Regioselective dicouplings: Application to differentially substituted pyrroles

Article abstract of DOI:10.1021/ol0530981

In an effort to develop a more concise route to differentially substituted pyrroles (such as that found in the lamellarins), a completely regioselective one-pot double Suzuki coupling has been discovered. The key feature is the use of a ligand-free pallad

Full text of DOI:10.1021/ol0530981

ORGANIC
LETTERS
2
006
Vol. 8, No. 8
537-1539
Regioselective Dicouplings: Application
to Differentially Substituted Pyrroles
1
Scott T. Handy*,† and Jesse J. Sabatini
Department of Chemistry, State UniVersity of New York at Binghamton,
Binghamton, New York 13902
shandy@mtsu.edu
Received December 21, 2005 (Revised Manuscript Received March 8, 2006)
ABSTRACT
In an effort to develop a more concise route to differentially substituted pyrroles (such as that found in the lamellarins), a completely regioselective
one-pot double Suzuki coupling has been discovered. The key feature is the use of a ligand-free palladium catalyst under optimized conditions,
which results only in coupling of the C5 bromide. At this point, addition of a second boronic acid and a phosphine ligand enables coupling
at the remaining C4 bromide.
As part of our ongoing interest in the synthesis of the
lamellarin family of natural products, we were interested in
methods that could reduce the overall length of our first
The idea of regioselective couplings of polyhaloheteroaro-
matics has been attracting greater attention in recent years,
so although it had not been studied in the context of pyrroles,
1
2
generation synthesis. In examining this synthesis, it became
it appeared promising. However, the concept of conducting
apparent that the most promising way to reduce the number
of steps was to focus on the coupling and halogenation steps.
Because each one of the three aryl subunits is installed via
a Suzuki coupling, this required three separate halogenations
and three separate couplings. As such, just the installation
of the aryl subunits accounted for over half of the total length
of the synthesis (6 out of 11 steps).
In contemplating ways to reduce the number of halogena-
tion/coupling steps, we found that one potential solution
would be to explore the option of regioselective couplings
on polyhalopyrroles and to conduct more than one coupling
in the same pot (one-pot polycouplings). Taking this ap-
proach to its extreme, we could reduce the six steps for the
installation of the aryl subunits to two steps - one triple
halogenation followed by one triple coupling. As a result,
our lamellarin synthesis would be reduced from 11 steps to
more than one coupling in the same reaction vessel is quite
rare. Indeed, within the area of heteroaromatic systems, there
are only two existing examples.
3
To explore the potential of this regioselective polycoupling
approach, we began our efforts with pyrrole aldehyde 1. This
substrate was selected for several reasons. First, for successful
regioselectivity in the coupling of polyhaloheteroaromatics,
there is a clear requirement for some degree of electronic
4
difference between the different halogenated centers. For
the pyrrole ring system, this can be most readily achieved
by substitution with an electron-withdrawing group. Further,
because previous studies had indicated that the pyrrole
nitrogen would need to be protected to avoid extensive
reductive dehalogenation at C4 and to simplify product
analysis by avoiding deprotection under the reaction condi-
(
2) Schroeter, S.; Stock, C.; Bach, T. Tetrahedron 2005, 61, 2245-
267.
3) Kaswasaki, I.; Yamashita, M.; Ohta, S. J. Chem. Soc., Chem.
7
- a tremendous improvement.
2
(
†
Current address: Department of Chemistry, Middle Tennessee State
Commun. 1994, 2085-2086. Duan, X.-F.; Li, X.-H.; Li, F.-Y.; Huang, C.-
University, Murfreesboro, Tennessee 37132. Phone: 615-494-8655.
1) Handy, S. T.; Zhang, Y.; Bregman, H. J. Org. Chem. 2004, 69, 2362-
366.
H. Synthesis 2004, 2614-2617.
(
(4) Fauvarque, J.-F.; Pfleuger, F.; Troupel, M. J. Organomet. Chem. 1981,
208, 419-427.
2
1
0.1021/ol0530981 CCC: $33.50
© 2006 American Chemical Society
Published on Web 03/22/2006

5
tions, an N-alkyl group was selected. Ultimately, pyrrole
both the 4- and 5-positions, this was selected as the second
catalyst. Thus, after TLC indicated complete consumption
of starting material, a second boronic acid was added, along
with tetrakis(triphenylphosphine) palladium(0). Much to our
delight, this did lead to the formation of the desired double-
coupled product 4 in 45% yield (Scheme 1).
aldehyde 1 was prepared via a slight modification of the
literature procedure.6
Treatment of aldehyde 1 with p-methoxyphenyl boronic
acid under standard Suzuki conditions did afford a separable
mixture of products from which monocoupled product 2 and
dicoupled product 3 could be isolated in 32 and 29% yield,
respectively (Table 1, entry 1). Armed with this encouraging
Scheme 1. First Regioselective Double Couplings
Table 1. Regioselective Monocoupling Studies
Although the yield may seem modest, it is an improvement
over the three-step synthesis starting from 4-bromo-1-ethyl-
pyrrole-2-carbaldehyde, which affords 4 in a 39% yield over-
all. Using this stepwise approach (coupling, halogenation,
coupling) also enabled us to confirm the suspected regio-
chemical outcome of the first coupling with dibromide 1.
The next question was whether the addition of two
different catalysts was necessary. On the surface, it appeared
that there might be a different approach to go from a ligand-
free system to a phosphine-ligated system - simple addition
of a phosphine. As a result, the first stage of the coupling
was performed as before, but now, along with the second
boronic acid, two equivalents of triphenylphosphine (relative
to palladium acetate) were added. Gratifyingly, this resulted
in an improved 48% yield of double-coupled product 4
entry
catalyst
base
solvent
DMF
DMF
DMF
DMF
DMF
DMF
DMF
% yield (2/3)
1
2
3
4
5
6
7
8
9
Pd(Ph3P)4
Pd(Ph3P)4
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Na2CO3
K2CO3
Na2CO3
K2CO3
K2CO3
Cs2CO3
Ba(OH)2
K2CO3
K2CO3
32/29
42/5
45/18
56/<5
28/7
31/5
complex mix
48/8
complex mix
dioxane
H2O/acetone
result, several modifications with respect to the palladium
catalyst, activator, and solvent were studied. Very rapidly,
it was noted that a ligand-free palladium catalyst dramatically
reduced the amount of dicoupled product that was observed.
In conjunction with this catalyst, potassium carbonate
afforded the best selectivity, although a number of other bases
were effective but afforded lower conversions or simply
lower isolated yields (Table 1, entries 3 and 5-7). Finally,
shifting to an ethereal solvent proved to slightly increase the
amount of double coupling when using potassium carbonate
as the base (Table 1, entry 8), and an acetone/water mix
afforded a very complex mixture of products (Table 1, entry
(Table 2, entry 2). Indeed, an even better result was obtained
Table 2. Additional Regioselective Double Couplings
9
).
The end result of these optimizations was a catalyst system
palladium acetate, potassium carbonate, DMF) that com-
entry
Pro
Et
Et
Et
Et
Et
R
R′
additive
% yield
(
pletely avoided any coupling at the 4-position. Indeed,
coupled product 2 could be resubjected to the coupling
conditions, and no trace of dicoupled product was observed
even after 24 h at 100 °C. Thus, high regioselectivity could
be achieved. At the same time, the question remained as to
how to now accomplish the second coupling.
1
2
3
4
5
6
p-MeOPh p-FPh
p-MeOPh p-FPh
p-MeOPh p-FPh
p-FPh
p-MeOPh stryenyl
Pd(Ph P)
44
48
58
51
56
54
3
4
Ph3P
tBu3P/HBF4
p-MeOPh tBu3P/HBF4
tBu3P/HBF4
tBu3P/HBF4
MEM p-MeOPh p-FPh
The first solution was to add a more active catalyst with
the second boronic acid. Because tetrakis(triphenylphosphine)
palladium(0) had already been shown to affect coupling at
by employing the stabilized phosphonium salt form of tri-
tert-butylphosphine popularized by Fu. Now the isolated
7
yield was 58% (Table 2, entry 3).
(
5) Handy, S. T.; Bregman, H.; Lewis, J.; Zhang, Y. Tetrahedron Lett.
003, 44, 427-430.
6) Anderson, H. J.; Lee, S.-F. Can. J. Chem. 1965, 43, 409-414.
2
(
(7) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295-4298.
Org. Lett., Vol. 8, No. 8, 2006
1
538

These same reaction conditions were equally effective with
other boronic acids, including alkenyl boronic acids (Table
great benefit in the rapid and convergent synthesis of substi-
tuted pyrroles. The extension of this method to pyrroles with
other substitution patterns and even to other heteroaromatic
systems is underway and will be reported in due course.
2, entry 5). The alkyl group on the pyrrole nitrogen could
be modified as well to other potentially more readily
removable groups such as the MEM group (Table 2, entry
8
Acknowledgment. We thank the Research Foundation
and the Leukemia Research Foundation for financial support
of this research.
6). This observation now opens the door for numerous
synthetic applications of this one-pot, double-coupling route
to substituted pyrroles such as the lamellarins, Lipitor, and
the prodigiosins.9
In conclusion, we have reported the ability to regioselec-
tively couple 4,5-dibromopyrrole aldehydes and to carry out
two couplings in the same reaction pot by simply adding a
phosphine ligand to generate a more active catalyst for the
second coupling. This approach has the potential to be of
Supporting Information Available: Full spectral data
and experimental procedures for all new compounds. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL0530981
(
8) Interestingly, allyl and benzyl protecting groups caused the reaction
(9) Lamellarins: Handy, S. T.; Zhang, Y. Org. Prep. Proced. Int. 2005,
37, 411-445. Prodigiosins: Fuerstner, A. Angew. Chem., Int. Ed. 2003,
42, 3582-3603. Lipitor: Li, J. J.; Johnson, D. S.; Sliskovic, D. R.; Roth,
B. D. Contemporary Drug Synthesis; Wiley-Interscience: New York, 2004;
pp 113-124.
to only proceed part of the way to completion and, in the case of the allyl
protecting group, completely shut down the second coupling, even under
the added phosphine conditions. The origin of this change in behavior is
the object of current study.
Org. Lett., Vol. 8, No. 8, 2006
1539