Sustainable synthesis of thioimidazoles via carbohydrate-based multicomponent reactions

Article abstract of DOI:10.1021/ol502845h

The synthesis of diversely functionalized thioimidazoles through a modern variant of the Marckwald reaction is presented. This new protocol utilizes unprotected carbohydrates as well as simple amine salts as sustainable and biorenewable starting materials. Importantly it was discovered that a bifurcated reaction pathway results from using aldoses and ketoses respectively, yielding distinct reaction products in a highly selective manner.

Full text of DOI:10.1021/ol502845h

Letter
pubs.acs.org/OrgLett
Sustainable Synthesis of Thioimidazoles via Carbohydrate-Based
Multicomponent Reactions
Marcus Baumann and Ian R. Baxendale*
Department of Chemistry, Durham University, South Road, Durham, DH1 3LE, United Kingdom
S
* Supporting Information
ABSTRACT: The synthesis of diversely functionalized thioimida-
zoles through a modern variant of the Marckwald reaction is
presented. This new protocol utilizes unprotected carbohydrates as
well as simple amine salts as sustainable and biorenewable starting
materials. Importantly it was discovered that a bifurcated reaction
pathway results from using aldoses and ketoses respectively, yielding
distinct reaction products in a highly selective manner.
Scheme 1. Representative Imidazole-Forming
Multicomponent Reactions
ulticomponent reactions represent some of the most
Mversatile chemical transformations converting three or
more simple starting materials into a complex molecule in a
single cascade.1 As such, these reactions continue to play a
pivotal role in the assembly of complex molecular
architectures for both natural product and medicinal chemistry
programs.2 Despite the structural diversity resulting from
these atom- and step-economical reactions,3 sourcing the
appropriately prefunctionalized substrates can be cumbersome
as this commonly adds extra steps to the final synthetic route.
We therefore sought to address this generic problem by
making use of diverse yet highly abundant biorenewable
starting materials such as carbohydrates and amino acid
derivatives.4 We reasoned that combining these readily
available inputs with potassium isothiocyanate (KSCN)
would yield structurally diverse sets of thioimidazoles based
on the overlooked Marckwald thioimidazole synthesis.5
Importantly, this would allow for an efficient and highly
sustainable access into the important imidazole motif, which
can be found in numerous biologically active molecular
structures.6,7
We set out preparing several trisubstituted thioimidazoles
using different benzylamine hydrochloride salts (1), KSCN
(2), and dihydroxyacetone dimer (3, DHA) which are the
common inputs for this multicomponent reaction.8 Pleasingly,
when heating the combined starting materials in wet
acetonitrile (1 M, 2−5% water content) and glacial acetic
acid (1 equiv) a light colored suspension resulted allowing for
a very high yielding synthesis of products 4 (Scheme 1).
It was found that the heterogeneity of this reaction mixture
was crucial for a successful outcome as otherwise competing
Maillard processes9 would inevitably lead to numerous
undesired condensation products, whereas in the heteroge-
neous scenario the desired product would precipitate once
formed enabling simple product isolation through filtration.
Early investigations into the nature of the amine salt
component had also revealed that other salts are tolerated
although either lower yields (HOAc salts) or ester hydrolysis/
lactonization (H2SO4 salts) might result. Encouraged by these
results we decided to study the scope of this reaction by
varying the nature of the amine salt component and found
that various benzylamine derivatives as well as aliphatic
Received: September 30, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ol502845h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
amines as their corresponding HCl salts give the desired
thioimidazole products in high isolated yields. More
importantly, employing several amino ester salts was found
to also deliver the desired thioimidazole products (Figure 1).
Scheme 2. Polyhydroxyl-thioimidazoles Generated from
Different Aldoses
Scheme 3. Proposed Reaction Mechanism for Aldoses
(Shown for D-(+)-Xylose)
Figure 1. Extended scope of Marckwald thioimidazole synthesis (1−
can undergo an Amadori rearrangement10 allowing a
subsequent cyclocondensation to furnish the thioimidazole
scaffold (6e) upon elimination of water. Based on this
reasoning and confirmed by polarimetric experiments, aldose-
derived Marckwald reaction products are optically active
entities.
Intriguingly, when subjecting ketoses to the same reaction
conditions a very different outcome was observed delivering
bicyclic 6-hydroxytetrahydro-1H-furo[2,3-d]imidazole-2(5H)-
thiones (Figure 2, 7a−f) as major products. It is noteworthy
that although the bicyclic framework of compounds 7a−f is
precedented in carbohydrate literature,11 its conventional
synthesis typically starts from protected 2-aminosugars and
various isothiocyanates representing a more laborious and
cost-intensive synthesis than the transformation reported here
via a one-step multicomponent route.
150 mmol scale).
While the variation of the amine component had reliably
delivered the desired thioimidazole products tolerating many
(hetero)aromatic moieties as well as potentially labile esters
and carbamates (e.g., 4l), we decided to next challenge this
transformation by substituting dihydroxyacetone with highly
abundant carbohydrates such as tetroses, pentoses, and
hexoses. As such we chose to submit different amine
hydrochloride species (1) to our multicomponent protocol
with L-(+)-erythrulose, D-(+)-xylose, D-(−)-ribose, D-(+)-gal-
actose, or D-(−)-fructose as the carbohydrate component.
Pleasingly, the various thioimidazole structures were
obtained in high yield as reaction products when using
aldoses (e.g., D-(+)-xylose, D-(−)-ribose, and D-(+)-galactose)
as carbohydrate components (Scheme 2). Further analysis of
2D-NMR data (e.g. HMBC, see SI for details) revealed the
presence of a different regioisomeric thioimidazole product in
these cases.
Being a much more attractive route into these bicyclic
structures we decided to further study the mechanism of this
unexpected substrate-specific transformation.
The fact that these reactions consistently produced single
diastereomers (5a−f) prompted us to consider a mechanism
in which epimerization through enolization pathways would
only affect C1 and C2 of the acyclic carbohydrate substrates
(Scheme 3). The assumption that C3 is not undergoing
We propose that under the mildly acidic reaction conditions
both erythrulose and fructose exist in their acyclic keto-forms
analogous to the aldoses. This allows for a ketimine product
8a to form with the amine component prior to the reaction of
this species with KSCN (Scheme 4).
Once the resulting ternary adduct 8b is generated two
separate Amadori rearrangement pathways can occur leading
to either aldehyde 8c or ketone 8e. Aldehyde 8c can undergo
subsequent ring closure furnishing the cyclic thiourea 8d,
which provides the main reaction product 7a−c via 5-exo-trig
1
epimerization is furthermore supported by the fact that the H
and 13C NMR data for C3-epimeric product pairs such as 5a
and 5b are not identical (see Supporting Information for
details). We thus propose that an imine species 6b is initially
formed prior to its reaction with KSCN. The transient adduct
B
dx.doi.org/10.1021/ol502845h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
transformation as well as further green applications of related
chemistries.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, spectroscopic characterization of the
products, and CIF data for 7b (CCDC 1020070) and
(OAc)3-7f (CCDC 1020071). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: i.r.baxendale@durham.ac.uk.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We would like to acknowledge financial support of this
project through the Royal Society. Furthermore, we are very
grateful to Dmitry Yufit and Andrej Batsanov (Department of
Chemistry, Durham University) for solving the X-ray crystal
structures.
Figure 2. Ketose-based reaction products and X-ray structures of 7b
and (OAc)3-7f.
Scheme 4. Proposed Reaction Mechanism for the
Formation of Bicyclic Reaction Products from Ketoses
(Shown for L-(+)-Erythrulose)
REFERENCES
■
(1) (a) Domling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 3168.
̈
(b) Tietze, L. F. Chem. Rev. 1996, 96, 115.
(2) (a) Rikimaru, K.; Mori, K.; Kan, T.; Fukuyama, T. Chem.
Commun. 2005, 394. (b) Hulme, C.; Gore, V. Curr. Med. Chem.
2003, 10, 51. (c) Fayol, A.; Zhu, J. Org. Lett. 2005, 7, 239. (d) Kolb,
J.; Beck, B.; Domling, A. Tetrahedron Lett. 2002, 43, 6897. (e) Shoda,
̈
M.; Harada, T.; Kogami, Y.; Tsujita, R.; Akashi, H.; Kouji, H.;
Stahura, F. L.; Xue, L.; Bajorath, J. J. Med. Chem. 2004, 47, 4286.
(3) (a) Trost, B. M. Angew. Chem., Int. Ed. 1995, 34, 259.
(b) Sheldon, R. A. Pure Appl. Chem. 2000, 72, 1233.
(4) (a) Hollingsworth, R. I.; Wang, G. Chem. Rev. 2000, 100, 4267.
(b) Winterton, N. Chemistry for Sustainable Technologies: A
Foundation; Royal Society of Chemistry: Cambridge, U.K., 2010.
(5) (a) Marckwald, W. Ber. Dtsch. Chem. Ges. 1892, 25, 2354.
(b) Jones, R. G. J. Am. Chem. Soc. 1949, 71, 644.
(6) (a) Brackeen, M. F.; Stafford, J. A.; Feldman, P. L.; Karanewsky,
D. S. Tetrahedron Lett. 1994, 35, 1635. (b) Grimmett, M. R. Sci.
Synth. 2002, 12, 325. (c) Molina, P.; Conesa, C.; Desamparados, V.
M. Synthesis 1996, 12, 1459.
(7) (a) Baumann, M.; Baxendale, I. R.; Ley, S. V.; Nikbin, N.
Beilstein J. Org. Chem. 2011, 7, 442. (b) Grimmett, M. R. In
Comprehensive Heterocyclic Chemistry II, Vol. 3; Shinkai, I., Katrizky,
A. R., Rees, C. W., Scriven, E. F. V., Eds.; Pergamon: 1996; Chapter
3.02, pp 79−202.
(8) (a) Maligres, P. E.; Waters, M. S.; Weissman, S. A.;
McWilliams, J. C.; Lewis, S.; Cowen, J.; Reamer, R. A.; Volante, R.
P.; Reider, P. J.; Askin, D. J. Heterocycl. Chem. 2003, 40, 229.
(b) Hadizadeh, F.; Tafti, F. I. J. Heterocycl. Chem. 2002, 39, 84.
(c) Carini, D. J.; Duncia, J. V.; Johnson, A. L.; Chiu, A. T.; Price, W.
A.; Wong, P. C.; Timmermans, P. B. M. W. J. Med. Chem. 1990, 33,
1330. (d) Dener, J. M.; Zhang, L.-H.; Rapoport, H. J. Org. Chem.
1993, 58, 1159.
(9) (a) Hellwig, M.; Henle, T. Angew. Chem., Int. Ed. 2014, 53, 2.
(b) Ames, J. M. The Maillard Reaction. In Biochemistry of Food
Proteins; Hudson, B. J. F., Ed.; Elsevier Applied Science: London,
1992; pp 99−153.
(10) (a) Amadori, M. Atti. Reale accad. Nazl. Lincei 1925, 2, 337.
(b) Isbell, H. S.; Frush, H. J. J. Org. Chem. 1958, 23, 1309. (c) Kuhn,
R.; Weygand, F. Ber. Dtsch. Chem. Ges. A/B 1937, 70, 769.
cyclization. Alternatively, competing pathways (shown in gray)
can generate isolable byproducts such as 8g and 8h indicating
the presence of different concurrent reaction pathways.12
Overall, the dynamic nature of this proposed mechanism
explains the observed product outcome and appears to be
controlled by thermodynamic elements.
In conclusion, we have developed a modern variant of the
overlooked Marckwald multicomponent reaction that can
deliver diverse sets of thioimidazole products on multigram
scale. Importantly, the developed procedures rely on
biorenewable starting materials such as various carbohydrates
and amino acid derivatives providing a cheap and sustainable
access to these structures. During the course of our studies,
we discovered and exploited the substrate-specific generation
of either monocyclic thioimidazoles or bicyclic 6-hydroxy-
tetrahydro-1H-furo[2,3-d]imidazole-2(5H)-thiones in a proc-
ess starting from aldoses and ketoses, respectively. We believe
that our simple and high yielding synthetic routes to these
interesting entities will spark renewed interest in this
(d) Gallas, K.; Pototschnig, G.; Adanitsch, F.; Stutz, A. E.; Wrodnigg,
̈
T. M. Beilstein J. Org. Chem. 2012, 8, 1619.
C
dx.doi.org/10.1021/ol502845h | Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(11) (a) Avalos, M.; Babiano, R.; Cintas, P.; Jimenez, J. L.; Palacios,
J. C.; Valencia, C. Tetrahedron 1994, 50, 3273. (b) Avalos Gonzalez,
M.; Jimenez Requejo, J. L.; Palacios Albarran, J. C.; Ramos Montero,
M. D. Carbohydr. Res. 1987, 161, 49.
(12) These minor products were isolated by column chromatog-
raphy of the crude material obtained after evaporating the mother
liquors, from which the main product had been triturated.
D
dx.doi.org/10.1021/ol502845h | Org. Lett. XXXX, XXX, XXX−XXX