Switching between reaction pathways by an alcohol cosolvent effect: SmI2-ethylene glycol VS SmI2-H2O mediated synthesis of uracils

Article abstract of DOI:10.1021/ol502775w

A chemoselective switch between reaction pathways by an alcohol cosolvent effect in a general SmI₂-mediated synthesis of uracil derivatives is described. The method relies on the use of coordinating solvents to increase the redox potential of S

Full text of DOI:10.1021/ol502775w

Letter
pubs.acs.org/OrgLett
Switching between Reaction Pathways by an Alcohol Cosolvent
Effect: SmI2−Ethylene Glycol vs SmI2−H2O Mediated Synthesis of
Uracils
Michal Szostak,*,† Malcolm Spain,‡ Brice Sautier,‡ and David J. Procter*,‡
†Department of Chemistry, Rutgers University, 73 Warren Street, Newark, New Jersey 07102, United States
‡School of Chemistry, University of Manchester, Oxford Road, Manchester, M13 9PL, United Kingdom
S
* Supporting Information
ABSTRACT: A chemoselective switch between reaction pathways by
an alcohol cosolvent effect in a general SmI2-mediated synthesis of
uracil derivatives is described. The method relies on the use of
coordinating solvents to increase the redox potential of Sm(II) and
results in a chemoselective 1,2-reduction (SmI2−H2O) or 1,2-
migration via in situ generated N-acyliminium ions (SmI2−ethylene
glycol, EG). This work exploits the mild conditions of the SmI2-
mediated monoreduction of barbituric acids and offers an attractive protocol for the synthesis of uracil derivatives with biological
activity from readily accessible building blocks.
racil derivatives are ubiquitous pharmacophores in bio-
logically active compounds and pharmaceuticals.1,2 As
direct homologues of primary nucleobases, uracils have appeared
as selective modulators of ionotropic glutamate receptors2a and
exhibit activity against thymidine phosphorylase,2b hepatitis C
virus,2c and HIV-1 integrase2d (Figure 1). Thus, it is not
open-shell intermediates.10 Typically, strongly coordinating
alcohols (e.g., H2O and MeOH)11 are used to increase the
redox potential of Sm(II) and accelerate the otherwise slow
electron transfer and/or protonation steps.12 This may result in a
fully chemoselective reduction or cyclization depending on the
choice of alcohol cosolvent.11,12 Recently, multicomponent
reagents based on dual activation of SmI2 by alcohols and
Lewis bases have also emerged to direct SmI2-mediated
processes toward the reduction pathway;13 however, these
systems suffer from a prohibitively high redox potential.14 With
few exceptions,15 the development of alcohol-controlled selectivity
in SmI2-mediated reactions that lead to divergent, synthetically
useful products, while controlling the inherent preference of the
substrate by a simple change of an additive under mild reaction
conditions that tolerate sensitive functional groups,8 remains an
unmet challenge in reductive electron transfer chemistry.
U
Figure 1. Examples of pharmacologically active uracils.
Herein, we report a general SmI2-mediated synthesis of 5- and
6-substituted uracils from the same synthetic precursors in which
control of the reaction pathway is governed by the alcohol
cosolvent (Figure 2A−B).8,9 To date only a few examples of
additive-controlled selectivity in radical electron transfer
reactions have been reported.15 This new process enables an
operationally simple and diverse synthesis of functionalized
uracils1−4 from modular barbituric acid building blocks5 under
very mild single electron transfer conditions.6a Of general
interest is the first application of SmI2−ethylene glycol to the
synthesis of novel targets.16 Importantly, our results suggest that
surprising that the selective synthesis of 5- and 6-substituted
uracils has been the subject of numerous investigations.3,4 While
several approaches based on dearomatization, electrophilic
substitution, and condensation have been reported, these
methods are limited to specific substitution, require preassembly
of the uracil framework prior to the formation of desired
analogues, or proceed under harsh conditions.3,4 A powerful
approach to increase the diversity of accessible uracil analogues
would involve a direct conversion of hemiaminals obtained from
modular barbituric acids5 to divergent products from the same
synthetic precursor; however, until recently these hemiaminals
remained inaccessible due to the lack of methods for the
monoreduction of barbituric acids.6
12c,f
the use of ethylene glycol as a coordinating ligand for SmI2
will have broad applications in organic synthesis due to its
beneficial selectivity over SmI2−H2O.11a Mechanistic data
Selective SmI2-mediated reductive transformations7,8 in which
selection of a reaction pathway9 is governed by the choice of
alcohol cosolvent have a profound impact on the synthesis of
complex molecules and pharmaceutically relevant motifs via
Received: September 19, 2014
Published: October 24, 2014
© 2014 American Chemical Society
5694
dx.doi.org/10.1021/ol502775w | Org. Lett. 2014, 16, 5694−5697

Organic Letters
Letter
a
Table 1. Optimization of the Synthesis of 5-Alkyl Uracils
bc
,
b
entry
conditions
time
conv (%)
yield (%)
d
1
SmI2−H2O
SmI2−H2O
[O] to Sm(III)
HCl (1.0 N)
HCl (1.0 N)
BF3·Et2O
60 s
60 s
2 h
2 h
2 h
2 h
2 h
2 h
92
>95
39
34
59
23
72
77
75
75
77
e
2
f
3
g
4
91
h
5
>95
>95
>95
>95
h
6
h
7
TiCl4
h
8
p-TsOH
a
Barbituric acid, SmI2 (4 equiv), H2O (1000 equiv), THF, 23 °C.
b
c
Determined by 1H NMR. Refers to hemiaminal; in all entries >95%
d
e
f
conv of 1a. H2O (50 equiv). H2O (100 equiv). Oxidized to Sm(III).
g
h
Added in situ. Added after aqueous workup.
a
Table 2. Scope of the Synthesis of 5-Alkyl Uracils
Figure 2. (a) Previous study: the first selective reduction of barbiturates
enabled by Sm(II). (b) This study: chemoselective synthesis of uracils
via Sm(II)−solvent effect (SmI2−H2O vs SmI2−EG).
suggest that reaction with SmI2−EG proceeds via a rate-
determining conjugate reduction of uracils, which may find
applications in the chemoselective α,β-reduction of other
substrates.17 Finally, application to the synthesis of bicyclic
uracils18 is described.
We recently reported the first general reduction of barbituric
acids to the corresponding hemiaminals using SmI2−H2O as the
key reagent system (Figure 2A).6a We have also reported the first
generation of N-acyliminium ions6b and their vinylogs6c derived
from barbituric acids (not shown). During the development of
this process, we noticed that under certain conditions hemi-
aminals resulting from the monoreduction of barbituric acids
with SmI2−H2O underwent dehydration to give 5-substituted
uracils in trace quantities. We recognized that further
optimization of the reaction would allow for a streamlined
synthesis of 5-substituted uracils.1−4 Optimization studies were
conducted using barbituric acid 1a (Table 1). Pleasingly, the use
of SmI2 at a low concentration of the water additive afforded the
desired product, albeit in low yields due to nonselective
reduction of barbituric acids, consistent with previous studies
on the effect of water concentration on the redox potential of
Sm(II)19 (entries 1−2). Oxidation to Sm(III) after completion
of the reduction to enhance Lewis acidity11b had a minor effect
on the reaction efficiency (entry 3). Interestingly, the addition of
a protic acid significantly improved the yield (entry 4), with the
optimum results obtained when the acid was added after the
aqueous workup to remove Sm(III) salts, in a one-pot process
(entry 5). Importantly, a variety of other acids gave the desired 5-
substituted uracil in high yields under mild conditions (entries
6−8).
a
Barbituric acid, SmI2 (4 equiv), H2O (1000 equiv), THF, 60 s, 23 °C.
Then, Conditions A: HCl (1.0 N), 2 h. Conditions B: BF3·Et2O (3
equiv), 2 h. Conditions C: p-TsOH (3 equiv), 2 h. Conditions D: D2O
b
instead of H2O, p-TsOH, 2 h. >98% D1.
Furthermore, substrates bearing halide functional handles gave
the uracil product with no reduction in yield (entry 9).
Importantly, several of these functional groups are not
compatible with other SET conditions.7,8 Finally, by exploiting
the affinity of water for Sm(II),12c−f we demonstrated that the
protocol can be readily extended to the chemoselective synthesis
of 6-D1-uracils with >98% deuterium incorporation (entries 10−
11).20 The synthesis of these biologically important analogues
would be very difficult using current methods.
A wide range of barbituric acids was found to afford the
products in high yields (Table 2). Alkyl branched (entry 1),
linear (entries 2−4), and aromatic substrates (entry 5), including
barbituric acids with bulky substitution around the carbonyl
(entry 6), yielded the desired products with good efficiency for a
two-step process. Electron-donating and -withdrawing groups
were compatible with the developed protocol (entries 7−8).
Remarkably, during optimization of the reaction conditions
(Table 1), we established that the use of bidentate alcohols as
ligands for SmI2 resulted in the formation of 6-substituted-5,6-
5695
dx.doi.org/10.1021/ol502775w | Org. Lett. 2014, 16, 5694−5697

Organic Letters
Letter
a
dihydrouracils via a net reduction/iminium formation/1,2-
migration/conjugate reduction process (Table 3).16 This trans-
Table 4. Synthesis of Uracils Using SmI2−EG
a
Table 3. Optimization of the Synthesis of 6-Alkyl Uracils
b
entry
3
R1
R2
yield of 3 (%) selectivity
1
2
3a
3b
3c
3d
3e
3f
i-Bu
H
H
H
H
H
H
H
Et
76
85
83
85
84
89
78
77
89:11
97:3
91:9
90:10
93:7
93:7
93:7
(CH2)2Ph
bc
,
b
d
c
entry additive equiv
time
conv (%) yield (%) selectivity
3
CH2CH(Me)Ph
CH2Cy
4
5
6
7
1
2
3
4
5
6
7
8
9
−
H2O
−
3 h
29
94
7
41
84
81
17
49
10
32
6
97:3
73:27
93:7
92:8
63:37
91:9
71:29
95:5
89:11
−
(CH2)2-p-MeO-C6H4
(CH2)2-p-CF3-C6H4
(CH2)2-p-Br-C6H4
Ph
36
36
12
144
24
12
36
36
36
3 h
EG
3 h
>95
>95
82
3g
3h
EG
3 h
d
e
8
>95:5
EG
5 min
5 min
60 s
3 h
e
f
a
EG
>95
83
Barbituric acid, SmI2 (6 equiv), EG (36 equiv), THF, 3 h, 23 °C.
Selectivity refers to the ratio of rearrangement (3) vs dehydration (2)
product. 55:45 dr. 62:38 dr. Ph/Et migration selectivity.
b
EG
c
d
e
DEG
ED
47
3 h
29
olefin poised for further functionalization and are analogous to
xantine alkaloids,18a millipede metabolites,18b and pyrimidine
cross-linking models18c with important biological applications.
Several studies were conducted to gain insight into the
mechanism of the SmI2−EG promoted process (Scheme 1): (i)
10
DCH
3 h
24
<2
a
Barbituric acid, SmI2 (6 equiv), THF, 23 °C. Quenched with air after
b
c
1
the indicated time. Determined by H NMR. Refers to barbituric
d
acid; in all entries, <5% of hemiaminal. Selectivity refers to the ratio
e
of rearrangement (3) vs dehydration (2) product. SmI2 (4 equiv).
f
SmI2 (2 equiv).
Scheme 1. Studies Designed To Probe the Mechanism of
Rearrangement of Barbituric Acids Using SmI2−EG
formation was further evaluated using barbituric acid 1d.
Essentially, no reaction occurred in the absence of alcohol
(entry 1). The use of water at low concentration led to
decomposition and inconsistent results (entry 2). Notably, when
ethylene glycol16a was employed as the alcohol cosolvent, the
desired product was formed in high yield and with excellent
selectivity for 1,2-migration (cf. 1,2-reduction) (entry 3). A
reagent stoichiometry study revealed optimal conditions in terms
of yield and selectivity (entries 4−7): (i) at a high concentration
of EG oxidation to Sm(III) was observed, consistent with
previous studies (entry 5);14 (ii) at a lower SmI2 loading
incomplete conversion was observed (entries 6−7), consistent
with the presence of a reaction intermediate. Interestingly, other
chelating additives, including diethylene glycol (DEG), ethyl-
enediamine (ED), and trans-N,N′-dimethyl-1,2-cyclohexyl-
diamine (DCH), were less effective in promoting the reaction
(entries 8−10), which contrasts with studies on 5-exo-trig
cyclizations and suggests a unique role for ethylene glycol in this
process.16a
The synthesis of 6-substituted uracils from barbituric acids
using SmI2−EG is broad in scope and can accommodate an array
of substrates (Table 4). Thus, hindered (entries 1−4), electron-
rich (entry 5), electron-poor (entry 6), halide-containing (entry
7), and aryl substrates (entry 8) furnish the reductive 1,2-
migration products in good yields. Mono- and disubstitution are
tolerated (entries 1−8); however, at present, α,α-dialkyl
substituted barbituric acids are not viable substrates due to
steric hindrance around the carbonyl group (see Supporting
Information).19 Notably, the reaction selectivity is uniformly
high for all examples, favoring alkyl and aryl migration over
hydride and alkyl migration, respectively.21 Overall, the reaction
provides general access to 6-substituted uracils starting from the
same synthetic precursor as for the synthesis of 5-substituted
uracils by a simple change of the reaction conditions.
Deuterium incorporation studies demonstrate that anions are
generated and protonated in a series of electron transfer steps.16
Exchange of acidic protons with SmI2−ROH systems is a
common process.19b (ii) Determination of the kinetic isotope
effect suggests that the olefin reduction may be involved in the
rate-determining step;22 the KIE value is in an excellent
agreement with studies on the conjugate reduction of activated
acyclic olefins.12f (iii) Studies with a limiting amount of the
reagent led to the formation of a 6-hydrated intermediate. (iv)
Control experiments with α,β-unsaturated uracils using SmI2−
H2O and SmI2−EG demonstrate that the major reaction
pathway does not involve unsaturated 5-substituted uracils.
These results indicate the high levels of chemoselectivity possible
with the SmI2−EG system (cf. SmI2−H2O)11a for all stages of the
process. A mechanism that is consistent with the reactivity and
mechanistic studies outlined above is shown in Scheme 2.23
Furthermore, we have extended the reaction scope to include
bicyclic uracils prepared via reductive cyclization6a (Figure 2B
and Table 1-SI). The obtained products feature an endocyclic
5696
dx.doi.org/10.1021/ol502775w | Org. Lett. 2014, 16, 5694−5697

Organic Letters
Letter
(4) Selected examples: (a) Kopp, F.; Knochel, P. Org. Lett. 2007, 9,
1639. (b) Ulanenko, K.; Falb, E.; Gottlieb, H. E.; Herzig, Y. J. Org. Chem.
2006, 71, 7053. (c) Yu, Y. Y.; Georg, G. I. Chem. Commun. 2013, 49,
3694.
Scheme 2. Proposed Mechanism
(5) Bojarski, J. T.; Mokrosz, J. L.; Barton, H. J.; Paluchowska, M. H.
Adv. Heterocycl. Chem. 1985, 38, 229.
(6) (a) Szostak, M.; Sautier, B.; Spain, M.; Behlendorf, M.; Procter, D.
J. Angew. Chem., Int. Ed. 2013, 52, 12559. (b) Szostak, M.; Sautier, B.;
Procter, D. J. Org. Lett. 2014, 16, 452. (c) Szostak, M.; Sautier, B.;
Procter, D. J. Chem. Commun. 2014, 50, 2518.
(7) Recent reviews on SmI2: (a) Kagan, H. B. Tetrahedron 2003, 59,
10351. (b) Nicolaou, K. C.; Ellery, S. P.; Chen, J. S. Angew. Chem., Int.
Ed. 2009, 48, 7140. (c) Szostak, M.; Procter, D. J. Angew. Chem., Int. Ed.
2011, 50, 7737. (d) Szostak, M.; Fazakerley, N. J.; Parmar, D.; Procter,
D. J. Chem. Rev. 2014, 114, 5959.
(8) A review on chemoselective SmI2 reactions: Szostak, M.; Spain, M.;
Procter, D. J. Chem. Soc. Rev. 2013, 42, 9155.
In conclusion, we have described a mild and general, SmI2-
mediated method for the divergent synthesis of uracil derivatives.
This study provides one of the very few examples of alcohol
additive-controlled selectivity in SmI2-mediated reductive
processes that lead to synthetically useful products. Furthermore,
this study features the development of the SmI2−ethylene glycol
system as a mild and chemoselective reagent with redox
properties tailored to the desired transformation. We anticipate
that our findings will contribute to the development of new
cosolvent-controlled chemoselective SmI2 reactions. Studies in
this direction are underway in our laboratory.
(9) Reviews on chemoselective reactions: (a) Afagh, N. A.; Yudin, A. K.
Angew. Chem., Int. Ed. 2010, 49, 262. (b) Mahatthananchai, J.; Dumas,
A.; Bode, J. W. Angew. Chem., Int. Ed. 2012, 51, 10954.
(10) Reviews on metal-based radical transformations: (a) Gansauer,
A.; Bluhm, H. Chem. Rev. 2000, 100, 2771. (b) Szostak, M.; Procter, D. J.
Angew. Chem., Int. Ed. 2012, 51, 9238.
(11) Reviews on additives for SmI2: (a) Szostak, M.; Spain, M.; Parmar,
D.; Procter, D. J. Chem. Commun. 2012, 48, 330. (b) Dahlen
Hilmersson, G. Eur. J. Inorg. Chem. 2004, 3393.
(12) Selected examples: (a) Chopade, P. R.; Prasad, E.; Flowers, R. A.,
II. J. Am. Chem. Soc. 2004, 126, 44. (b) Prasad, E.; Flowers, R. A., II. J.
Am. Chem. Soc. 2005, 127, 18093. (c) Upadhyay, S. K.; Hoz, S. J. Org.
Chem. 2011, 76, 1355. (d) Yacovan, A.; Bilkis, I.; Hoz, S. J. Am. Chem.
Soc. 1996, 118, 261. (e) Tarnopolsky, A.; Hoz, S. J. Am. Chem. Soc. 2007,
129, 3402. (f) Amiel-Levy, M.; Hoz, S. J. Am. Chem. Soc. 2009, 131, 8280.
̈
́
, A.;
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and characterization data. This material
is available free of charge via the Internet at http://pubs.acs.org.
(13) (a) Dahlen
́
, A.; Hilmersson, G. Chem.Eur. J. 2003, 9, 1123.
(b) Dahlen, A.; Hilmersson, G. J. Am. Chem. Soc. 2005, 127, 8340.
́
(c) Ankner, T.; Hilmersson, G. Org. Lett. 2009, 11, 503. (d) Szostak, M.;
Spain, M.; Eberhart, A. J.; Procter, D. J. J. Am. Chem. Soc. 2014, 136,
2268.
AUTHOR INFORMATION
Corresponding Authors
■
(14) Szostak, M.; Spain, M.; Procter, D. J. J. Org. Chem. 2014, 79, 2522
and references cited therein.
(15) (a) Hutton, T. K.; Muir, K.; Procter, D. J. Org. Lett. 2002, 4, 2345.
(b) Hutton, T. K.; Muir, K. W.; Procter, D. J. Org. Lett. 2003, 5, 4811.
(16) (a) Sadasivam, D. V.; Teprovich, J. A., Jr.; Procter, D. J.; Flowers,
*E-mail: michal.szostak@rutgers.edu.
*E-mail: david.j.procter@manchester.ac.uk.
Notes
The authors declare no competing financial interest.
R. A., II. Org. Lett. 2010, 12, 4140. (b) Dahlen
Tetrahedron Lett. 2001, 42, 5565.
(17) Concellon, J. M.; Rodríguez-Solla, H. Eur. J. Org. Chem. 2006,
1613.
́
, A.; Hilmersson, G.
ACKNOWLEDGMENTS
́
■
We thank the EPSRC and Leverhulme Trust for support. M.S.
thanks Rutgers University for support during the preparation of
this manuscript.
(18) (a) Baraldi, P. G.; Tabrizi, M. A.; Gessi, S.; Borea, P. A. Chem. Rev.
2008, 108, 238. (b) Bergman, J.; Svensson, P. V. Tetrahedron 2010, 66,
4601. (c) Sun, G.; Fecko, C. J.; Nicewonger, R. B.; Webb, W. W.; Begley,
T. P. Org. Lett. 2006, 8, 681.
(19) (a) Szostak, M.; Spain, M.; Procter, D. J. J. Am. Chem. Soc. 2014,
136, 8459. (b) Szostak, M.; Spain, M.; Choquette, K. A.; Flowers, R. A.,
II; Procter, D. A. J. Am. Chem. Soc. 2013, 135, 15702.
(20) (a) Atzrodt, J.; Derdau, V.; Fey, T.; Zimmermann, J. Angew.
Chem., Int. Ed. 2007, 46, 7744. (b) For the synthesis of α,α-dideuterio
alcohols using SmI2−D2O, see: Szostak, M.; Spain, M.; Procter, D. J.
Org. Lett. 2014, 16, 5052.
REFERENCES
■
(1) (a) Bloomfield, V. A.; Crothers, D. M.; Tinoco, I. Nucleic Acids:
Structures, Properties and Functions; University Science Books: Sausalito,
CA, 2000. (b) Brunton, L.; Chabner, B.; Knollman, B. Goodman and
Gilman’s The Pharmacological Basis of Theraupeutics; McGraw-Hill: New
York, 2010.
(2) Selected examples: (a) Butini, S.; Pickering, D. S.; Morelli, E.;
Coccone, S. S.; Trotta, F.; De Angelis, M.; Guarino, E.; Fiorini, I.;
Campiani, G.; Novellino, E.; Schousboe, A.; Christensen, J. K.; Gemma,
S. J. Med. Chem. 2008, 51, 6614. (b) Kalman, T. I.; Lai, L. Nucleosides
Nucleotides Nucleic Acids 2005, 24, 367. (c) Liu, Y.; Lim, B. H.; Jiang, W.
W.; Flentge, C. A.; Hutchinson, D. K.; Madigan, D. L.; Randolph, J. T.;
Wagner, R.; Maring, C. J.; Kati, W. M.; Molla, A. Bioorg. Med. Chem. Lett.
2012, 22, 3737. (d) Embrey, M. W.; Wai, J. S.; Funk, T. W.; Homnick, C.
F.; Perlow, D. S.; Young, S. D.; Vacca, J. P.; Hazuda, D. J.; Felock, P. J.;
Stillmock, K. A.; Witmer, M. V.; Moyer, G.; Schleif, W. A.; Gabryelski, L.
J.; Jin, L.; Chen, I. W.; Ellis, J. D.; Wong, B. K.; Lin, J. H.; Leonard, Y. M.;
Tsou, N. N.; Zhuang, L. Bioorg. Med. Chem. Lett. 2005, 15, 4550.
(3) (a) Bardagi, J. I.; Rossi, R. A. Org. Prep. Proced. Int. 2009, 41, 479.
(b) Joule, J. A.; Mills, K. Heterocyclic Chemistry; Wiley: 2010.
(21) For a related 1,2-shift, see: Nguyen, Q.; Nguyen, T.; Driver, T. G.
J. Am. Chem. Soc. 2013, 135, 620.
(22) Simmons, E. M.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51,
3066.
(23) An alternative mechanism involving a concerted 1,2-shift/−OH
departure is also possible. For the reduction of iminium ions with Sm(II)
reagents, see: Kim, M.; Knettle, B. W.; Dahlen
́
, A.; Hilmersson, G.;
Flowers, R. A., II. Tetrahedron 2003, 59, 10397.
5697
dx.doi.org/10.1021/ol502775w | Org. Lett. 2014, 16, 5694−5697