A Convenient Route to New (Radio)Fluorinated and (Radio)Iodinated Cyclic Tyrosine Analogs

The use of radiolabeled non-natural amino acids can provide high contrast SPECT/PET metabolic imaging of solid tumors. Among them, radiohalogenated tyrosine analogs (i.e., [123I]IMT, [18F]FET, [18F]FDOPA, [123I]8-iodo-L-TIC(OH), etc.) are of particular interest. While radioiodinated derivatives, such as [123I]IMT, are easily available via electrophilic aromatic substitutions, the production of radiofluorinated aryl tyrosine analogs was a long-standing challenge for radiochemists before the development of innovative radiofluorination processes using arylboronate, arylstannane or iodoniums salts as precursors. Surprisingly, despite these methodological advances, no radiofluorinated analogs have been reported for [123I]8-iodo-L-TIC(OH), a very promising radiotracer for SPECT imaging of prostatic tumors. This work describes a convenient synthetic pathway to obtain new radioiodinated and radiofluorinated derivatives of TIC(OH), as well as their non-radiolabeled counterparts. Using organotin compounds as key intermediates, [125I]5-iodo-L-TIC(OH), [125I]6-iodo-L-TIC(OH) and [125I]8-iodo-L-TIC(OH) were efficiently prepared with good radiochemical yield (RCY, 51–78%), high radiochemical purity (RCP, >98%), molar activity (Am, >1.5–2.9 GBq/µmol) and enantiomeric excess (e.e. >99%). The corresponding [18F]fluoro-L-TIC(OH) derivatives were also successfully obtained by radiofluorination of the organotin precursors in the presence of tetrakis(pyridine)copper(II) triflate and nucleophilic [18F]F− with 19–28% RCY d.c., high RCP (>98.9%), Am (20–107 GBq/µmol) and e.e. (>99%).


Introduction
Amino acid transporters (AATs) are membrane proteins that supply cells with amino acids (AAs), and particularly essential AAs, to support their metabolism, growth and survival [1]. Christensen et al.'s pioneering work in the 1960s described and classified more than 20 AAT systems according to substrate specificity and transport mechanism [2][3][4][5][6]. More recently, AATs have been included in the Solute Carrier (SLC) superfamily of transporter systems [7][8][9]. AATs are currently included in 11 SLC families, with at least 66 different members. Because of the key functional role played by AATs in biological processes, expression alterations or dysfunctions have been linked with several human pathologies, such as neurodegenerative disorders, chronic kidney disease and cancer [10,11]. Tumor and a high proliferation rate, supported by increased AATs express [12,13]. This characteristic, leading to an intensive accumulation of A compared with normal cells, has been largely exploited in diagnosti Thus, the use of natural or non-natural radiolabeled AAs can provide h Photon Emission Computed Tomography (SPECT) or Positron Emi (PET) imaging of primary cancer lesions and distant metastases in num such as staging, treatment follow-up and early detection of recurren most upregulated AATs in cancer is the LAT1 system (L system, SLC ports large neutral AAs, such as branched-chain and aromatic amino a tophan, phenylalanine or tyrosine, for example) [16][17][18]. In this context radiohalogenated tyrosine analogs (i.e., [ 123 I]IMT, [ 18 F]FET, [ 18 F]FDO TIC(OH), Figure 1) have been developed and thoroughly investigated decades, mostly for imaging neuroendocrine, prostatic and brain tumo compared with their natural counterparts, non-natural AAs are often vivo. This reduces the production of radiometabolites, which could image analysis.   Among all existing halogen radionuclides, fluorine-18 is currently positron-emitting radionuclide for PET imaging due to its highly su nuclear characteristics. This PET radionuclide displays simple decay a erties with high positron abundance (97%). Fluorine-18 has relatively lo (maximum 635 keV) and a short positron linear range in tissue (2.3 mm high-resolution PET images. Its half-life (109.8 min) is long enough to a imaging protocols, therefore facilitating kinetic studies and high-quali yses. Moreover, from a radiochemical point of view, fluorine-18 allo thetic approaches lasting several hours. Finally, fluorine-18 can be rel produced at the multi-Curie level on widely implemented biomedica the well-characterized (p, n) nuclear reaction on an oxygen-18-enriche a relatively low-energy proton beam (e.g., 18 MeV). In addition to fluor radioisotopes of iodine offer a wide range of applications in nuclear me a convenient bridge between animal models and human clinical trials.  Among all existing halogen radionuclides, fluorine-18 is currently the most attractive positron-emitting radionuclide for PET imaging due to its highly suitable physical and nuclear characteristics. This PET radionuclide displays simple decay and emission properties with high positron abundance (97%). Fluorine-18 has relatively low positron energy (maximum 635 keV) and a short positron linear range in tissue (2.3 mm), which results in high-resolution PET images. Its half-life (109.8 min) is long enough to allow relatively long imaging protocols, therefore facilitating kinetic studies and high-quality metabolite analyses. Moreover, from a radiochemical point of view, fluorine-18 allows multi-step synthetic approaches lasting several hours. Finally, fluorine-18 can be reliably and routinely produced at the multi-Curie level on widely implemented biomedical cyclotrons, using the well-characterized (p, n) nuclear reaction on an oxygen-18-enriched water target with a relatively low-energy proton beam (e.g., 18 MeV). In addition to fluorine-18, the multiple radioisotopes of iodine offer a wide range of applications in nuclear medicine and provide a convenient bridge between animal models and human clinical trials. Indeed, iodine-125 (half-life: 59.41 d; γ emitter) is suitable for pre-clinical in vitro experiments and SPECT imaging on small animal models, while iodine-123 (half-life: 13.22 h; γ emitter) and iodine-124 (half-life: 4.17 d; β + emitter) are well suited for clinical SPECT and PET imaging purposes, respectively. The β − -emitting radioisotope, iodine-131 (half-life: 8.03 d), can be used for targeted radionuclide therapeutic approaches.
While radioiodinated derivatives of some tyrosine analogs ([ 123 I]IMT or [ 123 I]8-iodo-L-TIC(OH), for example, Figure 1) can be easily produced under mild conditions via well-known electrophilic aromatic substitution reactions, radiofluorinated aryl tyrosine analogs are particularly difficult to obtain. Indeed, direct radiofluorination of electron-rich aromatic structures from a [ 18 F]F − source is still a challenge for the scientific community of radiochemists, as evidenced by the number of new methods, starting from promising arylboronate, aryl organotin, aryl sulfonium or iodonium salt precursors that have been published in recent years [29][30][31][32]. The significant progress in methodology recently reported in the radiofluorination of [ 18 F]FDOPA, for example, is a perfect illustration of the new opportunities that are now available for producing radiofluorinated arenes that could not be routinely obtained even a few years ago [33][34][35][36][37][38][39][40][41].
Surprisingly, these methodological advances have not been applied to the synthesis of radiofluorinated derivatives of TIC(OH), a cyclic analog of tyrosine, which can be found in several compounds with a wide variety of biological activities ( Figure 2). JDTic (1) is a potent and selective κ opioid receptor antagonist [42], which displays antidepressant and anxiolytic effects and reduces the signs related to substance abuse in a rodent model [43,44]. ZYJ-34c (2) is a histone deacetylase inhibitor, which exhibits higher in vivo antitumor potency in human breast carcinoma and pulmonary metastasis mouse models than the FDA-approved drug, suberoylanilide hydroxamic acid [45]. Compounds 3a and 3b, possessing decarboxylated TIC(OH) cores, have also demonstrated anticancer effects in oestrogen-sensitive breast cancers. These compounds present in vitro dual action as selective oestrogen receptor modulators. Moreover, compound 3a inhibits steroidal sulfatase, while compound 3b targets the alkaline phosphatase enzyme [46]. Compound 4 has shown hypoglycemic and hypolipidemic effects in a mouse model via peroxisome proliferator-activated receptor γ agonism and protein-tyrosine phosphatase 1B inhibition. These results highlighted that compound 4 can be considered as a candidate drug for the treatment of diabetes [47]. The TIC(OH) scaffold is also found in several peptides, such as non-natural conformation-constrained tyrosine analogs. These include the macrocyclic pentapeptide 5 [48], a potent inhibitor of hepatitis C virus NS3 protease, and Skite and Htc-tide peptides, which have been described as peptidic reporters of protein tyrosine kinase activity due to their efficient phosphorylation and slower dephosphorylation than conventional tyrosine reporters [49,50].  [29][30][31][32]. The significant progress in methodology recently reported in the radiofluorination of [ 18 F]FDOPA, for example, is a perfect illustration of the new opportunities that are now available for producing radiofluorinated arenes that could not be routinely obtained even a few years ago [33][34][35][36][37][38][39][40][41]. Surprisingly, these methodological advances have not been applied to the synthesis of radiofluorinated derivatives of TIC(OH), a cyclic analog of tyrosine, which can be found in several compounds with a wide variety of biological activities ( Figure 2). JDTic (1) is a potent and selective κ opioid receptor antagonist [42], which displays antidepressant and anxiolytic effects and reduces the signs related to substance abuse in a rodent model [43,44]. ZYJ-34c (2) is a histone deacetylase inhibitor, which exhibits higher in vivo antitumor potency in human breast carcinoma and pulmonary metastasis mouse models than the FDA-approved drug, suberoylanilide hydroxamic acid [45]. Compounds 3a and 3b, possessing decarboxylated TIC(OH) cores, have also demonstrated anticancer effects in oestrogen-sensitive breast cancers. These compounds present in vitro dual action as selective oestrogen receptor modulators. Moreover, compound 3a inhibits steroidal sulfatase, while compound 3b targets the alkaline phosphatase enzyme [46]. Compound 4 has shown hypoglycemic and hypolipidemic effects in a mouse model via peroxisome proliferator-activated receptor γ agonism and protein-tyrosine phosphatase 1B inhibition. These results highlighted that compound 4 can be considered as a candidate drug for the treatment of diabetes [47]. The TIC(OH) scaffold is also found in several peptides, such as non-natural conformation-constrained tyrosine analogs. These include the macrocyclic pentapeptide 5 [48], a potent inhibitor of hepatitis C virus NS3 protease, and Skite and Htc-tide peptides, which have been described as peptidic reporters of protein tyrosine kinase activity due to their efficient phosphorylation and slower dephosphorylation than conventional tyrosine reporters [49,50]. Samnick et al. also obtained very promising results with the [ 123 I]8-iodo-L-TIC(OH) radiotracer for SPECT imaging of prostatic tumors in preclinical mouse models [51][52][53]. This compound presented high, rapid and long-lasting tumor uptake (>15% injected dose per gram, ID/g, at 15 min p.i.) associated with fast wash-out from non-target organs. However, the impact in terms of structure-activity relationship of the position of iodine substitution on this radiolabeled tyrosine derivative has never been explored. Furthermore, no radiofluorinated analogs of [ 123 I]8-iodo-L-TIC(OH) have been reported so far.
Based on these observations, it would be of great interest to develop radioiodinated isomers at positions 5 and 6 of the TIC(OH) scaffold and to adapt TIC(OH) for 18 F PET imaging. To this end, we have designed and developed a new synthetic pathway, using common organotin intermediates, to enable easy production of radioiodinated or radiofluorinated 5, 6 and 8 aryl-substituted TIC(OH) analogs. Since radiolabeling techniques are common in all iodine radioisotopes and considering the relatively low cost of iodine-  [51][52][53]. This compound presented high, rapid and long-lasting tumor uptake (>15% injected dose per gram, ID/g, at 15 min p.i.) associated with fast wash-out from non-target organs. However, the impact in terms of structure-activity relationship of the position of iodine substitution on this radiolabeled tyrosine derivative has never been explored. Furthermore, no radiofluorinated analogs of [ 123 I]8-iodo-L-TIC(OH) have been reported so far.
Based on these observations, it would be of great interest to develop radioiodinated isomers at positions 5 and 6 of the TIC(OH) scaffold and to adapt TIC(OH) for 18 F PET imaging. To this end, we have designed and developed a new synthetic pathway, using common organotin intermediates, to enable easy production of radioiodinated or radiofluorinated 5, 6 and 8 aryl-substituted TIC(OH) analogs. Since radiolabeling techniques are common in all iodine radioisotopes and considering the relatively low cost of iodine-125 and its favorable half-life and dosimetry, we decided to use this radionuclide as a model for our radioiodination protocols.

Chemistry
A convergent synthetic pathway (Scheme 1) was designed to easily produce the [ 125 I]iodo-L-TIC(OH) radioiodinated tracers, fluoro-L-TIC(OH) reference fluorinated derivatives and [ 18 F]fluoro-L-TIC(OH) radiofluorinated compounds from common key organotin intermediates. These organotin precursors were synthesized from iodinated analogs via a palladium catalyzed I/SnMe 3 exchange reaction. The non-radiolabeled iodo-L/D-TIC(OH) compounds used as references were prepared by hydrolysis of iodinated ethyl ester intermediates. The reference fluorinated derivatives were obtained by regioselective silver-catalyzed electrophilic aromatic fluorination of the key organotin compounds using the mild fluorinating reagent F-TEDA-PF 6 , while their radiofluorinated analogs, [ 18 F]fluoro-L-TIC(OH), were easily produced by copper-mediated nucleophilic aromatic radiofluorination from the organotin intermediates with [ 18 F]F − . Finally, the [ 125 I]iodo-L-TIC(OH) radiotracers were obtained by conventional electrophilic radioiododestannylation under mild conditions. For comparison and control of the enantiomeric excess, the corresponding non-radiolabeled iodinated and fluorinated derivatives from the D series were also synthesized. model for our radioiodination protocols. (R) and (S)-ethyl 7-hydroxy-5-iodo-1,2,3,4-tetrahydroisoquino ((R/S)-13) were successfully obtained in eight steps starting from com D-and L-phenylalanines, respectively (Scheme 2). Briefly, phenylala with an aqueous solution of formaldehyde in the presence of hydrobro to the Pictet-Spengler cyclization [54], to obtain tetrahydroisoquinolin 6. Regioselective nitration of the (R/S)-6 compounds at position 7 wit concentrated sulfuric acid [55] provided the (R/S)-7 compounds. It is that this reaction also led to the formation of minor by-products: the c rated isoquinoline compound and the 6-nitro derivative, which were r purification step. After esterification of the (R/S)-7 acids with thionyl the ethyl esters (R/S)-8 were regioselectively iodinated at position 5 w (R) and (S)-ethyl 7-hydroxy-5-iodo-1,2,3,4-tetrahydroisoquinoline-3-carboxylates ((R/S)-13) were successfully obtained in eight steps starting from commercially available D-and Lphenylalanines, respectively (Scheme 2). Briefly, phenylalanines were treated with an aqueous solution of formaldehyde in the presence of hydrobromic acid, according to the Pictet-Spengler cyclization [54], to obtain tetrahydroisoquinoline derivatives (R/S)-6. Regioselective nitration of the (R/S)-6 compounds at position 7 with sodium nitrate in concentrated sulfuric acid [55] provided the (R/S)-7 compounds. It is worth mentioning that this reaction also led to the formation of minor by-products: the corresponding saturated isoquinoline compound and the 6-nitro derivative, which were removed during the purification step. After esterification of the (R/S)-7 acids with thionyl chloride in ethanol, the ethyl esters (R/S)-8 were regioselectively iodinated at position 5 with iodine(I) trifluoromethanesulfonate generated in situ from N-iodosuccinimide and trifluoromethanesulfonic acid at room temperature [56]. The (R/S)-9 compounds then reacted with acetyl chloride [57] to obtain acetamides (R/S)-10 in good yields (72% and 90%, respectively). Reduction of the nitro group of (R/S)-10 using tin(II) chloride in refluxing ethanol [58] afforded amines (R/S)-11, which were diazotized with sodium nitrite in sulfuric acid [59] to produce phenols (R/S)-12. Finally, prolonged heating in hydrochloric acid [60] provided the fully deprotected compounds 5-iodo-TIC(OH) ((R/S)-14) or esters (R/S)-13 after an additional Fischer esterification step.        The Fischer esterification of (R/S)-15 then produced ethyl esters (R/S)-16 in good yields (82% and 73%, respectively). Monoiodinated compounds at position 6 ((R/S)-17) were obtained from (R/S)-16 by regioselective reduction with zinc/hydrobromic acid in ethanol [62]. The common intermediates (R/S)-16 were also hydrogenated using palladium on charcoal as a catalyst in the presence of trimethylamine [63] to afford the derivatives (R/S)-19. Finally, regioselective iodination at position 8 of (R/S)-19 with N-iodosuccinimide in dichloromethane under ultrasound activation [64] provided the (R/S)-20 compounds. It is worth mentioning that this regioselective monoiodination reaction has to be performed with 0.33 equivalents of N-iodosuccinimide at a concentration not exceeding 1.    phine)palladium(0) in refluxing 1,4-dioxane [33]. Under these conditions, organotin derivatives (S)-25 and (S)-26 were obtained in good yields (72% and 56%, respectively), while only small amounts of the 8-substituted organotin derivative (S)-27 were isolated (yield < 20%). For this compound, TLC monitoring of the reaction highlighted the formation of several byproducts, probably due to thermal decomposition of the organotin product. We therefore tried to synthesize this compound using a slightly modified protocol of Mentzel et al. [65] involving treatment of the iodinated starting material (S)-24 with an isopropylmagnesium chloride lithium chloride complex to generate an organomagnesium intermediate in situ, to enable a low temperature (−40 • C) reaction with trimethyltin chloride. Under these conditions, the organotin compound (S)-27 was successfully produced with a good yield (73%).

Chemistry
The organotin compounds (S)-25, (S)-26 and (S)-27 were then radiolabeled with high molar activity using a radioiodo-demetallation reaction with no-carrier-added [ 125 I]NaI in the presence of Chloramine-T as a mild oxidative agent. To prevent the tert-butoxycarbonyl protecting groups from hydrolyzing under the acidic conditions that are normally used in these radioiodination processes, the pH of the reaction mixtures was maintained at 7 using a phosphate buffer. After optimization of the reaction parameters (i.e., Chloramine-T concentration, temperature, etc.), the corresponding radioiodinated intermediates were successfully produced in 5 min with high labeling efficiencies (>95%). We then turned our attention to the deprotection of these compounds to obtain the desired [ 125 I]iodo-L-TIC(OH) derivatives. The removal of all protecting groups was performed under mild conditions using a two-step deprotection, consisting of saponification of the ethyl ester function, followed by acidic cleavage of the Boc groups. Under these conditions, hydrolysis of the ester function was achieved after 1 h using 10 M aqueous sodium hydroxide solution with limited deiodination (<5%, determined by radio-TLC and analytical radio-HPLC monitoring). The Nand O-Boc protecting groups were then removed by adding an excess of trifluoroacetic acid at 0  Figure 3), molar activities (>1.5-2.9 GBq/µmol) and enantiomeric excess (>99%, Figure S1 in ESI). Interestingly, these three radioiodinated derivatives were found to be stable in the formulation medium for up to 7 days after production.

Synthesis of [ 19 F]fluoro-D/L-TIC(OH) References and [ 18 F]fluoro-L-TIC(OH) Radiotracers
The synthetic pathway to obtain the [ 19   were efficiently isolated with good radiochemical yields (RCY, 51-78%), high radiochemical purities (RCP, >98%, Figure 3), molar activities (>1.5-2.9 GBq/µmol) and enantiomeric excess (>99%, Figure S1 in ESI). Interestingly, these three radioiodinated derivatives were found to be stable in the formulation medium for up to 7 days after production.   [66], according to the procedure described by Tang et al. [67]. This reaction was chosen for its broad substrate scope and large functional group tolerance, which allows late-stage fluorination. However, the silver-catalyzed electrophilic fluorination reaction turned out to be very moisture sensitive and, as a result, the hydrodestannylated by-product was generally formed. When applied to (R/S)-34, (R/S)-35 and (R/S)-36, these reaction conditions led to a mixture of F/H derivatives in a ratio ranging from 38/62 to 93/7, determined by 1 H NMR. Moreover, despite several attempts, the desired fluorinated products could not be separated from this by-product due to their similar structural properties. We therefore used a mixture of fully protected [ 19 F]fluoro-D/L-TIC(OH) and the hydrodestannylated by-product in the next deprotection step without further purification. All our attempts to deprotect the phenolic and amino acid functions in a single step with concentrated aqueous acids, such as hydrogen chloride or hydrogen bromide, and at high temperatures, unfortunately led to by-product formation. Deprotection was therefore performed in two successive steps, involving cleavage of the EOM and N-Boc groups with hydrogen chloride in dioxane, followed by ester saponification with lithium hydroxide, both at room temperature.  [35]. Briefly, radioactive fluorides were trapped on a QMA carbonate cartridge, previously washed with an aqueous solution of sodium triflate, and efficiently eluted to the reaction vessel using an aqueous solution of sodium triflate containing a small amount (50 µg) of potassium carbonate (recovery: >96.3%, n = 17). After azeotropic drying using acetonitrile, the [ 18 F]NaF was diluted in anhydrous N,N-dimethylacetamide (DMA) for use as a solvent in the subsequent radiofluorination step. This process enabled copper-mediated nucleophilic radiofluorination of the organotin compounds (S)-34, (S)-35 and (S)-36 in the presence of small amounts of base, preventing any degradation of the precursors in the reaction mixture. Aliquots of the resulting solution (200-300 µL) were used to optimize the reaction parameters. As stated by Bowden et al., the amount of organotin precursor used, the nature of copper complex introduced and the molar ratios of reagents are often critical parameters for the copper-mediated radiofluorination [68].
A screening of reaction parameters was therefore conducted with the precursor (S)-35 to determine the optimal conditions of radiofluorination for the 18 F-TIC(OH) analogs. First, we tested Cu(OTf) 2 (2 or 4 molar equivalents (eq.) relative to the precursor) and anhydrous pyridine as an additive (2.5 to 25 molar eq. relative to the precursor) ( Table 1) while maintaining constant the total volume of DMA (550 µL) and the temperature (110 • C, 10 min). The most promising results were obtained with 20 µmol of the organotin precursor (S)-35 in the presence of 2 molar eq. of copper complex and 4 or 5 molar eq. of pyridine (entries 4 and 5). Under these conditions, RCY of around 30% could be achieved (determined by radio-TLC analyses of the reaction mixture). Interestingly, higher amounts of pyridine were always detrimental for this radiofluorination step and resulted in very low RCY (entries 6-8). Increasing the amount of copper complex and/or precursor (entries [9][10][11][12][13][14] or extending the reaction time (data not shown) did not have a significant effect on radiolabeling efficiency. At this step, no radiochemical impurities were identified during the TLC monitoring of the reaction mixture, and only the desired radiofluorinated intermediate and the remaining [ 18 F]fluoride were visible. A second set of experiments investigated the influence of the nature of the copper complex (i.e., Cu(OTf) 2 /pyridine vs. tetrakis(pyridine)copper(II) triflate, Cu(OTf) 2 (py) 4 ) ( Table 1, entries [15][16][17][18]. Different amounts of Cu(OTf) 2 (py) 4 were used while keeping constant the quantity of precursor (S)-35 (20 µmol), the total volume of DMA (550 µL), the temperature (110 • C) and reaction time (10 min). Interestingly, higher RCY yields were obtained compared with the in situ combination of Cu(OTf) 2 and pyridine described above (i.e., 37, 54, 41 or 36% when using 1, 1.5, 2 or 2.5 eq. of Cu(OTf) 2 (py) 4 , respectively). Longer heating at 110 • C did not induce a significant increase in labeling efficiency, proving that optimal RCY can be obtained in a very short time. With a view to facilitate further purification steps of the reaction mixture, the condition using small amounts of Cu(OTf) 2 (py) 4    . Semi-preparative RP-HPLC purifications were performed on a Perkin Elmer system equipped with a Flexar LC autosampler, a Series 200 pump, a Peltier column oven, a vacuum degasser, a Photodiode Array Detector (PDA) and a GabiStar detector (Raytest). The separation was carried out on a Nucleodur C-18 H-Tec column (Macherey-Nagel, 5 µm, 10 × 250 mm) using the following conditions: isocratic elution, flow rate = 1.2 mL/min, ammonium formate 20 mM/EtOH (80/20, v/v), λ = 254 and 288 nm. Analytical HPLC measurements were performed on a system consisting of a HP1100 (Hewlett Packard, Les Ulis, France) and a Flo-one A500 Radiomatic detector (Packard, Canberra, Australia). The separation was carried out on a C-18 column (Agilent Zorbax, 5 µm, 4.6 × 150 mm) using the following conditions: isocratic elution, flow rate = 0.7 mL/min, ammonium formate 20 mM/EtOH (95/5, v/v), λ = 254 and 288 nm. For the determination of the enantiomeric excess of each radiolabeled compound, the analytical HPLC measurements were performed on a Reprosil Chiral-AA 8 µm column (250 × 4.6 mm; 8 µm; CIL Cluzeau; Sainte-Foy-la-Grande, France) using the following conditions: water/ACN (30/70, v/v) as isocratic eluent mixture and a flow rate of 1 mL/min. Oasis MCX Plus extraction short cartridges (225 mg, 60 µm) were purchased from Waters. All radiolabeled compounds were compared by TLC or analytical HPLC to the authentic non-radioactive material and to be free of significant UV-absorbing chemical and radiochemical impurities. The radio TLC strips (Merck neutral aluminum oxide 60F 254 plates) were developed with dichloromethane/ethanol (97/3, v/v) and measured on an AMBIS 400 (Scanalytics, CSPI, San Diego, CA, USA). Semi-preparative RP-HPLC purifications were performed on a Perkin Elmer system equipped with a Flexar LC autosampler, a Series 200 pump, a Peltier column oven, a vacuum degasser, a Photodiode Array Detector (PDA) and a GabiStar detector (Raytest). The separation was carried out on a Nucleodur C-18 H-Tec column (Macherey-Nagel, 5 µm, 10 × 250 mm) using the following conditions: isocratic elution, flow rate = 1.2 mL/min, ammonium formate 20 mM/EtOH (80/20, v/v), λ = 254 and 288 nm. Analytical HPLC measurements were performed on a system consisting of a HP1100 (Hewlett Packard, Les Ulis, France) and a Flo-one A 500 Radiomatic detector (Packard, Canberra, Australia). The separation was carried out on a C-18 column (Agilent Zorbax, 5 µm, 4.6 × 150 mm) using the following conditions: isocratic elution, flow rate = 0.7 mL/min, ammonium formate 20 mM/EtOH (95/5, v/v), λ = 254 and 288 nm. For the determination of the enantiomeric excess of each radiolabeled compound, the analytical HPLC measurements were performed on a Reprosil Chiral-AA 8 µm column (250 × 4.6 mm; 8 µm; CIL Cluzeau; Sainte-Foy-la-Grande, France) using the following conditions: water/ACN (30/70, v/v) as isocratic eluent mixture and a flow rate of 1 mL/min. Oasis MCX Plus extraction short cartridges (225 mg, 60 µm) were purchased from Waters. All radiolabeled compounds were compared by TLC or analytical HPLC to the authentic non-radioactive material and to be free of significant UV-absorbing chemical and radiochemical impurities. The reaction mixture was stirred at room temperature for 5 min and cooled to 0 • C in an ice bath before addition of a cold aqueous sodium hydroxide solution (10 M, 50 µL). The reaction mixture was then stirred at room temperature for 1 h. After cooling to 0 • C, trifluoroacetic acid (500 µL) was added to the reaction mixture, which was further stirred at 60 • C in a sealed vial for 30 min. After cooling to 0 • C, the solution was neutralized by careful addition of an aqueous sodium hydroxide solution (10 M, 660 µL) and diluted with an aqueous citrate buffer solution (0.1 M, pH 5, 18 mL) before passing through an anionic cartridge (MCX Plus extraction short cartridges, Waters). The latter was successively washed with an aqueous formic acid solution (2% vol., 500 µL) and methanol (500 µL). Then, the recovery of the radioactive compound was performed by elution of the cartridge with a solution of ammonia in methanol (5/  . Radio thin layer chromatography (radio-TLC) was performed on silica pre-coated TLC sheets (Alugram ® Xtra Sil G/UV 254 , Macherey-Nagel, Hoerdt, France), eluted with a mixture of ethyl acetate/cyclohexane (3/7, v/v) and measured on a miniGITA Dual radio-TLC instrument (Elysia-Raytest, Liège, Belgium). Analytical HPLC measurements were performed on a system consisting of an Agilent HP series 1100 (Hewlett Packard, Les Ulis, France) combined with a Flo-one A500 Radiomatic detector (Packard, Canberra, Australia). A Reprosil Chiral-AA 8 µm column (250 × 4.6 mm; 8 µm; CIL Cluzeau; Sainte-Foy-la-Grande; France) was employed for the determination of the enantiomeric excess of the produced radiotracers using the following solvent conditions: isocratic elution with a mixture of water/ACN (70/30, v/v) and a flow rate of 1 mL/min. For the determination of the radiochemical purities (RCP), a Zorbax Extend-C18 analytical column (4.6 × 150 mm, 5 µm, Agilent, Les Ulis, France) was employed using the following solvent conditions: water containing 0.1% of trifluoroacetic acid (solvent A) and methanol containing 0.1% of trifluoroacetic acid (solvent B); 0 to 3 min: isocratic elution 95% A; 3 to 15 min: gradient elution 95% → 5% A; 15 to preconditioned with 10 mL of ethanol, 10 mL of an aqueous solution of sodium trifluoromethanesulfonate (90 mg/mL) and 10 mL of water (without drying) following the protocol of Makaravage et al. [35]. The radioactivity was eluted to the reactor using a solution of potassium carbonate (50 µg) and sodium trifluoromethanesulfonate (10 mg) in water (550 µL) before the addition of anhydrous acetonitrile (1 mL). The resulting solution was dried by azeotropic distillation under reduced pressure and He flow at 100 • C for 12 min. After cooling to 30 • C, anhydrous N,N-dimethylacetamide (DMA, 2-2.5 mL) were added to the reactor. The reaction mixture was stirred at 110 • C for 10 min. After cooling to 0 • C and addition of water (3 mL), the reaction mixture was passed through a Sep-Pak light C18 cartridge, preconditioned with 10 mL of water, 10 mL of methanol and 10 mL of water. The cartridge was washed with water (1 mL) and flushed with air before elution of the desired 18 F-labeled intermediate with methanol (750 µL). After gentle evaporation of the solvent under argon flow at 70 • C, a concentrated hydrobromic acid solution (48 wt%, 200 µL) was added. The reaction mixture was heated at 110 • C for 10 min. After cooling to 0 • C, the mixture was diluted with 2 mL of a mixture of 20 mM aq. ammonium bicarbonate/ethanol (99/1, v/v) and purified by RP-HPLC. The collected fractions were diluted in saline and analyzed by analytical radio-HPLC to determine the radiochemical purity, enantiomeric excess and specific activity.

Conclusions
A novel and convenient pathway, involving common organotin intermediates, was successfully developed for the production of radioiodinated or radiofluorinated TIC(OH) analogs halogenated at positions 5, 6 or 8. This could pave the way for producing new radiohalogenated derivatives of well-known medicinal compounds containing the TIC(OH) scaffold, which could, in turn, lead to a broad set of applications not only in nuclear medicine but also in pharmaceutical and medicinal chemistry.