A general 11C-labeling approach enabled by fluoride-mediated desilylation of organosilanes

Carbon-11 (11C) is one of the most ideal positron emitters for labeling bioactive molecules for molecular imaging studies. The lack of convenient and fast incorporation methods to introduce 11C into organic molecules often hampers the use of this radioisotope. Here, a fluoride-mediated desilylation (FMDS) 11C-labeling approach is reported. This method relies on thermodynamically favored Si-F bond formation to generate a carbanion, therefore enabling the highly efficient and speedy incorporation of [11C]CO2 and [11C]CH3I into molecules with diversified structures. It provides facile and rapid access to 11C-labeled compounds with carbon-11 attached at various hybridized carbons as well as oxygen, sulfur and nitrogen atoms with broad functional group tolerance. The exemplified syntheses of several biologically and clinically important radiotracers illustrates the potentials of this methodology.

Calculated mass: 146.0. 1 H and 13 C NMR data matched with previously reported results (2). 4-Methoxy-4-oxobut-2-ynoic acid ( 12 C standard 7i) The title compound was prepared according to literature method (3,4): To a dried Schlenk-flask under nitrogen, equipped with a magnetic stirrer and a septum, methyl propiolate (0.84 mL, 10 mmol) was added in 25 mL dry THF and cooled to -78 ºC. n-BuLi (2.5 M in hexane, 4.0 mL, 10 mmol) was added dropwise via a syringe. After being maintained at -78 ºC for 30 min, CO2 was infused into reaction mixture by purging the flask with CO2 (balloon) three times. The reaction mixture was slowly warmed up to r.t. and maintained at r.t. for overnight. The solvent was removed in vacuo. Water (30 mL) was added and the mixture was extracted with hexane (30 mL x 2). The aqueous layer was cooled with an ice water bath and acidified with aqueous HCl (2 M, 20 mL), and then extracted with ethyl acetate (20 mL x 3). The combined organic layers were washed with water and brine, dried over Na2SO4 and filtered. The solvent was removed under vacuum. The residue was purified via flash chromatography (DCM : MeOH = 90 : 10) to provide the product (0.48 g, 37%) as a slightly brown liquid. Because the product owes low boiling point and the vacuum drying method could not entirely remove solvent (MeOH and acetone) from product. We did see the contaminated MeOH and acetone in NMR spectra in final product and the practical yield of product was 35.1% (calculated based upon the integration values of proton NMR). 1 H NMR (500 MHz, CDCl3): δ = 7.14 (brs, 1H), 3.87 (s, 3H). 13 C NMR (125 MHz, CDCl3): δ = 154. 1, 152.1, 76.1, 74.3, 53.6. ESI (+) = 129.0 (M+H) + ; Calculated mass: 128.0. 1 H and 13 C NMR data matched those previously reported results (3,4).

Ethyl 2-(4-isobutylphenyl)acetate
The title compound was prepared according to literature method (14): To a 250-mL round-bottom flask, anhydrous ethanol (80 mL) was added. It was cooled with an ice water bath, and thionyl chloride (1.83 mL, 25 mmol) was slowly added. After 15 min, 2-(4-isobutylphenyl)acetic acid (1.92 g, 10 mmol) was added. The reaction mixture was stirred at r.t. for 3 h and the solution was concentrated to get the title product ethyl 2-(4-isobutylphenyl)acetate (2.20 g, quant.) as a colorless liquid. (The NMR spectra indicated that the final product contains a small amount of EtOH). 1 (15).
The reaction mixture was stirred at r.t. for overnight, making sure to keep it out from direct exposure of light.

General Information for Radiochemistry Experiments
The reagents and solvents used for the synthesis process were generally purchased from Sigma-Aldrich (MO, USA), Thermo Fisher Scientific and TCI America with a minimum of ACS reagent grade, or from ABX Advanced Biochemical Compounds and used without further purification. Solid phase extraction (SPE) cartridges were obtained from Waters (Waters Corporation, MA, USA). Reaction vials were purchased from Biotage (Biotage ® Microwave reaction Kit, 2-5 mL or 0.5 -2 mL). High and low level radioactivity measurements were performed using a Capintec CRC-15 DUAL PET radioisotope dose calibrator (Capintec Inc., NJ, USA). Semi-prep radio-HPLC purification and analytical radio-HPLC analysis was performed using either a VARIAN ProStar HPLC system (Agilent Technologies) equipped with dual Model 210 pumps and a ProStar UV-vis 325 or 1260 DAD VL dual ultraviolet-visible light (UV-vis) detector, or RefractoMax 520 RI detector; Or Agilent 1200 series equipped with Interface35900E multichannel interference. The radioactivity signal was measured with Eckert & Ziegler Flow-Count radio-HPLC detecting system equipped with Bioscan B-FC-3400 PIN diode or a NaI PMT detector.
The HPLC chromatographic data were collected using a ProStar data acquisition system, Galaxie Chromatography Data system or OpenLab system. The desired products were identified by comparing the retention time of peaks from radio-HPLC chromatography with the UV peak coming from co-injected product standard; the retention time had been previously verified using identical conditions.
[ 11 C]CO2 production [ 11 C]CO2 was generated by bombarding N2 gas (360 psi 99.9999% pure N2 doped with 0.5% O2) via the 14 N(p,α) 11 C nuclear reaction using a EBCO TR-19/9 cyclotron. General bombardment conditions: 2 -40 min beam time with 25 µA current (3.7 -44.4 GBq; 100 -1200 mCi). After the bombardment, target gas containing radioactivity was released and delivered to a home-made automated [ 11 C]CO2 purification box for controlled trap and release of [ 11 C]CO2, where the [ 11 C]CO2 was first trapped by a molecular sieve (MS) furnace at room temperature (200 mg Molecular Sieve 13X, 100/120 mesh, SUPELCO). Next, the furnace was heated to 190 o C and [ 11 C]CO2 was released and delivered to the reaction vial using helium flow (10 mL/min). Once the radioactivity collected in the reaction vial plateaued, the delivery was stopped and [ 11 C]CO2 production and collection was done. It took 3 -4 min from end of bombardment (EOB) to finish the collection of [ 11 C]CO2 in the reaction vial.
General procedures of 11 C-carboxylation reaction, determination of radiochemical purities (RCP) and radiochemical yields (RCY) When the [ 11 C]CO2 was ready, to a reaction vial equipped with outlet line (an ascarite trap was attached at the end for trapping escaped [ 11 C]CO2) the [ 11 C]CO2 delivery line with a 4 inch needle was inserted into the anhydrous fluoride reagent and solvent reaction mixture. After confirmation of a stable helium flow (10 mL/min), the organosilane precursor was immediately added. The reaction vial was placed in a dose calibrator for measuring the collected radioactivity. Once the increase of radioactivity in the reaction vial plateaued, the [ 11 C]CO2 delivery line and airflow outlet line were removed immediately. The total activity trapped in the reaction vial was checked again, and recorded as starting radioactivity A0. When dimethylformamide (DMF), dimethylacetamide (DMA) and dimethyl sulfoxide (DMSO), as well as its mixture with tetrahydrofuran (THF) (1/1, v/v) were used as solvents for reactions, the trapping of radioactivity [ 11 C]CO2 was generally efficient and the escaped [ 11 C]CO2 collected by an ascarite trap attached to the reaction vial was usually less than 10%. When only THF or THF/dichloromethane (DCM) (3/1, v/v) was used as the reaction solvent, a cooling bath was used (-70 -0 o C) to keep reaction vial cool and leaked radioactivity was also minimal (< 10%). After maintaining the reaction mixture at the desired temperature with stirring for a certain time frame (normally 2 -10 minutes), the reaction was quenched by addition of an acidic solution (1 mL, CH3CN/H2O/formic acid, 90/9/1; or 0.1 M HCl aqueous solution). The unreacted [ 11 C]CO2 was excluded from the reaction vial by purging with a gentle stream of helium (2 -5 psi) and it was trapped in a second ascarite trap. When the radioactivity collected in this ascarite trap became constant (Aleak), the remaining radioactivity in the reaction vial was again measured and recorded as Aleft. After that, a small portion of solution was removed from reaction vial (0.1 -0.2 mL) and diluted in a sample vial pre-loaded with an acidic solution (1 mL, CH3CN/1% formic acid, 90/10). Next, an analytical sample, which was a mixture of an aliquot of sample solution (usually 10 µL) and a product standard solution (usually 10 µL, 1 mg/mL solution), was injected into HPLC for analysis. The percentage of the radio-peak in radiochromatogram coincident with product reference UV peak was regarded as radiochemical purity (RCP). The radiochemical yield (RCY) was calculated by the equation × 0 × 100%. The A0 and Aleft were decay corrected values. If the reaction mixture was submitted for the purification process (solid phase extraction, anion/cation resins exchange method, semi-prep HPLC, or the combination of two of these methods), the total amount of radioactivity of purified product was recorded as Aprod. The radiochemical yield (RCY, decay corrected) was calculated as 0 × 100%. Molar activity values (Am), decay corrected back to EOB and recorded in GBq/µmol, were determined from the carbon-11 activity in the HPLC product peak and the mass of compound.
Total synthesis times were calculated from time point of finished collection of [ 11 C]CO2 to the end of radiotracer purification process.
3). To a pre-dried R.B. flask (50 mL), cooled with acetone-dry ice bath, MeLi (3.9 mL 1.6 M) was added. After that, isopropenyl acetate (3.12 mmol, 0.34 mL) was added dropwise into flask, with caution. Once this was finished, the reaction solution was maintained at -78 o C for 1 h and was then diluted with THF (4.4 mL) to give fresh prepared lithium enolate reagent (0.376 M).

Radiolabeling and purification:
1) To a pre-dried reaction vial cooled with ice-water bath, enolate reagent (0.5 mL, 0.376 M) was added. When [ 11 C]CO2 was ready, it was delivered by bubbling into the enolate solution using helium flow (10 mL/min).
When the radioactivity in reaction vial plateaued, the reaction vial was removed from the ice-water bath and maintained at r.t. for 6 min while stirring.
2) The reaction solution was diluted with H2O (15 mL) and the formed mixture was passed through DOWEX and AG 1X-8 resin columns consecutively. The column containing AG 1X-8 resin was further washed with H2O (10 mL) and the radioactivity trapped in it was eluted with citrate buffer (2.5 mL) and the acidic crude product solution was collected in a plastic tube (15 mL).
3) Unreacted [ 11 C]CO2 dissolved in crude product solution was excluded from citrate elute by flushing this solution with a stream of N2 (5 psi) until no radioactivity increase was found in the ascarite trap (usually ≤ 3 min) attached at the end of an air-flow line. The radioactivity left in this plastic tube was measured and regarded as the final product. If necessary, this solution was passed through a 0.22 µm Millex GV filter for sterilization and the collected solution in the sterile vial was the final drug product (FDP).

Reaction parameters investigated in preliminary research:
Optimized synthetic protocol for synthesis of [ 11 C]3: General procedure of pre-labeling work: 1). CsF stock solution: A stock solution was prepared by dissolving CsF( 1.52 g, 10 mmol) in H2O (12 mL) and stored in a FALCON ® plastic tube (15 mL).
2). Azeotropic drying of CsF: The CsF stock solution (0.3 mL, 0.25 mmol) and CH3CN (1 mL) was added to a microwave glass vial (Biotage, 2 -5 mL, RV01) and the solvent was evaporated by heating at 120 °C under a mild argon stream. After most of the solvent was evaporated, CsF reagent was further dried azeotropically with CH3CN (2 × 1 mL). Next, RV01 was heated at 140 o C under vacuum for 0.5 h to remove trace amount of H2O from CsF. Once the drying process was complete, RV01 was cooled to room temperature under the argon atmosphere, solvent (0.25 mL DMF + 0.25 mL THF) was added and the glass vial was subjected to vigorously vortex (15 s), and it was ready for radiolabelling reaction.

Radiolabeling and purification:
1) To a reaction vial (RV01) pre-cooled with a dry ice-acetone bath, (isopropenyloxy)trimethylsilane stock solution (0.25 mL, 0.25 mmol) was added. Next, [ 11 C]CO2 was delivered into this vessel using helium flow (10 mL/min). When the radioactivity in RV01 plateaued, the dry ice-acetone bath was removed and the reaction mixture was heated at 40 o C for 5 min with stirring.
2) The reaction mixture was diluted with H2O (15 mL) and was passed through DOWEX and AG 1X-8 resin columns consecutively. The column containing AG 1X-8 resin was further washed with a portion of H2O (10 mL) and the radioactivity trapped in it was eluted with citrate buffer (8 mL). After being passed through an Al-N plus cartridge, the acidic crude product solution was collected in a plastic tube (15 mL

2-Phenylacetic-1-[ 11 C]acid, ([ 11 C]13c) synthesis
Following the procedures of pre-labeling work and 11 C-carboxylation reaction described above, [ 11 C]PAA, [ 11 C]13c, was synthesized (see above scheme for details) and the detailed results are listed in the table below:

General procedures of 11 C-methylation reaction, determination of radiochemical purities (RCP) and radiochemical yields (RCY)
The FDSM 11 C-methylation experimental process was the same as FDSM 11 C-carboxylation reactions except for that there was no unreacted [ 11 C]CO2 exclusion process since unreacted [ 11 C]CH3I or by-product [ 11 C]CH3OH dissolved well in the reaction mixture and there was no leakage of radioactivity detected during the sampling process after the 11 C-methylation reaction.
When [ 11 C]CH3I or [ 11 C]CH3OTf was ready, to a reaction vial equipped with outlet line, the radioactivity delivery line with a 4" needle was inserted into the anhydrous fluoride reagent and solvent reaction mixture. The organosilane precursor was immediately added. The reaction vial was placed in a dose calibrator for measuring the collected radioactivity. Once the increase of radioactivity in the reaction vial plateaued, the [ 11 C]CH3I or [ 11 C]CH3OTf delivery line and airflow outlet line were removed. The total activity trapped in the reaction vial was checked again as starting radioactivity A0. After keeping the reaction mixture stirring under desired temperature for a certain time (generally 2 -10 minutes), the reaction mixture was measured again to have total radioactivity A0' (The value of A0' was supposed to equal as A0 after decay correction since unreacted CH3OH dissolved well in the reaction mixture. However, it was found that the value of A0' was slightly less than the value of A0 after decay correction in some experiments. It is most likely because of the small leakage of  C]CH3OH from reaction vial.) A small portion of reaction mixture (~ 0.1 mL) was removed and diluted with an acidic solution (CH3CN/1% formic acid, 90/10) in a septa cap sealed glass vial; the radioactivity of this sample was counted and recorded. Next, an analytical sample, which was a mixture of an aliquot of sample solution (usually 10 µL) and a product standard solution (usually 10 µL, 1mg/mL solution), was analyzed by analytical HPLC (A-HPLC). The percentage of radio-peak in radio-chromatogram coincident with product reference UV peak was regarded as radiochemical purity (RCP) and also as radiochemical yield (RCY, decay corrected) if A0' = A0, or RCY = RCP × [A0'/A0] at the case of A0' < A0. If the reaction mixture was submitted for the purification process (solid phase extraction, semi-prep HPLC, anion/cation exchange method, or the combination of two of these methods), the total amount of radioactivity of purified product was recorded as Aprod.
The radiochemical yield (RCY, decay corrected) was calculated as 0 × 100%. Molar activity values (Am), decay corrected back to EOB and recorded in GBq/µmol, were determined from the carbon-11 activity in the HPLC product peak and the mass of compound. Total synthesis times were calculated from EOB to the end of radioactive product collection after the purification.

1-(Methyl-11 C)-1H-imidazole, ([ 11 C]16m) synthesis
Following the procedures of pre-labeling work and 11 C-methylation reaction described above, [ 11 C]N-Me-Imi ([ 11 C]16m) was synthesized (see above scheme for details) and the detailed results are listed in the table below:
Since the amount of [ 12 C]13g in final product solution (100 µL injection per sample) couldn't be determined under the above conditions ( Figure S67), a higher concentration of the analyte was required. Thus, the decayed product fraction (1.5 mL) was basified with NaOH (0.1 M, 300 µL), and was concentrated by lyophilization. The white solid residue was re-dissolved with HCl (0.6 M, 150 µL). The obtained aqueous solution was analyzed by analytical HPLC (sample injected 100 µL) using the isocratic method described above. The UV absorbance was compared to a standard curve of UV absorbance versus mass of authentic [ 12 C]13g. The standard curve was generated by integration of the UV absorbance signal of three different known amounts of authentic