Decomposition of 2-(1-hydroxybenzyl)thiamin. Ruling out stepwise cationic fragmentation.

Article abstract of DOI:10.1021/ol005927i

[reaction: see text] The rapid fragmentation of 2-(1-hydroxybenzyl)thiamin (1) is initiated by transfer of a proton from C2alpha to give an enamine. The subsequent irreversible process can be written as a concerted (or stepwise) rearrangement involving migration of the hydroxyl hydrogen to the methylene bridge. An attractive alternative is internal addition of C2alpha to the pyrimidine, generating a carbocation. However, addition of azide to the reaction solution, which could trap the carbocation, has no effect on the rate or products of reaction.

Full text of DOI:10.1021/ol005927i

ORGANIC
LETTERS
2
000
Vol. 2, No. 14
035-2036
Decomposition of
-(1-Hydroxybenzyl)thiamin. Ruling Out
2
2
Stepwise Cationic Fragmentation
Ian F. Moore and Ronald Kluger*
Lash Miller Laboratories, Department of Chemistry, UniVersity of Toronto, Toronto,
Ontario, Canada M5S 3H6
rkluger@chem.utoronto.ca
Received April 11, 2000
ABSTRACT
The rapid fragmentation of 2-(1-hydroxybenzyl)thiamin (1) is initiated by transfer of a proton from C2r to give an enamine. The subsequent
irreversible process can be written as a concerted (or stepwise) rearrangement involving migration of the hydroxyl hydrogen to the methylene
bridge. An attractive alternative is internal addition of C2r to the pyrimidine, generating a carbocation. However, addition of azide to the
reaction solution, which could trap the carbocation, has no effect on the rate or products of reaction.
In neutral solution the adduct of thiamin (vitamin B1) and
benzaldehyde, 2-(1-hydroxybenzyl)thiamin (HBzT 1, R )
H, an intermediate in the formation of benzoin), undergoes
a surprising fragmentation reaction that splits the molecule
into its pyrimidine and thiazole components. This over-
whelms reversion to thiamin and benzaldehyde. The reaction
occurs with oxidation at C2R and is promoted by a positive
charge at N(1′).1 We have previously reported results that
establish that the loss of the C2R proton, which generates
Scheme 1
,2
3
an enamine, is the initial step in the process. This can be
rate-determining, and subsequent steps compete with repro-
tonation. We have determined that the rate constant for
mentation of the C2R carbanion is shown in Scheme 2. It is
a variation of that discovered by Zoltewicz for the sulfite-
5
-1
fragmentation of the enamine is 1 × 10 s and have
catalyzed fragmentation of thiamin, which also applies to
proposed an electrocyclic mechanism, Scheme 1.
4
2
-(1-hydroxyethyl)thiamin (HET). The proposed mechanism
Although the first step in the mechanism is clearly proton
abstraction, the nature of subsequent steps has not been
established. An intriguing alternative mechanism for frag-
involves intramolecular attack of the C2R carbanion at the
′ position of the pyrimidine ring. Expulsion of the thiazole
follows to give the resonance-stabilized cation 2 (Scheme
). This cation cleaves at C2R-C6′ to give the pyrimidine
6
2
(
(
1) Kluger, R.; Lam, J. F.; Kim, C.-S. Bioorg. Chem. 1993, 21, 275.
2) Kluger, R.; Lam, J. F.; Pezacki, J. P.; Yang, C.-M. J. Am. Chem.
and the phenyl thiazole ketone (Scheme 2).
Soc. 1995, 117, 11383.
3) Kluger, R.; Moore, I. F. J. Am. Chem. Soc. In press.
(
(4) Zoltewicz, J. A.; Kauffman, G. M. J. Am. Chem. Soc. 1977, 99, 3134.
1
0.1021/ol005927i CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/08/2000

the reaction mixture by extraction. Analysis of the extract
by NMR indicated the presence of only the phenyl thiazole
ketone. The freeze-dried aqueous layer contained the N(1′)-
Scheme 2
5
methylpyrimidine but no azidopyrimidine (compared with
6
a synthesized sample).
Azide ion is an efficient cation trapping reagent with
9
-1 -1
second-order rate constants ranging from 5 × 10 M
s
5
-1 -1
7,8
to 5 × 10 M
s
(the latter for stable cations). Using
the lower limit for trapping 2, we obtain a predicted first-
6
-1
order rate constant of 1 × 10 s under our experimental
conditions. This is fast enough to compete with the frag-
5
-1 3
mentation step (k ) 1 × 10 s ). Since no azido product
is formed, either intermediate 2 is not formed or it breaks
6
down with a rate constant that is much larger than 1 × 10
-
1
s . If the latter were the case, then the two steps preceding
1 2
the breakdown of 2 (k and k in Scheme 2) would have to
5
-1
have rate constants of 1 × 10 s or more, which is
inconsistent with the reactivity of this type of species: the
rate constant for cleavage of 2-1-(hydroxyethyl)thiamin by
-
1
-1
sulfite is 0.12 M s , with the addition of sulfite and
elimination of the thiazole partially rate-limiting.3
,9
To distinguish the two mechanisms, we designed an
experiment based on Zoltewicz’s observation that azide ion
can trap a cationic intermediate following addition of a
nucleophile to the pyrimidine. Thus, cation 2 would be
trapped by addition of azide, giving the azido-substituted
pyridinium salt (3) and the reduced phenyl thiazole (4,
Scheme 3).
Thus, our results establish that the fragmentation avoids
carbocationic intermediates, which would be trapped by
azide. The large rate constant and the lack of trapping
indicate that an electrocyclic route for fragmentation should
1
0
be considered as being likely.
Acknowledgment. We thank the Natural Sciences and
Engineering Research Council of Canada for support through
an operating grant and John Pezacki for helpful discussions.
Ian Moore is the recipient of an Ontario Graduate Scholarship
in Science and Technology.
Scheme 3
OL005927I
(
(
5) Singlet at δ 4.6 for (CH2N3).
6) Synthesized by reaction of N(1′)-methylthiamin and sodium sulfite
(
0.1 equiv) in the presence of sodium azide (2.5 equiv). We have estimated
the minimum amount detectable by NMR to be 0.1 mg or <0.5% based on
NMHBzT.
(
7) McClelland, R. A. Tetrahedron 1996, 52, 6823.
(8) Richard, J. P.; Toteva, M. M.; Crugeiras, J. J. Am. Chem. Soc. 2000,
N(1′)-MethylHBzT (NMHBzT, R ) methyl) (0.040 g) was
dissolved in 4.0 mL of 3.0 M sodium azide. The solution
was kept at pH 8 (pH-stat). After 3 h, the solution was
brought to pH 5. The thiazole fragment was separated from
1
22, 1664.
(9) Uray, G.; Celotto, C.; Ibovnik, A.; Zoltewicz, J. A. J. Org. Chem.
989, 54, 3941.
1
(10) Less, A.; Nendel, M.; Wilsey, S.; Houk, K. N. J. Am. Chem. Soc.
1999, 121, 4524 and references therein.
2036
Org. Lett., Vol. 2, No. 14, 2000