Easy access to the family of thiazole N-oxides using HOF·CH 3CN

Article abstract of DOI:10.1039/b602594c

An efficient procedure for transferring an oxygen atom to thiazole-containing compounds, resulting in the corresponding N-oxides, was developed by using HOF·CH₃CN; mild reaction conditions, high yields and easy purification are the main features of this novel route, while X-ray structural analysis reveals a hydrogen bond between the N-oxide functionality and a water molecule. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b602594c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Easy access to the family of thiazole N-oxides using HOF?CH3CN{
Elizabeta Amir and Shlomo Rozen*
Received (in Cambridge, UK) 21st February 2006, Accepted 5th April 2006
First published as an Advance Article on the web 25th April 2006
DOI: 10.1039/b602594c
facilitates the transfer of the oxygen atom to most nucleophilic
sites under mild conditions (0 uC or lower, in seconds or minutes).
This complex has been instrumental in transforming azides and
amines, including vicinal ones, into their corresponding nitro
derivatives,10 in the epoxidation of practically every conceivable
olefin,11 in the unprecedented oxidation of oligothiophenes to their
corresponding S,S-dioxides,12 and much more.13 Exploring the
possibilities offered by this reagent in transforming thiazole rings
directly into their desired corresponding N-oxide derivatives was
therefore quite attractive, and the results were rewarding.
An efficient procedure for transferring an oxygen atom to
thiazole-containing compounds, resulting in the corresponding
N-oxides, was developed by using HOF?CH3CN; mild reaction
conditions, high yields and easy purification are the main
features of this novel route, while X-ray structural analysis
reveals a hydrogen bond between the N-oxide functionality and
a water molecule.
The thiazole ring appears in a wide range of natural and synthetic
products, such as epothilones and thiopeptide antibiotics,
oligosaccharides, C-glycosides, and more.1 Dynamic interest lay
behind the development of many modifications of this heterocycle,
but mainly concentrated on the carbon atoms of the ring. Changes
around the heteroatoms were relatively rare, with the notable
exception of the synthesis of N-oxide derivatives, which captured
the attention of the chemical and especially pharmaceutical
communities.
Treating 2,4-dimethylthiazole (1a) with
2 equivalents of
HOF?CH3CN for 20 min at room temperature resulted in the
formation of the previously unknown N-oxide 2a in 91% yield.
While none of the double bonds of the thiazole moiety were
affected by this reaction, a small amount of water soluble
N,S,S-trioxide 3a was also formed and identified. It is worth
mentioning that normal MS methods could not detect the
molecular ion of the fully oxygenated 3a, and we had to resort
to Amirav’s supersonic GC-MS, developed in our department, to
detect it.14
A few examples underline the reasons for this interest. b-Lactam
antibiotics, such as cephalosporins, bearing thiazole N-oxide
derivatives, display unusual and effective antibacterial properties.2
Thiazole N-oxide analogues of epothilones A and B serve as novel
anticancer agents.3 Recent studies on the design and synthesis of
oligonucleotides containing thiazole N-oxide disclose that this
functionality is unique in the sense of introducing a pronounced
directional dipole into molecules that is capable of forming specific
and strong hydrogen bonds.4 Such biologically important findings
have triggered intensive efforts aimed at the preparation of
the thiazole N-oxide moiety, but the built-in limitations of the
orthodox routes used did not result in conspicuous success. The
synthesis, via cyclization of v-(dialkylamino-oximinomercapto)-
acetophenone, involves a multi-step reaction and is restricted to
specific starting materials,5 while direct oxidation using meta-
CPBA or permaleic acid requires very long reaction times, leading
to thiazole N-oxides in yields of 15 to 50% only.3,4,6,7 There is only
one example describing the formation of thiazole N-oxide
C-nucleoside in good yield using trifluoroacetic anhydride and
hydrogen peroxide–urea complex.4 Clearly, a general and efficient
methodology for the preparation of this important family is
needed.
Steric hindrance is not a compelling factor in this reaction, so
replacing the methyl group at the C-2 position with a bulkier
isopropyl one, as in 1b, did not change the outcome, and the new
2-isopropyl-4-methylthiazole N-oxide (2b) was formed after 30 min
in 87% yield. It was important to find out whether an
unsubstituted thiazole at C-2 could also be transferred to the
corresponding N-oxide, since it could open the way for further
chemical transformations. 4,5-Dimethylthiazole (1c) and 4-methyl-
5-thiazolylethyl acetate (1d) served as test compounds, and both
were converted in excellent yields to their respective N-oxides 2c
and 2d. The latter also demonstrated that protected alcohols will
not interfere with the oxygen transfer process. The N-oxide moiety
did not prevent reactions at C-2, such as bromination with NBS,
which formed 2-bromo-4-methyl-5-thiazolylethyl acetate N-oxide
(2d–Br) after 30 min in 91% yield. Obviously, introducing a
bromine atom at the 2-position means that a host of other
reactions become feasible.
As in many biologically important compounds the thiazole ring
is often fully substituted, it was of interest to find out if such
trisubstituted heterocycles would undergo an oxygen transfer
process. Compounds 1e, 1f and 1g served as model molecules and
were subjected to the HOF?CH3CN complex. In all these cases, a
three-fold excess of the oxidant was required, but the new major
products 2,4,5-trimethylthiazole N-oxide (2e), 2-ethyl-4,5-
dimethylthiazole N-oxide (2f) and 2,4-dimethyl-5-acetylthiazole
N-oxide (2g) were obtained in high yield. It should be noted that in
all of these reactions, small amounts (5–10%) of the N,S,S-trioxo
derivatives (3e–3g) were also formed, but their separation was
straightforward since they are all soluble in water. While
The readily made HOF?CH3CN complex8 is considered to be
one of the best oxygen transfer agents in chemistry.9 Unlike all
other oxygen transfer reagents, HOF?CH3CN is a unique source
of a permanent electrophilic oxygen species since it is bound to
fluorine, the only atom more electronegative than itself. This
School of Chemistry, Tel-Aviv University, Tel-Aviv, Israel 69978.
E-mail: rozens@post.tau.ac.il; Fax: +972 3 640 9293;
Tel: +972 3 640 8378
{ Electronic supplementary information (ESI) available: Experimental
procedures and full spectral data. See DOI: 10.1039/b602594c
2262 | Chem. Commun., 2006, 2262–2264
This journal is ß The Royal Society of Chemistry 2006

View Article Online
HOF?CH3CN is known to react with the p-electrons of aromatic
rings, such as benzene or phenanthrene derivatives, resulting in the
corresponding epoxides,15 it reacts much faster with the nitrogen
atom of thiazole rings. This is demonstrated by the reaction of
2-methylnaphtho[1,2-d]thiazole (1h) with HOF?CH3CN, which
produced the corresponding N-oxide 2h in 78% yield without
affecting the naphthalene core.
One of the advantages of HOF?CH3CN is that the origin of its
electrophilic oxygen atom is water, which is the most convenient
source of all oxygen isotopes. H18OF?CH3CN, prepared by
passing diluted fluorine through H218O and acetonitrile, was
reacted with 1a, leading to thiazole 2a with a N–18O functionality.
This functionality is difficult to introduce by other routes, and can
serve as a probe in degradation and other studies. The heavy
oxygen in the N–18O moiety is not interchangeable with the
common 16O isotope when treated with regular water or exposed
to air, as evident from its HRMS (CI) (m/z): calc. for C5H8N18OS
132.036906 (MH)+, found 132.036609.
Fig. 1 X-Ray crystal structure of 2g.
Additional evidence for N-oxide vs. S-oxide formation can be
1
extracted from the H–15N heteronuclear NMR shift correlation
experiment we performed on 2f and its starting material 1f. The
results reveal a strong long-range correlation from the methyl
hydrogens at the C-4 carbon, as well as from the methylene
hydrogens of the ethyl moiety at the C-2 position, to the N-3
3
nitrogen. While the JNH correlation for 1f is observed at
318.7 ppm,19 the corresponding value for the product 2f is at
276.6 ppm (liquid ammonia was used as a chemical shift reference
standard). A similar shift to higher field of about 40 ppm for the
thiazole N-oxide 2g relative to its non-oxygenated precursor 1g
(from 321.8 to 283.4 ppm, respectively) was also recorded.
Another important feature of the thiazole N-oxide family can be
revealed from their UV-vis spectra. There are two well-defined
differences between the starting materials 1 and the respective
products 2. While the spectra of the thiazoles present a single
absorption maximum around 250 nm (see the ESI{), the
corresponding N-oxides display two major peaks. The first has a
value close to the one of the starting material, while the second
maximum is red-shifted by 24–76 nm. These shifts and the
consequent narrowing of the HOMO–LUMO energy gap, derived
from the absorption maxima, point to the greater electron
delocalization and electron affinity of the thiazole N-oxides, a
very important feature for organoelectronic devices.12
In conclusion, it is clear that the HOF?CH3CN complex is a
powerful reagent for the preparation of almost any thiazole
N-oxide in fast and high yielding reactions. Despite the super-
stitious reluctance of some chemists to use F2 (needed for the
reagent preparation), working with this halogen is no more
complicated than working with chlorine. Moreover, the commer-
cial availability of pre-diluted fluorine makes this procedure very
simple indeed.
In the past we had found that there was no distinct preference
by the electrophilic oxygen in HOF?CH3CN towards nitrogen or
sulfur atoms.16 Despite the aromaticity of the ring, which should
encourage attack at the nitrogen, there is always the remote
possibility of the oxygen reacting with the sulfur atom to form a
sulfoxide. However, X-ray structural analysis performed on 2g
(Fig. 1) clearly shows that an N-oxide moiety is obtained.17 The
thiazole ring remains planar and there is a hydrogen bond
between the oxygens of two thiazole N-oxide molecules and the
hydrogens of water, an important feature of thiazole N-oxide-
containing nucleosides when forming strong contacts with DNA
polymerases.18
This work was supported by the Israel Science Foundation.
Notes and references
1 K. C. Nicolaou, A. Ritzen and K. Namoto, Chem. Commun., 2001, 17,
1523–1535; A. Dondoni and A. Marra, Chem. Rev., 2004, 104,
2557–2599; M. C. Bagley, J. W. Dale, E. A. Merritt and X. Xiong,
Chem. Rev., 2005, 105, 685–714; K. C. Nicolaou, B. S. Safina, M. Zak,
S. H. Lee, M. Nevalainen, M. Bella, A. A. Estrada, C. Funke, F. J. Zecri
and S. Bulat, J. Am. Chem. Soc., 2005, 127, 11159–11175.
2 E. Perrone, M. Alpegiani, F. Giudici, F. Buzzetti, G. Nannini,
G. Meinardi, S. Grasso, A. Bianchi and I. De Carneri, J. Antibiot.,
1984, 37, 1423–1440.
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 2262–2264 | 2263

View Article Online
3 G. Hofle, N. Glaser, T. Leibold and M. Sefkow, Pure Appl. Chem.,
1999, 71, 2019–2024.
12 E. Amir and S. Rozen, Angew. Chem., Int. Ed., 2005, 44, 7374–7378.
13 For the latest review on HOF?CH3CN, see: S. Rozen, Eur. J. Org.
Chem., 2005, 12, 2433.
14 A. B. Fialkov and A. Amirav, J. Chromatogr., A, 2004, 1058,
233–242.
15 M. Kol and S. Rozen, J. Org. Chem., 1993, 58, 1593–1595; S. Rozen,
Y. Bareket and J. Blum, Tetrahedron Lett., 1997, 38, 2333–2334.
16 S. Rozen and Y. Bareket, J. Org. Chem., 1997, 62, 1457–1462.
17 C14H20N2O5S2, M = 360.45, monoclinic, space group C2/c, a =
25.7986(15), b = 6.0590(3), c = 10.8001(6) s, b = 99.7930 (10)u, U =
1663.60(16) s3, T = 293 K, Z = 4, m(Mo-Ka) = 0.346 mm21, 8836
reflections measured, 1807 independent (Rint = 0.0176) which were used
in all calculations. The final R1 was 0.0419 and wR(F2) was 0.1065 (all
data). CCDC 289613. For crystallographic data in CIF or other
electronic format see DOI: 10.1039/b602594c.
4 T. J. Miller, H. D. Farquar, A. Sheybani, C. E. Tallini, A. S. Saurage,
F. R. Fronczek and R. P. Hammer, Org. Lett., 2002, 4, 877–880.
5 K. Hirai and T. Ishiba, Heterocycles, 1978, 9, 1223–1231.
6 N. Honjo, T. Niiya and Y. Goto, Chem. Pharm. Bull., 1982, 30,
1722–1730.
7 E. Perrone, M. Alpegiani, F. Giudici, A. Bedeschi, R. Pellizzato and
G. Nannini, J. Heterocycl. Chem., 1984, 21, 1097–1111.
8 A detailed procedure for the preparation of HOF?CH3CN from either
commercially pre-diluted or pure (95%) fluorine can be found in:
S. Dayan, Y. Bareket and S. Rozen, Tetrahedron, 1999, 55, 3657–3664.
9 S. Rozen and S. Dayan, Angew. Chem., Int. Ed., 1999, 38, 3471–3473.
10 E. Golan and S. Rozen, J. Org. Chem., 2003, 68, 9170–9172; S. Rozen
and M. Carmeli, J. Am. Chem. Soc., 2003, 125, 8118–8119; G. K. S.
Prakash and M. Etzkorn, Angew. Chem., Int. Ed., 2004, 43, 26–28.
11 S. Rozen and E. Golan, Eur. J. Org. Chem., 2003, 1915–1917; E. Golan,
A. Hagooly and S. Rozen, Tetrahedron Lett., 2004, 45, 3397–3399.
18 J. R. Kiefer, C. Mao, J. C. Braman and L. S. Beese, Nature, 1998, 391,
304–307.
19 G. E. Martin and C. E. Hadden, J. Nat. Prod., 2000, 63, 543–585.
2264 | Chem. Commun., 2006, 2262–2264
This journal is ß The Royal Society of Chemistry 2006