isolated from photochemical studies using the parent acid of 1,
i.e. N-hydroxy-4-(p-chlorophenyl)thiazolethione, as hydroxyl
radical source for DNA-strand break in photobiological stud-
ies.12
In conclusion, we have demonstrated that
L
-cysteine-deriva-
tives, e.g. -CysOEt, or the tripeptide GSH can be applied as
L
useful radical traps for the stereoselective formation of
disubstituted tetrahydrofurans 3 via an alkoxyl radical pathway
in aqueous solvents. These investigations point to the feasibility
of O-radical reactions using N-alkoxythiazole-2(3H)-thiones,
e.g. derivatives of 1, under biomimetic conditions. Further work
is in progress in order to pursue O-radical chemistry in water as
the sole solvent.
This work was generously supported by the Deutsche
Forschungsgemeinschaft (Project Ha1705/3–2). Also, we ex-
press our gratitude to Dipl.-Chem. Philipp Schmidt for
providing samples of thiazolethione 1a and Dr Hideki Okamoto
for helpful discussions.
Scheme 2 Formation of thiazoles 5–8 from N-alkoxythiazolethiones 1.
Notes and references
† Satisfactory analytical data were obtained for all new compounds in this
study: thiazolethiones 1a, 1c, and 1d, and tetrahydrofurans 3c and 3d.
‡ For comparison, yields of tetrahydrofurans 3 from photoreactions of 1 and
Bu3SnH in C6H5CF3 were determined. Figures in brackets denote the cis–
trans ratios: 3a: 93% (30+70), 3b: 60% (88+12), 3c+97% (90+10), 3d+88%
( < 2+ > 98).
trahydrofuran 3c from the corresponding alkenoxyl radical 2c
(entry 3, cis+trans = 90+10) is considered to originate from
beneficial steric effects of the tert-butyl substituent. This
interpretation is supported by a noteworthy 2,3-trans-selectivity
for the formation of 3-tert-butyl-2-methyltetrahydrofuran 3d
from the corresponding thiazolethione 1d (entry 4).7,10,11
According to the data which are reported in Scheme 1 it is
obvious that heterogeneous and homogeneous conditions 1 and
§ In a typical run thiazolethione 1 (1 mmol) was dissolved in the organic
solvent (C6H5CF3 or 1,4-dioxane, 20 ml).
L-CysOEt·HCl (10 mmol) and a
base (9 mmol, see Scheme 1), or GSH, or
L-cysteine (10 mmol) were
dissolved in water (5 mL). Both solutions were combined while stirring to
afford the reaction mixture which was photolyzed at ambient temperature in
a Rayonet® photoreactor (l = 350 nm). Upon complete consumption of 1
(ca. 30 min), the colorless to yellowish precipitate was filtered-off and the
remaining solution was worked up as follows. For reactions with CysOEt in
2 (Scheme 1), which demand -CysOEt·HCl as hydrogen atom
L
donor and a suitable base, are superior to the GSH for preparing
tetrahydrofurans 3 from thiones 1. It is worth mentioning, that
GSH requires the presence of water in order to deliver its
hydrogen atom to cyclized O-radicals, although GSH is
partially soluble in pure dioxane. Similar observations were
C6H5CF3–H2O, 2
M HCl was added with agitation to adjust to pH 2 in the
aqueous phase. Subsequently, phases were separated and the aqueous layer
was extracted with diethyl ether (2 3 10 mL). The combined organic phases
were dried with MgSO4 and concentrated in vacuo. The product 3 was
purified by column chromatography (SiO2, petroleum ether–Et2O = 2+1,
v/v). If dioxane is used as solvent, the organic solvent is first evaporated
from the reaction mixture. Subsequently, diethyl ether was added (20 mL)
to the aqueous phase and tetrahydrofurans 3 were isolated as described
above.
made for free amino acid -cysteine as hydrogen atom donor
L
which affords almost identical yields and selectivities of 3 as
GSH does (not shown in Scheme 1). Further, syntheses of
tetrahydrofurans 3 fail if photolyses of thiones 1 and GSH are
performed in heterogeneous mixtures. For example, the photo-
reaction of radical precursor 1b and GSH in C6H5CF3–H2O
affords 2-phenylpent-4-en-1-ol (34%) and the corresponding
aldehyde 2-phenylpent-4-enal (21%) as sole products.
1 J. Hartung and R. Kneuer, Eur. J. Org. Chem., 2000, 1677.
2 D. C. Ayres and J. D. Loike, Lignans—Chemical, Biological, and
Clinical Properties, Cambridge University Press, Cambridge, 1990, pp
269–302
3 V. Ullrich and R. Brugger, Angew. Chem., 1994, 106, 1987; Angew.
Chem., Int. Ed. Engl., 1994, 33, 1911.
4 E. J. Corey, C. Shih, N. Y. Shih and K. Shimoji, Tetrahedron Lett.,
1984, 25, 5013.
5 N. A. Porter, M. O. Funk, D. Gilmore, R. Isaac and J. Nixon, J. Am.
Chem. Soc., 1976, 98, 6000.
6 J. Hartung, M. Schwarz, I. Svoboda and H. Fuess, Eur. J. Org. Chem.,
1999, 1275.
7 H. Strittmatter, A. Dussy, U. Schwitter and B. Giese, Angew. Chem.,
1999, 111, 238; Angew. Chem., Int. Ed. Engl., 1999, 38, 135.
8 J. Hartung and F. Gallou, J. Org. Chem., 1995, 60, 6706.
9 J. Hartung, M. Hiller and P. Schmidt, Chem. Eur. J., 1996, 2, 1014.
10 J. Hartung, Eur. J. Org. Chem., 2001, 619.
In almost every photochemical run, formation of a colorless
to yellowish precipitate was observed. This material was
soluble in acetone, but less soluble in diethyl ether. It was shown
to be a mixture of strongly UV-absorbing compounds which
were characterized as different primary and secondary photo-
products presumably of the starting thiones 1 (Scheme 2).
Expected addition products of either glutathionyl or cysteinyl
radicals to the thiocarbonyl group in parent thiones 1a–d were
surprisingly absent in the reaction mixtures.6 According to
control experiments it is likely that the disulfide 5 is formed as
primary product which then undergoes further absorption of UV
light to afford both 4-(p-chlorophenyl)thiazole-2(3H)-thione (6)
and thiazole 7 besides the bisthiazole 8. Evidence for this
assumption is derived from products obtained after photo-
decomposition of disulfide 5 at l = 350 nm in dioxane–water
(4+1, v/v) which affords 43% of thiazole 7 and 16% of
thiazolethione 6. It is interesting to note that the thiazole-
derived photoproducts 5–8 (Scheme 2) are identical to those
11 B. Giese, N. Porter and D. P. Curran, Stereochemistry of Radical
Reactions, VCH, Weinheim, 1995.
ˆ
12 W. Adam, J. Hartung, H. Okamoto, C. R. Saha-Möller and K. Spehar,
Photochem. Photobiol., 2000, 72, 619.
800
Chem. Commun., 2001, 799–800