd(PPh3)4 and 100% Mo(CO)6 provided the desired product 1 (94% isolated yield; entry 1) after 15 hours (entry 1). Lowering the quantity of Mo(CO)6 to 50% gave similar results (95% yield; entry 2) after 15 hours. These conditions were the most optimal and were consequently used throughout the study with aryl iodides. Decreas-ing the time, the loading of catalyst, or the loading of Mo(CO)6 gave lower yields (14-88%; entries 3-8). These conditions were also appropriate when the reaction was scaled up to 2 mmol, four times the initial amount providing 92% yield (entry 9).

It is also worth mentioning the easy purification of this compound. The crude mixture was simply evaporated to remove the solvent, unreacted base and n-butylamine and, after processing through a short column (30 mm wide, 5 cm long), the pure product was obtained within minutes.

The maximum pressure was measured without any reagents in C2. The evolved gas was measured using a gas-tight syringe; the pressure was ap-proximately 2 bar when 0.5 mmol of Mo(CO)6 was used.

Table 2. Optimization of reaction conditions using the two-chamber vial.

Entry

(n) Time

(hours)Pd(PPh3)4 / Mo(CO)6

(mol%)Yield (%)a

1 15 5 / 100 94

2 15 5 / 50 95

3 5 3 / 50 61

4 15 5 / 30 88

5 5 5 / 30 52

6 5 5/ 50 69

7 15 3 / 100 75

8 15 1 / 50 14

9 15 5 / 50 92b

Reagents and conditions: C1: 4-iodoanisole (0.5 mmol), n-butylamine (1.0 mmol), Et3N (1 mmol), Pd(PPh3)4 (1-5 mol% to aryl iodide). C2: Mo(CO)6 (30-100 mol%), DBU (0.75 mmol). 1,4-dioxane (3+3 mL), 5-15 hours, 65 °C. aIsolated yields, >95% purity (GC-MS and 1H-NMR). b2.0 mmol reaction scale.

26

Scope of the Carbonylative Reaction The optimized reaction conditions were applied to a range of electronic and sterically diverse substrates to evaluate the scope and limitations of the ami-nocarbonylation process (Table 3).

Table 3. Scope of the two chamber palladium catalyzed carbonylation of aryl iodides using primary amides.

Reagents and conditions: C1: aryl iodide (0.5 mmol), amine (1.0 mmol), Et3N (1.0 mmol), Pd(PPh3)4 (5 mol% to aryl iodide). C2: Mo(CO)6 (0.25 mmol), DBU (0.75 mmol). 1,4-Dioxane (6 mL, 3 in each vial), 65 °C, 15 hours. Isolated yields, >95% purity (GC-MS and 1H-NMR). aAll reagents were placed in the same vial. bReaction performed at a 2.0 mmol scale.

To evaluate the strength of the two-step procedure in separating the CO-releasing step from the aminocarbonylation, a reaction was performed with all reagents in the same vial. Although the conversion of the aryl iodide was complete, this resulted in 20% isolated yield of the carbonylated product together with unknown by-products. Aryl iodides with para, meta and ortho-methoxy substituents provided corresponding benzamides 1-3 in 91-95%

27

isolated yield when n-butylamine was used as the nucleophile. When the same nucleophile was used for the electron-neutral (entries 4-5) and elec-tron-deficient (entry 6) aryl iodides this gave benzamides 4-6 in 83-97% yield.

Table 4. Scope of the two chamber palladium catalyzed carbonylation of aryl iodides using anilines and secondary amines as nucleophiles.

Reagents and conditions: same as in Table 3 but 1 mmol of DMAP was added to the reaction vial (C1). Isolated yields, >95% purity (GC-MS and 1H-NMR). aReaction performed at 85 °C.

Further, nitro-containing aryl iodides furnished good yields of the para-, meta- and ortho-substituted carbonylated products 7-9 (78-92%; entries 7-9). No trace of the corresponding reduced products was detected in the reaction mixture. On the other hand, when all reagents were mixed in the same vial, 1-iodo-4-nitrobenzene gave conversion but only a 12% yield of isolated product. Cyclopentylamine provided products 10 and 11 in 92% and 87% yield, respectively. When the less nucleophilic (by resonance stabilization)

28

aniline was used, a yield of 46% was obtained for 12 under these conditions (Table 4; entry 1). Both unreacted 4-iodoanisole and dehalogenated starting material (anisole) were found in the reaction mixture, which suggested that CO-insertion or the nucleophilic attack was sluggish. The temperature was increased to 85 °C and 2 equivalents of the acylation catalyst 4-dimethylaminopyridine (DMAP) were added to overcome this problem, resulting in 87% yield of product 12. DMAP was used for secondary amines (13-17; 69-89% yield) and for anilines (18-20; 71-77% yield) which also required increased temperature (85 °C) to give full conversion of the starting aryl iodides. The product 16 contained the intact bromine suggesting that 65 °C was not a sufficient temperature for the oxidative addition of the less active halide to occur.

Aryl bromides were evaluated next (Table 5). Because a higher reaction temperature was required (vide supra), n-hexylamine (bp 131.5 °C) was used as the standard nucleophile. When the temperature was increased to 85 °C, 4-bromoanisole gave the corresponding benzamide 21 in 55% yield using Pd(PPh3)4 (entry 1). When the catalyst was changed to the bidentate Pd(dppf)Cl2, the product was isolated in 72% yield. The crude mixture re-vealed traces of N,N-dihexylurea, possibly derived from a competing Pd(II)-mediated oxidative carbonylation taking place at the elevated temperature.124 This impurity disappeared when the vial was flushed with nitrogen prior to heating and the yield was increased to 88%. To further investigate the advan-tages of this two-chamber setup, two reactions with all reagents in the same vial were performed. When 4-bromoanisole was employed, 16% of product was obtained after purification despite full conversion of the starting materi-al (no starting material could be detected by LC-MS). Noteworthy, when 1-bromo-4-nitrobenzene was employed, only the reduced product was detected in the crude sample and purification gave no product.

In terms of scope and limitations, primary amines were generally ac-cepted and gave high isolated yields. Aryl bromides bearing either para- or meta-substituents of electron-donating or electron-withdrawing character furnished products in 78-89% isolated yields (21-24). Sterically hindered (ortho-substituted) aryl bromides provided products 25-27 in 32-55% yields. 2-Bromonaphtalene gave product 28 in a good 93% yield. The more sluggish aniline provided products 29 in 27% yield at 85 °C and when DMAP was added this increased to 83%. Importantly, sensitive functional groups such as aldehyde, chlorides and vinyl groups were intact after the carbonylative reac-tion providing corresponding products 30-33 in 69-81% isolated yield. Final-ly, 2-bromopyridine and 3-bromofuran provided the corresponding hetero-cyclic products 34 and 35 in 41% and 79%, respectively.

To conclude, a two-chamber reaction vial was developed and utilized for the preparation of benzamides using a palladium(0)-catalyzed reaction with aryl halides, CO from the solid source Mo(CO)6, and amines. In the setup, Mo(CO)6 was separated from the reaction chamber which enabled the use of

29

reduction-prone functional groups. Aryl halides containing both electron-donating and electron-donating substituents were found to be suitable sub-strates in the carbonylative reaction.

Table 5. Scope of the two chamber palladium catalyzed carbonylation of aryl bro-mides using amines as nucleophiles.

Reagents and conditions: C1: aryl bromide (0.5 mmol), amine (1.0 mmol), Et3N (1.0 mmol), Pd(dppf)Cl2 (5 mol% to aryl bromide). C2: Mo(CO)6 (0.25 mmol), DBU (0.75 mmol). 1,4-Dioxane (6 mL, 3 in each vial), 85 °C, 15 hours. Isolated yields, >95% purity (GC-MS and 1H-NMR). aAll reagents were placed in the same cham-ber. b5 mol % of Pd(PPh3)4 was used. cWithout nitrogen flush prior to heating. d2 equivalents of DMAP was added to C2.

30

In total 20 secondary and tertiary benzamides were produced from aryl iodides and 15 from aryl bromides. The isolated yields spanned from 71-97% and 41-93%, respectively. All compounds were pure as deduced by gas and liquid chromatography with mass spectrometry (GC-MS/LC-MS) and nuclear magnetic resonance (NMR). Aryl chlorides were also subjected to aminocarbonylation but these attempts were unsuccessful.

Synthesis of N-Cyanobenzamides (Paper II) Background and Aim The N-acylcyanamide moiety is interesting from several perspectives. It serves as a carboxylic acid isostere with pKa ranging from 2-4.125 The acyl-cyano functionality could also be subjected to cyclization to form, for exam-ple, tetrazoles. Other transformations are also possible.126,127 Typically, N-acylcyanamides have been produced by N-cyanation of amines or by using reactive acid chlorides or esters128 and cyanamide. This precludes late-stage incorporation of the N-acylcyanamide moiety or N-cyanobenzamide into a multi-step synthesis.

It was proposed that N-cyanobenzamides might be synthesized from an aryl halide precursor by adopting the three-pot methodology using an aryl halide, CO, and cyanamide (Scheme 8).

Scheme 8. The three component carbonylative synthesis of N-cyanobenzamides from aryl halides.

In our laboratory we have previously attempted to develop a one-pot gen-eral methodology for the synthesis of N-cyanobenzamides with Mo(CO)6. However, this approach has not been successful due to troublesome workup and extensive byproduct formation. The aim of this project was to evaluate the two-chamber reaction vial as a method for palladium-catalyzed carbo-nylative synthesis of N-cyanobenzamides. This would provide a new route to this interesting class of N-C-N compounds.

Results and Discussion Optimization of Reaction Conditions As a starting point for the investigation, the conditions from paper I using Pd(PPh3)4 as the catalyst, were adopted. When 50 mol% of Mo(CO)6 (0.25 mmol) and 2 equivalents of cyanamide was used, the desired N-

31

cyanobenzamide 36 was formed in 68% isolated yield (Table 6; entry 1). According to 1H- and 13C-NMR, the isolated product was obtained as the Et3N salt. By washing with 10% aqueous HCl prior to column chromatogra-phy the free benzamide was obtained and this method was used throughout the paper. When the amount of CO-source (100 mol% Mo(CO)6) was in-creased, similar yield was obtained (entry 2) and increasing the amount of nucleophile provided some increasing yield (entry 3; 76%). When the reac-tion time was extended to 20 hours, complete conversion of the starting ma-terial was observed and 83% of 36 was isolated after workup and purifica-tion (entry 4).

Table 6. Optimization of reaction conditions.

Entry

(n) Time

(hours)Pd(PPh3)4/ Mo(CO)6

(mol%)

H2NCN (equiv.)

Yield (%)

1 15 5 / 50 2 68

2 15 5 / 100 2 70

3 15 5 / 100 3 76

4 20 5 / 100 3 83

Reagents and conditions: C1: aryl iodide (0.5 mmol), cyanamide, Et3N (1.0 mmol), Pd(PPh3)4. C2: Mo(CO)6, DBU (1.5 mmol). 1,4-Dioxane (6 mL, 3 in each vial), 65 °C. Isolated yields, >95% purity (1H-NMR).

Scope of the Carbonylative Reaction

Initially, aryl iodides were used in the carbonylative synthesis and the sepa-rated system was compared to a single vial system. When all reagents were mixed in the same vial, only 13% of N-cyanobenzamide 36 was obtained from 4-iodoanisole after purification (Table 7; entry 1). When 3-nitro-iodobenzene was employed in a one-chamber reaction, only traces of the desired product were obtained, despite full conversion of the starting aryl iodide. However, when the two-chamber system was used, 80% of 3-nitro-N-cyanobenzamide 37 was obtained. The electron-withdrawing meta and para-nitro- and cyano- aryl iodides furnished products 38-40 in 54-88% yield. The yield of 40 was low due to a competitive hydrolysis to primary acylurea.129 When electron-neutral and electron-rich substituents were em-ployed the corresponding products 41-43 were isolated in 79-84% yield. The ortho-substituted N-cyanobenzamide 44 was obtained in 58% yield, presum-ably due to sterical hindrance. The tandem carbonylation of 1,4-diiodobenzene produced N,N-dicyanoterephthalamide 45 in 78% yield. In

32

the reaction, double the amount of cyanamide, Pd-catalyst and Mo(CO)6/DBU were used. When this reaction was performed with all rea-gents in the same vial, only small amounts of the product could be obtained despite full conversion of 1,4-diiodobenzene. From 4-iodobenzoic acid, the diacidic 4-(cyano-carbamoyl)benzoic acid 46 was isolated in 64% yield.

Table 7. Scope of aryl iodides in the carbonylative synthesis of N-cyanobenzamides using the two-chamber vial.

Reagents and conditions: C1: aryl iodide (0.5 mmol), cyanamide (1.5 mmol), Et3N (1.0 mmol), Pd(PPh3)4 (5 mol%). C2: Mo(CO)6 (0.5 mmol), DBU (1.5 mmol). 1,4-Dioxane (6 mL, 3 in each vial). aAll reagents were placed in the same vial. Isolated yields, >95% purity (1H-NMR).

Next, the scope was expanded to include aryl bromides (Table 8). When aryl bromides were employed the reaction temperature was increased to 85 °C. When 4-bromoanisole was used in a single vial, no product was found in the post-reaction mixture (entry 1). Separating the release and capture of CO

33

furnished 46% of isolated product using Pd(PPh3)4. Switching to the more active bidentate ligand dppf [Pd(dppf)Cl2] increased the product yield to 68% (entry 1). A one-pot reaction with 1-bromo-3nitrobenzene resulted in no detectable product in the post-reaction mixture.

Table 8. Scope of aryl bromides in the carbonylative N-cyanobenzamide formation using the two-chamber vial.

Reagents and conditions: C1: aryl bromide (0.5 mmol), cyanamide (1.5 mmol), Pd(PPh3)4, Et3N (1.0 mmol). C2: Mo(CO)6 (0.5 mmol), DBU (1.5 mmol). 1,4-Dioxane (6 mL, 3 mL in each vial). aPd(PPh3)4 (5 mol%) was used. bAll reagents in the vial of the chamber. cPd(dppf)Cl2 (5 mol%) was used. Isolated yields, >95% purity (1H-NMR).

When the two chamber system was employed with Pd(PPh3)4, 66% of 37 was isolated (entry 2). Using Pd(dppf)Cl2 gave a similar yield of 69%. This provided further confirmation that a more reactive Pd-catalyst was needed for electron-donating (deactivated) aryl bromides. Substrates compromising electron-withdrawing substituents (meta and para cyano/nitro) provided products 38-40 in 42-71% yield. Electron-neutral aryl bromides furnished N-cyanobenzamides 41-43 in 72-85% yield using Pd(PPh3)4. The sterically hindered 1-bromo-4-methoxy-2-nitrobenzene provided product 44 in 47% using Pd(PPh3)4 and 55% using Pd(dppf)Cl2. When 1,4-dibromobenzene was subjected to a tandem carbonylation using the Pd catalysts, the desired prod-uct was isolated in 72% and 77% of 45, respectively. Using the same proto-col for a one-pot reaction did not give any product but only a complicated reaction mixture. The heterocyclic 3-bromobenzo[b]thiophene furnished

34

corresponding N-cyanobenzamide 47 in 79% using Pd(PPh3)4 and 81% using Pd(dppf)Cl2.

Finally, the carbonylative synthesis of N-cyanobenzamides using pressu-rized CO gas in an autoclave was explored (Table 9). This would provide an opportunity to compare the two chamber reaction vial to a CO-reactor. Fur-ther, the synthesis could be performed on a larger scale than with the two chamber reaction vial. The CO pressure in the reactions was 2 atm (~30 psi). When 4-iodoanisole was used, the isolated yield of 36 was in the same range as in the two chamber system (76%; entry 1).

Table 9. Scope of aryl bromides in the carbonylative N-cyanobenzamide formation using a CO-reactor.

Reagents and conditions: aryl halide (0.5 mmol), cyanamide (1.5 mmol), Pd(PPh3)4 (5 mol%), Et3N (1 mmol). CO-autoclave (2 atm). 1,4-dioxane (3 mL, 10 mL for 5 mmol scale). 65 °C for aryl iodides, 85 °C for aryl bromides, 20 hours. aReaction conducted on a 5 mmol scale.

Aryl iodides with nitro (meta, para) or cyano (para) also provided corres-ponding N-cyanobenzamides 37-40 in the same range as with the two cham-ber vial (58-76%). Electron-neutral aryl bromides and iodides furnished products 41, 42, 44 and 49 in 69-79% yield. When 1-bromo-4-chlorobenzene was used as the substrate, 4-chloro-N-cyanobenzamide (48) was obtained in 78% yield, suggesting that the reaction was highly regioselective. Three aryl iodides, 4-iodoanisole, 4-phenyl-iodobenzene and 1,4-diiodobenzne were

35

reacted on a 5 mmol scale which furnished the corresponding N-cyanobenzamides 36, 42 and 45 in 81%, 84% and 77%, respectively. The reactions were conducted in 10 mL of solvent with the same stoichiometry as for the reactions on 0.5 mmol scale.

In conclusion, a palladium-catalyzed protocol for the carbonylative syn-thesis of N-cyanobenzamides from aryl halides using Mo(CO)6 as a solid source of CO has been developed. Both aryl iodides and aryl bromides were used as substrates and the reaction proceed smoothly at 65 °C and 85 °C, respectively. In order for the reaction to proceed efficiently, a two chamber reaction vial which separates the CO-releasing chamber from the CO-accepting/reaction chamber was used. Aryl halides with both electron-donating and electron-withdrawing substituents furnished N-cyano-benzamides with isolated yields spanning from 55 to 88%. The methodology was further used in a CO-reactor operating at 2 atm of CO. Ten aryl halides were synthesized with yields spanning from 58 to 85%. Three aryl iodides were further synthesized on a 5 mmol, a tenfold increase in scale from the two chamber reaction vial, which gave products in high yields (77-84%).

36

Pd-mediated 11C-Carbonylative Synthesis of 11C-N-Cyanobenzamides (Paper III)

Background and Aim The previously developed three-component synthesis of N-cyanobenzamides inspired us to investigate the possibility of an analogous 11C carbonylative method for the synthesis of 11C-N-cyanobenzamides from aryl halides. N-cyanobenzamides serve as a bioisostere for carboxylic acid, which is an im-portant functional group in many bioactive molecules.130 The aim of this project was to develop a method of labeling N-cyanobenzamides with 11C. The three-component reaction setup below was used (Scheme 9).

Scheme 9. The three component setup for the carbonylative synthesis of 11C-N-cyanobenzamides.

The main parameters of interest for optimization such as temperature, media-tor system (Pd/L) and solvent of the reaction were evaluated. The effective-ness of the reaction was determined by using the parameters trapping effi-ciency (TE), radiochemical purity (RCP) and analytical radiochemical yield (RCY) as outlined in Equation 1.

TE × RCP = RCY

Equation 1. Relationship between the trapping efficiency (TE) and radiochemical purity (RCP) with analytical radiochemical yield (RCY).

• The TE is defined here as the fraction of [11C]CO retained in the reaction mixture. This is measured by removal of unreacted [11C]CO in the crude post-reaction mixture through a nitrogen flush.

• The RCP is defined here as the fraction of the product in relation to other labeled entities as analyzed by radio-HPLC.

• The RCY is defined as the amount of isolated product in relation to the starting amount of radioactivity. This could also be estimated using ana-lytical radio-HPLC and TE.

37

The corresponding non-radioactive N-cyanobenzamides were synthesized either through the procedure in paper II or by peptide coupling of the corres-ponding benzoic acid with cyanamide using benzotriazol-1-yl-oxytri-pyrrolidinophosphonium hexafluorophosphate (PyBOP). The reference compounds were used to identify the product through co-elution on analyti-cal radio-HPLC. All isolated RCYs were decay-corrected to the end of bom-bardment (EOB) throughout the thesis.

Results and Discussion Evaluation of Reaction Parameters The appropriate reaction conditions were screened by optimizing tempera-ture, reaction time and catalytic system. A small solvent screen concluded that DMF readily dissolved all reagents properly. For this optimization study, the electron-neutral iodobenzene (20 µmol) was chosen as model sub-strate and Pd(dba)2 (8 µmol, 40 mol-%) as the palladium source with either monodentate (20 µmol) or bidentate (10 µmol) ligand, a large excess (20 equivalents 400 µmol) of cyanamide were used throughout the study. As a starting point 120 °C was used as the reaction temperature.

Initially, a ligand-free protocol was employed furnishing [11C]50 in 85% TE, a RCP of 88% RCP and RCY of 75%. (Table 10; Entry 1). The mono-dentate ligand triphenylphosphine furnished almost a complete conversion of [11C]CO into non-volatile products (95% TE; entry 2). The lower RCP (75%) was due to a byproduct with similar chromatographic behavior (Fig-ure 9; A). This by-product was identified as [11C]benzoic acid by co-elution using benzoic acid.

Figure 9. Crude analytical radio-HPLC chromatograms (radio-trace). A) PPh3, B) Xantphos, C) DPEphos, D) dppf.

Next, the bidentate Xantphos (bite angle 111 °) was used as the ligand, which resulted in a complete conversion of [11C]CO into non-volatile prod-ucts. However, a highly lipophilic entity was detected (Figure 9, B) which resulted in a lower RCP at the 120-150 °C interval (61-72%; entries 3-4). When the ligand was changed to the more flexible bidentate ligand DPEphos

A B

C D

[11C]50

[11C]50

[11C]50

[11C]50

38

(bite angle: 104 °) this resulted in a lower TE at the interval 90-150 °C (76-85%) but slightly higher RCP (Table 10; 79-88%; entries 5-7, Figure 9; C). When the ligand was changed to dppf (bite angle: 99 °), a complete TE (>99%) and high RCP (Table 10; 94±2%; entry 8, Figure 9; D) was ob-served. The analytical RCY was 93±2%. When a higher (150 °C) or lower (90 °C) temperature was used, the observed quantity of TE and/or RCP was lower (RCY 72-87%; entries 9-10).

Table 10. Screening parameters for the synthesis of [11C]N-cyanobenzamide.

Entrya Ligand Temp (°C) TE (%)b RCP (%)c Analytical RCY (%)d

1 - 120 85 88 75

2 PPh3 120 95 75 71

3 Xantphos 120 >99 61 61

4 Xantphos 150 >99 72 72

5 DPEphos 90 85 88 75

6 DPEphos 120 76 79 60

7 DPEphos 150 82 86 71

8e dppf 120 >99 94±2 94±2

9 dppf 150 >99 87 87

10 dppf 90 84 86 72

11f dppf 120 84 87 73

12e,f dppf 140 96±2 93±3 89±5

13f dppf 150 93 90 84

14g dppf 170 63 74 47

15h dppf 120 89 88 78

aReagents and conditions: iodobenzene (20 µmol). H2NCN (400 µmol), Pd(dba)2 (8 µmol), ligand (monodentate 20 µmol, bidentate 10 µmol), DMF (400 µL), 5 mi-nutes. bFraction of immobilized [11C]CO after N2 purge. cDetermined by analytical radio-HPLC. dSee equation 1. eAverage of three experiments. fPh-Br. gPh-Cl. hPh-OTf.

The use of bromobenzene resulted in a lower TE (84%) and RCP (87%) at the choosen temperature. Increasing the temperature to 140 °C resulted in a higher TE (96±2%) and RCP (93±3%; entry 12). The analytical RCY was 89±5%. The conversion did not increase when the temperature was raised to 150 °C (entry 13). When less reactive chlorobenzene was used the TE was 63% and the RCP 74% (entry 14). Finally, the conditions used for the iodo-

39

benzene were applied for phenyl trifluoromethane-sulfonate (triflate) result-ing in a TE of 89% and a RCP of 88% (entry 15), thus indicating that this method was applicable to the most common substrate classes.

Scope of the 11C-Carbonylative Reaction The scope and limitation of the method was evaluated by using aryl halides with different electronic (electron-withdrawing and donating) and steric properties. The reactions were performed starting with 10 GBq of [11C]CO2 at the EOB. Aryl iodides were examined initially.

Iodobenzene produced [11C]N-cyanobenzamide [11C]50 in 72% RCY af-ter semi-preparative radio-HPLC (Table 11; entry 1).

Table 11. Scope of the 11C-carbonylative reaction using aryl iodides/triflate.

Reagents and conditions: aryl iodide (20 µmol). H2NCN (400 µmol), Pd(dba)2 (8 µmol), dppf (10 µmol), DMF (400 µL), 120 °C, 5 minutes. aValues with ± indicates the average of three reactions. bFraction of immobilized [11C]CO. cIsolated yield after semi-preparative radio-HPLC, the RCP was higher than 95% for all isolated products.

40

The same protocol was used for phenyl triflate which provided [11C]50 in 72% RCY. Aryl iodides with electron-donating (ortho/para-methoxy) and electron-neutral (para-phenyl) substituents provided products [11C]51-53 in 64-73% RCY. Aryl iodides substituted at the para-position with electron-withdrawing groups furnished products [11C]54-56 in 41-62% RCY. The sterically hindered 2-trifluoromethyl-iodobenzene provided [11C]57 in 28% RCY (entry 8). The SA for [11C]50, [11C]52 and [11C]54 was measured to 266, 337 and 283 GBq/µmol, respectively (at EOS).

Next, different aryl bromides were evaluated (Table 12).

Table 12. Scope of the 11C-carbonylative reaction using aryl bromides/chloride.

Reaction conditions: aryl bromide (20 µmol). H2NCN (400 µmol), Pd(dba)2 (8 µmol), dppf (10 µmol), DMF (400 µL), 140-170 °C, 5 minutes. aValues with ± indi-cates the average of three reactions. bFraction of immobilized [11C]CO. cIsolated after semi-preparative radio-HPLC, the RCP was higher than 95% for all isolated products.

41

Bromobenzene was reacted at 140 °C, providing [11C]50 in 72% RCY (entry 1). This was similar to the result when aryl iodides or aryl triflate was used. Applying the same conditions to chlorobenzene, but at elevated temperature (170 °C) furnished the same product in 34% RCY (Entry 1). Aryl bromides containing the electron-donating para-methoxy or electron-withdrawing para-nitrile gave corresponding 11C-N-cyanobenzamides [11C]51 and [11C]52 in 67 and 68% RCY, respectively (entry 2-3). The 2-bromonaphthyl produced compound [11C]53 in 51% RCY (Entry 4). Heteroaromatic bromo substrates were also subjected to the 11C-carbonylation and 2-bromobenzo-furane provided the corresponding [11C]59 in 48% RCY. Thiophenes substi-tuted at the 3 or 2 position furnished [11C]60 and [11C]61 in 59 and 64% RCY, respectively.

With a suitable protocol for the 11C-carbonylative synthesis in hand we sought to radiolabel bioactive molecules, thus simulating a delivery to the clinic with a tracer candidate containing the 11C-N-cyanobenzamide moiety. A literature survey of potent benzoic acid molecules was conducted. Our aim was to produce the 11C-N-cyanobenzamide analog using our developed me-thodology. Two different compounds were selected from the search, the non-steroidal anti-inflammatory drug flufenamic acid and the thromboxane syn-thase inhibitor Dazoxiben (Scheme 10).

Scheme 10. Retrosynthetic approach to 11C-N-cyanobenzamide analogs using aryl bromide precursors.

The synthetic route to the flufenamic acid analog in outlined in Scheme 11.

Scheme 11. Reagents and conditions: (i) Pd2(dba)3, Xantphos, Cs2CO3, DMF, 60 °C, 12 hours; (ii) Pd(dba)2, dppf, [11C]CO, DMF, 140 °C, 5 minutes.

The precursor synthesis started from readily available 1-bromo-2-iodo-benzene which was subjected to a palladium-catalyzed amination using 3-

42

(trifluoromethane)aniline as the nucleophile with Pd2dba3/Xantpos as the catalytic system providing 63 in 46% yield.131,132 The reference compound 64 could be synthesized from flufenamic acid and cyanamide using PyBOP as the peptide coupling reagent. The bromo precursor 63 was then subjected to the 11C-carbonylation starting with 20 GBq of 11CO2. The protocol for aryl bromides was applied and after semi-preparative radio-HPLC, 2.7±0.1 GBq (54±2% RCY, n=2) of [11C]64 (RCP> 95%) was obtained. The total synthe-sis time was 40 minutes with an SA of 382±21 GBq/µmol at the EOS.

The synthesis of dazoxibene started from 1-bromo-4-(2-chloroethoxy)-benzene which was reacted with imidazole to give the bromo precursor 65 in 78% yield (Scheme 12). The corresponding reference compound 66 was synthesized using the two-chamber vial system described in paper I-II. The conditions from the aryl bromide reactions were applied in the 11C-carbonylation to yield 3.9 GBq (71% RCY, n=1) of [11C]66 40 minutes after EOB. The SA was 477 GBq/µmol at the EOS.

Scheme 12. Reagents and conditions: (i) NaH, 1,4-dioxane, 0-6 °C, 5 hours; (ii) Pd(dba)2, dppf, [11C]CO, DMF, 140 °C, 5 min.

In conclusion, the 11C-carbonylative radiosynthesis of 11C-N-cyanobenz-amides from aryl halides (I, Br, Cl, OTf) was described. An initial screening concluded that the carbonylations proceeded at 120 °C for iodides and trif-lates, 140 °C for bromides and 170 °C for chlorides. Aryl halides (eight iodides, one triflate, eight bromides and one chloride) containing electron-donating and electron-withdrawing substituents furnished the corresponding [11C]N-cyanobenzamides in 34-79% RCY after semi-preparative radio-HPLC. Two analogs of bioactive compounds containing benzoic acids (flu-fenamic acid and dazoxibene) were synthesized containing the [11C]-N-cyanamide moiety. The flufenamic acid and dazoxibene analogs will be sent for biological evaluation to further evaluate the N-cyanobenzamide as bio-isostere for the carboxylic acid moiety. This biological study is, however, outside the scope of this thesis.

43

Amyloid Diseases - Enzyme Inhibitors and PET Tracers

Amyloid Diseases The aggregation of soluble proteins into structured aggregates of insoluble deposits which disrupt normal organ and tissue function is known as amyloi-dosis.133 In the process, extracellular deposits of proteins are adopting an insoluble secondary structure of fibrillar type, similar to that of a β-sheet (Figure 10).134

Figure 10. Aggregation of peptides into higher ordered fibrils.

Depending on the distribution of deposits the amyloidosis is classified as localized (one organ or tissue is affected) or systemic (multiple organ af-fected).135 The origin of amyloidosis may depend on genetic (familial) or environmental factors, however a majority of the cases are sporadic in na-ture. The pathological background is often unclear; for example, protein aggregation may occur as a result of increased protein concentrations or of modified structure stability due to mutations or proteolysis.136

Many proteins have been identified as capable to form aggregates and they are characterized based upon the precursor protein. Currently, the Inter-national Society of Amyloidosis have listed thirty non-related proteins capa-ble of aggregate formation.137 Localized amyloidosis is associated with ma-jor disorders such as Alzheimer's [AD, involving amyloid-β (Aβ)] disease with amyloid-β aggregates in the brain and diabetes type II (involving islet amyloid polypeptide; IAPP).138 The systemic forms of amyloidosis of partic-

44

ular interest in this thesis are139 immunoglobulin light chain amyloidosis (AL)140 where the light-chain derives from a subunit of an antibody which is produced abnormally by the bone marrow and Transthyretin (ATTR), a normal transport protein found in plasma.141 The AL amyloidosis causes amyloid deposits in several organs and the most severe are those found in the heart and the prognosis for the patients with heart involvement is very poor.142 In ATTR amyloidosis the stable tetrameric form is by unknown or genetic reason disrupted and the protein aggregates in different organs in-cluding the heart.143,144

Alzheimer´s Disease, Amyloid-β and BACE-1 More than a century ago, the German psychiatrist Alois Alzheimer first de-scribed what later come to be known as AD. In 1901 he was investigating a patient named Auguste Deter who was suffering from severe, strange beha-vioral symptoms. After her death five years later, pathological investigation of the brain by Alzheimer revealed “clumps of filaments” in the cerebral cortex.145 In 1911 he published a more comprehensive summary of the study together with illustrations of his findings.146,147

AD is a progressive, neurodegenerative disorder that currently has no cure. The disease can be inherited genetically or occur sporadically. The World Health Organization has estimated that almost 8 million people de-veloped dementia in 2010 and that over the next few decades this will in-crease to over 20 million annually.148 It is estimated that approximately 37 million people are currently suffering from AD and that this number will have doubled by 2030.149 Since most people with dementia have some form of AD pathology,150 there is an urgent need for better therapeutic and diag-nostic tools to understand and fight the disease.151

The "clumps" noticed by Alzheimer were aggregates of Aβ, a 36-43 ami-no acid peptide which is the major constituent of extracellular amyloid depo-sits in the brain. Aβ is created when membrane-bound amyloid precursor protein (APP) is sequentially cleaved by the aspartic proteases β-secretase (also known as BACE-1) and γ-secretase. The enzyme α-secretase is also able to cleave APP, but this leads to the non-amyloidogenic p3 fragment, Aβ17-42 (Figure 11, A). The Aβ42 peptide fragment is especially prone to amyloid formation and consequently forms a major part of the deposits.152

The inhibition of β- and γ-secretase has been identified as a plausible drug target with the aim of decreasing Aβ levels in the CNS. However, since β-secretase is the rate-limiting step, it seems the most promising therapeutic target.153 Aspartic proteases cleave peptide bonds though catalytic hydrolysis with high selectivity. In the process, a tetrahedral transition state (TS) is formed by the activation of a water molecule, and an amine and a carboxylic acid are formed thereafter (Figure 11, B). One strategy in the attempts to inhibit this aspartic protease has been to mimic the TS. The activity of the

45

enzyme can be decreased by replacing the cleaved amide with a stable, non-cleavable TS isostere.

Figure 11. A: Cleavage of APP by either α- or β-secretase and subsequently by γ-secretase into amyloidogenic Aβ or to non-amylodogenic p3. B: Proposed mechan-ism of aspartic proteases.

This approach has previously been used in the development of antiviral agents for HIV.154 Identification of the BACE-1 enzyme in 1999 and devel-opment of the first potent inhibitor (Figure 12, OM99-2) one year later laid the foundation for development of TS isosteres as an approach to BACE-1 inhibition and a possible Alzheimer´s therapy.155 Since then, a plethora of BACE-1 inhibitors, both peptide-like and non-peptide-like, have been devel-oped (Figure 12, KMI-429,156 BSI-IV157).158

Figure 12. Selected BACE-1 inhibitors.

46

Synthesis of BACE-1 Inhibitors Comprising α-Phenylnorstatine and α-Benzylnorstatine as Transition State Isosters (Paper IV) Background and Aim As described in the previous section, the processing of APP into amyloid-β by BACE-1 is an important stage in the development of Alzheimer’s disease. Inhibition of the aspartic protease BACE-1 is one approach to decreasing the burden of Aβ associated with neurodegeneration and AD. Since the discov-ery of the BACE-1 structure, many academic and industrial groups have developed inhibitors containing peptidomimetic TS isosteres aiming to block the proteolytic cleavage of APP.

Our lab has synthesized TS isosteres containing tertiary alcohols in order to increase the permeability of the HIV-1 protease membrane to inhibitors, while preserving the inhibitory properties.159,160 This motivated us to develop TS isostere inhibitors targeting BACE-1.161,162 Several TS-isostere moieties such as norstatine and hydroxyethylamine (HEA), have been successfully developed. Kiso and coworkers have employed the phenylnorstatine as a TS-isostere.156,163 The aim of this project was to synthesize and evaluate a series of α-phenylnorstatine- and α-benzylnorstatine-containing BACE-1 inhibitors (Figure 13).

Figure 13. Phenylnorstatine investigated by Kiso et al. (Figure 12, KMI-429) and four proposed TS-isosters investigated in paper IV.

The synthetic strategy was to use a combinatorial approach with solid-phase peptide synthesis (SPPS). SPPS is the sequential synthesis of peptides from amino acid fragments; the process was pioneered by Robert Merrifield in the 1960s.164 Initially, a protected (using groups such as Fmoc, tert-butyl, etc.) amino acid is coupled to an insoluble polystyrene bead. The protecting group is then removed and the next protected amino acid is coupled to the first. This sequence can then be repeated until the desired number of amino acids are coupled to the resin.165 The formed peptide is then released from the resin and further purified by HPLC. In this project, a 2-chlorotrityl chlo-ride resin was used with Fmoc-protected acid, built from the C-terminus.166,167

The non-peptide core fragments containing α-phenyl-norstatine, α-benzylnorstatine and isoserine were synthesized; the peptide β-alanine was

47

commercially available. The backbone of the inhibitors is outlined in Figure 14.

Figure 14. The backbone of the inhibitors with domains a-e.

Results and Discussion Synthesis and Evaluation of Inhibitors The fragments containing α-phenylnorstatine and α-benzylnorstatine were synthesized according to Scheme 13. The acrylic acid 67 was reacted with tert-butyl-3-aminobenzoate in a peptide-coupling reaction to give a 36% yield of amide 68.

Scheme 13. Reagents and conditions: i) tert-butyl-3-aminobenzoate, HATU, DI-PEA, DCM, 40 °C, 14 hours; (ii) mCPBA, DCM, 80 °C, 24 hours; (iii) KOH, ETOH, rt; (iv) TFA, DCM, rt, 5 hours; (v) 2-chlorotrityl chloride resin, DIPEA, DCM; (vi) NH3, 0 °C, 4 hours; (vii) PyBOP, DIPEA, DMF or DMSO, rt, 15 hours; (viii) TFA (95% aq.), Et3SiH, rt, 2 hours.

The racemic epoxide 69 and enantiomers (R)-70 and (S)-70 were hydrolyzed and peptide-coupled with tert-butyl-3-aminobenzoate to give 37-46% yields

48

of the epoxides 71, (R)-72 and (S)-72. The amide 68 was epoxidized to give the racemic (rac)-72. The compounds were treated with aqueous TFA and then immobilized to give 73, (R)-74, (S)-74 and (rac)-74. Epoxide opening with ammonia gave the primary amines, which were peptide-coupled in four consecutive cycles using Fmoc-protected amino acids and finally with tetra-zole-5-carboxylic acid. The peptide was released using TFA and purified using HPLC, giving inhibitors 75-85.

The synthesis of isoserine as the core fragment is outlined in Scheme 14. Acryloyl chloride was reacted with tert-butyl-3-aminobenzoate to give amide 86. Epoxidation gave compound 87, which was subsequently opened with aqueous ammonia to give the primary amine 88 in quantitative yield. The amine was Fmoc-protected, hydrolyzed and coupled to the resin (2-chlorotrityl chloride). The immobilized, Fmoc-protected 89 was deprotected and subjected to four consecutive amino acid coupling cycles, after which tetrazole-5-carboxylic acid was attached. The peptide was released with TFA and purified by HPLC, giving inhibitors 90-95.

Scheme 14. Reagents and condtions: (i) tert-butyl-3-aminobenzoate, DCM, 0 °C, 2 hours; (ii) mCPBA, DCM, 80 °C, 24 hours; (iii) NH3 (25% aq.), 60 °C, quant.; (iv) Fmoc-Cl, NaHCO3, 1,4-doxane, 0 °C; (v) TFA, DCM, rt, 5 hours; (vi) 2-chlorotrityl chloride resin, DIPEA, DCM; (vii) 20% piperidine in DMF, rt, 25 minutes; (viii) DIPEA, DMF or DMSO, rt, 15 hours; (ix) TFA (95% aq.), Et3SiH, rt, 2 hours.

Inhibitors lacking the hydroxyl moiety were also synthesized with the commercially available β-alanine as the central core. 3-(Fmoc-amino)benzoic acid was immobilized onto a 2-chlorotrityl chloride resin and, after five peptide couplings, the inhibitors (R)/(S) 96-97 were released using aqueous TFA/Et3SiH, and then purified using HPLC (Scheme 15). The crystal structure of (R)-77 (vide infra) revealed a novel binding mode to the enzyme in which the key interaction with the two aspartic acids was via the N-terminal moiety situated between domains a and b of the inhibitor and not

49

via the prospective hydroxyl moiety in domain d. For this reason, two inhibi-tors truncated at the C-terminal (c and c,d domains removed) were also syn-thesized.

Scheme 15. SPPS route to inhibitors with β-alanine TS isosteres.

Inhibitors containing the tertiary hydroxyl moiety are outlined in Table 13. Moving the benzyl moiety in domain d from the β-carbon to the α-carbon resulted in a more than hundred-fold decline in activity against the enzyme [3.9 nM for KMI-429 versus 0.43 µM for 75 (epimeric mixure)].

Table 13. Structures and inhibition data (IC50) for BACE-1 inhibitors containing tertiary hydroxyl moiety.

aDiastereomeric mixture (1:1)

50

The similar potency was obtained for the one-carbon elongated 76. Of the corresponding diastereomers (R)-77 and (S)-77, the R-isomer was the more potent, with inhibition data (IC50) of 0.19 and 1.4 µM, respectively. The amino acids in domains a-d were interchanged in order to find inhibitors with higher potency. A slight loss of potency resulted from exchanging the leucine to L-α-cyclohexylglycine in domain c (78, IC50 = 0.68 µM). Loss of activity was also seen on elongating or truncating the valine in domain b with L-phenylalanine [(R)-79, (S)-79)], L-norleucine (80), L-leucine (81) or β-leucine (82) (IC50 >10 µM). Modification of the N-terminal by inversion of the stereochemistry (83), truncation (84), or shifting to a serine analog [(R)-85 and (S)-85)] rendered inactive compounds (IC50 >10 µM).

Table 14. Structures and inhibition data (IC50) for BACE-1 inhibitors compromising secondary hydroxyl and aliphatic moieties.

aDiastereomeric mixture (1:1)

Inhibitors without the phenyl and benzyl moiety are outlined in Table 14. The isoserine-containing 90 was equally as potent as the benzyl-containing 75 (IC50 0.44 vs 0.43 µM) suggesting that the phenyl and benzyl moieties did not interact with the active site of the enzyme. This was further strengthened by the crystallography data. Truncation of the b domain resulted in loss of activity (91, IC50 = 3 µM). Modification of the N-terminal by truncation (92), introduction of a serine isostere (93 and 94) or stereochemical inver-sion (95) resulted in inactive compounds (IC50 >10 µM). Inhibitors with β-alanine as the TS iostere were also synthesized. Inhibitor (R)-96 had an IC50

51

of 0.78 µM whereas (S)-96 was inactive (IC50 >10 µM). Modification of the N-terminal with a serine analog resulted in inactive inhibitors (R)-97 and (S)-97 (IC50 >10 µM). Two C-terminal truncated inhibitors were also synthe-sized, with IC50 values of 6.4 µM (98) and 4.4 µM (99).

Figure 15. Co-crystallization of (R)-77 with BACE-1.

The most potent inhibitor, (R)-77, was co-crystallized with BACE-1 (Figure 15, PDB-code: 3KYR). The crystal structure revealed that the amine of the N-terminal forms a hydrogen bond with Asp228, one of the catalytic aspartic acids. Moreover, the tetrazole (a) interacts with Arg235, and the valine (b) interacts with Gly34 and Thr198. The leucine (c) interacts with Pro70 in the flap and Thr198 from the carbonyl. The phenyl moiety (d) ex-

52

tends outside the surface of the enzyme, thus explaining the retention of potency for inhibitors with reduced d domains (75 and 76 versus 90). The C-terminal (e) interacts with Thr68, Lys75 and Arg128.

In summary, BACE-1 inhibitors containing α-phenylnorstatine, α-benzylnorstatine, isoserine and β-alanine as the TS isostere were synthesized using SPPS with Fmoc-protected amino acids using 2-chlorotrityl chloride connected to the solid phase. The overall yield over four or five steps was between 3 and 35% after HPLC purification. The most potent inhibitor [(R)-77, IC50 = 0.19 μM] was co-crystallized with BACE-1 to reveal a mode of binding to the two aspartic acids (Asp32 and Asp228) that was different from that of the similar phenylnorstatine containing KMI-429 (IC50 = 3.9 nM). According to the X-ray data, the N-terminal, rather than the intended TS isostere, of (R)-77 was interacting with the catalytic site.

11C-Labeling of a Potent Hydroxyethylamine BACE-1 Inhibitor and Evaluation in vitro and in vivo (Paper V) Background and Aim The BACE-1 enzyme is responsible for the catalytic cleavage of APP into Aβ (see Figure 11). Increased activity of the BACE-1 enzyme is linked to increased production of Aβ. The in vivo imaging of BACE-1 using PET might help in further understanding the progression of the disease as well as in the development of new BACE-1 inhibitors.168 With this in mind, we sought to translate one high affinity BACE-1 ligand into a 11C PET tracer with the aim of quantifying the enzyme in vitro and in vivo.

In 2004, Merck published a study of an HEA BACE-1 inhibitor (abbre-viated here as BSI-IV) with high affinity to the BACE-1 enzyme.157 After a retrosynthetic analysis it was concluded that the compound could be ac-cessed using a palladium-mediated aminocarbonylation with an aryl halide precursor, [11C]CO and a commercially available chiral benzylamine (Fig-ure 16).

Figure 16. Retrosynthetic approach to [11C]BSI-IV.

53

Results and Discussion

Synthesis of Precursor and Reference Compounds The aryl iodide and the aryl bromide precursor were synthesized in parallel, starting from methyl-3-nitrobenzoate (Scheme 16). Halogenation gave 100a and 100b which were reduced to provide compounds 101a and 101b. Treat-ment with mesyl chloride gave compounds 102a and 102b.

Scheme 16. Reagents and conditions: (i) NBS or NIS, triflic acid, rt, 4 days; (ii) SnCl2, EtOH:THF, reflux, 30 minutes; (iii) MsCl, DCM:Pyridine, rt, 15 hours.

Methylation of 102a and 102b provided 103a and 103b (Scheme 17). Basic hydrolysis gave 104a and 104b in an overall yield of 28% and 29% over five steps, respectively. The chiral epoxide 105 was treated with cyclopropyla-mine and, after hydrolysis of the tert-butyloxycarbonyl (BOC) protecting group, 106 was isolated. The two precursors 107a and 107b were synthe-sized by peptide coupling of either iodide 104a or bromide 104b with 106 using PyBOP as coupling reagent.

Scheme 17. Reagents and conditions: (i) NaH, MeI, DMF, rt, 2 hours; (ii) NaOH, THF:MeOH, 50 °C, 15 hours. (iii) cyclopropylamine, i-PrOH, 50 °C, 2 hours; (iv) HCl (1M in Et2O), rt, 15 hours; (v) 106, 104a or 104b, PyBOP, DIPEA, 1,4-dioxane, rt, 15 hours; (vi) Boc2O, Et3N, 1,4-dioxane, rt, on.

From precursor 107a, a third precursor was also synthesized with a BOC protecting group on the secondary amine moiety. The reference compound was synthesized from precursor 107a and (R)‐1‐phenylethylamine using the

54

two-chamber vial discussed in papers I-II with Mo(CO)6 as the source of CO (Scheme 18). The reaction proceeded under the standard conditions for aryl iodide, producing the reference compound BSI-IV.

Scheme 18. Reagents and conditions: (i) Pd(PPh3)4, Et3N, Mo(CO)6, DBU, 1,4-dioxane, 65 °C, 15 hours, two chambers.

Radiosynthesis As a starting point, Pd(PPh3)4 was used as the precatalyst in the carbonyla-tive reaction. These conditions resulted in only a small amount of [11C]BSI-IV (2-10% RCY) as deduced by analytical radio-HPLC (Table 15; entry 1). Changing the precatalyst to dichlorobis(tri-o-tolylphosphine)-palladium(II) increased the RCY somewhat to 13±8% (entry 2, isolated after semi-preparative radio-HPLC).

Table 15. Evaluation of reaction conditions and the palladium/ligand system.

Entry Precursor Pd-source Temp (°C) TE (%) RCY (%)

1 107a Pd(PPh3)4 70-150 45±8 2-10

2 107a Pd(o-tol3)2Cl2 120 54±14 13±8

3 107a Pd(dppf)Cl2 120-140 41±23 11±6

4 107a Pd(Xantphos)Cl2 120 63±9 29±12

5 107b Pd(Xantphos)Cl2 150 55±15 7±5

6 107b Pd(PPh3)4 150 35±12 3-5

7 107c Pd(Xantphos)Cl2 120 61 27

Reagents and conditions: Precursors (10 µmol), Pd-source (10 µmol) (R)‐1‐phenylethylamine (50 µmol), THF (400 µL), 5 minutes reaction time.

Two bidentate ligands, dppf and Xantphos, were also employed in the evalu-ation. Pd(dppf)Cl2 produced [11C]BSI-IV with 11±6% isolated RCY using temperatures of 120 °C or 140 °C (entry 3). When Pd(Xantphos)Cl2 was employed at 120 °C, the decay-corrected isolated yield increased to 29±12% (entry 4). Pd(Xantphos)Cl2 and Pd(PPh3)4 were also used in the 11C carbony-lation of precursor 107b but this yielded only small amounts of the desired product (2-12%; entries 5-6). Precursor 107c was also subjected to the con-ditions in entry 4 with an isolated yield of 27% (entry 7). This was a similar yield to that of the unprotected precursor 107a and it was therefore con-cluded that the amine was not interfering with the reaction. For this reason,

55

only precursor 107a was used in the subsequent evaluation. [11C]BSI-IV was purified on a semi-preparative radio-HPLC system and after collection and evaporation of the solvents, the pure compound was formulated in phos-phate buffer (pH 7.4). The RCP was over 95% and a typical production gave approximately 0.5 GBq [11C]BSI-IV 45 minutes after EOB. The compound was identified using co-elution of the cold reference compound and LC-MS-MS analysis the day after synthesis. The MS-MS spectrum was in accor-dance with that of the cold reference compound (Figure 17).

Figure 17. Authentic MS-MS spectrum from decayed product (left) and reference spectrum (right) of BSI-IV [m/z 579 (M+H)].

In vitro and in vivo Evaluation In vitro autoradiography was performed on frozen sections of rat brain. The nonspecific binding was determined as the residual radioactivity after adding excess (10 µM) of a blocking substance (compound 7e, IC50 = 98 nM, see reference 169). The total binding was low and only a small amount was blocked (<20%), indicating a low degree of specific binding to rat brain tis-sue.

The ex vivo tissue distribution and kinetics of [11C]BSI-IV were studied using male Sprague-Dawley rats. Appropriately formulated [11C]BSI-IV was administered in the lateral tail vein and the rats were sacrificed 5, 15, 30 or 60 minutes post-injection (n=2 per time point). The organs of interest [blood, heart, lungs, liver, pancreas, spleen, adrenals, kidneys, intestines (with and without the contents), testis, and brain] were collected and the standardized uptake values (SUVs) were calculated according to Equation 2. The result is presented in Figure 18. SUV = Rad(Bq/g)InjRad(Bq)/TotBW

Equation 2. Standardized uptake value (SUV). Rad (Bq/g) is the concentration of radioactivity in the organ, InjRad (Bq) is the amount of radioactivity injected and TotBW (g) is the total body weight of the animal.

The clearance of [11C]BSI-IV from the blood was rapid; after five minutes the radioactivity in the blood (SUV) was 0.2. The uptake in all organs (ex-cept the brain) was high, particularly in organs involved in elimination. The

56

highest concentration of radioactivity was found in the intestine plus the contents, suggesting biliary clearance. The uptake was low in the brain (SUV < 0.05) signifying limited BBB penetration.

Figure 18. Ex vivo biodistribution of [11C]BSI-IV at various time points after injec-tion.

In vivo small-animal PET was performed in a PET-CT (Triumph, GE Healthcare). The animals were placed with field of view covering the whole head (centered over the brain) and [11C]BSI-IV was injected at the same time as the 60 minute PET acquisition was started (Figure 19). The radioac-tivity uptake was expressed as SUV. The data confirmed the in vitro experi-ments showing low uptake by the brain (SUV ≈ 0.1).

Figure 19. Transverse, coronal and sagittal plane from small animal PET scan (summarized 0-60 min) in a rat with [11C]BSI-IV.

In conclusion, a potent BACE-1 inhibitor was radiolabeled using [11C]CO through a palladium-mediated carbonylation reaction. The target compound [11C]BSI-IV was obtained in 29±12% decay-corrected radiochemical yield with a radiochemical purity of >99% after semi-preparative HPLC. The spe-cific radioactivity was 790±155 GBq/µmol at the EOS. Autoradiography showed a low degree of specific binding of [11C]BSI-IV in healthy rat brain.

Brain

Brain Brain

57

Ex vivo organ distribution studies in rats showed fast clearance into the small intestine and low brain uptake at all time points investigated. Small animal PET in four animals confirmed the low brain.

The brain concentration of BACE-1 has been determined to be 4.3 nM in raw rat brain tissue.170 This would give a theoretical binding potential of approximately 0.3 for [11C]BSI-IV. A higher affinity BACE-1 inhibitor would thus be required for successful in vivo PET imaging. However, in disease states such as after brain trauma (TBI) an up-regulation of BACE-1 has been found.171,172

11C and 18F Radiolabeling of Tetra- and Penta-tiophenes as PET-Ligands for Amyloid Deposits (Paper VI) Background and Aim Early detection of amyloid deposits will help in the understanding of the disease process and most importantly be a tool for the monitoring the effect of treatment. Amyloidosis is a progressive disease and early intervention is important before irreversible damage to tissue has occurred.

A non-invasive molecular imaging modality such as PET could be a use-ful way of visualizing and quantifying amyloid deposits in the brain, heart and other important organs affected by amyloid diseases. Over the last dec-ade a number of amyloid binding PET tracers for senile plaque (Aβ) in AD have been published (Figure 20).173-175

Figure 20. [18F]Florbetapir and [11C]PIB, two amyloid-β PET-ligands.

The most widely employed is Pittsburg compound B, [11C]PIB, which is referred to as the “gold standard” when it comes to visualization of Aβ in the brain. Amyloid deposits in other parts of the body have received less atten-tion; however, PIB was recently shown to also visualize amyloid deposits of transthyretin (ATTR) and amyloid light chain (AL) in the heart.176 This in-spired us to develop a new class of PET ligands based on luminescent conju-gated oligothiophenes (LCOs) (Scheme 21) targeting both localized and systemic amyloid deposits.177

LCOs have previously been evaluated in vitro by molecular imaging us-ing fluorescence and showed high affinity to Aβ deposits and were also found to penetrate the BBB.178 The aim of this project was to translate LCOs

58

into thiophene PET ligands for amyloid imaging in a major organ and to evaluate them using in vitro and in vivo methods associated with PET.

Figure 21. p-FTAA, a pentameric LCO.

The synthetic approach was to build a small library of tracer candidates with either four or five thiophene moieties radiolabeled with either 11C or 18F us-ing suitable methods such as carbonylation, cyanation or nucleophilic fluori-nation.179

Results and Discussion Synthesis of Precursors and Reference Compounds The precursor compounds 109 and 110 were synthesized by our collabora-tors at Linköping University (see paper VI, schemes 1-3 for detailed experi-mental procedures)180 from the compound 108 (Scheme 19).

Scheme 19. Tetra- and penta-thiophene precursors compromising a bromine LG.

A third precursor containing a tosyl group susceptible to nucleophilic fluori-nation was also synthesized from precursor 108 (Scheme 20).

59

Scheme 20. Pentathiophene precursor compromising a tosyl LG.

Radiolabeling Palladium-Mediated 11C-Carbonylation The precursors 109 and 110 where reacted in a palladium-mediated 11C-carbonylation to yield the pentameric thiophene [11C]TPHA or the tetramer-ic thiophene [11C]TPHC (Scheme 21). Two different hydroxyl sources were tested. Firstly, tetrabutylammonium hydroxide was used as a direct approach to form the acid. This method has previously been successful using thio-phenes.115 Secondly, a two-step reaction was employed, using methanol to form the ester, with subsequent hydrolysis to yield the acid. Despite exten-sive evaluation using different reaction parameters (solvent, temperature, Pd catalyst and additives), neither the TBAOH nor methanol produced the de-sired product in sufficient amounts from a bromine precursor.

Scheme 21. 11C-carbonylation of tetra- and pentathiophene bromine precursors.

Palladium-Mediated 11C-Cyanation The two bromine precursors were then subjected to a palladium-mediated 11C cyanation using Pd(Xantphos)Cl2 in DMF (Scheme 22).The synthesis produced [11C]TPHB (26±3% RCY) and [11C]TPHD (27±4% RCY). In a typical synthesis 0.5-1.5 GBq of 11C ligand (RCP> 95%) was obtained 50 minutes after EOB. The SA at EOS was 25-40 GBq/µmol, which corres-ponded to a substantially lower radioactivity to mass ratio than expected. This encouraged us to investigate the possibility that a pseudo-carrier was present in the collected fraction. An analytical run over 30 minutes revealed that the precursor was present in the product. Despite extensive chromato-graphy we were not able to separate the product from the precursor on a preparative scale. Disregarding the pseudo-carrier, the SA was 220-280 GBq/µmol, which also corresponded to the UV ratio of product versus the chemical impurity in the analytical radio-HPLC.

60

Scheme 22. Reagents and conditions: (i) NH4[

11C]CN, DMF, 130 °C, 5 minutes.

Nucleophilic 18F-Fluorination The tosyl precursor 111 was subjected to an [18F]F- nucleophilic fluorination using kryptofix2.2.2 as the phase transfer catalyst followed by hydrolysis of the protecting groups to produce [18F]TPHE (Scheme 23). The crude mix-ture was purified on a semi-preparative radio-HPLC providing [18F]TPHE. The RCY was 43±5% and the RCP >95%. The SA was 300-500 GBq/µmol which was within the expected range. A typical synthesis provided 1-2 GBq of [18F]TPHE 90 minutes after EOB.

Scheme 23. Reagents and conditions: (i) [18F]F-, MeCN, K2.2.2., 80 °C, 10 minutes; (ii) NaOH, 100 °C, 15 minutes.

In Vitro Binding Studies Autoradiography experiments were carried out on tissue from mouse Aβ (not shown) and human Aβ in brain tissue from patients with inherited (familial, PS1) AD and sporadic AD (Figure 22).

Figure 22. Autoradiograms showing specific and nonspecific binding of [11C]TPHD and [18F]TPHE (vide infra).

[11C]TPHD

61

Figure 22, continued.

[11C]TPHD and [18F]TPHE showed a high degree of specific binding (29-73%) to Aβ deposits when p-FTAA was used as the blocking substance. No specific binding was observed in control tissue without Aβ deposits. Surpri-singly, [11C]TPHB did not show any specific binding in the autoradiography studies.

A single autoradiography experiment in pancreas from a transgenic mouse expressing human IAPP (hIAPP) was performed (Figure 23). These mice de-velop IAPP amyloid after long term high-fat diet. Pancreatic islets of the stu-died mouse showed moderate uptake of [11C]TPHD superimposed on amyloid stained by Congo red. The congo red and [11C]TPHD staining co-localized. The binding was specific and could be blocked with excess of p-FTAA.

Figure 23. Autoradiogram showing specific binding of [11C]TPHD to two IAPP-containing islets in pancreatic tissue from mice. Congo red and hematoxylin/eosin staining (bottom left) and further magnification of one islet (bottom right).

Homogenate binding studies were performed on amyloid containing tissue from human brain (AD), heart (ATTR), spleen (AL-γ) and liver (AL-κ) (Table 16). In contrast to the results of the autoradiography, all three compounds in-cluding [11C]TPHB bound specifically to amyloid-containing tissue.

[18F]TPHE

62

Table 16. Results of homogenate binding in different amyloid-containing tissues using p-FTAA. Values are presented as percentage (%) specific binding.

Tissue [11C]TPHB [11C]TPHD [18F]TPHE

Heart, ATTR 88 99 73 Heart, control 64 42 27 Spleen, AL-λ 88 94 71 Liver, AL-κ 72 95 49 Brain, t ctx, Aβ 47 83 39

Brain, t ctx, control 64 47 36

The specific binding was determined by blocking with p-FTAA and PIB. Especially [11C]TPHD showed high specific binding to all investigated amy-loid-containing tissues. This indicates that the pseudo-carrier, the precursor, has a negligible or at least small effect on the binding.

When PIB was used as the blocking substance, a low degree of cross binding was observed. This could indicate different binding sites to the amy-loid deposits. Moreover, the Bmax of [3H]PIB181 has been determined as 10.4 pmol/mg tissue (ATTR, heart), whereas Bmax values for [11C]TPHD, [11C]TPHB and [18F]TPHE were estimated to be 270, 210 and 21 pmol/mg tissue (ATTR, heart), respectively. These values were obtained using only one concentration. However, this indicates that the thiophenes might have potential as PET ligands for visualization of amyloid deposits in vivo.

In vivo PET-CT Evaluation in Healthy Animals [11C]TPHD, [11C]TPHB and [18F]TPHE were administered to one female cynomolgus monkey on the same day and evaluated using PET-CT (Figure 24). The scans lasted 60 minutes per 11C-labeled compound, and 90 minutes for the 18F-labeled compound, with 60-90 minutes between scans. Dynamic brain scans were performed over 0-60 min followed by a static whole body scan (three 5-min scans) over a period of 60-75 min. No accumulation of radioactivity was found in organs typically affected by amyloid deposits such as the brain, heart, pancreas and spleen.

This verifies the in vitro binding assays and a low background uptake of tracer can be expected in healthy animals thus giving a high signal-to-noise in the image of the diseased state. In the liver and kidney, organs responsible for excretion of the ligands, some retention was found. The brain uptake was very low, indicating either no BBB penetration or no binding. [11C]TPHD was also found in the urinary bladder (UB), indicating partial renal excre-tion.

63

Figure 24. In vivo PET scans of [11C]TPHB, [11C]TPHD and [18F]TPHE, respec-tively. Summation images (top, 0-60 minutes) and summation of the static scan sessions (bottom, 60-75 minutes).

All three ligands showed slow blood kinetics with high blood radioactivity levels even after 60 minutes (Figure 25). The similar results were found when small animal PET was performed using healthy male SPD rats.

Figure 25. Time activity SUV curves of the uptake of the tracers in the blood.

In summary, a new class of PET-ligands for visualization of both loca-lized and systemic amyloid deposits has been described. Two oligothio-phenes were labeled with carbon-11 by through a palladium-mediated 11C-cyanation reaction using bromide as the leaving group. The decay corrected isolated yield was 23-31% with RCP >95%. One pentathiophene compound was labeled with fluorine-18 through a nucleophilic fluorination in 43±5% decay corrected yield, >95% RCP and a SA of 300-500 GBq/μmol. The in

Brain Brain

UB

Heart Heart

64

vitro evaluation showed high specific binding in tissue containing amyloid deposits in the CNS (Aβ) or in the peripheral organs (AIAPP, AL, ATTR). [11C]TPHD is the most promising ligand with respect to, blood kinetics and specific binding. In comparison to [11C]PIB, the current “golden standard” for amyloid deposits the results indicate that the thiophene derivative use a different binding site and that the number of binding sites probably is signif-icantly higher than compared to PIB binding sites. This is of importance in relation to sensitivity in imaging. However, to further evaluate these new potential amyloid-imaging PET-ligands, suitable in vivo studies using appro-priate animal disease models are required.

65

Conclusions

From the described work in this thesis the following results, conclusions and outlooks can be made:

• A two chamber reaction vial intended for small scale synthesis (0.5-2.0

mmol) was manufactured and employed in palladium(0)-catalyzed car-bonylative reactions. Carbon monoxide was readily released by molyb-denum hexacarbonyl through a ligand exchange with DBU in one cham-ber and incorporated into the palladium reaction in the other chamber. In this way, reduction-prone or otherwise sensitive functional groups could be utilized in the reactions. The workup was also simplified as no extra base or molybdenum is present in the post-reaction mixture. The reac-tions proceeded smoothly at 65 °C for aryl iodides and 85 °C for aryl bromides and may be accelerated using DMAP. From these starting ma-terials, various primary and secondary benzamides as well as N-cyanobenzamides were synthesized in high yields. This method has also been applied in the use of other nucleophiles and substrates outside this thesis.

• The carbonylative synthesis using low concentrations of [11C]CO to produce the carboxylic acid bioisostere 11C-N-Cyanobenzamide was eva-luated. The use of 1,1'-bis(diphenylphosphino)ferrocene, dppf, as the li-gand in the palladium(0)-mediated synthesis was crucial for the effective conversion of [11C]CO into product. Aryl iodides, bromides, a chloride and a triflate (120-170 °C, 5 minutes) were successfully converted into the corresponding 11C-N-Cyanobenzamides using cyanamide as the nuc-leophile. This protocol was also further applied to the synthesis of two analogs of bioactive molecules, flufenamic acid and dazoxibene which are currently being evaluated in preclinical studies.

• The SPPS of BACE-1 inhibitors containing α-phenylnorstatine, α-benzylnorstatine, iso-serine and β-alanine as transition state isosteres was described. The most potent inhibitor was crystallized in the active site of the enzyme revealing a new binding mode which may be used as a starting point for further evaluation.

66

• A potent BACE-1 inhibitor was successfully translated into the corres-ponding 11C-labeled compound through a palladium(0)-mediated carbo-nylation. The preclinical studies disclosed that the compound was cleared rapidly from the body of SPD rats. However, this compound may be useful as a PET tracer for study of BACE-1 expression following traumatic brain injury.

• Three thiophene PET Ligands (TPLs) were synthesized using

[11C]cyanide or [18F]fluoride and evaluated in vitro and in vivo for their amyloid binding properties. Specific binding was observed in vitro in brain, heart, spleen and liver tissue containing amyloid deposits. No up-take was found in healthy rats or monkey. These new PET ligands are thus promising in vivo tools for the study of amyloid deposits in different organs.

67

Acknowledgements

This work was carried out at the department of medicinal chemistry during the years 2009-2014. The first two years of my study were devoted to the synthesis of non-radioactive molecules at the division of organic pharma-ceutical chemistry. During the last years of my PhD I was introduced to the world of radiochemistry at the Platform of Preclinical PET and the Uppsala University Hospital PET centre. During all this time, I have had the pleasure of working with many devoted and skillful people without whom this thesis would not have been completed. My supervisors, Associate Professor Gunnar Antoni, Professor Mats Larhed and Associate Professor Luke Odell for all the inspiration, support, motiva-tion and fruitful discussions about PET, radiochemistry, organic chemistry and pharmaceutical chemistry, preclinical matters, palladium in general, carbonylations and life. Colleges and friends at OFK. 5th floor FTW. Ashkan, Dr Alejandro, Marc. Pd-group members, Drs Jonas×2, Fredrik, frequent "H-rör" users Anna, Dr Sanjay, Bobo, Jonas R.

Sorin for all IT related matters and Gunilla, for keeping everything running at the department. PhD students and scientists at the UU PET centre, Marc, Sara, Drs Ola, Jo-nas, Alf, Ulrika, Cyklotron-Jocke & SÅG. People at PPP for great collabora-tion, Dr Sergio, Professor Håkan, Veronica, Ram. Dr Gopal for great tutoring during my Pd-undergraduate work and Dr Fre-drik W for great introduction to BACE and to PhD studies during my first year. My roommates during the years, mostly postdocs that have visited the de-partment; Drs Mounir, Shamray, Matteu, Raju. Dr Aleh Yahorau for good collaboration with HRMS studies. Our collaborators at Linköping University, Aarhus University and UU.

68

Master and summer student Pontus, Pauline, Melad and Taraneh. I wish you all the best in your future careers. Tutors at Mälardalens Högskola for the first introduction to the wonderful world of organic chemistry, Drs Sarah, Simon, Lasse, Robert. Conference friends in US (Dr Per, Anna-karin), DK (Dr Alejandro, Dr Lin-da), GR and SUR (Marc, Sara and, Drs Gunnar, Ola, Jonas, Anna, Vladimir, Irina) and of course Smålands Nation AML Scholarship and Vinnova for making it possible to travel. I would also like to put on record, my sincere gratitude to everybody who in any way, directly and indirectly, has lent their helping hand in the work and creation of this thesis. Familjen i Eskilstuna, Pappa, Mamma, Bror och Syster! Tack för allt stöd under alla år. Elin & Alba, jag älskar er av hela mitt hjärta.

69

References

1. Negishi, E.; Handbook of Organopalladium Chemistry for Organic Synthesis. Wiley-Interscience New York, 2002; Vol. 1, Chapter 1.

2. Smidt, J.; Hafner, W.; Jira, R.; Sieber, R.; Sedlmeier, J.; Sabel, A., Angew. Chem. Int. Ed. 1962, 1, 80-88.

3. Heck, R. F.; Nolley, J. P., J. Org. Chem. 1972, 37, 2320-2322. 4. Mizoroki, T.; Mori, K.; Ozaki, A., B. Chem. Soc. Jpn. 1971, 44,

581-581. 5. Yamamura, M.; Moritani, I.; Murahashi, S.-I., J. Organomet.

Chem. 1975, 91, 39-42. 6. Knappke, C. E. I.; Jacobi von Wangelin, A., Chem. Soc. Rev. 2011,

40, 4948-4962. 7. Sonogashira, K.; Tohda, Y.; Hagihara, N., Tetrahedron Lett. 1975,

16, 4467-4470. 8. Chinchilla, R.; Najera, C., Chem. Soc. Rev. 2011, 40, 5084-5121. 9. Negishi, E.; King, A. O.; Okukado, N., J. Org. Chem. 1977, 42,

1821-1823. 10. Jana, R.; Pathak, T. P.; Sigman, M. S., Chem. Rev. 2011, 111,

1417-1492. 11. Milstein, D.; Stille, J. K., J. Am. Chem. Soc. 1979, 101, 4992-4998. 12. Espinet, P.; Echavarren, A. M., Angew. Chem. Int. Ed. 2004, 43,

4704-4734. 13. Miyaura, N.; Yamada, K.; Suzuki, A., Tetrahedron Lett. 1979, 20,

3437-3440. 14. Hiyama, T.; Hatanaka, Y., Pure Appl. Chem. 1994, 66, 1471-1478. 15. Johansson-Seechurn, C. C. C.; Kitching, M. O.; Colacot, T. J.;

Snieckus, V., Angew. Chem. Int. Ed. 2012, 51, 5062-5085. 16. Seligson, A. L.; Trogler, W. C., J. Am. Chem. Soc. 1991, 113,

2520-2527. 17. Tolman, C. A.; Seidel, W. C.; Gerlach, D. H., J. Am. Chem. Soc.

1972, 94, 2669-2676. 18. Tolman, C. A., Chem. Rev. 1977, 77, 313-348. 19. Birkholz, M.-N.; Freixa, Z.; van Leeuwen, P. W. N. M., Chem.

Soc. Rev. 2009, 38, 1099-1118. 20. van Leeuwen, P. W. N. M.; Kamer, P. C. J.; Reek, J. N. H.;

Dierkes, P., Chem. Rev. 2000, 100, 2741-2770. 21. Beletskaya, I. P.; Cheprakov, A. V., Chem. Rev. 2000, 100, 3009-

3066.

70

22. Li, J.; Hua, R.; Liu, T., J. Org. Chem. 2010, 75, 2966-2970. 23. Seayad, A.; Jayasree, S.; Chaudhari, R. V., Org. Lett. 1999, 1, 459-

462. 24. Heck, R. F., J. Am. Chem. Soc. 1968, 90, 5546-5548. 25. Schoenberg, A.; Bartoletti, I.; Heck, R. F., J. Org. Chem. 1974, 39,

3318-3326. 26. Schoenberg, A.; Heck, R. F., J. Org. Chem. 1974, 39, 3327-3331. 27. Crabtree, R. H., The Organometallic Chemistry of the Transition

Metals. John Wiley & Sons, 2009. 28. Brennführer, A.; Neumann, H.; Beller, M., Angew. Chem. Int. Ed.

2009, 48, 4114-4133. 29. Wu, X.; Neumann, H.; Beller, M., Chem. Soc. Rev. 2011, 40, 4986-

5009. 30. Korsager, S.; Taaning, R. H.; Lindhardt, A. T.; Skrydstrup, T., J.

Org. Chem. 2013, 78, 6112-6120. 31. Cao, H.; McNamee, L.; Alper, H., J. Org. Chem. 2008, 73, 3530-

3534. 32. Munday, R. H.; Martinelli, J. R.; Buchwald, S. L., J. Am. Chem.

Soc. 2008, 130, 2754-2755. 33. Wu, X.-F.; Neumann, H.; Beller, M., Chem-Eur J. 2012, 18, 3831-

3834. 34. Wu, X.-F.; Neumann, H.; Beller, M., Chem-Asian J. 2012, 7, 282-

285. 35. Grigg, R.; Mutton, S., Tetrahedron 2010, 66, 5515-5548. 36. Guan, Z.-H.; Chen, M.; Ren, Z.-H., J. Am. Chem. Soc. 2012, 134,

17490-17493. 37. Rajeshkumar, V.; Lee, T.-H.; Chuang, S.-C., Org. Lett. 2013, 15,

1468-1471. 38. Sigma-Aldrich Phosphine Ligand Application Guide,

www.sigmaaldrich.com, (accessed December 19, 2013). 39. Wu, X.-F.; Schranck, J.; Neumann, H.; Beller, M., ChemCatChem

2012, 4, 69-71. 40. Kumar, K.; et. al, Org. Lett. 2003, 6, 7-10. 41. Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Reek, J. N. H., Acc.

Chem. Res. 2001, 34, 895-904. 42. Wu, X.-F.; Zhang, M.; Jiao, H.; Neumann, H.; Beller, M., Asian J.

Chem. 2013, 2, 135-139. 43. Wu, X.-F.; Neumann, H.; Beller, M., Chem-Eur. J. 2012, 18, 419-

422. 44. Martinelli, J.; Watson, D.; Freckmann, D.; Barder, T.; Buchwald,

S., J. Org. Chem. 2008, 73, 7102-7107. 45. Ozawa, F.; Kubo, A.; Hayashi, T., Chem. Lett. 1992, 21, 2177-

2180. 46. Amatore, C.; Jutand, A.; M'Barki, M. A., Organometallics 1992,

11, 3009-3013.

71

47. Strem Chemicals Inc., The Strem Chemiker, 2014, Vol. XXVII. 48. Whitcombe, N. J.; Hii, K. K.; Gibson, S. E., Tetrahedron 2001, 57,

7449-7476. 49. Jutand, A.; Mosleh, A., Organometallics 1995, 14, 1810-1817. 50. Portnoy, M.; Milstein, D., Organometallics 1993, 12, 1665-1673. 51. Barrios-Landeros, F.; Hartwig, J. F., J. Am. Chem. Soc. 2005, 127,

6944-6945. 52. Garrou, P. E.; Heck, R. F., J. Am. Chem. Soc. 1976, 98, 4115-4127. 53. Martinelli, J. R.; Clark, T. P.; Watson, D. A.; Munday, R. H.;

Buchwald, S. L., Angew. Chem. Int. Ed. 2007, 46, 8460-8463. 54. Odell, L. R.; Sävmarker, J.; Larhed, M., Tetrahedron Lett. 2008,

49, 6115-6118. 55. Wiȩckowska, A.; Fransson, R.; Odell, L. R.; Larhed, M., J. Org.

Chem. 2011, 76, 978-981. 56. Wannberg, J.; Larhed, M., J. Org. Chem. 2003, 68, 5750-5753. 57. Hartwig, J. F.; Richards, S.; Barañano, D.; Paul, F., J. Am. Chem.

Soc. 1996, 118, 3626-3633. 58. Paulik, F. E.; Roth, J. F., Chem. Commun. 1968, 24, 1578. 59. Spencer, M. S., Top. Catal. 1999, 8, 259-266. 60. Ueda, T.; Konishi, H.; Manabe, K., Org. Lett. 2012, 14, 3100-

3103. 61. Schareina, T.; Zapf, A.; Cotté, A.; Gotta, M.; Beller, M., Adv.

Synth. Catal. 2010, 352, 1205-1209. 62. Morimoto, T.; Kakiuchi, K., Angew. Chem. Int. Ed. 2004, 43,

5580-5588. 63. Shang, R.; Fu, Y.; Li, J.-B.; Zhang, S.-L.; Guo, Q.-X.; Liu, L., J.

Am. Chem. Soc. 2009, 131, 5738-5739. 64. Cunico, R. F.; Maity, B. C., Org. Lett. 2003, 5, 4947-4949. 65. Ueda, T.; Konishi, H.; Manabe, K., Org. Lett. 2012, 14, 5370-

5373. 66. Wan, Y.; Alterman, M.; Larhed, M.; Hallberg, A., J. Org. Chem.

2002, 67, 6232-6235. 67. Sawant, D. N.; Wagh, Y. S.; Bhatte, K. D.; Bhanage, B. M., J. Org.

Chem. 2011, 76, 5489-5494. 68. Hermange, P.; Lindhardt, A.; Taaning, R.; Bjerglund, K.; Lupp, D.;

Skrydstrup, T., J. Am. Chem. Soc. 2011, 133, 6061-6071. 69. Hermange, P.; Gogsig, T.; Lindhardt, A.; Taaning, R.; Skrydstrup,

T., Org. Lett. 2011, 13, 2444-2447. 70. Friis, S.; Taaning, R.; Lindhardt, A.; Skrydstrup, T., J. Am. Chem.

Soc. 2011, 133, 18114-18117. 71. Friis, S. D.; Andersen, T. L.; Skrydstrup, T., Org. Lett. 2013, 15,

1378-1381. 72. Nielsen, D. U.; Neumann, K.; Taaning, R. H.; Lindhardt, A. T.;

Modvig, A.; Skrydstrup, T., J. Org. Chem. 2012, 77, 6155-6165.

72

73. Lindhardt, A. T. et al., J. Labelled. Compd. Rad. 2012, 55, 411-418.

74. Odell, L.; Russo, F.; Larhed, M., Synlett 2012, 5, 685-698. 75. Borhade, S. R.; Sandström, A.; Arvidsson, P. I., Org. Lett. 2013,

15, 1056-1059. 76. Skogh, A.; Fransson, R.; Sköld, C.; Larhed, M.; Sandström, A., J.

Org. Chem. 2013, 78, 12251-12256. 77. Motwani, H. V.; Larhed, M., Eur. J. Org. Chem. 2013, 22, 4729-

4733. 78. The Nobel Organizations. The Noble Prize in Chemistry 1943,

www.nobelprize.org (Accessed December 19, 2013). 79. IUPAC, Tracer, www.iupac.org (Accessed December 19, 2013). 80. Bailey, D. L.; Townsend, D. W.; Valk, P. E.; Maisey, M. N.,

Positron Emission Tomography: Basic Sciences. Springer, 2005. 81. Anderson, C., Science 1932, 76, 238-239. 82. Anderson, C. D., Phys. Rev. 1933, 43, 491-494. 83. Lawrence, E. O., Method and Apparatus for the Acceraton of Ions,

Patent, 1,948,384, 1934. 84. Brownell, G. L., Physics Research Laboratory, Massachusetts

General Hospital, 1999. 85. Fowler, J. S.; Ido, T., Semin. Nucl. Med. 2002, 32, 6-12. 86. IAEA, Cyclotron Produced Radionuclides: Principles and

Practice, Vienna, 2008, Vol. Technical Report Series no. 465. 87. von Schulthess, G. K.; Steinert, H. C.; Hany, T. F., Radiology

2006, 238, 405-422. 88. Judenhofer, M. S. et al., Nat. Med. 2008, 14, 459-465. 89. Kaufmann, P. A.; Camici, P. G., J. Nucl. Med. 2005, 46, 75-88. 90. Mintun, M. A.; Raichle, M. E.; Kilbourn, M. R.; Wooten, G. F.;

Welch, M. J., Ann. Neurol. 1984, 15, 217-227. 91. Zhang, Y.; Fox, G. B., J. Biomed. Res. 2012, 26, 69-76. 92. Syvänen, S.; Eriksson, J., ACS Chem. Neurosci. 2012, 4, 225-237. 93. Sandell, J. et al., J. Labelled. Compd. Rad. 2000, 43, 331-338. 94. Hicks, R. J.; Dorow, D.; Roselt, P., Cancer Imag. 2006, 6, S102. 95. Bergström, M.; Grahnen, A.; Långström, B., Eur. J. Pharmacol.

2003, 59, 357-366. 96. Wagner, C. C.; Langer, O., Adv. Drug. Deliver. Rev. 2011, 63,

539-546. 97. Bergström, M.; Långström, B., Pharmacokinetic studies with PET.

In Imaging in Drug Discovery and Early Clinical Trials, Springer: 2005; pp 279-317.

98. European Medicines Agency. ICH guideline M3(R2) on non-clinical safety studies for the conduct of human clinical trials and marketing authorisation for pharmaceuticals. EMA/CPMP/ICH/ 286/1995

73

99. Långström, B.; Itsenko, O.; Rahman, O., J. Labelled. Compd. Rad. 2007, 50, 794-810.

100. Iwata, R.; Ido, T.; Takahashi, T.; Nakanishi, H.; Iida, S., Int. J. Rad. Appl. Inst. Part A. App. Rad. Isot. 1987, 38, 97-102.

101. Scott, P., Angew. Chem. Int. Ed. 2009, 48, 6001-6004. 102. Lidström, P.; Kihlberg, T.; Långström, B., J. Chem. Soc. Perk. T.

1. 1997, 18, 2701-2706. 103. Kealey, S.; Miller, P. W.; Long, N. J.; Plisson, C.; Martarello, L.;

Gee, A. D., Chem. Commun. 2009, 25, 3696-3698. 104. Kealey, S.; Husbands, S. M.; Bennacef, I.; Gee, A. D.; Passchier,

J., J. Labelled. Compd. Rad. 2013, (doi: 10.1002/jlcr.3151) 105. Audrain, H.; Martarello, L.; Gee, A.; Bender, D., Chem. Commun.

2004, 5, 558-559. 106. Eriksson, J.; van den Hoek, J.; Windhorst, A. D., J. Labelled.

Compd. Rad. 2012, 55, 223-228. 107. Miller, P. W. et al., Angew. Chem. Int. Ed. 2007, 46, 2875-2878. 108. Miller, P. W.; Audrain, H.; Bender, D.; deMello, A. J.; Gee, A. D.;

Long, N. J.; Vilar, R., Chem-Eur J. 2011, 17, 460-463. 109. Kihlberg, T.; Långström, B. Method and Apparatus for Production

and use of [11C]carbon monoxide in Labelling Synthesis. Patent, PCT/SE02/01222, 2002.

110. Barletta, J.; Karimi, F.; Långström, B., J. Labelled. Compd. Rad. 2006, 49, 429-436.

111. Kihlberg, T.; Karimi, F.; Långström, B., J. Org. Chem. 2002, 67, 3687-3692.

112. Åberg, O.; Långström, B., J. Labelled. Compd. Rad. 2012, 55, 477-483.

113. Åberg, O. et al., J. Labelled. Compd. Rad. 2010, 53, 616-624. 114. Karimi, F.; Långström, B., Org. Biomol. Chem. 2003, 1, 541-546. 115. Karimi, F.; Långström, B., J. Chem. Soc. Perk. T. 1. 2002, 20,

2256-2259. 116. Takashima-Hirano, M.; Ishii, H.; Suzuki, M., ACS Med. Chem.

Lett. 2012, 3, 804-807. 117. Dahl, K.; Schou, M.; Amini, N.; Halldin, C., Eur. J Org. Chem.

2013, 7, 1228-1231. 118. Wannberg, J. et al., J. Comb. Chem. 2005, 7, 611-617. 119. Armarego, W. L. F.; Chai, C., Purification of Laboratory

Chemicals, Butterworth-Heinemann: 2009. 120. Lagerlund, O.; Larhed, M., J. Comb. Chem. 2006, 8, 4-6. 121. Lagerlund, O.; Mantel, M.; Larhed, M., Tetrahedron 2009, 65,

7646-7652. 122. Kuethe, J. T.; Davies, I. W., Tetrahedron 2006, 62, 11381-11390. 123. Söderberg, B. C. G.; Banini, S. R.; Turner, M. R.; Minter, A. R.;

Arrington, A. K., Synthesis 2008, 2008, 903-912.

74

124. Gabriele, B.; Salerno, G.; Mancuso, R.; Costa, M., J. Org. Chem. 2004, 69, 4741-4750.

125. Rönn, R. et al., Bioorg. Med. Chem. 2007, 15, 4057-4068. 126. Maestri, G.; Larraufie, M.-H.; Ollivier, C.; Malacria, M.;

Fensterbank, L.; Lacôte, E., Org. Lett. 2012, 14, 5538-5541. 127. Vieira, E.; Huwyler, J.; Jolidon, S.; Knoflach, F.; Mutel, V.;

Wichmann, J., Bioorg. Med. Chem. Lett. 2005, 15, 4628-4631. 128. Hase, C. Process for the preparation of alkali metal salts of Acyl

cyanamides. Patent US4595782, 1984. 129. Xiao, Z.; Yang, M. G.; Tebben, A. J.; Galella, M. A.; Weinstein,

D. S., Tetrahedron Lett. 2010, 51, 5843-5844. 130. Ballatore, C.; Huryn, D.; Smith, A., ChemMedChem 2013, 8, 385-

395. 131. Guram, A. S.; Buchwald, S. L., J Am. Chem. Soc. 1994, 116, 7901-

7902. 132. Paul, F.; Patt, J.; Hartwig, J. F., J. Am. Chem. Soc. 1994, 116,

5969-5970. 133. Merlini, G.; Bellotti, V., New Engl. J. Med. 2003, 349, 583-596. 134. Eisenberg, D.; Jucker, M., Cell 2012, 148, 1188-1203. 135. Merlini, G.; Westermark, P., J. Intern. Med. 2004, 255, 159-178. 136. Picken M, M.; Dogan, A.; Guillermo, H. A., Amyloid and Related

Disorders, 2012. 137. Sipe, J. D. et al., Amyloid 2012, 19, 167-170. 138. Westermark, P.; Wernstedt, C.; Wilander, E.; Sletten, K., Biochem.

Bioph. Res. Co. 1986, 140, 827-831. 139. Kyle, R. A. et al., Blood 1992, 79, 1817-1822. 140. Glenner, G.; Terry, W.; Harada, M.; Isensky, C.; Page, D., Science

1971, 172, 1150. 141. Costa, P.; Figurera, A.; Bravo, F., Proceedings of the National

Academy of Sciences of the United States of America 1978, 75, 4499-4503.

142. Sanchorawala, V., Clin. J. Am. Soc. Nephrol. 2006, 1, 1331-1341. 143. Wiseman, R. L.; Powers, E. T.; Kelly, J. W., Biochemistry 2005,

44, 16612-16623. 144. Johnson, S. M.; Wiseman, R. L.; Sekijima, Y.; Green, N. S.;

Adamski-Werner, S. L.; Kelly, J. W., Acc. Chem. Res. 2005, 38, 911-921.

145. Stelzmann, R. A.; Norman Schnitzlein, H.; Reed Murtagh, F., Clin. Anat. 1995, 8, 429-431.

146. Möller, H. J.; Graeber, M. B., Eur. Arch. Psyc. Clin. Neurosci. 1998, 248, 111-122.

147. Alzheimer, A., History of Psychiatry 1991, 2, 74-101. 148. Batsch, N. L.; Mittelman, M. S. World Alzheimer Report 2012,

Overcoming the Stigma of Dementia; 2012.

75

149. World Health Organization, Dementia: a public health priority. UK, 2012.

150. Jellinger, K. A., J. Alzheimers Dis. 2006, 9, 61-70. 151. Selkoe, D., Science 2012, 337, 1488-1492. 152. Haass, C.; Selkoe, D., Nat. Rev. Mol. Cell Bio. 2007, 8, 101-112. 153. De Strooper, B.; Vassar, R.; Golde, T., Nat. Rev. Neurol. 2010, 6,

99-107. 154. Huff, J. R., J. Med. Chem. 1991, 34, 2305-2314. 155. Ghosh, A. K.; Shin, D.; Koelsch, G.; Lin, X.; Ermolieff, J.; Tang,

J., J. Am. Chem. Soc. 2000, 122, 3522-3523. 156. Kimura, T. et al., Bioorg. Med. Chem. Lett. 2005, 1, 211-215. 157. Stachel, S. J. et al., J. Med. Chem. 2004, 47, 6447-6450. 158. Silvestri, R., Med. Res. Rev. 2009, 29, 295-338. 159. Ekegren, J. K.; Unge, T.; Safa, M. Z.; Wallberg, H.; Samuelsson,

B.; Hallberg, A., J. Med. Chem. 2005, 48, 8098-8102. 160. Ekegren, J. K. et al., J. Med. Chem. 2006, 49, 1828-1832. 161. Russo, F.; Wångsell, F.; Sävmarker, J.; Jacobsson, M.; Larhed, M.,

Tetrahedron 2009, 65, 10047-10059. 162. Wångsell, F.; Russo, F.; Sävmarker, J.; Rosenquist, Å.;

Samuelsson, B.; Larhed, M., Bioorg. Med. Chem. Lett. 2009, 19, 4711-4714.

163. Hamada, Y. et al., Bioorg. Med. Chem. Lett. 2008, 18, 1649-1653. 164. Merrifield, R. B., J. Am. Chem. Soc. 1963, 85, 2149-2154. 165. Lloyd-Williams, P.; Albericio, F.; Giralt, E., Tetrahedron 1993, 49,

11065-11133. 166. Barlos, K. et al., Tetrahedron Lett. 1989, 30, 3943-3946. 167. Barlos, K.; Chatzi, O.; Gatos, D.; Stavropoulos, G., Int. J Pept.

Prot. Res. 1991, 37, 513-520. 168. Lang, H.; Huang, X.; Yang, Y., J. Alzheimer Dis. 2012, 29, 351-

359. 169. Stachel, S.; et al., Bioorg. Med. Chem. Lett. 2006, 16, 641-644. 170. Honer, M.; et al., A novel radiotracer for in vitro and in vivo

visualization of BACE1 in the rodent and baboon brain, XIth International Conference on Quantification of Brain Function with PET. Shanghai, China, 2013.

171. Blasko, I. et al., J. Neur. Trans. 2004, 111, 523-536. 172. Walker, K. R.; Kang, E. L.; Whalen, M. J.; Shen, Y.; Tesco, G., J.

Neurosci. 2012, 32, 10423-10437. 173. Svedberg, M. M.; Rahman, O.; Hall, H., Nuc. Med. Biol. 2012, 39,

484-501. 174. Wong, D. et al. J. Nucl. Med. 2010, 51, 913-920. 175. Klunk, W. et al., Ann. Neurol. 2004, 55, 306-319. 176. Antoni, G. et al., J. Nucl. Med. 2013, 54, 213-220. 177. Åslund, A. et al., ACS Chem. Biol. 2009, 4, 673-684. 178. Klingstedt, T. et al., Org. Biomol. Chem. 2011, 9, 8356-8370.

76

179. Miller, P. W.; Long, N. J.; Vilar, R.; Gee, A. D., Angew. Chem. Int. Ed. 2008, 47, 8998-9033.

180. Åslund, A. et al., Bioconjugate Chem. 2007, 18, 1860-1868. 181. Hellström-Lindahl, E.; Westermark, P.; Antoni, G.; Estrada, S.,

Amyloid 2013 (doi: 10.3109/13506129.2013.860895)

Acta Universitatis UpsaliensisDigital Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Pharmacy 183

Editor: The Dean of the Faculty of Pharmacy

A doctoral dissertation from the Faculty of Pharmacy, UppsalaUniversity, is usually a summary of a number of papers. A fewcopies of the complete dissertation are kept at major Swedishresearch libraries, while the summary alone is distributedinternationally through the series Digital ComprehensiveSummaries of Uppsala Dissertations from the Faculty ofPharmacy.

Distribution: publications.uu.seurn:nbn:se:uu:diva-213863

ACTAUNIVERSITATIS

UPSALIENSISUPPSALA

2014