Robust synthesis of trifluoromethionine and its derivatives by reductive trifluoromethylation of amino acid disulfides by CF3I/Na/Liq.NH 3 system

Article abstract of DOI:10.1016/j.jfluchem.2011.01.001

We disclose the reductive trifluoromethylation of chemically stable homocystine and cystine to provide corresponding trifluoromethyl ethers by the CF₃I/Na/Liq.NH₃ system. Both non-protected and protected homocystines can be nicely converted into trifluoromethylated methionines under the same condition. The method described offers a robust synthesis of pharmaceutically important trifluoromethionine, suitable for multigram synthesis. Pentafluoroethylation of homocystine was also achieved by the CF ₃CF₂I/Na/Liq.NH₃ system.

Full text of DOI:10.1016/j.jfluchem.2011.01.001

Journal of Fluorine Chemistry 132 (2011) 186–189
Contents lists available at ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Robust synthesis of trifluoromethionine and its derivatives by reductive
trifluoromethylation of amino acid disulfides by CF3I/Na/Liq.NH3 system
*
Hiroyuki Yasui, Takeshi Yamamoto, Etsuko Tokunaga, Norio Shibata
Department of Frontier Materials, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso, Showa-ku, Nagoya 466-8555, Japan
A R T I C L E I N F O
A B S T R A C T
Article history:
We disclose the reductive trifluoromethylation of chemically stable homocystine and cystine to provide
corresponding trifluoromethyl ethers by the CF3I/Na/Liq.NH3 system. Both non-protected and protected
homocystines can be nicely converted into trifluoromethylated methionines under the same condition.
The method described offers a robust synthesis of pharmaceutically important trifluoromethionine,
suitable for multigram synthesis. Pentafluoroethylation of homocystine was also achieved by the
CF3CF2I/Na/Liq.NH3 system.
Received 22 December 2010
Received in revised form 28 December 2010
Accepted 4 January 2011
Available online 8 January 2011
Keywords:
ß 2011 Elsevier B.V. All rights reserved.
Reductive trifluoromethylation
Trifluoromethionine
Amino acid
Disulfide
Antiamebic drug
Pentafluoroethylation
1. Introduction
proteins [6]. Honek and co-workers studied the incorporation of
trifluoromethionine into peptides and proteins, and gained
Amebiasis is an infectious disease caused by the enteric
protozoan parasite Entamoeba histolytica and is the second leading
cause of death from parasitic diseases after malaria. Metronidazole
and related compounds are commonly used against invasive
intestinal and extraintestinal amebiasis [1,2]. Although clinical
resistance against metronidazole has not yet been demonstrated,
sporadic cases of treatment failure have been reported [1]. In
addition, it has been shown that this parasite easily adapts to
therapeutic levels of metronidazole in vitro [3]. Resistance to
metronidazole is also acquired easily by many bacterial species as
well as Giardia intestinalis and Trichomonas vaginalis [1]. Therefore,
the development of a novel antiamebic drug is urgently required.
conformational information of protein structure, side chain
dynamics as well as ligand binding using 19F NMR spectroscopy
[7]. As
a part of our continuous research program in the
development of effective drugs for amebiasis [4], we required a
large amount of trifluoromethionine (1a) to start the clinical study.
Although several methods for the synthesis of 1 have been
reported [8], they require several synthetic steps difficult to
perform in a multi-gram scale, and are expensive with chemically
unstable homocysteine being required as a starting material. In
this paper, we report the facile synthesis of trifluoromethionine
(1a) by direct reductive trifluoromethylation of chemically stable
homocystine (2a) under the combinatorial condition, CF3I/Na/
Liq.NH3. This method is robust and can be perform at a large scale.
Direct reductive trifluoromethylation of protected homocystine 2b
and cystine (2c) also proceeded nicely by this method in good to
high yields (Scheme 1(b)).
Trifluoromethionine (1a, S-trifluoromethyl-L-homocysteine), a
trifluorinated analogue of the amino acid methionine, has emerged
as a promising lead for amebiasis due to its high toxicity against
bacteria highly related to amebiasis [4]. Recently, Nozaki and co-
workers within our group reported an efficient cytotoxic effect of
amide derivatives of trifluoromethionine against the enteric
protozoan parasite E. histolytica (Fig. 1) [4b,5].
2. Results and discussion
Another importance aspect of trifluoromethionine is its utility
as a probe to elucidate the biochemistry of methionine residues in
Among several synthetic methods available [8], the most
straightforward way to prepare 1a is the direct trifluoromethyla-
tion of homocysteine using trifluoromethyl iodide (CF3I) in liquid
NH3 under ultraviolet (UV) irradiation reported by Soloshonok
et al. (Scheme 1(a)) [8b]. Enantiomerically pure trifluoromethio-
nine can be easily obtained in a single step from homocysteine
* Corresponding author.
E-mail address: nozshiba@nitech.ac.jp (N. Shibata).
0022-1139/$ – see front matter ß 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2011.01.001

H. Yasui et al. / Journal of Fluorine Chemistry 132 (2011) 186–189
187
reported the reductive trifluoromethylation of disulfides with
CF3Br (4 bar) in the presence of dithionite or hydroxymethane-
sulfinate salts in DMF-H2O medium leads to the formation of
aliphatic and aromatic trifluoromethyl sulfides [10]. However, this
method is less efficient when using alkane–disulfides as substrates
and to make matters worse, CF3Br is known as an ozone depletion
chemical that could destruct ozone dramatically. For this reason,
substitute chemicals such as CF3I must be selected for consider-
ation to replace CF3Br. In 2004, Dolbier and co-workers reported an
atom-economic procedure for preparation of trifluoromethyl
thioethers from disulfide using CF3I in DMF in the presence of
tetrakis(dimethylamino)ethylene (TDAE) as the organic reducing
agent [11]. However, using expensive TDAE is a problem for large-
scale synthesis, and in fact, we failed to convert homocystine into
1a by CF3I/TDAE in the DMF system. After many attempts, we were
pleased to find that a combination of Birch reduction condition, Na/
Liq.NH3 with CF3I, efficiently produced 1a from homocystine.
Actually, as shown in Table 1, the use of Na (4.2 equiv.), CF3I
(2.5 equiv.) in liquid NH3 at À78 8C gave 1a in 80% yield from
homocystine (Entry 1). Furthermore, we successfully carried out
this reaction on a 10 g scale (Entry 2). It should be noted that N-
Boc-protected homocystine 2b is well-tolerated under this
condition to provide the desired N-Boc-protected trifluoromethio-
nine 1b in 70% yield (Entry 3). This method was also applicable for
the direct trifluoromethylation of cystine (2c) to trifluoromethyl
cysteine (1c) in 65% yield (Entry 4).
Fig. 1. Trifluoromethionine and its derivatives as promising drug candidates against
protozoan parasites [5].
without protection by the method. However, when we performed
this method in a large scale for a pre-clinical study of 1a [5], we
often suffered from difficulties of removing
a by-product,
homocystine (2a), through the use of ion-exchange chromatogra-
phy. Moreover UV-irradiation is also somewhat tedious in large-
scale synthesis. The instability of homocysteine is another
disadvantageous point of this transformation. To overcome these
problems, a different synthetic approach was executed. Robust
preparative methods are especially important for large-scale
synthesis, which must be highly reproducible. We attempted to
find out direct trifluoromethylation from chemically stable
homocystine (2a) instead of homocysteine into 1a without UV-
irradiation. Incidentally, trifluoromethyl thioethers can be synthe-
sized from disulfides by reaction of the trifluoromethyl anion [9];
however, the method suffers from the fact that half of the disulfide
is wasted in the process. In 1991, Wakselman and co-workers
It is interesting to note that when we used pentafluoroethyl
iodide (C2F5I) instead of CF3I, pentafluoroethionine 3 was obtained
from 2a in 67% yield (Scheme 2). The Birch condition was used for
reductive alkylation of disulfides, as reported more than 20 years
ago [12], but this method had not been applied to the fluoroalk-
ylation reaction of disulfides. As far as we know, this is the first
Table 1
Reductive trifluoromethylation of disulfides of amino acids, homocystine and cystine.
.
Entry
Substrate 2
Product 1
Yield (%)
1
80
2a
93
70
65
3
4
a
The experiment was performed on a 10 g scale.

188
H. Yasui et al. / Journal of Fluorine Chemistry 132 (2011) 186–189
Scheme 1. Synthesis of trifluoromethionine and its derivatives. (a) Trifluoromethylation of thiols by CF3I under UV irradiation reported by Soloshonok et al. (b) Reductive
trifluoromethylation of disulfide by CF3I/Na/Liq.NH3 (this work).
well with pure water, and the compound was eluted with 2%
ammonia. After evaporating all solvent, the pure product was
afforded as a white solid.
L-Trifluoromethionine (1a): reaction of L-homocystine (2a,
37 mmol, 10 g), sodium (160 mmol, 3.7 g) and CF3I (93 mmol,
Scheme 2. Reductive pentafluoroethylation of homocystine 2a to
pentafluoroethionine 3.
18 g) in about 150 mL of liquid ammonia gave 1a (14 g, 93%); m.p.
227–228 8C (H2O/methanol) [lit. 227–230 8C [8b]]; [
(c 2.01, 4 N HCl) [lit. [
25 = +24.1 (c 0.1, 4 N HCl) [8b]]; 1H NMR
(200 MHz, D2O)
a]
D
24 = +25.3
a
]
D
example of reductive fluoroalkylation reaction of disulfides under
the Birch condition.
d 4.09 (t, J = 6.6 Hz, 1H), 3.03 (t, J = 8.2 Hz, 2H),
2.40 À 1.95 (m 2H); 19F NMR (188 MHz, D2O)
d
À41.3 (s, 3F); 13C
NMR (75.5 MHz, D2O)
d 171.0, 130.6 (q, J = 306 Hz), 51.2, 30.2,
3. Conclusion
25.1; IR (KBr) 2917, 2849, 1557, 1506, 1417, 1101, 919 cmÀ1; MS
(ESI) m/z 204 (M+H)+.
In conclusion, extremely facile synthesis of trifluoromethionine
(1a) from readily available and chemically stable homocystine (2a)
was developed by direct trifluoromethylation using CF3I/Na/
Liq.NH3. The described method offers a robust synthesis of
trifluoromethionine (1a), suitable for multigram synthesis. The
methodology described here seems to be the simplest one for the
synthesis of these compounds and can be extended to the
perfluoroalkylation of disulfides such as the pentafluoroethylation
reaction.
N-Boc-S-trifluoromethyl-L-homocysteine (1b): N-Boc-L-Homo-
cystine (2b, 4.0 mmol, 1.87 g), sodium (16.8 mmol, 385 mg) and
CF3I (10.0 mmol, 1.96 g) reacted in about 30 mL of liquid ammonia.
After removed the ammonia, the residue was dissolved in 10 mL of
water. The aqueous solution was adjusted to pH 3.0. The mixture
was extracted with ethyl acetate (3Â 30 mL). The combined
organic layer was washed with 4 N NaS2O3 aq., brine, and then
dried over Na2SO4. After removing all solvent, the residue was
purified by silica gel column chromatography (CH2Cl2/MeOH = 9/
1) gave the pure product in 70% yield (5.60 mmol, 1.70 g); m.p. 53–
4. Experimental
54 8C (EtOAc/hexane);
25 = +16.2 (c 2.0, CHCl3) [13]]; 1H NMR (300 MHz, CDCl3)
d
[a]
24 = +14.2 (c 0.52, CHCl3) [lit.
D
[a
]
D
4.1. General
6.99 (br, 2/5H, rotamer), 5.11 (d, J = 6.9 Hz, 3/5H, rotamer), 4.45 (br,
3/5H, rotamer), 4.32 (br, 2/5H, rotamer), 2.98 (t, J = 7.5 Hz, 2H),
1H, 13C and 19F NMR spectra were recorded with Varian 200 or
300 spectrometers in CDCl3 or D2O solutions. TMS, residual
chloroform and CFCl3 were used as internal references for 1H, 13C
and 19F NMR in CDCl3 solution, respectively. Residual H2O was
used as an internal reference for 1H NMR in D2O solution. TFA-d
was used as an internal reference for 13C and 19F NMR in D2O
2.32 (br, 1H), 2.09 (br, 1H); 19F NMR (188 MHz, CDCl3)
d
À41.6, (s,
3F); 13C NMR (75.5 MHz, CDCl3)
d 175.9, 174.9, 157.0, 155.7, 130.9
(q, 306 Hz), 82.6, 80.7, 53.2, 52.3, 33.3, 33.0, 28.1, 30.0, 25.7; IR
(KBr) 2917, 2849, 1717, 1654, 1541, 1458, 1386, 1108, 670 cmÀ1
;
MS (ESI) m/z 302 (MÀH)À.
S-Trifluoromethyl-L-cysteine (1c): reaction of L-cystine (2c,
solution. Chemical shifts are expressed in ppm (
d
). Coupling
4.0 mmol, 961 mg), sodium (16.8 mmol, 385 mg) and CF3I
(10.0 mmol, 1.96 g) in about 30 mL of liquid ammonia gave 1c
(983 mg, 65%); m.p. 232–234 8C (H2O) [lit. 230–232 8C [8b]];
constants (J) values are in Hz. Optical rotations were measured on a
HORIBA SEPA-300. Infrared spectra were recorded on a JASCO FT/
IR-4100 spectrometer. Mass spectra were recorded on a SHIMAZU
LCMS-2010EV (ESI-MS). HRMS was recorded on a WATERS ESI/
Synapt G2 HDMS (ESI-MS).
[a
]
25 = À24.3 (c 0.52, H2O) [lit. [
a]
D
25 = +52.3 (c 0.1, H2O) [8b]].
D
Since the value of specific rotation observed here was very
different from the value reported in Ref. [8b]], we prepared 1c by
the procedure described in Ref. [8b]] to confirm this issue. The
specific rotation of 1c by the method [8b]] was ascertained to be
4.2. General experimental procedure for the reductive
trifluoromethylation of amino acid disulfides
[a]
26 = –25.0 (c 0.21, H2O), which is very close to our observed
D
value; 1H NMR (300 MHz, D2O)
d 4.33 (dd, J = 4.5, 7.2 Hz, 1H), 3.63
The disulfide 2 was dissolved in liquid ammonia in a flask
cooled to À78 8C, and sodium was slowly added in small pieces
until the reaction mixture turned blue. Trifluoromethyl iodide or
pentafluoroethyl iodide was then slowly added with a balloon.
After stirring for 20 min at this temperature, the cooling bath was
removed to evaporate the ammonia completely. The residue was
dissolved in 1 M sodium hydroxide aqueous solution. The resulting
mixture was applied to Dowex 50W (H+), the resin was washed
(dd, J = 4.5, 15.6 Hz, 1H), 3.48 (dd, J = 7.2, 15.6 Hz, 1H); 19F NMR
(282 MHz, D2O)
130.7 (q, J = 306 Hz), 54.3, 30.3; IR (KBr) 3427, 2926, 2855, 1615,
1505, 1403, 1346, 1094, 845, 759, 721 cmÀ1; MS (ESI) m/z 190
(M+H)+.
d
À41.4 (s, 3F); 13C NMR (150.9 MHz, D2O)
d 171.9,
S-Pentafluoroethyl-L-homocysteine (3): reaction of L-homocys-
tine (2a, 3.73 mmol, 1.0 g), sodium (15.7 mmol, 359 mg) and C2F5I
(9.33 mmol, 2.29 g) in about 30 mL of liquid ammonia gave 3

H. Yasui et al. / Journal of Fluorine Chemistry 132 (2011) 186–189
189
(1.27 g, 67%); m.p. 208–209 8C (H2O/methanol); [a]
24 = +23.2 (c
D
[5] D. Sato, S. Kobayashi, H. Yasui, N. Shibata, T. Toru, M. Yamamoto, G. Tokuro, V. Ali,
T. Soga, T. Takeuchi, M. Suematsu, T. Nozaki, Int. J. Antimicrob. Agents 35 (2010)
56–61.
[6] J.F. Honek, in: I. Ojima (Ed.), Fluorine in Medicinal Chemistry and Chemical
Biology, Effects of Fluorination on the Bioorganic Properties of Methionine,
Wiley-Blackwell, Chichester, 2009, pp. 447–462.
[7] (a) M.D. Vaughan, V.J. Robertson, J.F. Honek, J. Fluorine Chem. 128 (2007) 65–70;
(b) H.S. Duewel, E. Daub, V. Robinson, J.F. Honek, Biochemistry 40 (2001) 13167–
13176;
1.60, 4 N HCl); 1H NMR (200 MHz, D2O)
d 4.03 (t, J = 6.6 Hz, 2H),
3.01 (t, J = 7.4 Hz, 2H), 2.34 À 2.03 (m, 2H); 19F NMR (188 MHz,
D2O)
d
À83.3 (s, 3F), À91.8 (s, 2F); 13C NMR (75.5 MHz, D2O)
d
170.8, 121.4 (qt, J = 40.9, 286 Hz), 118.1 (tq, J = 36.5, 286 Hz), 51.1,
30.5, 23.8; IR (KBr) 2918, 2850, 1584, 1214, 1097, 984, 755 cmÀ1
;
MS (ESI) m/z 254 (M+H)+; HRMS (ESI) calcd for C6H9NO2F5S m/z
254.0274 (M+H)+, found: 254.0279. The fluorine signals deriving
from 1c in 19F NMR spectra exhibit little vicinal coupling, and they
appear as two singlet signals, as is exemplified by the fluorine NMR
spectrum of CF3CF2I [14].
(c) H. Duewel, E. Daub, V. Robinson, J.F. Honek, Biochemistry 36 (1997) 3404–
3416;
(d) M.D. Vaughan, P. Cleve, V. Robinson, H.S. Duewel, J.F. Honek, J. Am. Chem. Soc.
121 (1999) 8475–8478.
[8] (a) B. Langlois, D. Montegre, N. Roidot, J. Fluorine Chem. 68 (1994) 63–66;
(b) V. Soloshonok, V. Kukhar, Y. Pustovit, V. Nazaretian, Synlett (1992) 657–658;
(c) M.J. Houston, J.F. Honek, J. Chem. Soc. Chem. Commun. (1989) 761–762;
(d) T. Umemoto, A. Ando, Bull. Chem. Soc. Jpn. 59 (1986) 447–452;
(e) T. Umemoto, O. Miyano, Tetrahedron Lett. 23 (1982) 3929–3930;
(f) R.L. Dannley, R.G. Taborsky, J. Org. Chem. 22 (1957) 1275–1276.
[9] (a) D. Inschauspe, J.-P. Sortais, T. Billard, B.R. Langlois, Synlett (2003) 233–235;
(b) N. Roques, J. Fluorine Chem. 107 (2001) 311–314;
(c) G. Blond, T. Billard, B.R. Langlois, Tetrahedron Lett. 42 (2001) 2473–2475;
(d) T. Billard, N. Roques, B.R. Langlois, J. Org. Chem. 64 (1999) 3813–3820;
(e) J. Russell, N. Roques, Tetrahedron 54 (1998) 13771–13782;
(f) B. Quiclet-Sire, R.N. Saicic, S.Z. Zard, Tetrahedron Lett. 37 (1996) 9057–9060;
(g) T. Billard, B.R. Langlois, Tetrahedron Lett. 37 (1996) 6865–6868.
[10] (a) J.L. Clavel, B. Langlois, C. Wakselman, M. Tordeux, Phosphorus Sulfur Silicon
Relat. Elem. 59 (1991) 423–426;
Acknowledgements
This work was supported by KAKENHI (21390030, 22106515),
and aRigen Pharmaceuticals Incorporated. The authors are also
grateful to the TOSOH F-TECH, INC. for the generous gift of CF3I and
C2F5I.
References
(b) C. Wakselman, M. Tordeux, J.L. Clavel, B.R. Langlois, Chem. Commun. (1991)
993–994;
[1] V. Ali, T. Nozaki, Clin. Mirobiol. Rev. 20 (2007) 164–187.
[2] C.D. Freeman, N.E. Klutman, K.C. Lamp, Drugs 54 (1997) 679–708.
[3] C. Wassmann, A. Hellberg, E. Tannich, I. Bruchhaus, J. Biol. Chem. 274 (1999)
26051–26056.
[4] (a) D. Sato, W. Yamagata, S. Harada, T. Nozaki, FEBS J. 275 (2008) 548–560;
(b) T. Nozaki, T. Toru, N. Shibata, M. Tamamoto, WO-2007/077876.
(c) M. Tokoro, T. Asai, S. Kobayashi, T. Takeuchi, T. Nozaki, J. Biol. Chem. 278
(2003) 42717–42727;
(c) C. Wakselman, M. Tordeux, J. Org. Chem. 50 (1985) 4047–4051;
(d) C. Wakselman, M. Tordeux, Chem. Commun. (1984) 793–794.
[11] (a) C. Pooput, W.R. Dolbier Jr., M. Medebielle, J. Org. Chem. 71 (2006) 3564–3568;
(b) C. Pooput, M. Medebielle, W.R. Dolbier Jr., Org. Lett. 6 (2004) 301–304.
[12] (a) J. Jiracek, M. Collinsova, I. Rosenberg, M. Budesinsky, E. Protivinska, H.
Netusilova, T.A. Garrow, J. Med. Chem. 49 (2006) 3982–3989;
(b) G.S. Sheppard, J. Wang, M. Kawai, N.Y. BaMaung, R.A. Craig, S.A. Erickson, L.
Lynch, J. Patel, F. Yang, X.B. Searle, P. Lou, C. Park, K.H. Kim, J. Henkin, R.
Lesniewski, Bioorg. Med. Chem. Lett. 14 (2004) 865–868;
(d) M. Yoshimura, Y. Nakano, T. Koga, Biochem. Biophys. Res. Commun. 292
(2002) 964–968;
(e) G.H. Coombs, J.C. Mottram, Antimicrob. Agents Chemother. 45 (2001) 1743–
1745;
(f) H. Tanaka, N. Esaki, K. Soda, Enzyme Microb. Technol. 7 (1985) 530–537;
(g) W.A. Zygmunt, P.A. Tavormina, Can. J. Microbiol. 12 (1966) 143–148.
(c) A. Holy, I. Rosenberg, Collect. Czech. Chem. Commun. 50 (1985) 1514–1518.
[13] D. Vaughan, P.B. Sampson, E. Daub, J.F. Honek, Med. Chem. 1 (2005) 227–237.
[14] W.R. Dolbier Jr., Guide to Fluorine NMR for Organic Chemists, Wiley, New Jersey,
2009, pp. 177–210.