Syntheses of Precursors and Reference Compounds of the Melanin-Concentrating Hormone Receptor 1 (MCHR1) Tracers [11C]SNAP-7941 and [18F]FE@SNAP for Positron Emission Tomography

The MCH receptor has been revealed as a target of great interest in positron emission tomography imaging. The receptor′s eponymous substrate melanin-concentrating hormone (MCH) is a cyclic peptide hormone, which is located predominantly in the hypothalamus with a major influence on energy and weight regulation as well as water balance and memory. Therefore, it is thought to play an important role in the pathophysiology of adiposity, which is nowadays a big issue worldwide. Based on the selective and high-affinity MCH receptor 1 antagonist SNAP-7941, a series of novel SNAP derivatives has been developed to provide different precursors and reference compounds for the radiosyntheses of the novel PET radiotracers [11C]SNAP-7941 and [18F]FE@SNAP. Positron emission tomography promotes a better understanding of physiologic parameters on a molecular level, thus giving a deeper insight into MCHR1 related processes as adiposity.

In particular, this paper focuses on the synthesis of the novel MCHR1 PET tracers' 1a and 4a, non-radioactive reference compound FE@SNAP 4 as well as the precursors SNAP-acid 2 and TOE@SNAP 3, which represents the preliminary non-radioactive work paving the way for the subsequent radiosyntheses [15,16]. Compounds 2, 3, and 5 can either serve as precursors for radioactive labeling or regarding 3 for non-radioactive fluorination. The reference compounds 1, 4, and 6 serve as standards for the quality control of the radiosyntheses. Regarding the tracer [ 11 C]SNAP-7941 (1a), racSNAP-7941 1 [10][11][12][13][14] can be used as a reference compound. In-vivo studies, biodistribution, and micro PET investigations of the radiotracers [ 11 C]SNAP-7941 1a and [ 18 F]FE@SNAP 4a are going to be future challenges directly based on this work.

Results and Discussion
All SNAP derivatives and intermediates were produced as racemates, deviating from Borowsky et al [1]. The complete reaction sequence is depicted in scheme 1-14. Instead of using methoxymethyl acetoacetate as a starting material for the subsequent Biginelli cyclization, a series of different beta-ketoesters 8-13 carrying different protecting groups for easier cleavage was synthesized (Scheme 1).
As shown in Schemes 2, 4, 5, and 6, the synthesis of racSNAP-7941 1 was accomplished according to the literature without any modifications to the reaction conditions [17]. Derivatives 29-34 have been substituted with different protecting groups instead of the methyl ester moiety. The Biginelli cyclization reaction was conducted based on an alternative method of Murali Dhar et al. [18]. SNAP derivatives 29-32 were used for the synthesis of the precursor SNAP-acid 2, compounds 33 and 34 served as starting material for the hydroxyethyl derivative 35, as depicted in Scheme 10.
The syntheses leading to 2 and the allyl protected derivatives 11, 18, 25, and 32 were performed as already described by Philippe et al. [15], as were those of compounds 3 and 4 [16]. The syntheses of the already known compounds 1, 14, 21 and 28 were carried out according to Schönberger [17]. For completeness of this paper, they are depicted in Schemes 2 and 4-6 as well.
In the next step, a Biginelli reaction was performed using urea, the respective beta ketoesters 8-13 or methoxymethyl acetoacetate, and difluorobenzaldehyde as starting materials, followed by addition of copper oxide, acetic acid, and boron trifluoride diethyl etherate in THF. The mixtures were refluxed for 8 hours to give the seven different pyrimidinones 14-20 (Scheme 2).

Scheme 2. Biginelli cyclizations.
Figure 2/Graph 2 shows a comparison of the different yields of pyrimidinones 15-20 related to the protecting groups. Cyclization using the t-butylester 10, for example, gave only 30% of the corresponding pyrimidinone 17, whereas the best yields were accomplished using the allyl protected ester 11. Allyl pyrimidinone 18 was obtained in an excellent 90% yield. Unfortunately, while reacting the t-butyltrimethysilyloxyethyl protected ester 12 the protecting group was cleaved during the cyclization step, affording hydroxyethyl pyrimidinone 12a as shown in Scheme 3. Hence, the protecting group had to be reattached in an additional step.
Compounds 29-32 were subjected to cleavage reactions in order to obtain SNAP-acid 2. Unfortunately, only the t-butyl protected compound 31 and the allyl protected compound 32 could be converted into the free carboxylic acid 2 (Scheme 7).
In total, regarding the superior yields of 32 as shown in Figure 2, the synthesis of allyl ester 32 was established as the most effective route of preparing the PET precursor SNAP-acid 2. Additionally, allyl ester 32 served as starting material for the hydroxyethyl ester HE@SNAP 35 as well as for the hydroxypropyl ester HP@SNAP 36, which were subjected to tosylation for subsequent fluorination (Scheme 8). The tosylated compounds 3 and 5 were prepared as two alternative precursors of the desired target compounds 4 and 6, in order to compare the feasibility of fluorination of the tosyl ethyl derivative 3 to the tosyl propyl derivative 5.

Scheme 8. Synthesis of hydroxylethyl and hydroxypropyl esters 35 and 36.
The synthesis of 35 required three reaction steps, starting with the oxidation of the allyl protecting group using osmium tetroxide performed by adapting and combining different methods [19][20][21]. This yielded 2,3-dihydroxypropyl ester 32a, as depicted in Scheme 9.
Then, a glycol cleavage of the 2,3-dihydroxypropyl group was performed with sodium periodate adapting methods of Botti et al. [22] and Adam et al. [23] to yield aldehyde 32b, which was subjected to reduction under standard conditions with sodium borohydride [24] to give 2-hydroxylethyl ester 35.
The hydroxypropyl analogue 36 was synthesized in a one-pot two-step reaction as shown above in Scheme 8 adapting the methods of Heidecke/Lindhorst and Park et al. [25,26]. The cleavage to hydroxypropyl ester 36 was accomplished in an anti-Markovnikov reaction using a borane-tetrahydrofuran complex and hydrogen peroxide. Although the unconsumed starting material could be partially recovered by column chromatography, the reaction afforded only a moderate 26% yield. A second and third approach to HE@SNAP 35 was made accessible by the cleavage of the protecting groups of SNAP derivatives 33 and 34, respectively (Scheme 10). The protecting group of compound 33 was cleaved in a standard procedure [27] using tetrabutyl-ammonium fluoride, while the allyloxyethyl ester 34 had to be isomerized first with a Wilkinson catalyst as depicted in Scheme 11. Isomerization of the allyl group was conducted in the presence of diazabicyclooctane and the catalyst, adapting a method of Smith et al. [28]. Mercury-induced cleavage of the newly formed vinyl ether [29] give satisfying yields, regarding the feasibility of recycling the starting material. Scheme 11. Reaction sequence to hydroxyethyl ester 35.
The tosylated derivatives 3 and 5 were intended to be used for the following fluorination to afford the final compounds 4 and 6 (Scheme 13).

Scheme 13. Conversion of tosylated SNAP precursors 3 and 5.
Unfortunately, different fluorination methods such as reactions with tetrabutylammonium fluoride [30], crown ether Kryptofix® K2.2.2 and potassium fluoride [34], or tetrabutylammonium-(triphenylsilyl) difluorosilicate [35] were unsuccessful. Minor yields of 4 (2%-3%) could be obtained by fluorination with cesium fluoride [36] and tetrabutylammoniumhydrogen difluoride [37]. Conversion of compound 5 to fluoropropyl ester 6 under similar reaction conditions was confirmed by high resolution mass spectrometry (HRMS) analysis but purification and isolation could not be accomplished due to the probable instability of this product. Attempting to react the hydroxyethyl derivative 35 with diaminosulfur trifluoride by adapting a method of Shanab [30] did not provide the fluoroethyl ester 4 either.
Acid 2 was reacted with dicyclohexylcarbodiimide, 4-dimethylaminopyridine, and fluoroethanol or fluoropropanol, respectively, giving the fluoroethylated reference compound 4 in trace yields of 4% but again did not afford 6 in acceptable quantity, although the conversion was confirmed via mass spectrometry. Fluoropropyl ester 6 could not be isolated by different chromatographic purification methods. Therefore, the synthesis of the propylated compounds 36, 5, and 6 was not further pursued due to the better yields and superior purification properties of the ethylated compounds 35, 3, and 4.

General
All commercial chemicals and solvents used in the synthetic steps were purchased from Aldrich (Vienna, Austria) or Fisher Scientific (Vienna, Austria) and used as received. Reactions were monitored by thin layer chromatography (TLC) using appropriate developing solvents and pre-coated silica gel plates (UV 254 nm) purchased from Merck and Co. (Vienna, Austria). 1 H-NMR and 13 C-NMR spectra were recorded on a Bruker Avance DPX-200 spectrometer,a Varian UnityPlus 500 spectrometer or a Bruker Avance 500 spectrometer. Chemical shifts are reported in δ (ppm) relative to tetramethylsilane (TMS) as internal standard and multiplicities are given as singlet (s), doublet (d), quartet (q), multiplet (m) and broad singlet (br s). IR-spectra were recorded on a Perkin Elmer FT-IR Spectrum 1000 spectrophotometer. High resolution mass spectral data were obtained on a Finnigan MAT 8230 or on a Finnigan MAT 900 S. Elemental analyses were performed at the Mass Spectrometry Centre of the Faculty of Chemistry (University of Vienna).
To a solution of 32b (88 mg) in MeOH (3.0 mL), NaBH 4 (6.0 mg, 0.15 mmol) was added in portions under stirring, followed by stirring for another 45 min. The reaction was quenched with water, and the mixture extracted three times with Et 2 O. The organic layer was washed with water. After evaporation of the solvent the crude product was purified via column chromatography (silica gel, eluent: CH 2 Cl 2 /MeOH 10:1) to give 12 mg of 35 (13.3%) as a yellow oil.

Conclusions
Based on the increasing need for antiobesity drugs, two novel PET tracers for the MCHR1 have recently been developed, to investigate the role of the MCHR1 in terms of adiposity. The selective high-affinity MCHR1 antagonist SNAP-7941 1 was used as a promising basis for the development of the PET tracers [ 11 C]SNAP-7941 1a and [ 18 F]FE@SNAP 4a [1,[15][16][17]. This paper focuses on the synthesis of the non-radioactive precursors and reference compounds of [ 11 C]SNAP-7941 1a and [ 18 F]FE@SNAP 4a.
While the racemic receptor antagonist 1 [10][11][12][13][14] itself served as a reference compound for the preparation of [ 11 C]SNAP-7941 1a, a new reference compound had to be synthesized for the novel fluoroethyled tracer [ 18 F]FE@SNAP 4a as already published by Philippe et al. [15,16]. The carboxylic acid derivative SNAP-acid 2 served as precursor for the 11 C-methylation to afford the PET tracer [ 11 C]SNAP-7941, while [ 18 F]FE@SNAP was meant to be obtained in a first approach by 18 F-fluorination of the newly prepared tosylate 3.
The synthesis of these polyfunctional SNAP derivatives comprised many commonly used syntheses, including numerous methods attempting to cleave the methyl ester of 1 to prepare SNAP-acid 2, which unfortunately were unsuccessful. Therefore, new approaches to obtain the desired derivatives accessible had to be established. A rationale for the failure of these demethylation methods could lay: (1) in the electron density which is distributed from the adjacent nitrogen to the carbonyl carbon through the conjugated double bond and thus, hinders the attack of nucleophiles (like OH − ) and (2) in the circumstance that acidic conditions may affect the amide bonds. In summary, due to the failure of cleaving the methyl ester 1, four different protecting groups (carboxyl esters) were chosen, leading to SNAP derivatives 29-32. Finally, the precursor SNAP-acid 2 could be prepared through cleavage of the allyl protecting group of compound 32 [15,16]. Allyl-SNAP 32 was not only obtained in excellent yields compared with the other three protected derivatives, but was also used as starting material for two more SNAP derived compounds 35 and 36.
As precursor for the tosylated compound 3, HE@SNAP 35 was synthesized using three different methods, starting from either 32 or from the new derivatives 33 or 34, respectively. Compound 35 was reacted to tosylate TOE@SNAP 3, which could not be fluorinated in satisfying yields to furnish fluoroethyl ester FE@SNAP 4. Additionally, in order to increase yields and feasibility of the fluorination step, a series of propylated compounds was prepared. Allyl ester 32 was reacted to the hydroxypropyl ester HP@SNAP 35, followed by tosylation giving tosylpropyl ester TOP@SNAP 5. Similarly to fluoroethyl ester FE@SNAP 4, fluorination of 5 provided fluoropropyl ester FP@SNAP 6 only in low yields. The conversion was proved by mass spectrometry, but the isolation of 5 was hampered by decomposition during column chromatography.
Finally, fluoroethylation of the free SNAP-acid 2 (but not fluoropropylation) was achieved via Steglich esterification using DCC and DMAP. Thus, SNAP-acid 2 finally served as precursor for the radiosynthetically produced tracer [ 11 C]SNAP-7941 1a as well as for the non-radioactive reference compound FE@SNAP 4 instead of tosylate 3, which is used as precursor for the tracer [ 18 F]FE@SNAP 4a [15,16]. After radioactive labeling at the Medical University of Vienna [15,16], biodistribution and micro PET experiments will be the next step of this ongoing project, as recently shown by Philippe et al. [40].