A cobalt-catalyzed multicomponent approach to novel 2,3-Di-and 2,2,3-trisubstituted 3-methoxycarbonyl-γ-butyrolactones

Article abstract of DOI:10.1002/ejoc.201000698

A one-pot, three-component synthesis of the title butyrolactones starting from aryl bromides, dimethyl itaconate, and either aldehydes or ketones is described. The cobalt-catalyzed domino process formally involves the in situ metalation of an aromatic bromide, conjugate addition onto dimethyl itaconate, an aldolization reaction with a carbonyl compound and a final cyclization into a five-membered lactone. This procedure is applied to the concise synthesis of a range of functionalized γ-butyrolactones with a methyl paraconate subunit.

Full text of DOI:10.1002/ejoc.201000698

FULL PAPER
DOI: 10.1002/ejoc.201000698
A Cobalt-Catalyzed Multicomponent Approach to Novel 2,3-Di- and 2,2,3-
Trisubstituted 3-Methoxycarbonyl-γ-butyrolactones
Camille Le Floch,[a] Erwan Le Gall,*[a] Eric Léonel,[a] Jihen Koubaa,[a] Thierry Martens,[a]
and Pascal Retailleau[b]
Keywords: Lactones / Multicomponent reactions / Domino reactions / Cobalt / Zinc
A one-pot, three-component synthesis of the title butyro-
lactones starting from aryl bromides, dimethyl itaconate, and
either aldehydes or ketones is described. The cobalt-cata-
lyzed domino process formally involves the in situ metalation
of an aromatic bromide, conjugate addition onto dimethyl
itaconate, an aldolization reaction with a carbonyl compound
and a final cyclization into a five-membered lactone. This
procedure is applied to the concise synthesis of a range of
functionalized γ-butyrolactones with a methyl paraconate
subunit.
Introduction
Results and Discussion
As a starting point for the study, we anticipated that a
range of 2,3-polysustituted paraconic esters should be ac-
cessible through a sequence involving the formation of three
single bonds by means of conjugate addition, aldol reaction
and cyclization (Scheme 1). We also considered the ad-
ditional opportunity of turning the synthesis into a multi-
component procedure. This possibility relied on an attract-
ive feature of some organometallic reagents: their addition
to carbonyl compounds is generally slow.[9] Thus, we con-
ceived that the conjugate addition of a suitable organome-
tallic reagent, generated in situ from an organic bromide 2
(Barbier-like conditions), on dimethyl itaconate as the
Michael acceptor could be the faster pathway. The resulting
enolate would then add to the carbonyl compound 1 to in-
duce the formation of a transient alcoholate,[10] which
might undergo an intramolecular transesterification to pro-
vide the expected γ-butyrolactone 3–6.
The γ-butyrolactone subunit is a characteristic feature of
many natural products displaying significant biological ac-
tivities.[1] The substitution pattern of the five-membered
ring defines several classes of compounds to which para-
conic acids[2] belong. These compounds, bearing a carbox-
ylic acid function at the β position, constitute an important
group of γ-butyrolactones that displays antitumor and anti-
biotic biological activities, and also represents convenient
building blocks for the synthesis of compounds of pharma-
ceutical interest.[3] Although these interesting properties
have made them attractive synthetic targets for the organic
chemist,[4] the preparation of 2,3-polysubstituted γ-butyro-
lactones through multicomponent procedures[5,6] has been
only scarcely reported.[7]
Recently, we disclosed preliminary results regarding the
multicomponent synthesis of some paraconic acid ana-
logues from aryl bromides, aromatic aldehydes and di-
methyl itaconate through a domino metalation/conjugate
addition/aldol coupling/cyclization process.[8] In this paper,
we both confirm the efficient formation of 2,3-disubstituted
3-methoxycarbonyl-γ-butyrolactones and further expand
the scope of the procedure by applying the strategy to the
preparation of 2,2,3,3-tetrasubstituted γ-butyrolactones
through the use of ketones as carbonyl compounds. We also
discuss a putative reaction pathway.
Scheme 1. Retrosynthetic scheme for the multicomponent synthesis
of γ-butyrolactones.
[a] Électrochimie et Synthèse Organique, Institut de Chimie et des
Matériaux Paris Est, ICMPE, UMR 7182 CNRS – Université
Paris Est Créteil Val-de-Marne,
A preliminary experiment, which was conducted in ace-
tonitrile with benzaldehyde (1a), dimethyl itaconate, 4-bro-
moanisole (2a), cobalt(II) bromide as a catalyst and zinc
dust as a reducer, indicated that the formation of the ex-
pected five-membered ring lactone 3a was achievable in rea-
sonable time and yield. This encouraging result prompted
us to optimize the reaction conditions. Thus, in a first set
View this journal online at
2–8 rue Henri Dunant, 94320 Thiais, France
Fax: +33-1-49781148
E-mail: legall@glvt-cnrs.fr
[b] Laboratoire de Cristallochimie, Institut de Chimie des
Substances Naturelles, ICSN CNRS,
Bât 27, Avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex,
France
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5279–5286
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5279

E. Le Gall et al.
FULL PAPER
of experiments, we investigated the influence of several pa- workup conditions should also be noted; comparison of re-
rameters, such as: temperature, dimethyl itaconate and co- sults reported in entries 3 and 10 of Table 1, for which the
balt bromide amounts, zinc dust activation method, and conditions only differed in a hydrolysis step, suggested that
work-up conditions, on the reaction efficiency. To this end, the lactone is water-sensitive.
4-bromoanisole (2a; 1.5 equiv.) was allowed to react with
Consequently, it appears that the most general and reli-
dimethyl itaconate and benzaldehyde (1a; 1 equiv.) in the able conditions regarding the global efficiency of the reac-
presence of zinc dust (4.6 equiv.) and cobalt bromide, in tion, both in terms of yield and atom economy, are those
acetonitrile for 1–3 h;[11] the results are presented in Table 1. reported in entry 3 of Table 1. Standard reaction conditions
thus involved acetonitrile as solvent, zinc dust (4.6 equiv.,
activated by using the joint assistance of TFA and 1,2-di-
bromoethane) as a reductive metal, CoBr2 (13 mol-% vs.
Table 1. Optimization of the experimental conditions.[a]
ArBr) as a catalyst, and the bromide (1.5 equiv.), the alde-
hyde (1 equiv.), and dimethyl itaconate (5 equiv.), reacting
at 60 °C.
We then investigated the scope of the reaction system
under these optimized conditions. In an initial series of ex-
periments, dimethyl itaconate and 4-bromoanisole (2a),
which were taken as model compounds, were allowed to
react with a range of benzaldehyde derivatives for 1–2 h; the
results are reported in Table 2.
Entry Itaconate
amount
CoBr2 Temp.
Zn dust
activation[c] up[d]
Work- Yield [%][e]
amount
[°C]
[mmol] [mol-%][b]
1
2
3
4
5
6
7
8
9
10
15
30
50
50
50
50
50
50
50
50
13
13
13
–
6.5
100
13
13
13
13
60
60
60
60
60
60
20
60
60
60
A
A
A
A
A
A
A
–
A
A
A
A
A
A
A
A
A
B
13
84
95
–
64
99
71
51
77
43
Table 2. Scope of benzaldehyde derivatives.[a]
B
A
[a] Reagents and conditions: 2a (2.7 g, 15 mmol), 1a (1.1 g,
10 mmol), zinc dust (3 g, 46 mmol), acetonitrile (20 mL). [b] Calcu-
lated relative to 2a. [c] Method A: TFA and BrCH2CH2Br;
Method B: TFA and allyl chloride, see refs.[14b,14c] for details. [d]
Method A: Filtration over Celite followed by chromatography over
silica gel; Method B: Aqueous work-up followed by chromatog-
raphy over silica gel. [e] Isolated yield.
Entry
FG
3
Yield [%][b]
1
2
3
4
5
6
7
8
9
10
H
3a
3b
3c
3d
3e
3f
95
41
51
68
79
71
61
98
42
56
4-OMe
4-SMe
4-Me
3-Me
2-Me
3-Me, 4-Me
4-F
3g
3h
3i
These results indicate that the reaction works smoothly
under a wide range of conditions. However, it should be
mentioned that the use of an excess of dimethyl itaconate 11
2-F
2-CF3
4-CN
4-NO2
3j
[c]
–
12
–
is crucial for the efficiency of the reaction (Table 1, en-
tries 1–3).[12] The amount of cobalt(II) bromide can be
raised to 100 mol-% (Table 1, entry 6) to induce quantita-
tive formation of lactone 3a. In contrast, lower amounts of
CoBr2 leads to a reduction in the reaction yield (Table 1,
entry 5) and no coupling is observed in its absence (Table 1,
entry 4), indicating its critical role in the process. Another
interesting observation concerns the ability to complete the
[a] Reagents and conditions: 2a (2.7 g, 15 mmol), benzaldehyde de-
rivative (10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc dust
(3 g, 46 mmol), CoBr2 (0.44 g, 2 mmol), acetonitrile (20 mL). [b]
Isolated yield. [c] Degradation of the γ-butyrolactone upon chro-
matographic purification.
Yields generally range from moderate to excellent and it
reaction at ambient temperature (Table 1, entry 7) with only can be noted that the reaction tolerates a wide variety of
a slight decrease of yield. This set of conditions constitute functionalized aromatic aldehydes bearing, in particular,
very reliable reaction parameters for syntheses involving electron-donating groups and/or sterically hindering
heat-sensitive substrates. It can also be noted that the groups. In the case of p-cyanobenzaldehyde, degradation of
method of activation of zinc dust also plays a significant the corresponding butyrolactone upon chromatographic
role. Indeed, the overall reaction is much more efficient un- purification did not allow the expected product to be iso-
der combined trifluoroacetic acid (TFA)/1,2-dibromoethane lated (Table 2, entry 11). It should be noted that a limitation
activation (Table 1, entry 3) than under TFA/allyl chloride of the process concerns the impossibility of employing ni-
activation (Table 1, entry 9) or without activation (Table 1, tro-containing aldehydes under such reductive conditions
entry 8). The unexpected importance of the reaction (Table 2, entry 12).
5280
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5279–5286

Substituted 3-Methoxycarbonyl-γ-butyrolactones
To extend the study, a range of heteroaromatic aldehydes Table 4. Scope of aryl bromides.[a]
were submitted to the reaction conditions; the results are
summarized in Table 3.
Table 3. Scope of heteroaromatic aldehydes.[a]
Entry
FG
Temp. [°C]
3
Yield [%][b]
1
2
3
4
5
6
7
8
H
60
60
60
60
25
60
60
60
60
60
25
60
60
60
25
25
3k
3a
3l
64
95
44
4-OMe
3-OMe
2-OMe
2-OMe
4-SMe
4-Me
4-CF3
3-CF3
2-CF3
2-CF3
4-CO2Me
3-CO2Me
2-CO2Me
2-CO2Me
2-CN
–
3m
3n
3o
3p
3q
71[c]
52
69
54
48
9
10
11
12
13
14
15
16
–
3r
3s
3t
3u
3u
55[c],[d]
99
99
57
79[c]
[e]
–
[a] Reagents and conditions: Aryl bromide (15 mmol), 1a (1.1 g,
10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc dust (3 g,
46 mmol), CoBr2 (0.44 g, 2 mmol), acetonitrile (20 mL). [b] Iso-
lated yield. [c] Reaction conducted with 100 mol-% CoBr2. [d] The
reaction time was 2 d. [e] No reaction.
[a] Reagents and conditions: 2a (2.7 g, 15 mmol), heteroaromatic
aldehyde (10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc dust
(3 g, 46 mmol), CoBr2 (0.44 g, 2 mmol), acetonitrile (20 mL). [b]
Isolated yield.
The corresponding γ-butyrolactones were obtained in reaction conditions, these compounds did not undergo cou-
satisfactory to good yields, except when pyridine-3-carbal- pling. This inhibition of the reaction may be attributed to
dehyde was used as starting material, for which no three- a possible poisoning of the catalyst by the heteroarene, thus
component coupling was observed even after 24 h heating preventing the formation of organometallic species and the
at 60 °C (Table 3, entry 4). It can be noted that the major overall associated process.
product of the reaction was the alcohol resulting from the
In further experiments, we undertook the three-compo-
direct addition of the organozinc onto the carbonyl func- nent coupling starting from ketones; the results are pre-
tion.
In a subsequent set of experiments, we examined the
scope of aryl bromides; the results are reported in Table 4.
sented in Table 5.
Although, as expected, the yields were generally lower
than with aromatic aldehydes, it should be noticed that such
Very good yields were generally obtained with both elec- couplings can provide rapid entry to some novel 2,2,3,3-
tron-rich and electron-deficient aryl bromides. We were tetrasubstituted γ-lactones bearing, in particular, a spiranic
pleased to observe that even hindered aryl bromides under- system (Table 5, entry 6).
went the coupling (Table 4, entries 5, 11, and 15). However,
We then examined the scope of the procedure further by
it should be noted that an adaptation to the procedure was conducting some additional experiments; the results are
required when starting from an ortho-substituted bromo- presented in Table 6.
arene. In this case, the reaction was conducted at room tem-
An examination of these latter results, as well as those
perature to limit the formation of the biaryl, which resulted reported above, reveals that aromatic aldehydes may consti-
from the heat-promoted homocoupling of the starting ha- tute the most suitable carbonyl compounds. Indeed, it can
lide. This assertion is illustrated by comparing the results be seen that ketones (Table 5) are less reactive than benzal-
presented in entries 4 and 5, 10 and 11, or 14 and 15 of dehyde derivatives (Table 2). Furthermore, other carbonyl
Table 4. It should be noted that, in these cases, the presence compounds, such as aliphatic aldehydes (Table 6, entry 1),
of a large amount of cobalt bromide in the medium was paraformaldehyde (Table 6, entry 2), or ethyl glyoxylate
crucial for the reaction to reach completion in reasonable (Table 6, entry 3) are either less efficient in the coupling or
times. In our opinion, this rather unexpected ortho-effect of not reactive at all. Benzyl bromide did not undergo the cou-
the substituent is interesting from a synthetic point of view pling, even in the presence of additional cobalt bromide
and would be worthy of further investigation.
(Table 6, entry 4). On the other hand, vinylic bromide is al-
Additional experiments were conducted starting from most as reactive as an aryl bromide, furnishing the corre-
heteroaromatic bromides (bromopyridines, bromofurans, sponding lactone in reasonable time and yield (Table 6, en-
and bromothiophenes). Unfortunately, under the standard try 5).
Eur. J. Org. Chem. 2010, 5279–5286
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5281

E. Le Gall et al.
FULL PAPER
Table 5. Scope of carbonyl compounds.[a]
We also investigated the influence of the halogen atom
connected to the phenyl moiety; the results are reported in
Table 7.
Table 7. Effect of the halogen atom connected to the phenyl moi-
ety.[a]
Halogen
X
[b]
Entry
Temp. [°C]
CoBr2
Yield [%][c]
1
2
3
4
5
6
I
I
I
Br
Cl
Cl
60
20
60
60
60
60
13 mol-%
13 mol-%
–
13 mol-%
13 mol-%
100 mol-%
40
45
–
64
–
–
[a] Reagents and conditions: Aryl bromide (15 mmol), carbonyl
compound 1 (10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc
dust (3 g, 46 mmol), CoBr2 (0.44 g, 2 mmol), acetonitrile (20 mL).
[b] Isolated yield.
[a] Reagents and conditions: Halobenzene (2.26 g, 15 mmol), 1a
(1.1 g, 10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc dust
(3 g, 46 mmol), acetonitrile (20 mL). [b] The amount of CoBr2 is
calculated relative to the halobenzene. [c] Isolated yield.
Table 6. Miscellaneous experiments.[a]
[a] Reagents and conditions: Organic bromide 2 (15 mmol), carbonyl compound 1 (10 mmol), dimethyl itaconate (7.9 g, 50 mmol), zinc
dust (3 g, 46 mmol), CoBr2 (0.44 g, 2 mmol), acetonitrile (20 mL). [b] Isolated yield. [c] Caution! The vigorous reaction was conducted
without activation of the zinc dust.
5282
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5279–5286

Substituted 3-Methoxycarbonyl-γ-butyrolactones
It should be noted that iodobenzene was almost as ef- Reaction Mechanism
ficient as bromobenzene in the coupling reaction, even at
room temperature (Table 7, entry 2), albeit in the required
Additional experiments were devised to examine the re-
presence of cobalt bromide (Table 7, entry 3). On the other action mechanism. Preliminary experiments revealed that
hand, chlorobenzene did not undergo the reaction, even in organozinc reagents, preformed in acetonitrile, can be em-
the presence of increased amounts of cobalt bromide ployed in the reaction, hence suggesting that these organo-
(Table 7, entry 6). Taken together, these results indicate that metallics might be nucleophilic initiators of the domino
bromoarenes constitute very reliable starting halides, both process. However, the requisite presence of cobalt bromide
in terms of reactivity, commercial availability and price.
in the reaction medium during the arylzinc synthesis step
The efficiency of catalysts other than CoBr2 was also as- also implies the continuous presence of transient organoco-
sessed. Inferior results were obtained by using CoBr2bpy balt species that might also play additional roles in the fol-
instead of CoBr2, and nickel-based catalysts such as NiBr2 lowing three-component coupling reaction. In addition,
or NiBr2bpy were also inefficient in the process.
control experiments revealed that the presence of cobalt
In previous papers dealing with the Mannich-like, three- bromide was essential (Table 1, entry 4 and Table 7, en-
component coupling of organozinc reagents with aldehydes try 3). We then planed to replace the zinc dust by other
and amines,[13] the crucial role of the solvent in the reaction reducing metals in order to evaluate their activity as initia-
efficiency was recognized. In particular, it was noticed that tors of the domino process. These experiments, which were
common solvents such as tetrahydrofuran (THF) can dra- conducted with a range of metallic powders (Sn, Fe, Cr,
matically slow down the reaction rate, probably by stabiliz- Mn, Al, and Mg), indicated that although most metals were
ing the organozinc reagent. Hence, we chose to also investi- unable to initiate the reaction, manganese powder consti-
gate the influence of THF on the present reaction. To this tuted a very convenient reducing agent that was almost as
end, we replaced acetonitrile with THF in the Barbier-like efficient as zinc dust.
procedure; preliminary experiments resulted in failure, with
On the basis of these results, a reaction mechanism can
no coupling products being detected in the reaction mix- be proposed. The first point to note concerns the possibility
ture. This result is consistent with previous works by Gos- of conducting the reaction with reducers other than zinc
mini[14] that indicated that the zinc dust/cobalt bromide sys- dust, which indicates that the formation of an organozinc
tem is able to activate aryl halides and produce organome- compound is not required for the reaction to proceed. The
tallic reagents, provided that the reaction is conducted in use of manganese as a suitable reducer of cobalt bromide
acetonitrile. Consequently, the effect of THF was assessed has been described previously by Gosmini and co-workers
by using preformed[15] or commercial[16] organozinc rea- in a study on the conjugate addition of aryl halides onto
gents. Thus, organozinc reagents were allowed to react with activated olefins.[17] The authors mention the potential for-
benzaldehyde (1a) and dimethyl itaconate for 24 h at 60 °C. mation of an arylcobalt species that can react with an elec-
These experiments indicated that although the reaction can trophilic olefin to produce a cobalt enolate, which can be
be conducted in acetonitrile to furnish the expected lactone further protonated in situ by water. Consequently, we as-
in moderate yield (48%), no coupling products are observed sume that an organocobalt compound A could also be the
in the presence of THF, even when the reaction involves key intermediate in this process. Such a species could favor
organozinc that was preformed in acetonitrile. Further ad- conjugate addition and the further aldolization–cyclization
dition of cobalt bromide into the reaction mixture did not process leading to the formation of the five-membered ring
result in improved couplings, which supports the probable lactone, as depicted in Scheme 2, with Zn as the reducing
inhibitory role of THF in the present process.
metal.
Scheme 2. Proposed mechanism for the metalation/conjugate addition/aldolization/cyclization process.
Eur. J. Org. Chem. 2010, 5279–5286
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5283

E. Le Gall et al.
FULL PAPER
Several additional observations support such a cobalt- nance of the Z or E enolate in the medium, thus giving
mediated process. The first point concerns the fact that the rise to the formation of the corresponding aldol coupling
reaction is much more efficient under Barbier-like condi- products in similar ratios. Further efforts will be directed
tions than with preformed arylzinc reagents. In the latter toward the design of reaction systems that provide a sub-
case, it is assumed that the reaction mechanism involves the stantial improvement of diastereoselectivities.
formation of an intermediate organocobalt species A, which
is entirely converted into the corresponding organozinc B
by a thermodynamically favored transmetalation.[18] Conse-
quently, the first observation accounts for the limited reac-
Crystallographic Analyses
For all experiments conducted with a benzaldehyde de-
rivative as the starting carbonyl compound, the cyclic char-
acter of the final product was unambiguously determined
by using NMR analysis and, in particular, HMBC experi-
ments, which revealed a scalar correlation between the car-
bon of the carbonyl and the proton linked to the benzylic
carbon (3JCH). However, an X-ray diffraction experiment
confirming the nature of the final product was realized with
a single crystal of compound 3q (Figure 1).
tivity of the organozinc species towards Michael acceptors,
and the reverse transmetalation into the corresponding or-
ganocobalt species A must be necessary for the reaction to
take place. Additionally, the existence of a transmetalation
equilibrium between the organozinc B and the organo-
cobalt A is supported by the observation that the reaction
yield increases with increasing amounts of cobalt bromide
(Table 1, entries 3, 4, 5, and 6), thus revealing that a plaus-
ible cobalt-promoted equilibrium shift to the reactive orga-
nometallic species A may occur. Under Barbier-like condi-
tions, the organocobalt species that is primarily formed in
the medium might continuously trap dimethyl itaconate to
produce the cobalt enolate (C and CЈ mixture) instead of
undergoing a transmetalation into the inactive organozinc
species. Undoubtedly, this would favor the subsequent dom-
ino process that leads to the final formation of lactone E.
On the basis of this proposed mechanism, one could also
imagine potential roles of THF that might substantially de-
crease the overall reaction rate by, for example, preventing
transmetallation from B to A and/or limiting the reactivity
of the organocobalt species A by chelation to the cobalt
atom.
The stereochemical outcome of the reaction was charac-
terized by a lack of selectivity; the diastereoisomeric ratio,
determined by GC analyses, was generally between 60:40
and 50:50. It is assumed that the stereoselectivity of the
Figure 1. ORTEP drawing for compound 3q.
Starting from ketones, the absence of a benzylic proton
aldol reaction is defined by the configuration of the enolate on the coupling product made NMR determination of the
in the Zimmerman–Traxler transition state D. In the pres- structure difficult. Thus, single-crystal X-ray analysis was
ent case, the almost identical steric bulk of both –CH2Ar1 also recorded for a diastereoisomer of compound 5b, which
and –CH2CO2Me would not permit a notable predomi- also revealed the cyclic character of the molecule (Figure 2).
Figure 2. ORTEP drawing for compound 5b.
5284
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5279–5286

Substituted 3-Methoxycarbonyl-γ-butyrolactones
3859; j) R. da Silva, G. H. B. de Souza, A. A. da Silva, V. A.
de Souza, A. C. Pereira, V. de A. Royo, M. L. A. e Silva, P. M.
Donate, A. L. S. de Matos Araújo, J. C. T. Carvalho, J. K.
Bastos, Bioorg. Med. Chem. Lett. 2005, 15, 1033–1037; k) N.
Petragnani, H. M. C. Ferraz, G. V. J. Silva, Synthesis 1986,
157–183.
Conclusions
In conclusion, we have demonstrated that novel 2,3-di-
and 2,2,3-trisubstituted 3-methoxycarbonyl-γ-butyrolac-
tones can be successfully obtained through a simple and
efficient, one-pot, three-component reaction between di-
methyl itaconate, aryl bromides, and carbonyl compounds.
The domino process, which involves the formation of an
organometallic reagent, a conjugate addition, an aldol cou-
pling, and a final cyclization in the same experimental step,
provides potential access to an important variety of γ-
butyrolactones derived from paraconic acid, making this
strategy suitable for parallel synthesis. The development of
intramolecular versions of this reaction is in progress.
[2]
a) M. Pohmakotr, W. Harnying, P. Tuchinda, V. Reutrakul,
Helv. Chim. Acta 2002, 85, 3792–3813; b) M. T. Barros, C. D.
Maycock, R. M. Ventura, Org. Lett. 2003, 5, 4097–4099; c) D.
Blanc, J. Madec, F. Popowyck, T. Ayad, P. Phansavath, V. Ra-
tovelomanana, J.-P. Genêt, Adv. Synth. Catal. 2007, 349, 943–
950; d) S. Bazin, L. Feray, N. Vanthuyne, D. Siri, M. P. Ber-
trand, Tetrahedron 2007, 63, 77–85; e) C. Forzato, G. Furlan,
P. Nitti, G. Pitacco, E. Valentin, E. Zangrando, P. Buzzini, M.
Goretti, B. Turchetti, Tetrahedron: Asymmetry 2008, 19, 2026–
2036; f) X. Huang, X. Chen, Y. Chen, A. Zhang, X. Li, Tetra-
hedron: Asymmetry 2008, 19, 2529–2535; g) F. Berti, C. For-
zato, G. Furlan, P. Nitti, G. Pitacco, E. Valentin, E. Zangrando,
Tetrahedron: Asymmetry 2009, 20, 313–321; h) S. Coriani, C.
Forzato, G. Furlan, P. Nitti, G. Pitacco, M. Ringholm, K.
Ruud, Tetrahedron: Asymmetry 2009, 20, 1459–1467; i) D.
Horhant, A.-C. Le Lamer, J. Boustie, P. Uriac, N. Gouault,
Tetrahedron Lett. 2007, 48, 6031–6033; j) H.-C. Kim, O.-S.
Park, Tetrahedron: Asymmetry 2008, 19, 896–899; k) M. Am-
ador, X. Ariza, J. Garcia, J. Ortiz, J. Org. Chem. 2004, 69,
8172–8175; l) A. Brecht-Forster, J. Fitremann, P. Renaud, Helv.
Chim. Acta 2002, 85, 3965–3974; m) For a recent review on the
synthesis of paraconic acids, see: R. Bandichhor, B. Nosse, O.
Reiser, Top. Curr. Chem. 2005, 243, 43–72.
a) S. B. Mahato, K. A. I. Siddiqui, G. Battacharya, T. Ghosal,
K. Miyahara, M. Sholichin, T. Kawasaki, J. Nat. Prod. 1987,
50, 245–247; b) M. M. Murta, M. B. M. De Azevedo, A. E.
Greene, J. Org. Chem. 1993, 58, 7537–7541; c) P. A. Jacobi, P.
Herradura, Tetrahedron Lett. 1996, 37, 8297–8300; d) B. K.
Park, M. Nakagawa, A. Hirota, M. Nakayama, J. Antibiot.
1988, 41, 751–758; e) M. S. Maier, D. I. G. Marimon, C. A.
Stortz, M. T. Alder, J. Nat. Prod. 1999, 62, 1565–1567; f) X.
Ariza, J. Garcia, M. Lopez, L. Montserrat, Synlett 2001, 120–
122; g) C. Böhm, O. Reiser, Org. Lett. 2001, 3, 1315–1318; h)
K. Müller, Appl. Microbiol. Biotechnol. 2001, 56, 9–16.
a) F. Berti, F. Felluga, C. Forzato, G. Furlan, P. Nitti, G. Pit-
acco, E. Valentin, M. T. Barros, Tetrahedron: Asymmetry 2006,
17, 2344–2353; b) A. M. Jacobine, W. Lin, B. Walls, C. K.
Zercher, J. Org. Chem. 2008, 73, 7409–7412; c) S. S. Sohn, E. L.
Rosen, J. W. Bode, J. Am. Chem. Soc. 2004, 126, 14370–14371;
d) Y. Li, Z.-A. Zhao, H. He, S.-L. You, Adv. Synth. Catal. 2008,
350, 1885–1890; e) S. N. Greszler, J. S. Johnson, Angew. Chem.
Int. Ed. 2009, 48, 3689–3691; f) T. H. Mäkelä, S. A. Kaltia,
K. T. Wähälä, T. A. Hase, Steroids 2001, 66, 777–784; g)
R. R. A. Kitson, A. Millemaggi, R. J. K. Taylor, Angew. Chem.
Int. Ed. 2009, 48, 9426–9454; h) T. Shono, H. Hamaguchi, I.
Nishiguchi, M. Sasaki, T. Miyamoto, M. Miyamoto, S. Fujita,
Chem. Lett. 1981, 1217–1220; i) F. DЈOnofrio, R. Margarita,
L. Parlanti, G. Piancatelli, M. Sbraga, Chem. Commun. 1998,
185–186; j) A. Méou, L. Lamarque, P. Brun, Tetrahedron Lett.
2002, 43, 5301–5304; k) A. S.-Y. Lee, Y.-T. Chang, S.-H. Wang,
S.-F. Chu, Tetrahedron Lett. 2002, 43, 8489–8492.
Experimental Section
General Procedure for the Synthesis of Lactones: A dried 100-mL
round-bottomed flask was flushed with argon and charged with
acetonitrile (20 mL). Dodecane (0.2 mL), zinc dust (3 g, 46 mmol),
dimethyl itaconate (7.9 g, 50 mmol), aromatic aldehyde (10 mmol),
and aryl bromide (15 mmol) were added whilst stirring. Cobalt bro-
mide (0.44 g, 2 mmol), trifluoroacetic acid (0.1 mL), and 1,2-dibro-
moethane (0.2 mL) were added successively to the mixture, which
was heated at 60 °C until consumption of the aryl bromide was
complete (45 min to 3 h, monitored by gas chromatography). The
reaction mixture was then filtered through Celite. The Celite was
washed several times with diethyl ether and the combined organic
fractions were concentrated in vacuo. The crude reaction product
was purified by flash column chromatography over silica gel (pent-
ane/diethyl ether, 1:0 to 0:1) to afford the lactone.
[3]
CCDC-753629 [for rac-5b, (RR/SS)] and -753630 [for rac-3q,
(RR/SS)] contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[4]
Supporting Information (see also the footnote on the first page of
this article): General experimental procedures, characterization
data and copies of NMR spectra for all compounds.
Acknowledgments
Financial support of this work by the Centre National de la Re-
cherche Scientifique (CNRS) and the University Paris Est Créteil
(PhD grant) is gratefully acknowledged.
[1] a) H. M. R. Hoffmann, J. Rabe, Angew. Chem. Int. Ed. Engl.
1985, 24, 94–110; b) O. Spring, K. Albert, W. Gradmann, Phy-
tochemistry 1981, 20, 1883–1885; c) N. H. Fischer, T. Lu, C. L.
Cantrell, J. Castaneda-Acosta, L. Quijano, S. Franzblau, Phy-
tochemistry 1998, 49, 559–564; d) T.-C. Wang, K.-H. Lee, Y.-
L. Chen, S.-S. Liou, C.-C. Tzeng, Bioorg. Med. Chem. Lett.
1998, 8, 2773–2776; e) A. Ortega, R. A. Toscano, E. Maldon-
ado, Phytochemistry 1998, 49, 1085–1090; f) Y. Higuchi, F. Shi-
moma, M. Ando, J. Nat. Prod. 2003, 66, 810–817; g) K.-H.
Lee, I. H. Hall, E. C. Mar, C. O. Starnes, S. A. Elgebaly, T. G.
Waddell, R. I. Hadgraft, C. G. Ruffner, I. Weidner, Science
1977, 196, 533–535; h) R. Romagnoli, P. G. Baraldi, M. A. Tab-
rizi, J. Bermejo, F. Estévez, M. Borgatti, R. Gambari, J. Med.
Chem. 2005, 48, 7906–7910; i) M. A. Hughes, J. M. McFadden,
C. A. Townsend, Bioorg. Med. Chem. Lett. 2005, 15, 3857–
[5]
[6]
J. Zhu, H. Bienaymé, in: Multicomponent Reactions, Wiley-
VCH, Weinheim, 2005.
For recent reviews on multicomponent reactions (MCRs) and
asymmetric MCRs (AMCRs), see: a) R. V. A. Orru, M.
De Greef, Synthesis 2003, 1471–1499; b) J. Zhu, Eur. J. Org.
Chem. 2003, 1133–1144; c) C. Hulme, V. Gore, Curr. Med.
Chem. 2003, 10, 51–80; d) G. Balme, E. Bossharth, N. Mon-
teiro, Eur. J. Org. Chem. 2003, 4101–4111; e) C. Simon, T. Con-
stantieux, J. Rodriguez, Eur. J. Org. Chem. 2004, 4957–4980; f)
A. Dömling, Chem. Rev. 2006, 106, 17–89; g) D. M. D’Souza,
T. J. J. Müller, Chem. Soc. Rev. 2007, 36, 1095–1108; h) N. Is-
ambert, R. Lavilla, Chem. Eur. J. 2008, 14, 8444–8454; i) B. A.
Arndtsen, Chem. Eur. J. 2009, 15, 302–312; j) G. Guillena, D. J.
Eur. J. Org. Chem. 2010, 5279–5286
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5285

E. Le Gall et al.
FULL PAPER
Ramón, M. Yus, Tetrahedron: Asymmetry 2007, 18, 693–700;
k) D. J. Ramón, M. Yus, Angew. Chem. Int. Ed. 2005, 44, 1602–
1634; l) A. Dömling, I. Ugi, Angew. Chem. Int. Ed. 2000, 39,
3168–3210.
Tetrahedron 2006, 62, 9953–9965; c) S. Sengmany, E. Le Gall,
C. Le Jean, M. Troupel, J.-Y. Nédélec, Tetrahedron 2007, 63,
3672–3681; d) C. Haurena, S. Sengmany, P. Huguen, E.
Le Gall, T. Martens, M. Troupel, Tetrahedron Lett. 2008, 49,
7121–7123; e) S. Sengmany, E. Le Gall, M. Troupel, Synlett
2008, 1031–1035; f) E. Le Gall, C. Haurena, S. Sengmany, T.
Martens, M. Troupel, J. Org. Chem. 2009, 74, 7970–7973; g) E.
Le Gall, A. Decompte, T. Martens, M. Troupel, Synthesis 2010,
249–254; h) C. Haurena, E. Le Gall, S. Sengmany, T. Martens,
M. Troupel, J. Org. Chem. 2010, 75, 2645–2650.
[14] a) H. Fillon, C. Gosmini, J. Périchon, J. Am. Chem. Soc. 2003,
125, 3867–3870; b) I. Kazmierski, C. Gosmini, J.-M. Paris, J.
Périchon, Tetrahedron Lett. 2003, 44, 6417–6420; c) C. Gos-
mini, M. Amatore, S. Claudel, J. Périchon, Synlett 2005, 2171–
2174.
[15] Organozinc reagents were prepared in acetonitrile by the
method of Gosmini, see Ref.[13] for details, or in THF by lith-
ium/halogen exchange and transmetallation with ZnBr2, see:
a) P. Knochel, P. Jones, in: Organozinc Reagents, A Practical
Approach, Oxford University Press, New York, 1999; b) J. E.
Milne, S. L. Buchwald, J. Am. Chem. Soc. 2004, 126, 13028–
13032.
[7] The synthesis of monosubstituted lactones (N-carbamo-
ylmethyl-α-aminobutyrolactones) by the Ugi multicomponent
reaction has been reported, see: a) S. J. Park, G. Keum, S. B.
Kang, H. Y. Koh, Y. Kim, Tetrahedron Lett. 1998, 39, 7109–
7112. b) The synthesis bicyclic lactones starting from pre-
formed alkylcopper reagents, Michael acceptors and aldehydes
has been reported, see: A. Sidduri, P. Knochel, J. Am. Chem.
Soc. 1992, 114, 7579–7581.
[8] C. Le Floch, C. Bughin, E. Le Gall, E. Léonel, T. Martens,
Tetrahedron Lett. 2009, 50, 5456–5458.
[9] For instance, the limited reactivity of arylzinc reagents (ob-
tained through the cobalt bromide/zinc dust system) towards
aromatic aldehydes was noticed on several occasions during
studies conducted in the laboratory.
[10] A related procedure has been applied successfully by Mont-
gomery to the synthesis of β-hydroxy ester derivatives, see: a)
K. Subburaj, J. Montgomery, J. Am. Chem. Soc. 2003, 125,
11210–11211; b) C. C. Chrovian, J. Montgomery, Org. Lett.
2007, 9, 537–540.
[11] All reactions presented in this study were monitored by gas
chromatography.
[12] The use of reduced amounts of dimethyl itaconate leads to in-
creased formation of the biaryl resulting from the cobalt-cata-
lyzed reductive coupling of the starting aryl bromide.
[13] a) E. Le Gall, M. Troupel, J.-Y. Nédélec, Tetrahedron Lett.
2006, 47, 2497–2500; b) E. Le Gall, M. Troupel, J.-Y. Nédélec,
[16] THF solutions of organozinc reagents are commercially avail-
able (e.g., Aldrich).
[17] M. Amatore, C. Gosmini, J. Périchon, J. Org. Chem. 2006, 71,
6130–6134.
[18] S. Seka, O. Buriez, J.-Y. Nédélec, J. Périchon, Chem. Eur. J.
2002, 8, 2534–2538.
Received: May 17, 2010
Published Online: August 3, 2010
5286
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5279–5286