Palladium-Catalyzed Multicomponent Coupling of Alkynes, Imines, and Acid Chlorides: A Direct and Modular Approach to Pyrrole Synthesis

Article abstract of DOI:10.1021/ja039152v

A new palladium-catalyzed method to prepare pyrroles directly from three basic building blocks - imines, alkynes, and acid chlorides - is described. This approach provides a straightforward method both to prepare pyrroles in one step and to diversify their structure by simple variation of any of the three starting materials. Mechanistic studies suggest that this process occurs via a complex series of eight individual steps, and this is discussed. Copyright

Full text of DOI:10.1021/ja039152v

Published on Web 12/23/2003
Palladium-Catalyzed Multicomponent Coupling of Alkynes, Imines, and Acid
Chlorides: A Direct and Modular Approach to Pyrrole Synthesis
Rajiv Dhawan and Bruce A. Arndtsen*
Department of Chemistry, McGill UniVersity, 801 Sherbrooke Street West, Montreal, Quebec H3A 2K6, Canada
Received October 21, 2003; E-mail: bruce.arndtsen@mcgill.ca
Scheme 1. Potential Mechanism for a Catalytic Pyrrole Synthesis
Of the various synthetic and naturally occurring heterocyclic
structures, the pyrrole nucleus is among the most prevalent. Pyrroles
are found in a tremendous range of natural products1 and bioactive
molecules,2 including the blockbuster drug Atorvastatin Calcium,2a
as well as important anti-inflammatants,2b antitumor agents,2c and
immunosuppressants.2d Similarly, polypyrroles are of growing
relevance in materials science as conjugated polymers.3 While this
broad utility has made pyrroles important synthetic targets,
traditional routes to their preparation are multistep reactions, as
illustrated by the Paal-Knorr cyclization of amines with pre-formed
1,4-diketones.4 This has stimulated interest in the design of
alternative metal-based routes to pyrroles.5 Examples include the
isomerization of unsaturated imines,5a isonitrile/ketone couplings,5c
and alkyne additions to chromium carbenes,5e many of which
provide structures not easily generated by classical routes.
Despite these advances, a common feature of most pyrrole syn-
theses is their requisite use of pre-assembled precursor(s) for cy-
clization.1,4,5 This not only imposes further steps on the overall
synthesis but also can complicate structural diversification. In prin-
ciple, a more ideal way to construct complex molecules would in-
volve their preparation in one step, directly from simple, readily
available and easily varied substrates, in a fashion similar to the
Pauson-Khand reaction for carbocycle synthesis6 and other multi-
component reactions.7 We report below our design of such a direct
synthesis of pyrroles. This has been done by assembling the pyrrole
core via metal catalysis from the basic building blocks shown in eq 1.
mechanistic studies on this process, with the goal of designing a
metal catalyst that could mediate this synthesis in a selective fashion.
While there are several stages in this cycle wherein the reaction
might be inhibited, a useful observation came from mechanistic
studies on the Mu¨nchnone synthesis itself, which show that the
oxidative addition of iminium salt 7 to Pd(0) (step B) is rate
determining. This is evidenced by kinetic studies (the formation of
1 is first order in 7) and consistent with in situ NMR data (the
catalyst resting state is not 8-10).10 Considering that catalyst 611
forms an unligated 14 e- intermediate 8 (-L) for carbonylation
(steps C-E), a slow oxidative addition is not surprising and suggests
that the inhibition of pyrrole formation may potentially arise from
the alkyne, further slowing this step in the catalytic cycle. Consistent
with this postulate, examination of the products of eq 2 reveals
significant quantities of 7 after catalysis.
A method to accelerate iminium salt oxidative addition (step B)
would be to add donor ligands (L) to stabilize 8. However, simple
phosphines were found to completely inhibit Mu¨nchnone formation,
even without alkynes (Table 1, entry 2-4), presumably due to tight
coordination to 8 blocking subsequent CO binding (step C).9 A
more favorable scenario would be a ligand that could stabilize 8
as well as be sufficiently labile to allow the subsequent catalysis
steps. As has been noted in other systems,12 these features can be
obtained by the use of sterically encumbered phosphines. In the
case of P(o-tolyl)3, this generates a catalyst that is several times
more reactive for Mu¨nchnone formation (entry 8). More impor-
tantly, this catalyst is also capable of mediating these steps in the
presence of alkyne (Scheme 1). Thus, the addition of 15 mol %
P(o-tolyl)3 to the 6a-catalyzed reaction of imine, acid chloride,
alkyne (3a-5a), 4 atm of CO, and EtNiPr2 results in the disap-
pearance of these reagents over 16 h and the formation of pyrrole
2a as essentially the only significant reaction product (81% yield,
Table 2).13 Overall, this optimized protocol provides a straightfor-
ward catalytic method to construct a pyrrole in one step from three
separate and readily available building blocks.
Our approach to this synthesis is based upon the ability of alkynes
to undergo 1,3-dipolar addition to 1,3-oxazolium-5-oxides (Mu¨nch-
nones, 1) to form pyrroles.8 While Mu¨nchnones are typically pre-
pared in situ from amino acid derivatives, we have recently reported
their generation by the palladium-catalyzed coupling of imines, acid
chlorides, and carbon monoxide.9 This suggested a potential path-
way to construct pyrroles, outlined in Scheme 1. By performing
these steps simultaneously, this would provide overall a method to
convert imines, alkynes, and acid chlorides directly into a pyrrole.
However, examination of this reaction under the conditions reported
to generate 19 formed only a trace of pyrrole 2a (5%, eq 2). Indeed,
while mechanistically plausible, this process involves four separate
reagents, base, and a catalyst, all proceeding selectively through
over eight separate steps. Considering the reactivity of the inter-
mediates formed in this cycle (1, 7-11), this suggests the potential
for numerous other reactions. Thus, we have undertaken a series of
In addition to its simplicity, a useful feature of this synthesis is
the nature of the substrates employed, each of which can be easily
varied. As shown in Table 2, catalysis is tolerant to a range of
functionalities (e.g., esters, indoles, halides, thioethers). Aryl, hete-
roaryl, and alkyl substituents can be incorporated into the pyrrole
9
468
J. AM. CHEM. SOC. 2004, 126, 468-469
10.1021/ja039152v CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Ligand Influence on Mu¨nchnone Formationa
of the Mu¨nchnone synthesis, which without P(o-tolyl)3 is inert
toward these imines.9 In general, while there are some limitations
brought on by the complex series of reactions occurring during
catalysis,14 this process provides a method to construct pyrroles
wherein each of the five substituents (R1-R5) can be independently
controlled and varied by modulation of the three substrates. A
method to accomplish the latter in a single step reaction is, to our
knowledge, previously unknown.
entry
ligand
yield (%)b
entry
ligand
yield (%)b
1
2
3
4
c
33
0
0
5
6
7
8
PtBu3
29
31
51
78
PCy3
PPh3
dppe
PtBu2(2-biphenyl)
P(1-naphthyl)3
P(o-tolyl)3
0
a See Supporting Information for details. b NMR yield. c One equivalent
of Bu4NBr.
Table 2. Palladium-Catalyzed Pyrrole Synthesis (Eq 1)a,b
As illustrated in eq 3, this process can also be useful in the
incorporation of further levels of product complexity (12) into the
pyrrole product with minimal steps. The members of this class of
multicyclic pyrroles are of utility as potential therapeutics and
retinoic acid regulators,15 in this case generated in three steps from
an aldehyde, alkyne, amine, and acid chloride.
In conclusion, these studies have shown that pyrroles can be
considered as the product of three basic building blocks coupled
via palladium catalysis, providing a modular method to construct
these heterocycles with facile diversity and high atom economy.
Experiments directed toward understanding of the role of P(o-tolyl)3
in this catalysis, as well as the application of this approach to other
heterocyclic targets, are underway.
Acknowledgment. We thank NSERC for their financial support.
R.D. thanks NSERC for a Postgraduate Fellowship.
Supporting Information Available: Synthesis and characterization
of 2 and 12 (PDF). This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Pyrroles, Part II; Jones, R. A., Ed.; Wiley: New York, 1992.
(2) (a) Thompson, R. B. FASEB J. 2001, 15, 1671. (b) Muchowski, J. M.
AdV. Med. Chem. 1992, 1, 109. (c) Cozzi, P.; Mongelli, N. Curr. Pharm.
Des. 1998, 4, 181. (d) Fu¨rstner, A.; Szillat, H.; Gabor, B.; Mynott, R. J.
Am. Chem. Soc. 1998, 120, 8305.
(3) Skotheim, T. A., Elsenbaumer, R. L., Reynolds, J. R., Eds. Handbook of
Conducting Polymers, 2nd ed.; Marcel Dekker: New York, 1998.
(4) (a) Knorr, L. Ber. 1884, 17, 1635. (b) Paal, C. Ber. 1885, 18, 367.
(5) Representative examples: (a) Kel’in, A. V.; Sromek, A. W.; Gevorgyan,
V. J. Am Chem. Soc. 2001, 123, 2074. (b) Gabriele, B.; Salerno, G.; Fazio,
A. J. Org. Chem. 2003, 68, 7853. (c) Takaya, H.; Kojima, S.; Murahashi,
S.-I. Org. Lett. 2001, 3, 421. (d) Lee, C.-F.; Yang, L.-M.; Hwu, T.-Y.;
Feng, A. S.; Tseng, J.-C.; Luh, T.-Y. J. Am. Chem. Soc. 2000, 122, 4992.
(e) Merlic, C. A.; Baur, A.; Aldrich, C. C. J. Am. Chem. Soc. 2000, 122,
7398. (f) Braun, R. U.; Zeitler, K.; Mu¨ller, T. J. Org. Lett. 2001, 3, 3297.
(g) Nishibayashi, Y.; Yoshikawa, M.; Inada, Y.; Milton, M. D.; Hidai,
M.; Uemura, S. Angew. Chem., Int. Ed. 2003, 42, 2681. (h) Deng, G. S.;
Jiang, N.; Ma, Z. H.; Wang, J. B. Synlett 2002, 1913. (i) Wang, Y.; Zhu,
S. Org. Lett. 2003, 5, 745. (j) Dieter, R. K.; Yu, H. Org. Lett. 2001, 3,
3855.
(6) Gibson, S. E.; Stevenazzi, A. Angew. Chem., Int. Ed. 2003, 42, 1800.
(7) Representative examples: (a) Beller, M.; Eckert, M. Angew. Chem., Int.
Ed. 2000, 39, 1010. (b) Montgomery, J. Acc. Chem. Res. 2000, 33, 467.
(c) Dghaym, R. D.; Dhawan, R.; Arndtsen, B. A. Angew. Chem., Int. Ed.
2001, 40, 3228. (d) Kamijo, S.; Jin, T.; Huo, Z.; Yamamoto, Y. J. Am.
Chem. Soc. 2003, 125, 7786. (e) Trost, B. M.; Pinkerton, A. B. J. Am.
Chem. Soc. 2000, 122, 8081. (f) Cao, C.; Shi, Y.; Odom, A. L. J. Am.
Chem. Soc. 2003, 125, 2880.
a Imine (0.7 equiv), acid chloride, alkyne (1.4 equiv), EtNiPr2, CO (4
atm), 5% 6, and 15% P(o-tolyl)3 in CH3CN/THF, 16 h, 65 °C.
b Pd2dba3‚CHCl3 or [Pd(allyl)Cl]2 are viable catalysts at ca. 10% lower yield.
c Alkyne added to preformed 1. d 75 °C, 1 equiv of LiOTf in CH3CN, 6f
catalyst. e Major isomer (5:1 ratio).
(8) (a) Gribble, G. W. In The Chemistry of Heterocyclic Compounds; Padwa,
A., Ed.; Wiley: New York, 2002; Vol. 59, p 681. (b) Coppola, B. P.;
Noe, M. C.; Schwartz, D. J.; Abdon, R. L.; Trost, B. M. Tetrahedron
1994, 50, 93.
from the acid chloride or imine nitrogen, as can a variety of bis-,
mono-, and unsubstituted alkynes. While unsymmetrical alkynes
can lead to mixtures, steric and electronic effects provide a rea-
sonable degree of selectivity (5:1 ratio in 2c).8b Even electron-rich
alkynes, typically less potent Mu¨nchnone trapping reagents,8a form
pyrroles in reasonable yield (2f,g). Considering the nature of the
substrates and number of bonds generated in one pot, these all
represent effective syntheses of 2a-n.
Interestingly, the phosphine-based catalyst system can also form
pyrroles of Mu¨nchnones not previously accessible by catalysis, as
shown by the synthesis of pyrroles of both C-aromatic and C-alkyl
imines (2c, i). The latter represents a significant expansion in scope
(9) Dhawan, R.; Dghaym, R. D.; Arndtsen, B. A. J. Am. Chem. Soc. 2003,
125, 1474.
(10) See Supporting Information for full details.
(11) 11 6 is formed by pre-treating Pd2dba3 with imine and acid chloride (ref 9).
(12) For example: Stambuli, J. P.; Bu¨hl, M.; Hartwig, J. F. J. Am. Chem. Soc.
2002, 124, 9346 and references therein.
(13) Pd2dba3 or [Pd(allyl)Cl]2 are also viable catalysts at ca. 10% lower yield.
(14) Enolizable R2 are not stable, though these units can be incorporated into
the 2-pyrrole position via the acid chloride (2d). Strong dipolarophiles
react with 7 and must be added to a preformed 1 (2b,m,n).
(15) Yoshimura, H.; Nagai, M.; Hibi, S.; Kikuchi, K,; Abe, S.; Hida, T.;
Higashi, S.; Hishinuma, I.; Yamanaka, T. J. Med. Chem. 1995, 38, 3163.
JA039152V
9
J. AM. CHEM. SOC. VOL. 126, NO. 2, 2004 469