Catalyst-free three-component synthesis of highly functionalized 2,3-dihydropyrroles

Article abstract of DOI:10.1039/c8gc00987b

An efficient synthesis of fully substituted 2,3-dihydropyrroles has been achieved in one step through the three-component reaction of amines, aromatic aldehydes and α-ketoamides. This atom-economical and catalyst-free reaction is highly stereoselective and generates underexplored heterocycles in a single step. These compounds were examined in an enzymatic assay that led to the identification of potent α-glucosidase inhibitors, thereby demonstrating the utility of this novel methodology in medicinal chemistry.

Full text of DOI:10.1039/c8gc00987b

Green Chemistry
View Article Online
View Journal
PAPER
Catalyst-free three-component synthesis of highly
functionalized 2,3-dihydropyrroles†
Cite this: DOI: 10.1039/c8gc00987b
Dong Wang, a Linna Li,a Hairong Feng,a Hua Sun, a Fabrice Almeida-Veloso,b
a,b,c
Marine Charavin,b Peng Yu a and Laurent Désaubry
*
An efficient synthesis of fully substituted 2,3-dihydropyrroles has been achieved in one step through the
three-component reaction of amines, aromatic aldehydes and α-ketoamides. This atom-economical and
catalyst-free reaction is highly stereoselective and generates underexplored heterocycles in a single step.
These compounds were examined in an enzymatic assay that led to the identification of potent
α-glucosidase inhibitors, thereby demonstrating the utility of this novel methodology in medicinal
chemistry.
Received 28th March 2018,
Accepted 13th May 2018
DOI: 10.1039/c8gc00987b
rsc.li/greenchem
Introduction
Dihydropyrroles are a common structural motif found in a
number of naturally occurring alkaloids and biologically active
substances.1 In particular, 2,3-dihydropyrroles have been
reported to be important precursors for the synthesis of a
variety of natural products2 and other complex molecules.3
Several elegant methods have been reported for the synthesis
of these compounds but each of these is restricted to limited
types of substrates, and consequently delivers 2,3-dihydro-
pyrroles with a limited variety of substituents (Scheme 1).4
Herein, we report a highly efficient, one-step, stereoselective
and green synthetic approach to 2,3-dihydropyrrroles that
display a novel pattern of substitutions using simple amines,
aldehydes and easily accessible pyruvic amides as starting
materials.
Three-component reactions are one-pot reactions that use
three reagents to generate a single product containing most of
the reagent atoms.5 Due to their atom-sparing and single-step
nature, these reactions are extremely valuable in medicinal
chemistry.6 The Doebner reaction is one of the oldest known
three-component reactions. It generates substituted cinchoni-
aSino-French Joint Lab of Food Nutrition/Safety and Medicinal Chemistry, China
International Science and Technology Cooperation Base of Food Nutrition/Safety and
Medicinal Chemistry, College of Biotechnology, Tianjin University of Science and
Technology, Tianjin 300457, China. E-mail: desaubry@unistra.fr
bLaboratory of Therapeutic Innovation (UMR 7200), University of Strasbourg-CNRS,
Faculty of Pharmacy, 67400 Illkirch, France
Scheme 1 Selected processes for the synthesis of 2,3-dihydropyrroles.
nic acids (or quinoline-4-carboxylic acids) from anilines, alde-
hydes and pyruvic acids (Scheme 2a).7 Since its discovery in
1887, only one article has been published describing this reac-
tion with a pyruvic amide. While the latter was immobilized
on a Rink polystyrene resin (Scheme 2b)8 the same reaction in
solution has never been described to our knowledge.
cLaboratory of Biomolecules (UMR7203), Sorbonne University – CNRS,
4 place Jussieu, 75005 Paris, France
†Electronic supplementary information (ESI) available: Detailed experimental
procedures, compound characterization, and NMR spectra. CCDC 1822869 and
1580851. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c8gc00987b
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Table 1 Reaction optimization for the three-component reactiona
Entry
Additive
M
Solvent
Yield (%)
1
2
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
TFA (0.4 ml)
PTSA (25 mg)
Proline (12 mg)
AcOH (0.2 ml)
AcOH (0.4 ml)
AcOH (0.6 ml)
AcOH (0.6 ml)
AcOH (0.4 ml)
0.05
0.02
0.01
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
0.02
EtOH
EtOH
EtOH
EtOH
n-PrOH
i-PrOH
CF3CH2OH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
N.A.
14
22
3
n/rb
13
4c
5
n/r
n/r
n/r
22
n/r
n/r
n/r
32
51
40
27
n/r
6
7
8d
9
Scheme 2 (a) A classical Doebner reaction, (b) the sole published
example of the Doebner reaction performed with a pyruvic amide8 and
(c) the unexpected formation of dihydropyrroles during an attempt to
perform the Doebner reaction with a pyruvic amide in solution.
10
11
12
13
14
15e
16f
Results and discussion
Unexpected divergent pathway of the Doebner reaction using a
EtOH
a Unless otherwise noted, all reactions were conducted at 0.025 M
(p-anisidine concentration) with p-anisidine (1.0 equiv.), p-tolualde-
hyde (2.0 equiv.), and pyruvic amide (40 mg, 2.0 equiv.) under reflux
conditions. b No reaction. c Reaction run with p-anisidine (2.0 equiv.),
p-tolualdehyde (2.0 equiv.), and pyruvic amide (40 mg, 2.0 equiv.)
under reflux conditions. d Reaction run in the dark. e Reaction run in
neat conditions. f Reaction run under microwave conditions, tempera-
ture: 100 °C.
pyruvic amide instead of pyruvic acid
When we attempted to perform the Doebner reaction in solu-
tion, we were surprised to find that a 2,3-dihydropyrrole 4 was
formed, rather than the expected cinchoninic amide
(Scheme 2c). It seems likely that this reaction has been
attempted during the last 130 years but the products obtained
were probably discarded because they were difficult to identify.
Indeed, we had to resort to X-ray crystallography to establish
the structure of the product unambiguously. Given the orig-
Reaction scope and limitation
inality of this reaction and its potential utility in both organic Using our optimized reaction conditions, we tested various
and medicinal chemistry, we decided to explore the scope and amines (1), aldehydes (2) and pyruvic amides (3) (Scheme 3).
limitations of this novel three-component reaction.
Dihydropyrroles 4 were obtained with modest to good yields
In our initial study, p-anisidine (1a), p-tolualdehyde (2a) and with excellent diastereoselectivity (cis methyl and hydro-
and pyruvic amide 3a were used to optimize the reaction con- gen) in all cases. Yields were highest for the chlorine-contain-
ditions. We found that the solvent and the concentration ing aromatic pyruvic amides 4a–c, whereas lower yields were
significantly influenced the outcome of the reaction (entries 1 obtained for the electron-rich or poor aromatic pyruvic
to 7, Table 1). Ethanol was shown to be superior to isopro- amides, 4d–g. On the other hand, this reaction was not
panol or trifluoroethanol. The pyruvic amide was not comple- effective with aliphatic pyruvic amides as the adducts 4h and
tely consumed, so additional p-anisidine was added, but this 4i were not observed. We then assessed the reactivity of
resulted in a lower yield (entry 4). The product yield (22%, various amines. The replacement of anisidine 1a with 4-benzyl-
entry 8) was similar in complete darkness, indicating that light oxy- or 4-phenyl aniline was slightly deleterious (4j: 37%, 4k:
has no effect on the reaction. We then investigated whether 26%), and the less nucleophilic 4-chloroaniline did not gene-
various acids could increase yields (entries 9 to 16). PTSA, TFA rate the expected adduct 4l. 1-Naphthylamine did not produce
and L-proline had deleterious effects, but a much higher the adduct 4m, indicating a sensitivity of this three-component
reaction yield was obtained with acetic acid (entry 13).
reaction to steric hindrance. We found that n-butylamine
Running the reaction under neat or microwave conditions efficiently yielded the expected product 4n (51%), demonstrat-
resulted in a large yield loss (entries 15 and 16). We therefore ing the suitability of aliphatic amines for this reaction.
concluded that the best ratio of p-anisidine, p-tolualdehyde and Benzylamine gave rise to 4o, but with a yield of only 28%, con-
3a for this three-component reaction was 1 : 2 : 2, in ethanol firming the sensitivity to steric hindrance of the amine. We
(0.025 M) containing HOAc (VEtOH : VAcOH = 12 : 1) under reflux were able to crystallize 4o and to confirm its structure by X-ray
(51%, entry 13). It is worth noting that only product 4a having crystallography (ESI,† part 4). For the aldehyde component 2,
the cis configuration (as confirmed by NOE NMR experiments, chloro- and methoxy-benzaldehydes generated adducts 4p–u
see the ESI† for details) was produced in this transformation.
with yields between 30% and 40%. Surprisingly, furfural gave
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
Scheme 4 Competitive condensation in presence of chalcone 5a.
demonstrate that an enone resulting from aldol condensation
between the pyruvic amide and the aldehyde is, indeed, an
intermediate in the synthesis of dihydropyrroles 4. We also
crystallized 6a and determined its structure unambiguously by
X-ray crystallography (Fig. 1).
Condensation of the sterically more demanding homopyru-
vic amide 3j and 3k with the chalcones 5b and 5c generated
dihydropyrroles 6b and 6c respectively, but phenylpyruvic
amide 3l did not take part in this reaction, probably due to
steric hindrance or to the requirement for an enolisable keto-
amide (Scheme 5).
In order to obtain some insight into the reaction mecha-
nism, we next examined whether Mannich base 7 could be an
intermediate in the course of this reaction (Scheme 6).
Heating together p-tolualdehyde, ketoamide 3a and Mannich
base 7 did not deliver the adduct 6e but instead gave 4a, prob-
ably via an elimination of 4-anisidine, with concomitant for-
mation of chalcone 5c and 4-anisidine, which were then
engaged in the three-component reaction. This observation
strongly suggests that Mannich bases, such as 7, are not
involved in the construction of the 2,3-dihydropyrroles.
Based on our observations, we speculate that the formation
of 4 involves imine 8 and enone 5 which react together to
afford intermediate 9 (Scheme 7). Condensation with aldehyde
Scheme 3 Substrate scope for the three-component reaction. Unless
otherwise noted, all reactions were conducted at 0.025 M concentration
with 1 (1.0 equiv.), 2 (2.0 equiv.), and 3 (100 mg, 2.0 equiv.) in EtOH–
AcOH (12 : 1) under reflux conditions. a The reaction was done without
any acetic acid.
Fig. 1 X-Ray structure of 6a.
rise to 4v with a yield of only 19%, and all attempts to use
cyclohexane carboxaldehyde were unsuccessful, suggesting
that aliphatic aldehydes are not suitable for the production of
cognate dihydropyrroles.
We extended the scope of this reaction by performing the
reaction displayed in Scheme 3 with the addition of chalcone
5a to the reaction medium (Scheme 4). The expected products,
Scheme 5 Attempted condensation reactions with homopyruvic and
4b and 6a, were generated in a 1 : 1 ratio. These findings phenylpyruvic amides 3j–l.
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Scheme 8 Classification of intramolecular dipolar cycloaddition of
azomethine ylides based on their connectivity.9
Scheme 6 Attempted condensation reactions with Mannich base 7 to
generate adduct 6e.
Scheme 7 Putative mechanism for the formation of dihydropyrroles 4.
Scheme 9 Attempted condensation reactions with allylic alcohols 14
and pivalic ester 15.
2 could lead to the 1,3-oxazine 11. Dehydration of this latter
intermediate would then generate the iminium 12 which, the insufficient electrophilic character of the ester. However,
upon deprotonation, affords the azomethine ylide 13 stabilized we were pleased to observe that replacement of this function
by π-stacking between the Ar1 and Ar2 moieties. This hypoth- by a nitro group successfully afforded the adduct 16b, in 19%
esis would explain the requirement for an aromatic unit in the yield, indicating that 10a is a likely intermediate in the mecha-
ketoamide 3 and the observed stereochemistry of the adduct 4. nism. Going one step further, we examined the pivalic ester 15
The intramolecular dipolar cycloaddition of azomethine ylides as a 1,3-nucleophile,10 which also delivered the adduct 16b,
to form pyrrolidines has been widely studied.9 This approach revealing that 11a is also
a
probable intermediate.
has found utility in simplifying synthetic strategies to complex Furthermore, the aldol product 5 was isolated as a byproduct
bioactive compounds. For this intramolecular process, the in the reaction. Altogether, these observations strongly support
dipolarophile has been tethered to one of the carbon atoms or the mechanism proposed in Scheme 7.
to the nitrogen atom of the azomethine ylide (respectively type
I or type II, Scheme 8). To the best of our knowledge, no cyclo- prompted us to examine whether it could be screened in bio-
addition between dipole and dipolarophile directly logical assays to identify new drug candidates. As a proof-of-
connected together without any tether has been reported yet. concept and based on our ongoing interest in the development
The drug-like character of our 2,3-dihydropyrrole library
a
a
To validate the proposed mechanism we explored the possi- of α-glucosidase inhibitors,11 we determined whether 2,3-di-
bility of directly condensing imine 8 with allylic alcohol 14a to hydropyrroles could inhibit this enzyme. Indeed, α-glucosidase
directly generate intermediate 10a (Scheme 9). This first experi- inhibitors, such as acarbose, are widely used to treat type II
ment did not deliver the expected adduct 16a, probably due to diabetes, even though these agents display poor bioavailability
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
materials that are commercially available or easily accessible.
Furthermore, the reaction was highly stereoselective, atom
economic, and eco-friendly. To confirm its green chemistry
character, we calculated that the atom economy and atom
efficiency of the synthesis of 4a are respectively of 92.9 and
47.4%.12 The reaction had only moderate overall yield but it
was nevertheless remarkably effective given that it generated
five new bonds (∼90% average yield per bond formation).
Further mechanistic and synthetic studies on this new three-
component reaction are underway. With the discovery that 4a
and 4u potently inhibit α-glucosidase, we have illustrated that
this methodology is likely to prove very useful for the gene-
ration of drugs harbouring a new scaffold.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
Financial support from the National Natural Science
Foundation of China (No. 81673296), the start-up Foundation
from Tianjin University of Science & Technology and Région
Alsace for a Fellowship to MC is gratefully acknowledged. We
thank Miss Haijuan Qin (TUST) for assistance with 2D NMR
experiments.
Fig. 2 Lineweaver–Burk plot of the inhibition of α-glucosidase by 4a
(A) and 4u (B). Activities were determined with 4-nitrophenyl α-D-gluco-
pyranoside (pNGP) as a substrate.
and some adverse effects, such as bloating, intestinal spasm, Notes and references
abdominal pains and hepatic dysfunctions.12 Thus, there is
1 (a) C. B. Cui, H. Kakeya and H. Osada, J. Antibiot., 1996, 49,
currently an urgent need to develop novel antidiabetic
α-glucosidase inhibitors. The synthesized 2,3-dihydropyrroles
were screened for the inhibition of α-glucosidase at the dose of
5 μM and two hits, 4a and 4u were found to inhibit this
enzyme. Kinetic constants were then determined and com-
pared to the standard inhibitor, acarbose.
Dihydropyrroles 4a and 4u were shown to be non-competi-
tive inhibitors with Ki’s of 2.4 and 3.5 μM, respectively (Fig. 2),
which blocked α-glucosidase more efficiently than the stan-
dard competitive inhibitor acarbose (Ki of 320 μM, ESI
Fig. S1†). This observation, associated with the drug-like char-
acter of 4a, suggest that these hits represent promising lead
compounds for the development of novel antidiabetic agents.
832–835; (b) J. D. Leber, J. R. E. Hoover, K. G. Holden,
R. K. Johnson and S. M. Hecht, J. Am. Chem. Soc., 1988,
110, 2992–2993; (c) I. V. Magedov, G. Luchetti,
N. M. Evdokimov, M. Manpadi, W. F. A. Steelant, S. Van
slambrouck, P. Tongwa, M. Y. Antipin and A. Kornienko,
Bioorg. Med. Chem. Lett., 2008, 18, 1392–1396;
(d) Y. Takeuchi, T. Inoue and M. Sunagawa, J. Antibiot.,
1993, 46, 827–832.
2 (a) S. B. Herzon and A. G. Myers, J. Am. Chem. Soc., 2005,
127, 5342–5344; (b) J. M. Humphrey, Y. Liao, A. Ali, T. Rein,
Y.-L. Wong, H.-J. Chen, A. K. Courtney and S. F. Martin,
J. Am. Chem. Soc., 2002, 124, 8584–8592; (c) R. Martin,
A. Jäger, M. Böhl, S. Richter, R. Fedorov, D. J. Manstein,
H. O. Gutzeit and H.-J. Knölker, Angew. Chem., Int. Ed.,
2009, 48, 8042–8046; (d) J. Wegner, S. V. Ley, A. Kirschning,
A.-L. Hansen, J. Montenegro Garcia and I. R. Baxendale,
Org. Lett., 2012, 14, 696–699.
Conclusions
In conclusion, we have discovered a useful reactivity of
α-ketoamides towards imines, generated in situ from amines
and aromatic aldehydes. Ketoamides have previously been
used in diverse multicomponent reactions13 but their reactivity
in the present fashion has never to our knowledge been
described. Remarkably, highly substituted dihydropyrroles
were prepared in a single step, using three varied starting
3 (a) C. Bressy, C. Menant and O. Piva, Synlett, 2005, 577–
582, DOI: 10.1055/s-2004-862386; (b) P. Evans, T. McCabe,
B. S. Morgan and S. Reau, Org. Lett., 2005, 7, 43–46;
(c) N. Gigant and I. Gillaizeau, Org. Lett., 2012, 14, 4622–
4625; (d) T. P. Pathak and M. S. Sigman, Org. Lett., 2011, 13,
2774–2777.
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
4 (a) B. Jiang, F.-F. Meng, Q.-J. Liang, Y.-H. Xu and T.-P. Loh,
Org. Lett., 2017, 19, 914–917; (b) H. Liang, F. Yan, X. Dong,
Q. Liu, X. Wei, S. Liu, Y. Dong and H. Liu, Chem. Commun.,
2017, 53, 3138–3141; (c) T. Miura, T. Tanaka, K. Hiraga,
S. G. Stewart and M. Murakami, J. Am. Chem. Soc., 2013,
135, 13652–13655; (d) P. A. Wender and D. Strand, J. Am.
Chem. Soc., 2009, 131, 7528–7529; (e) G. Zhang, Y. Zhang,
X. Jiang, W. Yan and R. Wang, Org. Lett., 2011, 13, 3806–
3809; (f) S. Meninno, A. Capobianco, A. Peluso and
A. Lattanzi, Green Chem., 2015, 17, 2137–2140; (g) E. Vedejs
(g) M. Koszytkowska-Stawinska and W. Buchowicz, Beilstein
J. Org. Chem., 2014, 10, 1706–1732; (h) M. A. Mironov, QSAR
Comb. Sci., 2006, 25, 423–431; (i) S. Shaaban and
B. F. Abdel-Wahab, Mol. Diversity, 2016, 20, 233–254.
7 (a) O. Döbner, Annalen, 1887, 242, 265–289; (b) O. Döbner,
Ber., 1887, 20, 277–280; (c) M. M. Heravi, S. Asadi and
F. Azarakhshi, Curr. Org. Synth., 2014, 11, 701–731.
8 A. Gopalsamy and P. V. Pallai, Tetrahedron Lett., 1997, 38,
907–910.
9 I. Coldham and R. Hufton, Chem. Rev., 2005, 105, 2765–2809.
and J. W. Grissom, J. Am. Chem. Soc., 1988, 110, 3238–3246. 10 (a) P. Knochel and D. Seebach, Tetrahedron Lett., 1982, 23,
5 (a) R. P. Herrera and E. Marques-López, Multicomponent
Reactions: Concepts and Applications for Design and
3897–3900; (b) D. Seebach, M. Missbach, G. Calderari and
M. Eberle, J. Am. Chem. Soc., 1990, 112, 7625–7638.
Synthesis, Wiley-VCH Verlag GmbH & Co. KGaA, 2015; 11 (a) K. Han, Y. Li, Y. Zhang, Y. Teng, Y. Ma, M. Wang,
(b) J. Zhu, Q. Wang and M.-X. Wang, Multicomponent
Reactions in Organic Synthesis, Wiley-VCH Verlag GmbH &
Co. KGaA, 2014; (c) C. de Graaff, E. Ruijter and
R. V. A. Orru, Chem. Soc. Rev., 2012, 41, 3969–4009;
(d) J. Yu, F. Shi and L.-Z. Gong, Acc. Chem. Res., 2011, 44,
1156–1171; (e) F. Shi, Z.-L. Tao, S.-W. Luo, S.-J. Tu and
L.-Z. Gong, Chem. – Eur. J., 2012, 18, 6885–6894; (f) F. Shi,
G.-J. Xing, R.-Y. i. Zhu, W. Tan and S. Tu, Org. Lett., 2013,
15, 128–131; (g) W. Dai, H. Lu, X. Li, F. Shi and S.-J. Tu,
Chem. – Eur. J., 2014, 20, 11382–11389; (h) W. Dai,
X.-L. Jiang, J.-Y. Tao and F. Shi, J. Org. Chem., 2016, 81,
185–192.
R. Wang, W. Xu, Q. Yao, Y. Zhang, H. Qin, H. Sun and
P. Yu, Bioorg. Med. Chem. Lett., 2015, 25, 1471–1475;
(b) H. Sun, W. Ding, X. Song, D. Wang, M. Chen, K. Wang,
Y. Zhang, P. Yuan, Y. Ma, R. Wang, R. H. Dodd, Y. Zhang,
K. Lu and P. Yu, Bioorg. Med. Chem. Lett., 2017, 27, 3226–
3230; (c) H. Sun, Y. Li, X. Zhang, Y. Lei, W. Ding, X. Zhao,
H. Wang, X. Song, Q. Yao, Y. Zhang, Y. Ma, R. Wang, T. Zhu
and P. Yu, Bioorg. Med. Chem. Lett., 2015, 25, 4567–4571;
(d) H. Sun, D. Wang, X. Song, Y. Zhang, W. Ding, X. Peng,
X. Zhang, Y. Li, Y. Ma, R. Wang and P. Yu, J. Agric. Food
Chem., 2017, 65, 1574–1581; (e) H. Sun, Y. Zhang, W. Ding,
X. Zhao, X. Song, D. Wang, Y. Li, K. Han, Y. Yang, Y. Ma,
R. Wang, D. Wang and P. Yu, Eur. J. Med. Chem., 2016, 123,
365–378.
6 (a) R. W. Armstrong, A. P. Combs, P. A. Tempest,
S. D. Brown and T. A. Keating, Acc. Chem. Res., 1996, 29,
123–131; (b) M. Colombo and I. Peretto, Drug Discovery 12 (a) Z. Liu and S. Ma, ChemMedChem, 2017, 12, 819–829;
Today, 2008, 13, 677–684; (c) A. Cores, C. Carbajales and
A. Coelho, Curr. Top. Med. Chem., 2014, 14, 2209–2230;
(b) H. A. Spiller and T. S. Sawyer, Am. J. Health-Syst. Pharm.,
2006, 63, 929–938.
(d) A. Domling, W. Wang and K. Wang, Chem. Rev., 2012, 13 (a) F. Roschangar, R. A. Sheldonb and C. H. Senanayakea,
112, 3083–3135; (e) C. Hulme and V. Gore, Curr. Med.
Chem., 2003, 10, 51–80; (f) G. Koopmanschap, E. Ruijter
and R. V. A. Orru, Beilstein J. Org. Chem., 2014, 10, 544–598;
Green Chem., 2015, 17, 752–768; (b) D. Bonne,
T. Constantieux, Y. Coquerel and J. Rodriguez,
Chem. – Eur. J., 2013, 19, 2218–2231.
Green Chem.
This journal is © The Royal Society of Chemistry 2018