ORIGINAL 

Annals of Nuclear Medicine Vol. 7, No. 3, 173-177, 1993 

High reactivity of [11C] CH3I with thiol group in the synthesis of C-11 labeled radiopharmaceuticals 

Yasuhiro MAGATA,* Hideo SAJI,** Taro TOKUI,** Yoshiro OHMOMO,*** Yoshihisa YAMADA,** Masahiko HIRATA,** Junji KONISHI and Akira YOKOYAMA** 

* Department of Nuclear Medicine, Faculty of Medicine and * *Department of Radiopharmaceutical Chemistry, Faculty of Pharmaceutical Sciences, Kyoto University * * * Department of Radiopharmaceutical Chemistry, Osaka University ofPharmaceutical Sciences 

High reactivity of [11C]-methyl iodide ([11C]CH3I) with the thiol group was demonstrated with cysteamine and other compounds containing a thiol and another functional groups in each structure. The methylation of the thiol group in cysteamine with [11C]CH3I was very rapid at 0'C with no catalyst, and gave a high radiochemical yield and purity without any detectable by-product. Moreover, this reaction was not disturbed by the other functional groups, such as -NH2, -OH and -COOH in the same structure. This S-methylation reaction is very useful for producing a new radiopharmaceutical labeled with the short lived positron emitting nuclide C-11 . 

Key words : reaction radio pharmaceutical, positron emitting nuclide, 11C, thiol group, labeling 

INTRODUCTION 

THE POSITRON EMITTlNG NUCLIDE is very useful in clinical nuclear medicine because of its high resolu-tion in Positron Emission Computed Tomography (PET). On this basis, many radiopharmaceuticals labeled with positron emitting nuclide C-1 1 have been developed. Numerous C-11 labeled compounds have been synthesized by the methylation reaction with [nC]-methyl iodide ([11C]CH3I) as a precursor. Examples are receptor binding ligands,1-3 3_O-[11C]-methyl-glucose,4 amino acids,5-8 and amine analogs.9,10 Each methylation reaction needs opti-mum labeling conditions such as labeling time, temperature, solvent and an appropriate catalyst for a good yield. These radiopharmaceuticals have been synthesized by N-methylation with [11C]CH3I, and thus the reaction conditions have since been studied systematically.n,12 
Thiol groups display high reactivity with alkyl halides.13 As a radiolabeled alkyl halide for N-methylation, [11C]CH3I is readily available via an automated synthetic system. For a precursor con-taining a thiol group in its structure, the labeling reaction with [11C]CH3I requires only a one-step reaction. Except for the synthesis of S-[11C]-methio-nine, evaluating labeling conditions of the S-methyla-tion reaction has not been hitherto documented. In addition, the effects of other functional groups in the same structure on a reaction between the thiol group and alkyl halide has not been discussed, although biogenic compounds have many functional groups, such as amino, hydroxy and carboxylic groups. 
Our present study attempted to develop new C-11 labeled radiopharmaceuticals. Five compounds, con-taining a thiol and another functional group in each structure, were selected as model compounds for the [11C]-S-methylation reaction. 

MATERIALS AND METHODS 

All the reagents were of guaranteed grade and used without recrystalization. Thin layer chromato-graphy (TLC) was carried out with silicagel plate Kieselgel 60 (Art 5553, Merck Co. Ltd., Tokyo). High Performance Liquid Chromatography (HPLC ; LC-4A, Shimadzu Co. Ltd., Tokyo) was performed with a refractive index detector and radioactivity detector. NMR spectra were taken by means of a JOEL PMX 60 (Nihon Bunko Co. Ltd., Tokyo) with tetramethylsilane as the internal standard. 

Production of [11C]CH3I C-11 was produced via the 14N(p,a)11C reaction with 11.3 MeV protons on a nitrogen gas target in an ultra compact cyclotron (CYPRIS Model 325, Sumitomo Heavy Industry Ltd.. Japan). The target batch nitrogen gas was bombarded and the [11C] C02 produced was transported into an automated [11C]CH3I synthesis system (CUPID, Sumitomo Heavy Industry Ltd.). The radioligand, [11C]CH3I, was synthesized according to the method of Comar et al.14 Briefiy, [11C]C02 was first reacted with LiAlH4 m a tetrahydrofuran solution in the reaction vessel. After evaporating the solvent, HI solution was added to the residue. The resulting [11C]CH31 was distilled at 80'C and subsequently trapped in acetone cooled by liquid C02 gas. 

Synthesis of authentic S-methylated compounds Five mercapto compounds (Table 3) were used in this study. The authentic compounds of S-methyl-mercaptopropionic acid and S-methyl-mercapto-ethanol were commercially available. S-methyl-cysteamine, S-methyl-thioglycolic acid and S-methyl-thioglycerol were synthesized according to the follow-ing general procedures. Briefiy, each thiol com-pound (O.1 mole) was dissolved in a mixture of 100 ml of methanol and 100 mL of I N NaOH. After cooling, 0.11 mole of CH31 was added to the mixture and stirred at O'C for 10 min prior to adjusting to an optimum pH value. The solution was extracted with chloroform, dried, evaporated to dryness, and then the products were obtained by distillation in vacuo. 

S-methyl-2-mercapto-acetic acid bp. 81.0-83.5'C (2.5 mmHg). Yield 61.5% Ele-mental analysis agreed with the calculated value for C3H602S: C, 33.97 %; H, 5.70%. Found C, 34.05 %; H, 5.78%. NMR (CDC13): (ppm) 2.15 (s, 3H), 3.08 (s, 2H), 7.93 (s, IH). 

S-methyl-3-mercapt0-3-deoxy-glycerol bp. 100.5-101.0'C (2.0 mmHg). Yield 51.0%¥ Ele-mental analysis agreed with the calculated value for C4Hl002S : C, 39.34 % ; H, 8.25 % ; S, 26.21%. Found C, 39 09%. H 8 49%; S 26.14%. NMR (CDC13): (ppm) 3.52 (m, 5H), 2.52(d, 2H), 2.07 (s, 3H). 

S-methyl-cysteamine bp. 48.0-48.5'C (14.5 mmHg). Yield 47.3%. Ele-mental analysis agreed with the calculated value for CloH17N03S2 (as tosylate (mp. 91.0-94.0'C)) : C, 4562%; H, 6.51%. N 5 32%; Found C 45 32%. H, 6.56%; N, 5.22%¥ NMR (CDC13): (ppm) 1.68 (s, 2H), 2.03 (s, 3H), 2.48-2.78 (m, 4H). In addition, S,N,N-tri-methyl- and S,N,N,N-tetra-methyl-cysteamine were synthesizedl5 as authentic compounds of methylated cysteamine analogs, res pectively . 

S, N.N- trimeth yl -c ysteam ine bp. 51 .0-51 .5'C (18.0 mmHg). Yield 295 mg (11 .4 %)-Elemental analysis was agreed with the calculated value for C5H14NSCI (as hydrochloride; mp. 156.0-158.0'C): C, 38.59%; H, 9.07%; N, 9.00%. Found C, 38.41 %; H, 9.27%; N, 9.01 %. NMR (CDC13): (ppm) 2.03 (s, 3H), 2.17 (s, 6H), 2.46 (s, 4H). 

S, N,N,N- tetramethyl-cysteamine hydroiodide mp. 229.0-231.0'C. Yield 750 mg (58.0 %)¥ Ele-mental analysis agreed with the calculated value for C6H16NSI: C, 27.60%; H, 6,18%; N, 5.36~. Found C, 27.79%; H, 6.33%; N, 5.34%. NMR (CDC13): (ppm) 2.06 (s, 3H), 2.37-2.43 (dd, 4H), 3.09 (s, 9H). 

Methylation reaction by [11C]CH3I Labeling conditions were determined with cysteamine as a model compound. Labeling parameters were tested by varying the reaction time (30 sec-10 min) and labeling temperature (0-50'C) with I x 10-5 mole cysteamine in 0.25 mL of I N NaOH and 0.1 mL of acetone containing [11C]CH3I. After the removal of non-reacted [11C]CH31, the mixture was analyzed by TLC with silicagel plates and HPLC with a reverse-phase column C-18 at a column temperature of 40'C. The developing solvent for TLC was chloroform/methanol/ammonia water=8/2/1. The eluting HPLC solvent was acetonitrile/water/methyl-amine=49/50/1 and the flow rate was 0.5 mL/min. The proportion of the product was calculated from the area of the radioactive peak on the TLC or HPLC chart. Data are presented as the mean for every 3 trials, Using the cold synthesized compounds, the developed TLC showed Rf values of 0.17-0.43, 0.50-0.71 and 0.00-0.10 for S-methyl-. S,N,N-tri-methyl- and S,N,N,N-tetramethyl-cysteamine, re-spectively. Retention time (Rt) HPLC intervals for S-methyl-, S,N,N-tri-methyl-. S,N,N,N-tetra-methyl-cysteamine and CH31 were 5.6, 7.7, 4.2 and 8.4 min, res pectively. 
The carrier effect of CH31 (3 x 10-5 and I x 10-4 mole) was also tested in the manner described above. Non-radioactive CH31 was added in the acetone solu-tion containing [11C]CH3I. 
The methylation of the other four compounds by [11C]CH3I was carried out under the optimal condi-tions determined previously with cysteamine. Struc-tures of the labeled compounds were confirmed by the respective Rf and Rt values for TLC and HPLC. 

RESULTS 

Methylation ofcysteamine by [11C]CH3I Various labeling parameters were first surveyed. As no differences were observed between O and 50'C, the labeling temperature of O'C was selected to pre-vent the vaporization of [11C]CH31 in finding the optimum reaction time. At this temperature, a 93 ~ yield was obtained after only 30 sec mixing, and plateaued after a 1-min reaction (Fig. l). The fol-lowing optimal conditions were selected : I x 10-5 mole cysteamine was dissolved in 0.25 mL of I N NaOH at O'C before adding 0.1 mL of acetone con-taining [11C]CH3I, and the mixture was stirred at O'C for I min. The proportion of the desired com-pound was over 98% assayed by TLC and HPLC. The radioactive peak detected with TLC was super-imposed over the stained spot of the authentic com-pound by iodide. The radioactive peak detected with HPLC indicated an Rt similar to that of the authentic com pound. 
Ethanolamine was [nC]-methylated under the same labeling conditions as those for cysteamine described above. Table I compares the results for both compounds under these conditions (80% or more remained as [11C]CH3I), although small quantities of N-methylated compounds were ob-served. The hydroxy group of ethanolamine was not methylated. 
Effect of carrier CH3I on the labeling efficiency of cysteamine by [11C]CH3I The effects of carrier CH3I on the labeling efficiency of cysteamine by [11C]CH3I were investigated (Table 2). By increasing the amount of CH31 carrier, the proportion of S-methyl-cysteamine was reduced to 81.6% and 35.8 % with 3 x 10-5 and I x 10-4 mole of CH3I respectively, at O'C. At 50'C, a higher yield of S-methyl-cysteamine (42.8%) was obtained and a lower content of non-reacted [11C]CH3I remained. The proportions of S,N,N-tri-methyl-cysteamine and S,N,N,N-tetra-methyl-cysteamine were 3.8% and 11.7 %, respectively. 

S-[11C]-methylation reaction offive thiol compounds The five thiol compounds in Table 3 were methylated in a similar manner to that described above with [11C]CH3I in a basic solution, and the products were assayed by TLC with the developing solvents shown in Table 3. Each compound was selectively S-methylated at both O and 50'C, whereas only the latter temperature was employed in 2-mercapto-ethanol. The proportion of the desired compound was lower (94%) at 50'C than at O'C (98 ~)-

DISCUSSION 

The reactivity and selectivity of [11C]CH3I to the thiol group, and the effects of other functional groups on S-methylation reaction were evaluated in this study. 
Before radioactive methylation reactions were attempted, non-radioactive methylation trials of various thiol compounds were conducted in 5 min at O'C without other by-products. This rapid cold reaction was performed to evaluate the reactivity of the radioactive precursor, FIC]CH3I, with the thiol group of cysteamine. High reactivity of the thiol group with [11C]CH3I was observed by stirring at O'C for only I min in a basic solution without a catalyst (Fig. 1). Hence, [11C]-methylation of the thiol group required only 1-min stirring with no catalyst. In addition, no N-methylated compound was detected under these conditions. Moreover, the N-methylation reaction did not proceed at 50'C. The difference between the cold synthesis and C-1 l reaction was attributed to the varied quantities of the CH31 carrier. In fact, in the PIC]-methylation reac-tion, N-methylated compounds were produced by increasing the quantity of the CH31 carrier (Table 2). Although the structure of ethanolamine is similar to' that of cysteamine, the amino group of ethanolamine was methylated under conditions similar to that of cysteamine (Table 1). . Consequently, these facts indicate that the reactivity of the thiol group with methyl iodide is much higher than that of the amino group. In addition, results obtained with the other compounds containing hydroxy groups or carboxylic groups showed high selectivity for the thiol group in the [nC]CH31 methylation. 
In conclusion, the usefulness of this reaction be-tween the thiol group and methyl iodide was readily achieved without any heating. Because most biologic-ally active substances have chemically unstable groups in their structures, it is important to reduce those side reactions in order to produce labeled compounds with a high radiochemical yield and purity. Consequently, the S-methylation reaction reported here can be used to synthesize readily available [nC]-1abeled radiopharmaceuticals. Fur-ther evaluations of novel drugs by means of the S-methylation reaction are currently in progress. 

ACKNOWLEDGMENT 

We thank Ms Yuka Iwamoto for her assistance. 

REFERENCES 

1. Kloster G, Roder E, Machulla HJ : Synthesis, chro-matography and tissue distribution of methyl-L-[11C]-morphine and methyl-[11C]-heroin. J Label!ed Compd Radiopharm 16 : 441-448, 1979 
2. Berger G, Maziere M, Knipper R, et al : Automated synthesis of [nC]-labelled radiopharmaceuticals : Imipramine, Chlorpromazine, Nicotine and Methio-nine. Int J Appl Radiat Isot 30: 393-399, 1979 
3. Arnett CD, Fowler JS, Wolf AP, et al : Specific bind-ing of [uC]Spiroperidol in rat brain in vivo. J Neurol Chem 40: 455-459, 1983 
4. Vyska K, Magloire JR, Freundliebl C, et al : In vivo determination of the kinetic pararneters of glucose transport in the human brain using [11C]-methyl-D-glucose (CMG) and dynamic positron emission tomography (dPET). Eur J Nucl Med 11 : 97-106, 1985 
5. 
Langstrom B, Lundqvist H: [11C]-methy iodide and its use [nC]-methyl-L-methionine. Int 27: 357-363, 1976 
The preparation of in the synthesis of J Appl Radiat Isot 
6. Syrota A, Comar D, Cerf M, et al : [11C]Methionine pancreatic scanning with positron emission computed 
tomogra phy. J Nucl Med 20 : \778-781 , 1979 
7. Hubner KF, Andrews GA, Buonocore E, et al : Carbon-11 labeled amino acids for the rectilinear and positron tomographic imaging of the human pancreas. J Nucl Med 20 : 507-513, 1979 
8. Henze E, Schelbert HR, Barrio JR, et al : Evaluation of myocardial metabo]ism, with N-13- and C-11-labeled amino acids and positron computed tomo-graphy. J Nucl Med 23: 671-681, 1982 
9. Winstead MB, Dischino DD, Munder NA, et al : Relationship of molecular structure to in vivo distribu-tion of carbon-11-1abeled compounds VI. Carbon-11-labeled aliphatic diamines. Eur J Nucl Med 5 : 165-169, 1980 
10. Jay M, Chaney JE, Digenis GA : Development of methodology for synthesis of [ce-C]-labeled phene-thylamine. Int J Appl Radiat Isot 32 : 345-348, 1981 
11 . Finn RD, Boothe TE, Vora MM, et al : Syntheses with isotopically labelled carbon. Methyl iodide, formaldehyde and cyanide. Int J Appl Radiat Isot 35 : 323-335, 1984 
12. Comar D. Crouzel C, Maziere B : Positron emission tomography : Standardisation of labelling procedures. Appl Radiat Isot 38: 587-596, 1987 
13. Landini D, Rolla F: A convenient synthesis of primary and secondary dialkyl and aryl alky sulfides in the presence of phasetransfer catalysts. Synthesis (Communications) : 565-566, 1974 
14. Crouzel C, Langstrom B, Pike VW, et al : Recom-mendations for a practical production of [nC]methyl iodide. Appl Radiat Isot 38: 601-603, 1987 
1 5. Clark RJH, McAlees AJ : The chemistry of methyl-titanium trichloride. II. Variable-temperature nuclear magnetic resonance and infrared spectra of some complexes of methyltitanium trichloride and of titanium tetrachloride with unsymmetrical bidentate ligands. Inorg Chem 1 1 : 342-348, 1972