_____________________________ _____________________________
Palladium-Promoted Synthesis of
Compounds Labelled with
11
C
Synthesis of
11C-Labelled Prostacyclin and Prostaglandin
Analogues
BY
MARGARETA BJÖRKMAN
ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2000
ABSTRACT
Björkman, M. Palladium-Promoted Synthesis of Compounds Labelled with 11C. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the
Faculty of Science and Technology 564. 39 pp. Uppsala. ISBN 91-554-4797-X.
Palladium-promoted reactions have been employed for the synthesis of compounds labelled with 11C (t½ = 20.3 min). The precursor [11C]methyl iodide was used in
palladium-promoted cross-coupling reactions with organostannanes. With this method, large molecules with several functional groups, that is prostacyclin analogues, have been synthesised in up to 54 % decay-corrected radiochemical yield, calculated from [11C]methyl iodide. However, since this method did not afford reproducible yields, a second method where copper(I) was used as a co-catalyst with palladium, was developed. In the second method, a lower reaction temperature could be used and more reproducible yields were obtained. Employing this method, a prostaglandin analogue was synthesised in 34 % decay-corrected radiochemical yield calculated from [11C]methyl iodide. The total synthesis time was 30 min and the radiochemical purity was higher than 95 %. The specific radioactivity of the compounds obtained with these two methods was approximately 100 GBq/µmol.
11C-Labelled aliphatic and aromatic alkenes were synthesised from [11C]methyl
iodide in a Wittig olefination reaction using a published method. The 11C-labelled alkenes were reacted with five aromatic halides in Heck coupling reactions, producing five [11C]stilbene analogues in 34-40 % decay-corrected radiochemical yield. The radio-chemical purity was higher than 95 % and the total synthesis time was 40 min.
11C-Labelled alkenes were also synthesised from 11C-labelled aldehydes. The 11
C-labelled aldehydes were obtained from [11C]carbon monoxide in a palladium-mediated formylation of aryl iodides in 51-87 % radiochemical yield, determined by analytical LC and corrected for trapping efficiency. A range of palladium catalysts and hydride reagents were investigated. The labelled aldehydes were used in a subsequent Wittig olefination reaction where various Wittig salts were employed to synthesise a variety of alkenes. The radiochemical yields were 30-76 %, determined by analytical LC.
Margareta Björkman, Department of Organic Chemistry, Institute of Chemistry, Uppsala University, P.O. Box 531, SE-751 21 Uppsala Sweden.
Margareta Björkman 2000 ISSN 1104-232X
ISBN 91-554-4797-X
This thesis is based on the following papers, referred to in the text by their Roman numerals:
I Synthesis of 11C/13C-Labelled Prostacyclins
Margareta Björkman, Yvonne Andersson, Hisashi Doi, Koichi Kato, Masaaki Suzuki, Ryoji Noyori, Yasuyoshi Watanabe and Bengt Långström
Acta Chem. Scand., 52, (1998), 635-640.
II Rapid Coupling of Methyl Iodide with Aryltributylstannanes Mediated by Palladium(0) Complexes: A General Protocol for the Synthesis of 11CH3
-Labeled PET Tracers
Masaaki Suzuki, Hisashi Doi, Margareta Björkman, Yvonne Andersson, Bengt Långström, Yasuyoshi Watanabe and Ryoji Noyori
Chem. Eur. J., 3, (1997), 2039-2042.
III Synthesis of a 11C-Labelled Prostaglandin F2αααα Analogue Using an Improved
Method for Stille Reactions with [11C]Methyl iodide
Margareta Björkman, Hisashi Doi, Bahram Resul, Masaaki Suzuki, Ryoji Noyori, Yasuyoshi Watanabe and Bengt Långström
J. Labelled Compd. Radiopharm., (2000), Accepted.
IV Functionalisation of 11C-Labelled Olefins via a Heck Coupling reaction Margareta Björkman and Bengt Långström
J. Chem. Soc., Perkin Trans. 1, (2000), In Press.
V [11C]Carbon Monoxide in a Palladium-Promoted Formylation Reaction of Aryl Halides Followed by a Wittig Reaction
Margareta Björkman, Tor Kihlberg and Bengt Långström
Preliminary manuscript
Reprints were made with permission from the publishers.
I have been responsible for planning and carrying out all the experimental work, analysis of the results and writing of papers I, III-V. Paper II was written by M. Suzuki and I took part in experiments, analysis and discussions in collaboration with H. Doi, who performed most of the experimental work.
ABSTRACT 2
PAPERS INCLUDED IN THE THESIS 4
LIST OF ABBREVIATIONS 6
1. INTRODUCTION 7
2. SHORT-LIVED RADIONUCLIDES IN LABELLING SYNTHESIS 8
Biological aspects of tracer development Synthetic aspects of tracer development
3. PALLADIUM-PROMOTED CARBON-CARBON 12
BOND FORMATIONS Stille reactions
Carbonylation reactions Heck reactions
Palladium-promoted [11C]carbon-carbon bond formations
4. STILLE REACTIONS WITH [11C]METHYL IODIDE 17
Synthesis of a 11C-labelled prostacyclin analogue Modifications of the Stille reaction with methyl iodide Synthesis of a 11C-labelled prostaglandin F2α analogue
5. HECK REACTIONS WITH 11C-LABELLED OLEFINS 23
Functionalisation of 11C-labelled olefins via a Heck coupling reaction
6. CARBONYLATION REACTIONS WITH [11C]CARBON MONOXIDE 27
Palladium-promoted formylation reaction of aryl halides with [11C]carbon monoxide
7. APPLICATIONS TO POSITRON EMISSION TOMOGRAPHY 31
A 11C-labelled prostacyclin analogue in PET
8. CONCLUSIONS 34
ACKNOWLEDGEMENTS 35
LIST OF ABBREVIATIONS AsPh3 Triphenylarsine Bu3SnH Tributylstannane DMF N, N-Dimethylformamide Et3SiH Triethylsilane GC Gas chromatography LC Liquid chromatography
LC-MS Liquid chromatography-mass spectrometry
LiAlH4 Lithium aluminum hydride
MeCN Acetonitrile
NMP N-Methyl-2-pyrrolidinone
NMR Nuclear magnetic resonance
o-DCB o-Dichlorobenzene
Pd2(dba)3 Tris(dibenzylideneacetone)dipalladium(0)
Pd(OAc)2 Palladium(II) acetate
Pd(PPh3)4 Tetrakis(triphenylphosphine)palladium(0)
Pd[P(o-Tol)3]2 Bis(tri(o-tolyl)phosphine)palladium(0)
PET Positron emission tomography
P(o-Tol)3 Tri(o-tolyl)phosphine
PMHS Poly(methylhydrosiloxane)
SPE Solid phase extraction
1. INTRODUCTION
Ideally, the behaviour of a tracer should reflect a biological event by participating in the system studied without perturbing it. The methodology of using isotopic modification to achieve this was first presented by Hevesy in 1923, when he studied the uptake of 212Pb in horse-bean plants.1 Compounds labelled with radioactive isotopes have been
especially attractive as tracers since there are methods to detect and quantify
radioactivity with high precision. The nuclides 3H and 14C have been extensively used in the natural sciences but the radiation emitted from these radionuclides (β-) has a short
range, requiring the sacrifice of the experimental animal to allow evaluation of the uptake and distribution of the tracer. In order to follow the fate of a radiotracer in vivo, the radiation emitted from the tracer must penetrate the body to be detected from the outside.
Positron emission tomography (PET) is a non-invasive imaging technique that permits external monitoring of radioactive compounds in vivo.2 In a PET investigation, a
compound labelled with a positron emitting radionuclide is introduced into the body. These radionuclides are neutron deficient and decay by conversion of a proton to a neutron and emission of a positron. The positron will travel no further than a few millimetres in tissue before it combines with its antimatter equivalent, the electron. During this subsequent annihilation process, two high energy gamma photons (511 keV) are emitted in opposite directions, escaping the body. When two photons are detected in coincidence by detectors on opposite sides of the PET scanner, an event is registered. A large number of events are collected and reconstructed into images representing the spatial distribution of the radioactive source in the body.
Positron emitting nuclides most frequently used in combination with PET are 15O, 13N
11C and 18F, with half-lives of 2.07, 9.97, 20.3 and 110 min respectively. There are
several reasons why these radionuclides are especially useful in PET:
• The short half-life of the radionuclide makes it possible to minimise the radiation exposure to patients and to perform repetitive studies with short time intervals. • Since most biomolecules consist of one or several of these elements, radiolabelled
molecules can be synthesised that are chemically and biologically indistinguishable from their stable counterparts, with the exception of a small kinetic isotope effect.3
• Compounds labelled with these radionuclides can be obtained in high specific radioactivity, defined as radioactivity per unit mass. Therefore, the risk for
disturbing the system studied, e.g. by saturation of receptors or by causing unwanted pharmacological effects, can be minimised.
A number of biochemical processes have been studied with PET, such as energy metabolism, blood flow and neurotransmission,4, 5 but the utility of PET has not only
been demonstrated in biomedical research but also in clinical practice.2 PET has also
become an interesting tool in drug development, since the technique offers a possibility to study pharmacokinetic and pharmacodynamic processes in vivo. Due to the expansion of this area there has been an increasing demand for new labelling methods.
In this thesis, the development of synthetic methods for labelling with 11C is discussed and in particular, the application of palladium-promoted reactions to rapid labelling synthesis. The 11C-labelling of a prostacyclin and a prostaglandin analogue is presented, (paper I and III) and the labelling method employed for the synthesis of these
compounds was examined in detail using stable methyl iodide (paper II). Model compounds were synthesised with the purpose of exploring new synthetic possibilities based on palladium-mediated reactions. Thus, the Heck reaction was investigated in order to functionalise 11C-labelled olefins synthesised from [11C]methyl iodide in a Wittig reaction (paper IV). Finally, [11C]carbon monoxide was used in a palladium-mediated formylation reaction of aromatic halides and the resulting 11C-labelled aldehydes were used in a Wittig reaction (paper V).
2. SHORT-LIVED RADIONUCLIDES IN LABELLING SYNTHESIS Biological aspects of tracer development
A radiotracer should be designed in such a way that it describes the system of interest without disturbing the biological process. In order to accomplish this several
requirements should be fulfilled:
• The selected tracer should have the appropriate properties to reach the organ of interest and participate in the process studied. For example, penetrate the blood-brain barrier and bind selectively to a certain receptor in blood-brain.
• The tracer should be labelled with a radionuclide that has a physiological half-life that matches the biological process of interest. If the half-life is too short, little information will be gained, whereas if the half-life is too long, consecutive studies are inconvenient to perform. Another aspect regarding the choice of radionuclide is that an analogue might have different biological properties than the original
compound, even though a very small part of the molecule has been altered, e.g. exchange of H or OH for 18F.
• It is necessary to account for metabolism of the radiotracer since the PET image does not provide any structural information of the compound that gives rise to the image. If metabolism of the tracer is rapid, the label should be positioned where it is not removed from the tracer in an early stage of metabolism, since limited information will be gained from such an investigation.
• Specific radioactivity can be of importance depending on the biological target. In investigations where labelled endogenous compounds are used as tracers, the amount of tracer is usually not important for the study. However, when
neurotransmitter receptors are studied, the specific radioactivity can be crucial and in order to avoid saturation or pharmacological effects, it should be as high as possible. The maximum theoretical specific radioactivity of 11C is 3.4 x 105 GBq/µmol, but in practice, the specific radioactivity of 11C is in the range of
100-500 GBq/µmol. This is due to contamination with stable carbon isotopes which dilute the radioisotope during radionuclide production and labelling synthesis.
Synthetic aspects of tracer development
In general, macro-scale synthesis has to be modified before it can be applied to labelling synthesis with short-lived radionuclides.6, 7 Primarily, this is due to the very small
amount of labelled precursor† used in the synthesis and the short half-life of the radionuclides.
Labelled precursors
Because of the short half-life, each radiotracer synthesis begins with production of the radionuclide. 11C is generally obtained via the 14N(p,α)11C nuclear reaction. This
reaction is performed in a target, filled with nitrogen gas (99.5 %) and oxygen (0.05%), which is irradiated with high energy protons produced in an accelerator. [11C]Carbon dioxide is the primary labelled precursor and it is formed during nuclide production when 11C atoms react with traces of oxygen in the target. The [11C]carbon dioxide can be converted into a number of secondary labelled precursors by simple on-line or batch procedures, Figure 1. In this thesis, the labelled precursors [11C]methyl iodide and [11C]carbon monoxide were employed in palladium-promoted reactions. [11C]Methyl iodide may be obtained from [11C]carbon dioxide via a reduction reaction with lithium aluminum hydride and subsequent substitution reaction with hydriodic acid, Figure 1 (1). [11C]Methyl iodide is one of the most commonly used 11C-labelled precursor and it is not only used as a precursor in different labelling reactions, e.g. alkylations on heteroatoms,18 but also for synthesis of other 11C-labelled intermediates, such as
[11C]nitromethane,12 [11C]methyllithium19 and [11C]methyltriphenylphosphoranes.20
[11C]Carbon monoxide can be obtained from [11C]carbon dioxide by reduction at 400 °C in a zinc furnace, Figure 1 (2).
Figure 1. Conversion of [11C]carbon dioxide into 11C-labelled one-carbon precursors by simple on-line or batch procedures. References for the transformations: 1,8-10 2,11 3,12
4,13 5,14 6,15 7,16 8.17
Synthesis time
The synthesis time has to be minimized due to the short half-life of the radionuclide; a rule of thumb is that the total synthesis time should not exceed three half-lives of the radionuclide. Thus, the label should be incorporated as late as possible in the synthetic sequence and reactions and purifications must be rapid. Simple on-line and one-pot procedures are preferred, if achievable. To obtain the radiotracer in as high
11CO 2 11CO 11COCl2 11CH 4 H11CN 11CNBr 11CH 3OH 11CH3NO2 11CH3OTf 11CH3I 1 1 3 4 6 5 2 2 2 7 8
radiochemical yield as possible, a compromise has to be struck between the chemical yield and the decay,21 as illustrated in Figure 2. It is usually not the chemical yield that
is important but rather the amount of radioactivity available at end of synthesis, i.e. the radiochemical yield.
Figure 2. A hypothetical labelling reaction with 11C.
Specific radioactivity
When aiming for high specific radioactivity, isotopic dilution has to be minimised in radionuclide production and synthesis. The problem with isotopic dilution is directly related to the abundance of the corresponding unlabelled molecule in the surrounding environment. This can be exemplified by the synthesis of [11C]methyl iodide from [11C]methane, Figure 1 (8), where specific radioactivities of 550 GBq/µmol can be achieved,17 compared to around 150 GBq/µmol when starting from [11C]carbon dioxide,
Figure 1 (1). The major reason for the relatively low specific radioactivity of
[11C]methyl iodide in the latter case is probably due to the efficient trapping of carbon dioxide in lithium aluminum hydride. This illustrates the importance of avoiding reagents that contribute to the isotopic dilution in the synthesis.
Microscale synthesis
As a result of the high specific radioactivity, small masses have to be handled in the synthesis of radiotracers. The small mass can be used to increase reaction rates in some syntheses, since a 1000 fold excess or more of the other reagents can be used, resulting in pseudo-first order kinetics.22 However, small amounts of labelled precursor can also
be disadvantageous, especially in reactions where an equimolar amount of reagent, or 0 25 50 75 100 0 10 20 30 40 Time (min) Yield ( % ) Carbon-11 Decay Chemical yield Radiochemical yield
less, is necessary in order to obtain the desired product. Also, in some reactions the small amount of labelled precursor is insufficient for the reaction to proceed and unlabelled precursor (carrier) has to be added to the reaction mixture.
Purification and identification
Purification of labelled compounds is usually performed using semi-preparative liquid chromatography (LC) and solid phase extraction (SPE). Volatile compounds can also be purified by distillation. There are several ways of verifying that the intended compound has been obtained, and the combination of more than one analytical technique is usually preferred to ascertain the identity of the labelled compound. The most commonly used method is the addition of authentic reference substance to the labelled tracer and co-elution on a LC column connected in series with a radioactivity and a UV detector. This method is convenient and can easily be performed for each synthesis. Another
possibility is to verify the product identity with liquid chromatography combined with mass spectrometry (LC-MS). LC-MS is also a valuable tool in determining the specific radioactivity of a radiotracer since this analytical technique often offer a higher
sensitivity than the UV-detector.23 In order to demonstrate the labelling position in a
molecule, a combined 11C/13C synthesis may be performed. After decay, the position of the label is determined using 13C nuclear magnetic resonance (NMR) spectroscopy.
Radiation safety
It is important to ensure radiation protection for the chemists working with
radionuclides. This can be achieved by minimising the manual operations and by the use of remote controlled systems. It is therefore important that technical improvements are developed in parallel with the synthetic progress to achieve automated synthetic
procedures. Automated systems are also preferable from a pharmaceutical point of view since contamination can be avoided and better control of the process can be
accomplished. Also, many automated procedures afford higher reproducibility which is important in the production of radiopharmaceuticals.
3. PALLADIUM-PROMOTED CARBON-CARBON BOND FORMATIONS The first isolated organotransition metal compound was a platinum olefin complex discovered by the Danish pharmacist W. C. Zeise in 1827. However, it was not until the middle of the 20th century that organometallic chemistry found a general use in
have the ability to function as homogenous catalysts which can facilitate energetically unfavourable synthetic transformations.24 As a result of its capacity to catalyse a variety
of organic reactions, palladium is one of the transition metals most widely used in organometallic chemistry.24-27 Preparations of new carbon-carbon bonds are especially
attractive reactions that are made feasible with palladium catalysis.28
Stille reactions
The Stille reaction,29, 30 (Scheme 1) is the palladium-mediated coupling of an
organoelectrophile with an organostannane. This reaction has since its discovery in the late 1970s, been explored extensively both with respect to mechanisms and its synthetic applicability. The development of efficient palladium ligands31 and co-catalytic systems
with copper(I) salts32 are examples of factors that have significantly improved the
reaction. RX Pd(0)L4 1 Pd X L L R Pd(0)L2 PdII L R R' PdII L R R' [S] + L2 R'SnR''3 XSnR''3 R R' 5 4 3 2 PdII R' R L L or
Scheme 1. Catalytic cycle in a Stille reaction. L = ligand; X = Br, I; S = solvent; Y = Cl, Br
In Scheme 1 some of the suggested reaction mechanisms in the Stille reaction are illustrated. These reaction steps can be rationalised by a number of fundamental reactions, common in transition metal chemistry,33 and they are presented below with
the numbers that they appear with in Scheme 1 and 3.
1. Generation of Pd(0)L2 is the first reaction step in palladium(0) mediated reactions
since the active species is thought to be the zerovalent coordinatively unsaturated palladium complex, Pd(0)L2. This complex is usually formed in situ from Pd(0)L4
complex by ligand dissociation or by reduction of the Pd(II)L2Y2 complex. The
reducing agent is usually an organometallic species or a phosphane.34 Palladium
catalysts can also be formed in situ from Pd2(dba)3 or Pd(OAc)2 and free ligands.30, 35, 36
2. Oxidative addition of aryl or alkyl halides to Pd(0)L2 is a common method to
generate an organo-palladium complex. The Pd(0) complex is oxidised to Pd(II), and the carbon-halide bond is cleaved and two new σ-bonds to palladium are formed. The palladium complex has to be coordinatively unsaturated for the oxidative addition to take place. The mechanism suggested for this reaction step is an SN2 mechanism,24 or a concerted insertion of the Pd(0)L2 into the carbon-halide
bond,34, 39
3. Transmetallation is the transfer of an aryl or alkyl group from a metal (a stannane in
Stille reactions) to palladium. The transmetallation step is regarded as the rate-limiting step in Stille reactions.30, 37 The result of two alternative mechanisms are
illustrated in Scheme 1. One is the cis-[Pd(II)RR’L] complex which is the result of an associative L for R’ substitution,38 and the other trans-[Pd(II)RR’LS] complex is
the result of a dissociative L for S substitution followed by transfer of R’ to palladium.30
4. Isomerisation is the rearrangement between cis and trans complexes which is
thought to proceed by assistance of either the solvent, a free ligand or a palladium-halide species. Isomerisation has been suggested to take place before reductive elimination from trans-[Pd(II)RR’LS] to cis-[Pd(II)RR’LS].
5. Reductive elimination is when two groups positioned cis to each other in the
[Pd(II)RR’L] complex form a new carbon-carbon bond.40 This is the reverse of an
palladium complex is reduced by two. The mechanism of reductive elimination has been proposed to proceed from a “T-shaped” unsaturated palladium complex.38, 41
In the reductive elimination, a cross-coupling product is formed and the palladium(0) complex is regenerated.
Carbonylation reactions
Insertion reactions of carbon monoxide into metal ligand σ-bonds, forming metal-acyl complexes, (Scheme 2) were developed in the mid 1960s.42 The palladium-acyl
complexes can be transformed into a wide variety of carbonyl compounds by reaction with different nucleophiles, e.g. amines, alcohols or stannanes. In the case of ketone-and aldehyde synthesis, the reaction mechanism follows the same path as for the Stille reaction, with the addition of insertion of carbon monoxide,43 see Scheme 1 and 2.
Pd X L L R Pd X L CO R CO L Pd X L C R O Oxidative
addition Trans-metallation
Heck reactions
The palladium-promoted coupling of organoelectrophiles with alkenes was discovered in the beginning of the 1970s.44, 45 The catalytic Heck reaction is thought to follow the
mechanistic pathway shown in Scheme 3 and in this catalytic cycle, the oxidative addition is believed to be the rate-determining step.46, 47
L RX Pd(0)L2 Base Pd(0)L4 + L2 2 8 1 Y Pd R L X Y 6 7 Y R HBaseX + Y H H R L Pd X Y R H H Pd X L Pd X L L R
Scheme 3. Catalytic cycle of a Heck reaction. Reaction steps 1 and 2 are the same as described for the Stille reaction. L = ligand; X = Br, I; S = solvent; Y = Cl, Br
6. ππππ-coordination is the coordination of an alkene to a palladium(II) complex, in this case the [Pd(II)RXL2] complex. This process involves donation of π-electrons from
the alkene to palladium, due to the enhanced electrophilicity of the palladium(II) complex, accompanied by back-donation from the palladium to the alkene.
7. Migratory insertion occurs when an aryl or alkyl group migrates from palladium to
a coordinating ligand, such as an olefin. Migration to olefins usually proceeds by 1,2-insertions where the ligand and the palladium end up at adjacent carbons. This is thought to proceed through a four-center transition state.47
8. ββββ-Hydride elimination is the reverse insertion reaction of an olefin. The hydrogen and palladium must be synperiplanar for the β-elimination to occur. Since the elimination is a reversible process, the thermodynamic product is often formed. β-Elimination is considered to be a fast reaction step, limiting the selection of substrates that can be used in palladium-mediated reactions. After the β-elimination, the olefin is released from the complex and a base regenerates the palladium(0) complex.
Palladium-promoted [11C]carbon-carbon bond formations
Several palladium-mediated reactions have been shown to be valuable in the
development of 11C-C bond formation. Stille and Suzuki couplings with [11C]methyl iodide,48 carbonylation reactions with [11C]carbon monoxide49, 50 and cyanation
reactions with hydrogen [11C]cyanide51, 52 are some examples. These reactions tolerate a
variety of different functional groups present in the substrates and can usually be performed by simple one-pot procedures. Such features are important in
radiopharmaceutical synthesis and therefore, the palladium-mediated reactions provide a good complement to other carbon-carbon bond forming reactions developed for labelling synthesis, such as the Wittig reaction with [11
C]methylenetriphenyl-phosphorane,20 condensation of [11C]nitromethane with aldehydes,53 cuprate-mediated
labelling synthesis54 and alkylations on stabilised carbanions.18 Of course, due to the
very small amounts of labelling precursor, the palladium-mediated reactions are
performed with excess amounts of palladium catalyst with respect to labelled compound and the short half-life of 11C sometimes requires strenuous reaction conditions to speed up the reactions. Furthermore, since no general protocol for different palladium catalyst and solvents has been developed, it is important to optimise these parameters for each new application since a reactive catalyst is necessary in rapid labelling synthesis.
4. STILLE REACTIONS WITH [11C]METHYL IODIDE (PAPERS I, II, III) The Stille reaction has proven very useful in 11C-labelling synthesis since it is a mild method, which tolerates different functional groups in the substrates, and the reagents are not particularly sensitive towards moisture or oxygen. These features make the reaction especially attractive in the labelling of highly functionalised substrates, since the use of protective groups can be avoided. There are few examples in the literature of the Stille reaction with methyl iodide55 which is not surprising considering that few
synthetic chemists, except those working with 11C, would select this reaction path when synthesising a molecule containing a methylphenyl group.
Since the published method for the synthesis of [11C]methylphenyl model compounds48
gave poor yields when applied to the synthesis of 11C-labelled prostacyclin analogues, the Stille reaction with [11C]methyl iodide was further investigated to increase the reactivity towards large and highly functionalised stannanes.
Synthesis of a 11C-labelled prostacyclin analogueI
A coordinatively unsaturated palladium complex was employed for the synthesis of 15R-16-(3-[11C]methylphenyl)-17,18,19,20-tetranorisocarbacyclin methyl ester, since it was observed that unreacted [11C]methyl iodide remained after the cross-coupling using the published method,48 Scheme 4. With this approach, the prostacyclin analogue could
be synthesised from the corresponding stannane in up to 55 % decay-corrected isolated radiochemical yield calculated from [11C]methyl iodide. The radiochemical purity of the isolated product was higher that 95 %. However, a major drawback with the method was that it did not give a reproducible yield.
OH CH3O O HO SnBu3 OH CH3O O HO 11CH 3 11CH3I, Pd2(dba)3, P(o-Tol)3 DMF, 130°C
Scheme 4. Labelling of a prostacyclin analogue with 11C.
The unsaturated palladium complex, used in the reaction, was formed in situ from Pd2(dba)3 and tri(o-tolyl)phosphine (1:4). This complex was used with the intention to
facilitate the oxidative addition step in the catalytic cycle.30 Several different palladium
ligands, described as being useful to increase the rates in Stille reactions,31 e.g.
tri(2-furyl)phosphine and triphenylarsine, were investigated but gave poor yields. One reason for these results could be that the reaction conditions employed were too strenuous for
the unsaturated palladium complexes formed with these ligands. Tri(o-tolyl)phosphine, a relative strong donor ligand with a large cone angle, gave a more stable unsaturated palladium complex in a polar solvent such as DMF. When using this complex, a high yield could be obtained but a high reaction temperature (130 °C) was necessary for the reaction to give a good yield. The isolated Pd[P(o-Tol)3]2 complex afforded a low yield,
probably on account low solubility of the complex in DMF.
Modifications of the Stille reaction with methyl iodideII
It was found that when the unsaturated palladium(0) complex used in the 11C-labelling of the prostacyclin analogues in Scheme 4 was used at a reaction temperature of 130 °C, the decomposition of the palladium complex was sometimes faster than the cross-coupling reaction, with subsequent reaction quenching as result. This explains the irreproducible results obtained with this method. To improve the reproducibility of the reaction employing an unsaturated palladium complex, the temperature had to be decreased without loss of reactivity.
There have been several reports concerning increased reactivity in Stille couplings using copper(I) salts.56-58 The mechanisms explaining this improvement have been suggested
to depend on the solvent and type of ligand used in the reaction. In non-polar solvents,
e.g. THF or diethyl ether, with strong donor ligands, such as triphenylphosphine, the
copper salt can bind to some of the ligands in the reaction mixture and decrease the ligand concentration around the palladium and therefore increase the reaction rate.32
However, in polar solvents, e.g. DMF or NMP, and with softer ligands, such as tri(o-tolyl)phosphine, the copper might facilitate and increase the reactivity in the
transmetallation step by a prior stannane-copper transmetallation.32 The aryl or alkyl
copper moiety formed is believed to have a much higher reactivity and would thus increase the rate of the transmetallation.
To investigate if the approach with a copper salt as co-catalyst to palladium could be useful, the synthesis of toluene from tributylphenylstannane and stable methyl iodide was examined. An excess of the stannane (40 equivalents) was used to mimic 11 C-labelling conditions, Scheme 5.
Pd(0), Cu(I) SnR3 + CH3I CH3 R = Bu, Me excess
Scheme 5. The model reaction used to investigate the Stille reaction with methyl iodide. As indicated in Table 1, the best results in the cross-coupling reaction were obtained when copper(I) chloride or copper(I) bromide was used along with potassium carbonate and an unsaturated palladium(0) complex with tri(o-tolyl)phosphine as ligand, entry 1 and 2. When only copper salt was used in the reaction, no coupling product could be detected, entry 3, and when only the palladium complex was used, the yield was lower, entry 4. Hence, the combination of palladium complex, potassium carbonate and copper salt was necessary to have an efficient reaction at 60 °C. Surprisingly, copper(I) iodide did not give a good result in the reaction, contrary to previous reports.32
Table 1. Entrya Pd(0)
complex
Ligand Pd:L Additiveb Solvent Yield of toluenec,d/% 1 Pd2(dba)3 P(o-Tol)3 1:2 CuCl/K2CO3 DMF 91 ± 3
2 Pd2(dba)3 P(o-Tol)3 1:2 CuBr/K2CO3 DMF 90 ± 3
3 - - - CuCl/K2CO3 DMF 0
4 Pd2(dba)3 P(o-Tol)3 1:2 - DMF 63 ± 3e
5 Pd2(dba)3 P(o-Tol)3 1:2 CuI DMF 3 ± 3
6 Pd(PPh3)4 - - CuCl/K2CO3 THF 23 ± 3
a40 mol of tributylphenylstannane and 1 mol of Pd (0) was used relative to methyl iodide. b2 mol of
additive relative to methyl iodide was used. cThe reactions were performed at 60 °C for five min. dThe
yield was determined by GC analysis of a sample withdrawn from the reaction mixture, using an internal standard. eThe reaction was performed at 80 °C in five min. n > 2
The optimal result obtained with tributylphenylstannane, entry 1, was investigated with four other model compounds to assess the compatibility with different functionalities and with some aromatic heterocyclic compounds, Figure 3. The model compounds 1, 2, 3 and 4 were synthesised from the corresponding tributylstannanes in >99, 92, 73 and 40 % yields respectively, determined by GC analysis of a sample withdrawn from the reaction mixture. HO CH3 OCH3 CH3 S CH3 O CH3 1 2 3 4
Figure 3. Model compounds investigated with the improved reaction conditions for the Stille reaction with methyl iodide.
When trimethylphenylstannane was used in the reaction, the yield of toluene was more than 100 % relative to methyl iodide. This indicated that a methyl group from the trimethylphenylstannane was participating in the reaction, forming coupling product. This lack of selectivity in the transfer of groups from the stannane can be circumvented by the use of aryltributylstannanes. A phenyl group is transferred 1.3 times faster from trimethylphenylstannane than from tributylphenylstannane but a methyl is transferred ten times faster than a butyl from tin to palladium.30 Thus, the selectivity is higher with
the aryltributylstannane and the reactivity is only marginally reduced. In addition, aryltributylstannanes are less toxic than the methyl analogues,59 which could be of
importance when synthesising compounds for in vivo use.
Synthesis of a 11C-labelled prostaglandin F2αααα analogueIII
Using the conditions developed for the model compounds, an unsaturated palladium(0) complex with two tri(o-tolyl)phosphine ligands, copper(I) chloride and potassium carbonate in DMF, a [11C]methylphenyl analogue of prostaglandin F2α was synthesised
in 34 % decay-corrected isolated radiochemical yield, Scheme 6. The radiochemical purity was higher than 95 % and the total synthesis time, including purification and formulation, was 35 min.
HO HO OH COOCH(CH3)2 SnBu3 HO HO OH COOCH(CH3)2 11CH 3 DMF, 60°C, 5 min 11CH3I, Pd2(dba)3, P(o-Tol)3, Cu(I)Cl, K2CO3
Scheme 6. Synthesis of 11C-labelled prostaglandin F2α analogue
In principle, the improved method for Stille couplings with stable methyl iodide discussed above, could be applied to labelling synthesis with [11C]methyl iodide. Unfortunately, the cross-coupling was quenched when all reagents were present in the vial where [11C]methyl iodide was collected after distillation. Despite several attempts to improve the purity of the [11C]methyl iodide, with extra drying towers, a one-pot approach had to be abandoned. In order to accomplish the copper assisted Stille coupling with [11C]methyl iodide, it was necessary to separate the trapping solution from the copper salt. Thus, [11C]methyl iodide was trapped in a solution containing the palladium complex and in a separate vial, the stannane precursor, the copper salt and the potassium carbonate were carefully prepared under argon atmosphere. After trapping of [11C]methyl iodide, the palladium solution was added to the precursor solution and the resulting mixture was heated at 60 °C. Despite no additional purification of the trapping solution before addition to the copper salt solution, a good result could be obtained. Moreover, with this method it was possible to synthesise the 11C-labelled prostacyclin analogue (structure presented in Scheme 4) in a moderate isolated decay-corrected radiochemical yield with a high reproducibility, 45 ± 8 % (n = 8).
As mentioned above, a drawback with the Stille reaction is the lack of selectivity that sometimes occurs in transferring an organic group from the stannane to the palladium. This is particularly problematic with heavily substituted aryltrialkylstannanes, since in these cases, the alkyl transfer can compete with the desired transfer of the unsaturated group.32 This problem may be overcome when a stannane with only one transferable
group is used, see Scheme 7.60 Furthermore, depending on the structural complexity and
the presence of functional groups, the synthesis and purification of stannanes might sometimes also be difficult. With the approach shown in Scheme 7, the synthesis of
large complicated stannanes could be avoided since the 11C-label would be introduced from the stannane in a Stille reaction. This work is under investigation by co-workers.
N Sn Cl N Sn 11CH 3 11CH 3Li
Scheme 7. New 11C-labelled stannane for use in palladium-mediated cross-coupling reactions.
In conclusion, with the modifications presented above, the Stille reaction has been shown to be a valuable tool in the labelling of large and highly functionalised
molecules, that is prostaglandin analogues. The simplicity of the procedure along with the good and highly reproducible yields makes the method particularly attractive in radiopharmaceutical synthesis.
5. HECK REACTIONS WITH 11C-LABELLED OLEFINS (PAPER IV) The Wittig reaction is well established as an efficient and general means of forming double bonds by reaction of aldehydes or ketones with phosphonium ylides.6111
C-Labelled terminal alkenes have been synthesised from [11C]methyl iodide employing the Wittig reaction, as illustrated in Scheme 8. This reaction was first used in the synthesis of [11C]styrene20 but has also been applied in the synthesis of 11C-labelled
carbohydrates.62 11CH 3I PPh3 Base 11CH 2 PPh3 RCHO 11CH2 C R H + 11CH 3PPh3I
Functionalisation of 11C-labelled olefins via a Heck coupling reactionIV
[β-11C]Styrene and [1-11C]pent-1-ene were synthesised from benzaldehyde and
n-butyraldehyde, respectively, in a Wittig reaction with [11
C]methyltri(o-tolyl)phosphonium iodide in 85-90 % radiochemical yield determined by analytical LC, Scheme 9. R' H O Pd 2(dba)3 P(o-Tol)3 I R R' = n-Pr, Ph
= Position of the 11C-label ∗ P(o-Tol)3 11CH 3I ∗ R' ∗ R R'
Scheme 9. Wittig olefination reaction followed by a Heck reaction.
The Wittig reactions were performed as described earlier20 with the exception of the use
of tri(o-tolyl)phosphine, which was used as Wittig reagent instead of the more
commonly used triphenylphosphine. Tri(o-tolyl)phosphine gave the same result in the olefination reaction as triphenylphosphine; the reason for using this phosphine will be discussed below. Epichlorohydrin was used as base precursor in the reaction, according to the published method.20 Using this method, one equivalent of base to [11C]methyl
iodide is formed in the reaction mixture.63 The labelled olefins were coupled with five
aromatic halides in a palladium-mediated Heck reaction, Scheme 9.
The palladium-mediated synthesis was performed by adding a solution of Pd2(dba)3 and
tri(o-tolyl)phosphine (1:4) and the aromatic halide in o-dichlorobenzene to a vial
containing crude 11C-labelled alkene and heating the resulting mixture at 150 °C for five minutes. Since good results have been reported employing microwave heating in Heck reactions64, 65 as well as in labelling synthesis,66, 67 this method was selected for
investigation. However, the results obtained with microwave heating were only marginally improved from those obtained with conventional heating.
Table 2
Entry Halide Yielda/% Unreacted
[β-11C]styrene/% Product 1 I 55 (38)b 2 * 2 I H2N 53 (40) b 0 NH2 * 3 I HO 43 3 HO * 4 I O OEt 49 5 O OEt * 5 I CH3 47 (34) b 1 CH3 *
a Radiochemical yield presented as the percentage of the total amount of radioactivity in the reaction
mixture, determined by analytical LC of samples withdrawn from the reaction mixture, n > 5. b Decay-corrected isolated radiochemical yield.
Tri(o-tolyl)phosphine was chosen as ligand since the palladium complex formed with this ligand has been reported to give good results in Heck reactions with aromatic halides.46, 68
It was not possible to perform the reaction sequence using a one pot procedure when triphenylphosphine was used in the Wittig reaction. This was due to the fact that when triphenylphosphine was employed, the same product, [11C]stilbene, was formed in the Heck reaction regardless of which aromatic halide was added to the reaction mixture. The reason for this was that an aryl exchange between the palladium and the phosphine was taking place, Scheme 10. Aryl exchange between phosphorous bound aryls and palladium bound aryl or alkyl groups within palladium complexes have been reported earlier.55, 69
In the present investigation, it was evident that it was triphenylphosphine that was coordinating to the palladium and causing the aryl exchange since the product formed from the aryl exchange process was [11C]stilbene, Scheme 10.
I R Pd[P(o-Tol)3]2 PPh3 P(o-Tol)3 Pd(PPh3)2 Pd I PPh3 PPh3 R R Pd I PPh2 PPh3 Ph PdPPh3 + -+ PPh3I R ∗ ∗
∗ = Position of the 11C-label
Scheme 10. Aryl exchange between a phosphorous-bound phenyl group and a palladium-bound aryl group.
The mechanism of the ligand interchange, illustrated in Scheme 10, has been suggested to involve the formation of a phosphonium salt from the oxidatively added moiety and a phosphine ligand via a reductive elimination. When the phosphonium salt is oxidatively added to palladium again, a different aryl-phosphorous bond may be broken.69, 70
When tri(o-tolyl)phosphine was used both as a ligand to palladium and as Wittig reagent, the ligand interchange could not be detected. This is consistent with previous reports where tri(o-tolyl)phosphine has been reported to suppress this side reaction.69, 71
Heck couplings with [11C]methyl iodide and styrene were investigated, since this reaction would afford a 11C-labelled methylvinyl structure in a simple one-step
procedure. Unfortunately, formation of the desired product in the coupling reaction with [11C]methyl iodide could not be detected.
In conclusion, an efficient two step procedure for the incorporation of a 11C-label in an alkene structure has been developed. This method has been shown to be applicable to
functional groups. The method presented can be of interest when alternative labelling positions are necessary since it allows introduction of the label into the backbone of the structure. For example, the antiepileptic carbamazepine, Figure 4, could be a possible target for this labelling method to introduce a 11C-label in the carbon-carbon double bond “a”.
N H2N O
a
Figure 4. Carbamazepine with a suggested 11C-labelling position, a.
6. CARBONYLATION REACTIONS WITH [11C]CARBON MONOXIDE
(PAPER V)
[11C]Carbon monoxide was one of the first tracers applied in PET experiments in humans.72 Even though [11C]carbon monoxide is a readily available precursor, it has
only been applied in a few labelling reactions,73-75 which might seem surprising in the
view of the synthetic possibilities such a tracer would provide.42 The reason has been
the difficulties in the technical handling of the [11C]carbon monoxide, which has a very low solubility and reactivity. To facilitate the use of [11C]carbon monoxide in labelling synthesis, a system with a micro-autoclave where [11C]carbon monoxide is enclosed at high pressure along with the other reagents has been developed. This system has, for example, been used for the palladium-mediated synthesis of 11C-labelled amides.50
Palladium-mediated reactions have also been applied to labelling with [11C]carbon monoxide in the synthesis of aromatic ketones.49, 76
Palladium-promoted formylation reaction of aryl halides with [11C]carbon monoxideV
The synthesis of [11C]benzaldehyde via palladium-mediated formylation of iodobenzene with [11C]carbon monoxide has been reported.77 This was further examined, with
emphasis on different palladium catalysts and different functional groups on the aromatic halides, using the micro-autoclave system. This work is of interest since aldehydes are useful synthetic intermediates which can be converted into many other
functionalities, e.g. imines, amines and olefins. Thus, the formylation of six aromatic halides employing different palladium catalysts and various hydride substrates were investigated, Scheme 11. X R 11C O H R Pd(0), 11CO, [H-]
Scheme 11. Formylation reaction with [11C]carbon monoxide.
A solution of the palladium catalyst and a halide was prepared in NMP. The hydride reagent was added to the solution just before transferring it into the micro-autoclave. The reaction was performed at 70 °C and the pressure in the micro-autoclave was 35 MPa or higher.
Unfortunately, since the optimal result for each aldehyde was achieved with different palladium-catalysts, no general conclusion could be made from these results regarding best choice of catalyst,. AsPh3 was one of the ligands that gave good results with several
substrates, but in the case of 1-iodo-3-nitrobenzene, no product could be obtained with this ligand, Table 3 entry 13.
Et3SiH was the preferred hydride reagent78 since a high yield could be obtained with
most substrates, and functional groups that easily could be reduced remained intact, entry 12. Bu3SnH has been reported to give good results in formylation reactions,43
reacted with the palladium-halide solution instantaneously under gas evolution and the LC analysis of the products revealed a number of side-products (10-20 %). This was consistent with all the halides investigated with this hydride reagent.
Poly(methylhydrosiloxane) (PHMS)79 was equally efficient as Et3SiH in some of the
formylation reactions, entry 1. However, a SPE purification of the 11C-labelled benzaldehyde formed with PHMS was conducted and 30% of the total amount of the radioactivity was retained on the SPE column after elution of the 11C-labelled product, as compared to 12% when employing Et3SiH.
Table 3.
aRadiochemical yield presented as the percentage of the total amount of radioactivity trapped in the
reaction mixture, determined by analytical LC of samples withdrawn from the reaction solution, n > 4. Trapping efficiency of [11C]carbon monoxide in parentheses. b6 µmol of 4-iodoaniline was used in the
reaction mixture. c20 µmol of 4-iodoaniline was used in the reaction mixture. n.d. = not detected.
For entry 6, 6 µmol of 4-iodoaniline was used as compared to 20 µmol halide in the other reactions. This smaller amount of halide was necessary in order to decrease the side-reactions taking place at higher concentrations of 4-iodoaniline. The products formed in the side-reaction were not identified but they may be the result of the amine acting as a nucleophile. Since bromo compounds often are more easily accessible than iodo compounds, a reaction with bromotoluene was conducted as a comparison to 4-iodotoluene. The results with this substrate were not as good as with 4-iodotoluene; the trapping efficiency was decreased and several side-products were formed in the
reaction, entry 11. Since the oxidative addition is slower with a bromoaryl compound than with the corresponding iodoaryl compound the reaction mixture containing halide and palladium catalyst was pre-heated. This did not improve the result achieved with the substrate. It is unclear whether the reason for the poor result in entry 15 and 16 was because the bromo compounds were employed.
Entry Halide Catalyst Hydride
reagent
Yield of aldehydea/% 1 Iodobenzene Pd2(dba)3: AsPh3 1:4 PMHS 87 (96)
2 Pd(PPh3)4 Et3SiH 87 (76)
3 Pd2(dba)3: dppe 1:2 Et3SiH 79 (48)
4 Pd2(dba)3: AsPh3 1:4 Et3SiH 68 (78)
5 Pd2(dba)3: AsPh3 1:4 NaOH 14 (65)
6 4-Iodoaniline Pd(PPh3)4 Et3SiH 65b (78)
7 Pd(PPh3)4 Et3SiH 10c (83)
8 Pd2(dba)3: AsPh3 1:4 PMHS 23 (95)
9 3-Iodotoluene Pd(PPh3)4 Et3SiH 97 (90)
10 Pd2(dba)3: dppe 1:2 Et3SiH 49 (77)
11 3-Bromotoulene Pd(PPh3)4 Et3SiH 23 (54)
12 1-Iodo-3-nitrobenzene
Pd2(dba)3: dppe 1:2 Et3SiH 76 (78)
13 Pd2(dba)3: AsPh3 1:4 Et3SiH n.d. (63)
14 Pd(dppf)Cl2 Et3SiH 56 (58)
15
Methyl-4-bromobenzoate
Pd(dppf)Cl2 Et3SiH 25 (36)
In order to obtain more complex structures and to investigate whether it was possible to use the 11C-labelled aldehydes in a second reaction, a Wittig reaction was performed on the crude 11C-labelled aldehyde, Scheme 12.
11C O H R 11C CHR' H R Base -R'CH2PPh+ 3X
Scheme 12. Olefination of 11C-labelled aldehydes. Table 4.
a Radiochemical yield presented as the percentage of the total amount of radioactivity in the reaction,
determined by analytical LC of samples withdrawn from the reaction mixture, n > 3. bDecay-corrected
isolated radiochemical yield calculated from trapped [11C]carbon monoxide. cBis(trimethylsilyl)lithium
amide was used as base.
The Wittig salt employed for the olefination reaction was dissolved in THF and one equivalent of n-butyl lithium was added, whereafter the solution was mixed with the crude 11C-labelled aldehyde. After heating for 4 min at 100 °C, a sample was withdrawn from the reaction mixture for LC analysis, or the reaction mixture was diluted and injected onto a semi-preparative LC.
Entry Aldehyde Wittig salt Product Yielda/%
1 [11C]Benzaldehyde CH3PPh3Br Ph11CH=CH2 76 (57b) 2 CH3OCH2PPh3Cl Ph11CH=CHOCH3 62 3 CNCH2PPh3Cl Ph11CH=CHCN 40 4 4-Amino-[11C]benzaldehyde CH3PPh3Br 4-NH2Ph11CH=CH2 47 5 3-Methyl-[11C]benzaldehyde CH3PPh3Br 3-CH3Ph11CH=CH2 84 (69b) 6 3-Nitro-[11C]benzaldehyde CH3PPh3Br 3-NO3Ph11CH=CH2 30c
To summarise, an efficient method for the synthesis of 11C-labelled aldehydes has been developed and applied to the synthesis of 11C-labelled alkenes. It is likely that not only alkenes can be synthesised employing this method, but also other compounds easily obtained from aldehydes such as imines and amines.
7. APPLICATIONS TO POSITRON EMISSION TOMOGRAPHY
Prostaglandins are physiologically potent substances with multiple and variable
functions in the body. They were first discovered in human semen 1935 by the Swedish physiologist Ulf von Euler,80 who hypothesised that they were secreted from the
prostate gland. The chemical structure of prostaglandins are based on the structure of prostanoic acid which contains a 5-carbon ring and two side chains, Figure 5. The classification of prostaglandins into groups and subgroups depends on unsaturation and functional groups on both the ring and the side chains. Prostaglandins are synthesised in
vivo from arachidonic acid by cyclooxygenase and the respective synthase. Depending
on the type and tissue, prostaglandins can stimulate smooth-muscle contraction, lower or raise blood pressure, decrease or increase the clotting ability of blood, enhance ion transport across membranes, stimulate inflammation and inhibit lipolysis.81
O COOH HO OH Prostacyclin Prostanoicacid COOH
Figure 5.Prostanoic acid and prostacyclin (PGI2).
At the time of their discovery, prostaglandins were thought to have a tremendous potential as therapeutic agents in a large number of diseases. However, these
expectations were dissatisfied by the fact that these substances are chemically instable, are rapidly metabolised and that their actions are not specific, wherefore they exert a number of side effects.82
In the late 1960s, Corey published a general route for the synthesis of several prostaglandins.83 This method has been extensively used in the synthesis of natural
prostaglandins as well as many prostaglandin analogues. Another useful synthetic approach, often referred to as the three component coupling, has also been employed.84
With these synthetic methods, analogues could be designed to overcome the disadvantages mentioned above.
A 11C-labelled prostacyclin analogue in PET
Prostacyclin (PGI2), Figure 5, is an important regulator of cardiovascular function, via
dilation of vascular smooth muscle and inhibition of platelet aggregation.85 It is not
known whether prostacyclin has any function in the central nervous system (CNS), but a prostacyclin metabolite has been detected in the brain after convulsions86 and it has
been suggested that prostacyclin could be useful for treatment of ischaemic neuronal damage due to its beneficial effect on cerebral circulation.87, 88
An analogue of prostacyclin, (15R)-16-(3-methylphenyl)-17,18,19,20-tetranor-isocarbacyclin acid has been shown to bind to a subtype of prostacyclin receptors in CNS.89, 90 In order to investigate the binding and distribution of this compound with
PET, three 11C-labelled derivatives RTA, RTB and RTC, illustrated in Figure 6, were synthesised. OH CH3O O HO 11CH 3 OH 11CH 3O O HO CH3 OH HO O HO 11CH3 RTA RTB RTC
Figure 6. Three derivatives of (15R)-16-(3-methylphenyl)-17,18,19,20-tetranor-isocarbacyclin acid.
RTB was synthesised as RTA, illustrated in Scheme 4, with the addition of a basic hydrolysis of the methyl ester. RTC was labelled in an esterification reaction where a tetrabutylammonium salt of the corresponding acid was prepared and reacted with [11C]methyl iodide.
In a triple tracer experiment with RTA, RTB and RTC, RTA showed a higher uptake in brain than RTB and the pattern of brain uptake was similar to localisation of
(15R)-[15-3H]-16-(3-methylphenyl)-17,18,19,20-tetranorisocarbacyclin acid in frozen section of
rat brain.89 The uptake of RTC in brain was high but the rapid de-esterification of the methyl ester released the 11C-label from the tracer. Thus, it was established that althoughthe methyl ester of
(3-methylphenyl)-17,18,19,20-tetranor-isocarbacyclin could penetrate the blood-brain barrier the active species was (15R)-16-(3-methylphenyl)-17,18,19,20-tetranorisocarbacyclin acid.
8. CONCLUSIONS
In this thesis, the significance of palladium-promoted reactions in rapid labelling synthesis has been demonstrated. Carbon-carbon bond forming methods using
palladium-promoted reactions have been developed and applied to the synthesis of 11 C-labelled compounds. The Stille reaction with [11C]methyl iodide was shown to be a versatile method for the labelling of complex molecules, such as prostaglandins. Furthermore, 11C-labelled alkenes have been synthesised from [11C]methyl iodide and from [11C]carbon monoxide. These two methods complement each other well in their different approaches for the synthesis of 11C-labelled olefins, Figure 7. Also, the method presented for the synthesis of 11C-labelled aldehydes may prove valuable in the
synthesis of many different 11C-labelled compounds since aldehydes are versatile synthetic intermediates.
Figure 7. Two methods for the synthesis of a 11C-labelled alkene. ∗ = Position of 11C-label R ∗ 11CO H O ∗ Formylation reaction Wittig reaction 11CH 3I ∗ R
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to my supervisor, Professor Bengt
Långström, for his enthusiasm and encouragement throughout these years.
I am very grateful to Professor Ryoji Noyori for inviting me to Nagoya University and to Professor Masaaki Suzuki for welcoming me to his lab group and in his home, making my stay in Japan especially memorable. I also wish to thank Professor
Yasuyoshi Watanabe for fruitful collaboration and for enjoyable discussions.
Drs. Gunnar Antoni, Göran Westerberg and Mattias Ögren are gratefully acknowledged for constructive criticism on this thesis. I would also like to thank Dr. Robert Moulder and Fil. Lic. Anna Bergman for linguistic corrections.
I am indebt to M. Sc. Hisashi Doi and Dr. Koichi Kato for providing me with
precursors, and even more for taking excellent care of me in the lab in Nagoya and for showing me the beautiful surroundings.
I would like to thank all past and present members of the 11C-group and the staff at the PET centre for support and friendship and for all the amusing moments in the coffee room.
I would also like to thank the colleagues and the staff at Kemicum, especially Leif
Jansson, Gunnar Svensson, Ingvar Wik and Katarina Israelsson, for all the help and for
always making me feel welcome.
I am deeply grateful to all the members of the Björkman family and the Ögren family who has encouraged and supported me, and to my friends who are always there for me. Finally, thank you Mattias for love and support and for making my life special.
REFERENCES
1. G. Hevesy, Biochem. J., 1923, 17, 439.
2. H. N. J. Wagner, Z. Szabo and J. W. Buchanan Principles of Nuclear Medicine, 2nd ed.; W. B. Saunders Co.: Philadelphia, 1995.
3. B. S. Axelsson, B. Långström and O. Matsson, J. Am. Chem. Soc., 1987, 109, 7233. 4. A. D. Nunn Radiopharmaceuticals: Chemistry and Pharmacology; Marcel Dekker
Inc.: New York, 1992.
5. T. Greitz, D. H. Ingvar and L. Widén The Metabolism of the Human Brain Studied
with Positron Emisssion Tomography; Raven Press: New York, 1985.
6. J. S. Fowler and A. P. Wolf. Positron Emitter-Labeled Compounds: Priorities an Problems. In Positron Emission Tomography and Autoradiography: Principles and
Applications for the Brain and Heart; M. Phelps,Mazziotta, J. and Schelbert, H.
Eds.; Raven Press: New York, 1986; pp. 391-450.
7. J. S. Fowler and A. P. Wolf, Acc. Chem. Res., 1997, 30, 181-188.
8. B. Långström and H. Lundqvist, Int. J. Appl. Radiat Isot., 1976, 27, 357-363. 9. B. Långström, et al., J. Nuc. Med., 1987, 28, 1037-1040.
10. D. Comar, J.-C. Cartron, M. Maziere and C. Marazano, Eur. J. Nucl. Med., 1976, 1, 11-14.
11. D. R. Christman, R. D. Finn, K. I. Karlstrom and A. P. Wolf, Int. J. Appl. Radiat
Isot., 1975, 26, 435-442.
12. K.-O. Schoeps, et al., J. Labelled Compd. Radiopharm., 1988, 25, 749-758. 13. D. M. Jewett, Appl. Radiat. Isot., 1992, 43, 1383-1385.
14. P. Landais and C. Crouzel, Appl. Radiat. Isot., 1987, 38, 297-300.
15. G. Westerberg and B. Långström, Acta Chem. Scand., 1993, 47, 974-978. 16. D. Roeda and G. Westera, Int. J. Appl. Radiat Isot., 1981, 32, 931-932.
17. P. Larsen, J. Ulin, K. Dahlström and M. Jensen, Appl. Radiat. Isot., 1997, 48, 153-157.
18. B. Långström, et al., Acta Radiologica Supplementum, 1990, 374, 147-151. 19. S. Reiffers, et al., Int. J. Appl. Radiat. Isot., 1980, 31, 535.
20. T. Kihlberg, P. Gullberg and B. Långström, J. Labelled Compd. Radiopharm., 1990, 28, 1115-1120.
21. B. Långström and G. Bergson, Radiochem. Radioanal. Letters, 1980, 43, 47-54. 22. B. Långström. Doctoral Thesis, Acta Universitatis Upsaliensis No. 555, Uppsala
1980.
23. B. Hyllbrant, N. Tyrefors, K. E. Markides and B. Långström, J. Pharm. Biomed.
24. J. P. Collman, L. S. Hegedus, J. R. Norton and R. G. Finke Principles and
Applications of Organotransition Metal Chemistry; University Science Books: Mill
Valley, California, 1987.
25. I. Omae Applications of Organometallic Compounds; John Wiley & Sons Ltd: Chichester, 1998.
26. R. F. Heck Palladium Reagents in Organic Synthesis; Academic Press Ltd.: London, 1985.
27. J. Tsuji, Synthesis, 1990, 739-749.
28. F. Diederich and P. J. Stang Metal-catalyzed Cross-coupling Reactions; Wiley-VCH: Weinheim, 1998.
29. J. K. Stille, Angew. Chem. Int. Ed. Engl., 1986, 25, 508-524.
30. V. Farina and G. P. Roth. Recent advances in the Stille reaction. In Advances in
Metal-Organic Chemistry; L. S. Liebeskind Ed.; JAI Press: Greenwich, UK, 1996;
Vol. 5; pp. 1-53.
31. V. Farina and B. Krishnan, J. Am. Chem. Soc., 1991, 113, 9585-9595. 32. V. Farina, et al., J. Org. Chem., 1994, 59, 5905-5911.
33. R. H. Crabtree The Organometallic Chemistry of the Transition Metals, 2nd ed.; Wiley & Sons Inc.: New York, 1994.
34. C. Amatore, A. Jutand and A. Suarez, J. Am. Chem. Soc., 1993, 115, 9531-9541. 35. C. Amatore, G. Broecker, A. Jutand and F. Khalil, J. Am. Chem. Soc., 1997, 119,
5176-5185.
36. C. Amatore and A. Jutand, Acc. Chem. Res., 2000, 33, 314-321.
37. J. W. Labadie and J. K. Stille, J. Am. Chem. Soc., 1983, 105, 6129-6137. 38. A. L. Casado and P. Espinet, J. Am. Chem. Soc., 1998, 120, 8978-8985. 39. A. L. Casado and P. Espinet, Organometallics, 1998, 17, 954-959. 40. A. Gillie and J. K. Stille, J. Am. Chem. Soc., 1980, 102, 4933-4941. 41. A. Moravskiy and J. K. Stille, J. Am. Chem. Soc., 1981, 103, 4182-4186.
42. H. M. Colquhoun, D. J. Thompson and M. V. Twigg Carbonylation; Plenum Press: New York, 1991.
43. V. P. Baillargeon and J. K. Stille, J. Am. Chem. Soc., 1986, 108, 452-461. 44. R. F. Heck and J. Nolley, J. P., J. Org. Chem., 1972, 37, 2320-2322. 45. T. Mizoroki, Bull. Chem. Soc. Jap., 1971, 44, 581.
46. W. A. Herrmann, et al., Angew. Chem. Int. Ed. Engl., 1995, 34, 1844-1847.
47. A. de Meijere and F. E. Meyer, Angew. Chem. Int. Ed. Engl., 1994, 33, 2379-2411. 48. Y. Andersson, A. Cheng and B. Långström, Acta Chem. Scand., 1995, 49, 683-688. 49. Y. Andersson and B. Långström, J. Chem. Soc., Perkin Trans. 1, 1995, 287-289.
50. T. Kihlberg and B. Långström, J. Org. Chem., 1999, 64, 9201-9205. 51. G. Antoni and B. Långström, Int. J. Appl. Radiat. Isot., 1992, 43, 903-905.
52. Y. Andersson and B. Långström, J. Chem. Soc., Perkin Trans. 1, 1994, 1395-1400. 53. K. Någren, et al., Appl. Radiat. Isot., 1994, 45, 515.
54. T. Kihlberg, H. Neu and B. Långström, Acta Chem. Scand., 1997, 51, 791. 55. D. K. Morita, J. K. Stille and J. R. Norton, J. Am. Chem. Soc., 1995, 117,
8576-8581.
56. L. S. Liebeskind and R. W. Fengl, J. Org. Chem., 1990, 55, 5359-5364. 57. G. P. Roth and V. Farina, Tettrahedron Lett., 1995, 36, 2191-2194. 58. K. C. Nicolaou, et al., Angew. Chem. Int. Ed. Engl., 1996, 35, 889-891.
59. N. I. Sax Dangerous properties of Industrial Materials, 6th Ed. ed.; Van Nostrand Reinhold Comp. Inc.: New York, 1984.
60. E. Vedejs, A. R. Haight and W. O. Moss, J Am. Chem. Soc., 1992, 114, 6556-6558. 61. B. E. Maryanoff and A. B. Reitz, Chem. Rev., 1989, 89, 863-927.
62. M. Ögren. Doctoral Thesis, Acta Universitatis Upsaliensis No. 292, Uppsala 1997. 63. J. Buddrus and W. Kimpenhaus, Chem. Ber., 1973, 106, 1648.
64. Á. Díaz-Ortiz, P. Prieto and E. Vázquez, Synlett, 1997, 269-270. 65. M. Larhed and A. Hallberg, J. Org. Chem., 1996, 61, 9582-9584.
66. D.-R. Hwang, L. Moerlein and M. J. Welch, J. Chem. Soc. Chem. Commun., 1987, 1799.
67. S. Stone-Elander, et al., J. Labelled Compd. Radiopharm., 1994, 34, 949. 68. R. F. Heck, Acc. Chem. Res., 1979, 12, 146-151.
69. F. E. Goodson, T. I. Wallow and B. M. Novak, J. Am. Chem. Soc., 1997, 119, 12441-12453.
70. B. E. Segelstein, T. W. Butler and B. L. Chenard, J. Org. Chem., 1995, 60, 12-13. 71. C. Ziegler, B. Jr. and R. F. Heck, J. Org. Chem., 1978, 43, 2941-2946.
72. T. Lawrence, C. A., F. Roughton, J., W.,, W. Rooth, J., and M. Gregerson, I.,, Am.
J. Physiol., 1945, 145, 253.
73. D. Y. Tang, et al., J. Labeled Compd. Radiopharm., 1979, 16, 435-440. 74. P. J. Kothari, et al., Int. J. Appl. Radiat. Isot., 1985, 36, 412-413.
75. M. R. Kilbourne, P. A. Jarabek and M. J. Welch, J. Chem. Soc., Chem. Commun., 1983, 861-862.
76. P. Lidström, T. Kihlberg and B. Långström, J. Chem. Soc., Perkin Trans. 1, 1997, 2701-2706.
77. P. Lidström. Doctoral Thesis, Acta Universitatis Upsaliensis No. 304, Uppsala 1997.
78. K. Kikukawa, T. Totoki, F. Wada and T. Matsuda, J. Organomet. Chem., 1984,
270, 283-287.
79. I. Pri-Bar and O. Buchman, J. Org. Chem., 1984, 49, 4009-4011. 80. U. S. von Euler, Arch. Exp. Pathol. Pharmacol., 1934, 175, 78. 81. J. R. Vane, Angew. Chem., Int. Ed. Engl., 1983, 22, 741-804. 82. P. W. Collins and S. W. Djuric, Chem. Rev., 1993, 93, 1533-1564.
83. E. J. Corey, N. M. Weinshenker, T. K. Schaaf and W. Huber, J. Am. Chem. Soc., 1969, 91, 5675.
84. M. Suzuki, et al., Tetrahedron, 1990, 46, 4809-4822. 85. S. Moncada, Br. J. Pharmacol., 1982, 76, 3-31.
86. G. Hertting and A. Seregi Formation and function of eicosanoids in the central
nervous system; New York Academy of Science: New York, 1989; Vol. 559.
87. R. J. Grylewski, Biochem. Pharmacol., 1979, 28, 3161-3166. 88. V. T. Miller, et al., Neurology, 1984, 34, 1431-1435.
89. H. Takechi, et al., J. Biol. Chem., 1996, 271, 5901-5906. 90. Y. Watanabe, et al., J. Neurochem., 1999, 72, 2583-2592.
![[ C]CarbonMonoxideinRhodium-/Palladium-MediatedCarbonylationReactions 11](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAAAICRAEAOw==)