The facile synthesis of a pyrimidinyl sulfonamide ( N,N,N ,6- tetramethyl-2-(4-nitrophenylsulfonamido)pyrimidin-4-ammonium chloride) as a PET tracer precursor

We synthesized the novel sulfonamide derivative, N , N , N ,6-tetramethyl-2-(4-nitrophenylsulfonamido)pyrimidin-4-ammonium chloride ( 7 ), for use as a precursor in [ 18 F]fluoride chemistry. For this, 4-nitrobenzenesulfonyl guanidine was used as the starting material for condensation with ethyl acetoacetate. The 19 F-compound could be synthesized in four steps with moderate yields. The 18 F-labelled analogue, which has high potential as an intermediate for a positron emission tomography tracer, was produced with a radiochemical yield of 58±12% (n = 4, decay corrected).


Introduction
In order to develop an appropriate treatment for an infection, it is essential to identify the causing agent.Our aim is to develop PET tracers that can detect bacterial foci.An effective tracer should have appropriate pharmacokinetic properties (i.e, absorption, distribution, metabolism, and excretion) and a high target affinity. 1Antibacterial sulfonamides, which are second-line antibiotics for the treatment of urinary tract infections, meningitis, and toxoplasmosis, 2 may be suitable for 18 Flabeling experiments.Because of selective binding (i.e, to dihydropteroate synthetase during folic acid biosynthesis), the tracer would allow for the specific detection of bacteria.
Fluorine-18 is a commercially available isotope with ideal properties for PET experiments as it has a low maximum energy (yielding high resolution images) and an effective half-life of 109 minutes.In most cases, 18 F-labelled tracers are synthesized via nucleophilic substitution.The precursors are therefore synthesized with highly effective leaving groups, such as triflate, tosylate, or trimethylammonium halides.We focused on the short-route synthesis of such sulfonamide precursors and started with 2-amino-6-chloropyridine, isocytosine, and 2-amino-4-chloro-6methylpyrimidine as substrates for reaction with 4-nitrobenzenesulfonylchloride.The desired products have either a chlorine substituted heteroaromatic system or, in the case of isocytosine, an easily transformable OH-group.There are some examples [3][4][5][6] which describe the possibility of exchanging the chlorine on pyridine moieties or analogues with [ 18 F]F -in order to synthesize tracers for PET experiments (for a review of this chemistry, see Dollé 7 ).The nitro group of the 18 F-labelled compound will ultimately be reduced to an amino moiety to allow for binding to bacterial enzymes.

Results and Discussion
Synthesis from 4-nitrobenzenesulfonyl chloride and heteroaromatic compounds.We initially considered compounds which may easily be synthesized from commercially available starting materials.The derivative 1, which was synthesized from 2-amino-6-chlorpyridine and 4nitrobenzenesulfonyl chloride in pyridine, was the first candidate.Such -chloropyridines have already been transformed into their respective fluoro compounds 8 (for a review see the article of Gakh and the references herein 9 ).When 1 was reacted with fluoride under conditions for nucleophilic substitution (dry, 2,2,2-Cryptand, 180 °C, 0.5 h), no 19 F-substituted product was observed (Figure 1) and the starting material remained completely unconsumed (HPLC chromatogram not shown).The electron density of the pyridine ring may be too high to allow atom exchange under these conditions and therefore, heterocycles containing more than one nitrogen atom were considered because of the electron withdrawing effect.
We chose isocytosine as the pyrimidine derivative, as we thought it would allow chlorination in exchange of the hydroxy group once the starting material is coupled with the sulfonyl chloride.Instead of this, however, the isocytosine component reacted with pyridine to form the respective 1-(2-aminopyrimidin-4-yl)pyridinium chloride (2, Scheme 1).In analogy to this, attempts to couple 2amino-4-chloro-6-methylpyrimidine with 4-nitrobenzenesulfonyl chloride in pyridine did not result in the desired sulfonamide, but rather the 1-(2-amino-6-methylpyrimidin-4-yl)-pyridinium salt 3. To our knowledge, this phenomenon has only been reported by Price et al. who described bromopyrimidines that did not react to form the desired sulfanilyl amides. 10Scheme 1. Unexpected pyridinium products.
The structures of the unforeseen pyridinium products were confirmed by 2D NMR (Figure 2 and  3a-d).The atoms could be assigned, and all expected correlations were found.The C-H values for 4-nitrobenzenesulfonate were assigned via three-bond HMBC, and compared to values obtained by estimation software (ChemDraw, Cambridge/USA).As pyridine is often the solvent of choice for such reactions, and the analogous synthesis was described, 11,12 these products were unexpected.

Synthesis via condensation of the guanidyl compound
We adopted a new strategy and decided to use 4-nitrobenzenesulfonyl guanidine (4, Scheme 2) as the starting material in a condensation reaction.The initial attempts to synthesize 4 yielded only the guanidinium salt.We succeeded using a fourfold excess of guanidinium chloride and rapid heating of the mixture to reflux after mixing the components.In this way, it was possible to obtain good yields of the desired products.Consecutive steps were performed according to a standard reaction protocol 13 (Scheme 2).Using this precursor (7), 8 was synthesized in 100% yield as a "cold" reference Scheme 2. Synthesis of 8/[ 18 F]-8, a 4-fluorinepyrimidyl-N-sulfonamide derivative.

Semi automated synthesis of N-(4-[ 18 F]fluoro-6-methylpyrimidin-2-yl)-4nitrobenzenesulfonamide ([ 18 F]-8)
For the radiosynthesis a Hotbox III remote controlled synthesis module was used.Reaction times and temperature were adapted to optimize yield and purity of the product.
The overall reaction time for this semiautomated synthesis (Figure 4) of 18 F-labeled 8 was 45 min.The product was manually purified using RP18 cartridges.The average yield of [ 18 F]-8 was 58±12% (n = 4, RCY, decay corrected) in ethanol.The tracer can be dissolved in 0.9% NaCl and a small amount of PEG200 to avoid precipitation.Radiochemical purity and radiochemical identity were monitored via HPLC and TLC and the product was confirmed by overlapping the UV and radioactivity peaks, whereby the latter could be confirmed as slightly delayed due to the installation of detectors (Figure 5).TLC purity (89.2 ± 8.4%, n = 4) was in the same range of HPLC (94.9 ± 3.2%, n = 3).

Conclusions
Although the successful synthesis of sulfonamides has been described a number of times in recent decades, there are still obstacles in synthesizing functionalized halopyrimidines.We found that in isocytosine and 2-amino-4-chloro-6-methylpyrimidine, the OH-or Cl-substituted carbons quaternize more easily than formation of sulfonamide bonds occured.An alternative strategy involving the condensation of the N-guanidinium sulfonamide 4 with ethyl acetoacetate followed by chlorination was successful.Using this strategy, the precursor 7 for [ 19 F] and [ 18 F]fluoride nucleophilic substitution was synthesized with good yields.Both in vitro and in vivo studies of positron emitting compound [ 18 F]-8 are currently being investigated.To be used as a tracer for bacterial infection, the nitro group will need to be reduced.
All HPLC runs were performed under these conditions.Water was deionized and purified by a Purelab Ultra system (Elga Berkefeld GmbH, Celle, Germany).HPLC grade acetonitrile was from Merck (Darmstadt, Germany).Anhydrous pyridine was purchased from Sigma-Aldrich (Taufkirchen, Germany).Unless otherwise stated, chemicals and solvents were of analytical grade and used as received.

g
(1 mmol) 4nitrobenzenesulfonyl chloride were added to a stirred solution of 1.28 g (1 mmol) 2-amino-6chloropyridine in 50 ml pyridine.The mixture was stirred for 3 h at 85 °C and allowed to cool to ambient temperature.The pyridine was evaporated, and 50 ml water were added to remove residual pyridine.The residue was dissolved in 250 ml of dichloromethane and the organic phase was washed with sodium hydrogencarbonate (80 ml), brine (80 ml) and dried over sodium sulfate.TLC gave no evidence of starting material.(2-Amino-6-methylpyrimidin-4-yl)pyridinium 4-nitrobenzenesulfonate (3).A stirred solution of 0.71 g (0.5 mmol) 2-amino-4-chloro-6-methylpyrimidine in 30 ml pyridine was cooled to -20 °C, and 1.11 g (0.5 mmol) of 4-nitrophenylsulfonyl chloride were added.The solution was then allowed to warm to ambient temperature over 2 h and stirred for another 6 h, after which analytical HPLC showed complete consumption of the starting material.The pyridine was evaporated, and residual pyridine was removed by the addition of water (30 ml) and evaporation.This was repeated twice.The residue was mixed with 3 ml n-propanol, warmed to ambient temperature, and cooled to 0 °C.The precipitate was collected on a glass frit, washed with n-propanol, and dried at 10 mBar.The yield was 1.05 g (54.0%).HRMS: 187.0977 M + , calc.for C10H11N4 + 187.0978, m/z: 187.098 (100.0%),188.102 (10.8%), 188.095 (1.5%). 1

N-(4-Hydroxy-6-methylpyrimidin-2-yl)-4-nitrobenzenesulfonamide (5).
A mixture of 3.12 g (12.8 mmol) 4-nitrobenzenesulfonyl guanidine (4) and 15 ml ethyl acetoacetate was refluxed at 170-180 °C for 1.5 h.The condenser was removed after 30 min to evaporate alcohol and water formed during the reaction.The resulting red solution was cooled to ambient temperature and allowed to crystallize overnight.The precipitate was collected on a glass frit.The product was washed with a small amount of methanol and was evaporated to dryness.The reaction yielded 2.12 g (53.4%) of a white product.HRMS: 309.0296 [M-H] -, calc.for C11H9N4O5S 309.0299. 1

N-(4-Chloro-6-methylpyrimidin-2-yl)-4-nitrobenzenesulfonamide (6).
A mixture of 2.05 g (6.61mmol) N-(4-hydroxy-6-methylpyrimidin-2-yl)-4-nitrobenzenesulfonamide and 10 ml phosphoryl chloride was refluxed for 30 min with stirring.After cooling, 100 g of ice was added in portions.The red mixture became colourless and a precipitate was formed, which was collected on a glass frit.The precipitate was dissolved in 30 ml of 0.05 M sodium hydroxide.The brown solution was then cooled on ice and acidified with diluted acetic acid (10% v/v) with stirring.The precipitate was separated on a glass frit, washed thoroughly with water, and dried in a desiccator over potassium hydroxide yielding 1.925 g (88.5%) of the brownish product.HRMS: 326.9956 [M-H] -, calc.for C11H8ClN4O4S 326.9960.

Semiautomated synthesis of N-(4-fluoro-6-methylpyrimidin-2-yl)-4-nitrobenzenesulfonamide ([ 18 F]8).
For labelling experiments, a Scintomics synthesis module was used.Drying and labelling were controlled by Scintomics software adapted for this application.[ 18 F]fluoride in H2 18 O was purchased from f-con (Holzhausen, Germany) and was produced in the PET center of the Central Clinic in Bad Berka Germany.The activity was trapped on a QMA cartridge (SPE cartridge Chromafix 30-PS-HCO3, preconditioned with 5 ml of 0.5 M potassium carbonate and 2 x 5 ml of water).The activity was eluted from the resin with a solution of 3 mg K2CO3 and 20 mg Kryptofix 222 (synthesis grade; Merck) in 1 ml water/acetonitrile 0.3/0.7 (v/v) into a 5 ml conical glass vial (Wheaton V-Vial, 20 x 65 mm, purchased from Neolab, Heidelberg), which was preheated to 200 °C.The solvent was removed under a stream of nitrogen.After 2 min, a portion of 0.7 ml acetonitrile (Sigma-Aldrich, anhydrous) was added to the residue, and it was dried for another 3 min.A second portion of 0.7 ml acetonitrile was added, and after complete evaporation the system was cooled to 80 °C.Next, a solution of 1-2 mg 7 in 0.5 ml DMSO was added.This mixture was then heated (preset of 80 °C) in a closed vial and in 3 min intervals, a nitrogen stream mixed the solution.After a few minutes, the solution became visibly red.The overall reaction time was 20 min, and the mixture became dark red.The product was purified manually with a RP18 cartridge.The DMSO solution (~0.5 ml) was diluted with 5 ml water and the mixture was applied to a RP18 cartridge (Waters Corporation), which was activated with 10 ml of DMSO and 10 ml of a 1:9 (v/v) DMSO-water mixture.The cartridge was washed with 5 ml water.The product was eluted with 1.5 ml of ethanol.The radiochemical yield was up to 74% with radiochemical purity of 91%-99% (see HPLC conditions above).Radio TLC (on unmodified silica gel 60 F254, VWR, Germany) was performed using ethyl acetate/methanol 9:1 yielding a radioactive peak at RF=0.57.