Radical C–H 18 F-Difluoromethylation of Heteroarenes with [ 18 F]Difluoromethyl Heteroaryl-Sulfones by Visible Light Photoredox Catalysis

: The 18 F-labeling of CF 2 H groups has been recently studied in radiopharmaceutical chemistry owing to the favorable nuclear and physical characteristics of the radioisotope 18 F for positron emission tomography (PET). Following up on the reported efficiency of the F]difluoromethyl benzothiazolyl-sulfone ( [ 18 F]1 ) as a 18 F-difluoromethylating reagent, we investigated the influence of structurally-related [ 18 F]difluoromethyl heteroaryl-sulfones in the reactivity toward the photoredox C–H 18 F-difluoromethylation of heteroarenes under continuous-flow conditions. In the present work, six new [ 18 F]difluoromethyl heteroaryl-sulfones [ 18 F]5a – [ 18 F]5f were prepared and, based on the overall radiochemical yields (RCYs), three of these reagents ( [ 18 F]5a , [ 18 F]5c , and [ 18 F]5f ) were selected for the fully automated radiosynthesis on a FASTlab TM synthesizer (GE Healthcare) at high level of starting radioactivity. Subsequently, their efficiency as 18 F-difluoromethylating reagents was evaluated using the antiherpetic drug acyclovir as a model substrate. Our results showed that the introduction of molecular modifications in the structure of [ 18 F]1 influenced the amount of fac -Ir III (ppy) 3 and the residence time needed to ensure a complete C– H 18 F-difluoromethylation process. The photocatalytic C–H 18 F-difluoromethylation reaction with the reagents [ 18 F]5a , [ 18 F]5c , and [ 18 F]5f was extended to other heteroarenes. Radical-trapping experiments demonstrated the likely involvement of radical species in the C–H 18 F-difluoromethylation process. at position 5 or 6 of the benzothiazolyl ring, and modification of the original benzothiazolyl moiety to other heteroaryl rings ( isolated products. analyses were performed at 45 °C using an ACQUITY UPLC ® CSH TM C18 column (2.1 × 100 mm, 1.7 µ m; Waters, Milford, MA, USA) on an ACQUITY UPLC ® system with a mobile phase of MeCN and HCO 2 H/H 2 O (0.05%, v/v) in gradient mode at 0.5 mL·min − 1 (gradient A: from 100% HCO 2 H/H 2 O (0.05%, v/v) to 75% MeCN + 25% HCO 2 H/H 2 O (0.05%, v/v) in 6 min, and from 75% MeCN + 25% HCO 2 H/H 2 O (0.05%, v/v) to 100% H 2 O in 2 min; gradient B: from 100% HCO 2 H/H 2 O (0.05%, v/v) to 100% MeCN in 6 min, and from 100% MeCN to 100% HCO 2 H/H 2 O (0.05%, v/v) in 2 min). The UV signal of the newly synthesized 18 F-labeled compounds was measured at 254 nm with a photodiode array (PDA) UV detector (190–400 nm) controlled by the Empower software and connected to the UPLC system. A thallium-activated sodium iodide (NaI(Tl)) scintillation detector from Eberline (Eberline Instruments Corp, Miami, FL, USA) was used to monitor the radio-UPLC elution profile of the newly synthesized 18 F-labeled compounds. TLC analyses were carried out on silica gel Polygram ® SIL G/UV 254 pre-coated TLC-sheets (TLC eluent: methanol) (Macherey-Nagel, Düren, Germany). The TLC profile of 18 F-labeled compounds was then analyzed with a BertHold TLC scanner model AR2000 (BertHold, Bad Wildbad,


Introduction
The fluorine-18 ( 18 F) isotope has been regarded the "radionuclide of choice" due to its suitable physical and nuclear features for in vivo positron emission tomography (PET) imaging in living subjects [1][2][3][4]. The unique sensitivity of PET makes this technique appropriate for the study of absorption, distribution, metabolism, and excretion (ADME) properties of radiopharmaceuticals and the evaluation of their pharmacodynamic profile. In addition, PET technology has proven highly valuable in the observation of biochemical and physiological changes that may take place before the anatomical alterations of a certain disease are detected [5][6][7][8]. The suitability of the 18 F radioisotope in PET has encouraged radiochemists to invest much effort in the development of efficient 18 Ffluorination and 18 F-fluoroalkylation strategies [9][10][11][12][13][14][15][16][17][18][19].
Based on the effectiveness of the sulfone [ 18 F]1 as 18 F-difluoromethylating reagent [37,38], we intended to study the influence of certain molecular modifications in the structure of [ 18 F]1 on the reactivity towards the C-H 18 F-difluoromethylation of N-heteroaromatics. The molecular modifications consisted in the introduction of a single electron-donating (OCH3) (Figure 2A) or electron-withdrawing (NO2) substituent ( Figure 2B) either at position 5 or 6 of the benzothiazolyl ring and in the alteration of the original benzothiazolyl moiety to other heteroaryl rings (N-methylbenzimidazolyl and N-phenyl-tetrazolyl rings) ( Figure 2C). In this work, we opted to perform the radiosyntheses of structurally-related [ 18 F]difluoromethyl heteroaryl-sulfones and subsequently evaluate their efficiency in the photoredox C-H 18 F-difluoromethylation of heteroarenes, under continuous-flow conditions as reported previously [37]. To the best of our knowledge, the effectiveness of the non-radioactive references of these novel [ 18 F]difluoromethyl heteroaryl-sulfones in photoredox C-H difluoromethylation has never been described in the literature.

Synthesis of the Difluoromethyl Heteroaryl-Sulfones 5a-5f
We initially performed the organic synthesis of the difluoromethyl heteroaryl-sulfones 5a-5f as non-radioactive standards for confirmation of the identity of the 18 F-labeled compounds. In order to prepare the difluoromethyl heteroaryl-sulfones 5a-5f, we considered a two-step procedure involving the difluoromethylation of the heteroaryl-thiols 3a-3f to afford the difluoromethyl heteroarylsulfides 4a-4f. Oxidation of the sulfides 4a-4f would lead to the sulfones 5a-5f (Scheme 1).
Afterward, the oxidation of the cartridge-purified [ 18 Table 2]. These compounds were then selected for the investigation of their reactivity towards the photocatalytic 18 F-difluoromethylation of N-containing heteroarenes. In order to circumvent any potential radioprotection issues, a fully automated process involving the two-step radiosyntheses of [ 18 F]5a, [ 18 F]5c, or [ 18 F]5f and a high performance liquid chromatography (HPLC) purification was implemented on a FASTlab TM synthesizer (GE Healthcare) for the preparation of the purified 18 F-labeled compounds. The automated sequence for the radiosyntheses of  Table S23 and illustrated in Figure  3.     F]5f were isolated in 2.9 ± 0.1%, 5.7 ± 0.5%, and 8.0 ± 0.9% RCYs (decay-corrected at the SOS), respectively ( Table 3). The RCY of each automated radiosynthesis was determined based on the ratio between the radioactivity of the [ 18 F]5a, [ 18 F]5c, or [ 18 F]5f present in the DMSO solution (decay-corrected at the SOS) and the radioactivity trapped on the QMA carbonate cartridge at the SOS.
Obtaining a high molar activity still constitutes a major challenge for the radiosyntheses of [ 18 F]CF2H-bearing compounds, due to the unwanted 18 Table 3 were determined at the end of the synthesis (EOS)]. F]5f towards the C-H 18 F-difluoromethylation of N-containing heteroarenes, under irradiation with blue light-emitting diode (LED) (470 nm, 2 W). The C-H 18 Fdifluoromethylation reactions were performed in continuous-flow using an easy-to-use platform equipped with a 100 µL microreactor made from glass and a syringe that continuously pumps the reaction mixture into the microreactor at a given flow rate (FlowStart Evo, FutureChemistry, Nijmegen, The Netherlands) ( Figure S85). The use of a continuous-flow system assures an efficient irradiation of the reaction mixture during the photocatalytic processes and can potentially lead to an enhanced productivity in significantly reduced reaction times [37]. The reaction time is a relevant parameter in 18 F-radiochemistry. We initially explored the reactivity of the sulfones [ 18 F]5a, [ 18 F]5c, and [ 18 F]5f for the C-H 18 F-difluoromethylation of the antiherpetic drug acyclovir (7e) [48] under the reaction conditions recently reported in our laboratories (  (Table 4, Entry 5). Changing the amount of photocatalyst from 0.05 mol% to 0.1 mol % and the residence time from 2 min (flow rate = 50 μL·min −1 ) to 2.5 min (flow rate = 40 μL·min −1 ) enabled the complete consumption of the reagent [ 18 F]5f and resulted in the production of [ 18 F]8e in 56 ± 1% RCY ( Table 4, Entry 8). These results demonstrated that a single introduction of the -NO2 group (an electron-withdrawing substituent) at position 6 and the alteration of the benzothiazolyl ring of [ 18 F]1 to the N-phenyltetrazolyl moiety yielded 18 F-difluoromethylating reagents with lower reactivity for the C-H 18 Fdifluoromethylation of 7e in comparison with the sulfone [ 18 F]1. Overall, the introduction of molecular modifications in the structure of [ 18 F]1 can modulate the reactivity of the resulting 18 Fdifluoromethylating reagents, influencing the amount of the photocatalyst fac-Ir III (ppy)3 and the residence time necessary to assure a complete C-H 18 F-difluoromethylation reaction.  18 Fdifluoromethylating reagent, the ratio between the distinct isomers was not significantly changed. Besides the antiherpetic drug 7e [48], the C-H 18 F-difluoromethylation procedure was also extended to other heteroarenes of medicinal relevance, in particular to the demethylated derivative of the antihypertensive drug moxonidine (7f) [49] and to the xanthine derivative pentoxifylline (7g) [50].  (Figures S87-S103). To gain insights into the mechanism, the C-H 18 F-difluoromethylation of the substrate 7e was examined. The addition of the radical scavenger 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) to the reaction system completely inhibited the C-H 18 F-difluoromethylation of the substrate 7e, in the presence of the sulfones [ 18 F]5a, [ 18 F]5c, and [ 18 F]5f (Scheme 6, Entry 1). Furthermore, when the model reaction was performed without blue light irradiation (Scheme 6, Entry 2) and photocatalyst (Scheme 6, Entry 3), no desired product [ 18 F]8e was formed. These results suggest the involvement of radical intermediates in the present photocatalytic C-H 18 F-difluoromethylation reaction. On the basis of these observations and previous research works reporting photoredox C-H difluoromethylation reactions with the sulfones 1 [39][40][41][42][43][44] and 2 [45], a general and simplified reaction mechanism is shown in Figure 7. The proposed mechanism for the C-H 18 F-difluoromethylation of heteroarenes 7a-7g involved the reduction of the [ 18

General Procedure for the Synthesis of Difluoromethyl Heteroaryl-Sulfides (4a-4f)
The difluoromethylation of heteroaryl-thiols (3a-3f) was achieved following the slightly modified protocols [46,47]. Sodium chlorodifluoroacetate (915 mg, 6.0 mmol, 2.0 equiv.) and potassium carbonate (622 mg, 4.5 mmol, 1.5 equiv.) were added to a single-neck round-bottom flask with DMF (5 mL) and the resulting suspension was stirred at room temperature for 5 minutes. Afterwards, a solution of the heteroaryl-thiols 3a-3f (3.0 mmol, 1.0 equiv.) in DMF (5 mL) was slowly added. The reaction mixture was stirred at 95 °C for 15 minutes and then cooled down to room temperature. After dilution with H2O (10 mL), the crude product was extracted with DCM (3 × 20 mL). The organic layers were gathered and dried over anhydrous MgSO4. After filtration, the solution was concentrated under reduced pressure and the resulting crude product was purified by flash chromatography as described below. To a round-bottom flask containing the difluoromethyl heteroaryl-sulfides 4a-4f (1.0 mmol, 1.0 equiv.) in MeCN (2 mL) and CHCl3 (2 mL), a solution of sodium (meta)periodate (NaIO4) (1.07 g, 5.0 mmol, 5 equiv.) and ruthenium (III) chloride hydrate (RuCl3·xH2O) (10 mg, 0.05 mmol, 0.05 equiv.) in H2O (4 mL) was added to the reaction system. The resulting reaction mixture was stirred at room temperature for 1 h. After the completion of the reaction, the suspension was diluted with H2O (5 mL) and the crude product was extracted with DCM (3 × 25 mL). The combined organic layers were washed with saturated aqueous solution of NaHCO3 and subsequently dried over anhydrous MgSO4. After filtration, the solvent was evaporated under reduced pressure. The resulting crude product was then purified by flash chromatography (SiO2; heptane/EtOAc (90/10, v/v)) to afford the difluoromethyl heteroaryl-sulfones 5a-5f as pure compounds. The bromofluoromethylation of heteroaryl-thiols (3a-3f) was carried out on the basis of a previously described procedure [37] with slight modifications. A solution of KOH (1.68 g, 30.0 mmol, 10.0 equiv.) in H2O (4 mL) was placed in a single-neck round-bottom flask and stirred at 0°C. Afterwards, a solution of the heteroaryl-thiols 3a-3f (3.0 mmol, 1.0 equiv.) in THF (3 mL) was added and the resulting mixture was allowed to stir at room temperature for 20 min. A solution of dibromofluoromethane (0.713 mL, 9.0 mmol, 3.0 equiv.) in THF (1 mL) was slowly introduced in the reaction system, and the resulting mixture was stirred at room temperature for 15-20 min. The suspension was subsequently quenched by addition of H2O (20 mL), and the crude product was extracted with DCM (3 × 30 mL). The combined organic layers were gathered and were dried over anhydrous MgSO4. After filtration, the solvent was removed under reduced pressure. The purification of the concentrated crude product was performed by flash chromatography (SiO2; heptane/EtOAc (95/5, v/v)) to furnish the bromofluoromethyl heteroaryl-sulfides 6a-6f as pure compounds.  The difluoromethylated heteroarenes 8a-8g were synthesized from the heteroarenes 7a-7g and characterized according to the formerly reported procedures [37,51].

Conclusions
In the present work, the two-step radiosyntheses of the [ 18  F]5f were isolated in reproducible 2.9 ± 0.1%, 5.7 ± 0.5%, and 8.0 ± 0.9% RCYs, respectively (decay-corrected at the SOS). The use of automated synthesizers in radiochemical processes involving the production of multiple GBq of labeled compounds is required to assure the minimization of the radiation exposure to workers.
The great probability of 18  Interestingly, these newly synthesized compounds were revealed to be competent reagents for the C-H 18 F-difluoromethylation of the antiherpetic drug 7e under the reaction conditions recently reported in our laboratories [37]. Still, none of the new reagents performed as good as the original sulfone [ 18 F]1 in the radiosynthesis of the labeled compound [ 18 F]8e. Additionally, the sulfones [ 18 F]5c and [ 18 F]5f exhibited a lower reactivity towards the C-H 18 F-difluoromethylation of 7e, in comparison with the sulfone [ 18 F]1. Overall, the introduction of molecular modifications in the structure of [ 18 F]1 can modulate the reactivity of the resulting 18 F-difluoromethylating reagents, influencing the amount of photocatalyst and the residence time necessary to assure a complete C-H 18 F-difluoromethylation process. Delightfully, the labeled compounds [ 18 F]5a, [ 18 F]5c, and [ 18 F]5f were suitable for the developed flow photoredox C-H 18 F-difluoromethylation of a scope of heteroarenes, demanding a low amount of fac-Ir III (ppy)3 (0.05-0.5 mol %) and short residence times (2-4 min). Radical-scavenging experiments suggested the participation of radical intermediates in the present photocatalytic C-H 18 F-difluoromethylation reaction. To the best of our knowledge, the effectiveness of the nonradioactive references of [ 18 F]5a, [ 18 F]5c, and [ 18 F]5f as difluoromethylating reagents has never been described in visible light photoredox catalysis.