Metal-promoted variants of the Passerini reaction leading to functionalized heterocycles.

Article abstract of DOI:10.1021/ol025877c

[reaction: see text]. The effect of replacing the Bronsted acid in classic Passerini reactions with a mild Lewis acid has been studied. Triggered by metal-promoted silylation, condensations of aliphatic or aromatic carbonyl compounds with appropriately su

Full text of DOI:10.1021/ol025877c

ORGANIC
LETTERS
2002
Vol. 4, No. 9
1631-1634
Metal-Promoted Variants of the
Passerini Reaction Leading to
Functionalized Heterocycles
Qian Xia and Bruce Ganem*
Department of Chemistry & Chemical Biology, Baker Laboratory, Cornell UniVersity,
Ithaca, New York 14853-1301
bg18@cornell.edu
Received March 14, 2002
ABSTRACT
The effect of replacing the Brønsted acid in classic Passerini reactions with a mild Lewis acid has been studied. Triggered by metal-promoted
silylation, condensations of aliphatic or aromatic carbonyl compounds with appropriately substituted isonitriles capable of neighboring group
donation afford r-hydroxyamides, substituted oxazoles, and other useful heterocycles.
The Ugi and Passerini reactions play a central role in the
development of combinatorial chemistry.1 These and other
multiple-component condensation reactions make possible
versatile synthetic routes to collections of chemical sub-
stances sharing common structural features.
The normal Passerini reaction of a carbonyl compound 1,
an isonitrile, and a carboxylic acid leads to an R-acyloxy-
carboxamide 2 (Scheme 1). Over the past 35 years, several
acidic conditions, a recently developed modification employs
TFA in the presence of mild bases and is more compatible
with acid-sensitive functionality.5
We have been interested in developing milder Lewis acids
to expand the scope and generality of the Passerini reaction.
Here we report that use of Zn(OTf)2 and chlorotrimethyl-
silane (TMSCl) can direct the condensation to afford either
R-hydroxycarboxylic acid amides 3 or disubstituted oxazoles
4 in synthetically useful yields, depending on the choice of
isonitrile. Moreover, structure/reactivity studies provide fresh
insights into the role of the isonitrile component in these
metal-promoted Passerini variations.
Scheme 1
Although not widely appreciated, most Passerini reactions
are sluggish and afford products in low yields unless either
strong carboxylic acids (CF3CO2H, HCO2H) or unusually
electrophilic carbonyl compounds are used. For example,
2-(morpholinoethyl)isonitrile 5 and benzaldehyde fail to
(1) Do¨mling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 3168-3210.
(2) Hagedorn, I.; Eholzer, U. Chem. Ber. 1965, 98, 936-940.
(3) BF3-etherate: (a) Mu¨ller, E.; Zeeh, B. Liebigs Ann. Chem. 1968, 715,
47-51. (b) Mu¨ller, E.; Zeeh, B. Liebigs Ann. Chem. 1966, 696, 72-80.
(4) TiCl4: (a) Carofiglio, T.; Cozzi, P. G.; Floriani, C.; Chiesi-Villa,
A.; Rizzoli, C. Organometallics 1993, 12, 2726-2736. (b) Seebach, D.;
Adam, G.; Gees, T.; Schiess, M.; Weigand, W. Chem. Ber. 1988, 121, 507-
517.
two-component variants have been developed in which a
Brønsted2 or Lewis acid3,4 replaces the carboxylic acid to
form R-hydroxyamides 3. As an alternative to these highly
(5) Semple, J. E.; Owens, T. D.; Nguyen, K.; Levy, O. E. Org. Lett.
2000, 2, 2769-2772.
10.1021/ol025877c CCC: $22.00 © 2002 American Chemical Society
Published on Web 04/04/2002

undergo the Passerini reaction, even with excess acetic acid.
By contrast, the corresponding four-component Ugi reactions
of 5, which involve more electrophilic iminium ions formed
in situ, are known.6 These findings underscore the importance
of carbonyl activation.
Scheme 2
Consistent with that observation, Seebach et al. obtained
hydroxyamide 6a (R) H, R′) Ph, 67% yield) from the
reaction of 5 with benzaldehyde using stoichiometric TiCl4.4b
Although initially thought to activate the isonitrile compo-
nent, the titanium reagent was later shown to promote
condensation by enhancing the electrophilicity of the car-
bonyl compound.4a Thus, it became of interest to examine
the effect of other carbonyl-activating reagents on Passerini
reactions of simple carbonyl compounds.
We first investigated whether silylation of the carbonyl
component might promote Passerini condensations in the
absence of a protic acid. Silylations using TMSCl can be
markedly enhanced by various tin(II) and zinc(II) salts. In
fact, the combination of zinc triflate and TMSCl, a mixture
known to promote glycoside formation,7 successfully trans-
formed 5 to the desired benzaldehyde adduct 6a. Control
experiments showed that no reaction took place between 5
and benzaldehyde either in the presence of stoichiometric
quantities of TMSCl alone or in the presence of zinc triflate
alone.
of hydroxyamide 6d from octanal was significantly higher
(entry 7) using Zn(OTf)2/TMSCl. Similar advantages of
Zn(OTf)2/TMSCl as a silylating agent were also noted in
earlier work.7c
To our surprise, none of the condensations shown in Table
1 occurred when 5 was replaced with cyclohexylisonitrile.
While reminiscent of limitations in the classical Passerini
reaction, the failure of simple isonitriles to react by either
Method A or B led us to suspect that the morpholine ring in
5 may have played some role in promoting successful
condensations.
The highest yield of 6a (77%, Table 1) was obtained using
0.3 equiv of Zn(OTf)2 and 3 equiv of TMSCl in CH2Cl2 (24
Specifically, nitrilium ion 7 (Scheme 3) generated by the
condensation of 1 with 5 gains added stabilization through
Table 1. Metal-Promoted Passerini Condensations Using 5
Scheme 3
entry
carbonyl cmpd
product
% yield
1
2
3
4
PhCHOa
6a R ) H, R′ ) Ph
6a
6b R ) H, R′ ) Ph(CH2)2
77
68
74
61
PhCHOb
PhCH2CH2CHOa
(E)-PhCHdCHCHOa 6c R ) H,
R′ ) (E)-PhCHdCH-
5
6
7
8
(E)-PhCHdCHCHOb 6c
73
60
24
57
CH3(CH2)8CHOa
CH3(CH2)8CHOb
cyclohexanonea
6d R ) H, R′ ) n-C7H15
6d
6e R, R′ ) (CH2)5
a Method A, using Zn(OTf)2 (0.3 equiv) and TMSCl (3 equiv).
h, rt, followed by aqueous workup to promote desilylation).
As indicated (entries 1, 3, 4, 6, 8), the method proved to be
general in scope, successfully affording the desired conden-
sation product using a number of representative carbonyl
compounds.
We also tested whether Passerini reactions of 5 might be
promoted by trimethylsilyltriflate, which could be generated
in situ by the combination of Zn(OTf)2 and TMSCl. As
shown in Table 1 (entries 2, 5, and 7), preformed TMSOTf
(2.1 equiv) did afford the desired products, although the yield
lone pair donation by the proximal morpholine nitrogen, as
in 8, prior to hydrolytic workup. Attempts to enhance the
reactivity of simple isonitriles by adding a tertiary amine
were unsuccessful. No reaction occurred when benzaldehyde
and cyclohexyl isonitrile were stirred with N-ethylmorpholine
(NEM, 1 equiv) under the conditions of Method A.
The effect of other donor groups on the isonitrile was
probed. An ether analogue of 5, 2-(methoxyethyl)isonitrile,
failed to react with benzaldehyde (Method A). However,
ethyl isocyanoacetate 9 (Scheme 4) afforded a new product,
2-(R-silyloxyalkyl)-5-ethoxyoxazole 4a (R) Ph, R′) H) in
9% yield. The yield of 4 was improved to 53% by using 0.5
equiv of Zn(OTf)2 and 3 equiv of NEM. A plausible
(6) Tempest, P.; Ma, V.; Thomas, S.; Hua, Z.; Kelly, M. G.; Hulme, C.
Tetrahedron Lett. 2001, 42, 4959-4962.
(7) (a) Susaki, H.; Higashi, K. Chem. Pharm. Bull. 1992, 40, 2019-
2022. (b) Susaki, H.; Higashi, K. Chem. Pharm. Bull. 1993, 41, 201-204.
(c) Susaki, H. Chem. Pharm. Bull. 1994, 42, 1917-1918.
1632
Org. Lett., Vol. 4, No. 9, 2002

Scheme 4
Scheme 5
mechanism (Scheme 4) invokes a similar neighboring group
effect, leading to stabilization of the initial nitrilium ion 10
as oxonium ion 11.8 Deprotonation of 11 by N-ethyl-
morpholine would likely give 4a.
The scope and utility of this new condensation are shown
in Table 2. The process successfully afforded the expected
established the overall molecular weight and the presence
of two nitrogens. The yield of 15 could be increased to 45%
by boosting the quantity of 9 to three equiv.
Very recently, Sun et al. described a related route to 2-(R-
aminoalkyl)-5-aminooxazoles 17 by the reaction of iso-
cyanoacetamides 16 with aldehydes in the presence of
primary or secondary amines (Scheme 6).9 No promoter or
Table 2. Metal-Promoted Passerini Condensations Using 9; A
New Synthesis of 4-Alkoxyoxazoles
entry
carbonyl compd
PhCHO
PhCH2CH2CHO
(E)-PhCHdCHCHO 4c R ) H,
R′ ) (E)-PhCHdCH-
producta
% yield
1
2
3
4a R ) H, R′ ) Ph
4b R ) H, R′ ) Ph(CH2)2
53
58
54
Scheme 6
4
5
CH3(CH2)8CHO
cyclohexanone
4d R ) H, R′ ) n-C9H19
4e R, R′ ) (CH2)5
47
64
2-(R-silyloxyalkyl)-5-alkoxyoxazoles in good yield using a
range of aliphatic and aromatic carbonyl compounds.
In the case of cyclohexanone, formation of oxazole 4e was
accompanied by a minor product 15 (Scheme 5) representing
a 2:1 isonitrile/ketone adduct.
The formation of 15, an unusual substituted 2H-1,4-oxazin-
2-one, likely involved trapping of the initially formed
nitrilium ion 12 (Scheme 5) with a second molecule of
isonitrile 9 leading to 13. Cyclization of the carboethoxy
group in 13 to 14 followed by deprotonation gave 15. The
structure of 15 was assigned on the basis of spectroscopic
data. In particular, the electrospray ionization mass spectrum
of 15 exhibited an M + H+ peak at m/z 397, which
b Method B, using TMSOTf (2.1 equiv).
catalyst was needed, apparently because the in situ generated
iminium ions were more electrophilic than the uncharged
carbonyl compounds. In attempts to develop a hybrid
condensation for the synthesis of 2-(R-aminoalkyl)-5-alkoxy-
oxazoles, we replaced isonitrile 16 with 9 under the condi-
(8) A related mechanism has been proposed for the cyclization of
R-ketoimidoyl chlorides to 2-acyl-5-ethoxyoxazoles: Huang, W.-S.; Zhang,
Y.-X.; Yuan, C.-Y. Synth. Commun. 1996, 26, 1149-1154.
Org. Lett., Vol. 4, No. 9, 2002
1633

tions reported by Sun et al. However, only slow ammonolysis
of 9 was observed, and no oxazole was formed.
pharmaceutically interesting heterocycles.12 Moreover, Evans
et al. have recently reported that enantioselective aldol
condensations of 5-alkoxyoxazoles lead to the asymmetric
synthesis of syn and anti â-hydroxy-R-amino acids.13 It
therefore appears that substituted 5-alkoxyoxazoles are
broadly useful targets around which new combinatorial
libraries may be designed.
The results summarized in Tables 1 and 2 indicate that it
is possible to implement Passerini-like reactions under aprotic
conditions using mild silylation conditions instead of strong
Brønsted or Lewis acids to activate the carbonyl electrophile.
Such mild conditions will be of interest to pharmaceutical
chemists, for example, in preparing next-generation libraries
of aspartyl protease-directed norstatine derivatives.10 With
the appropriate choice of isonitrile, these variants can also
lead to the synthesis of novel heterocyclic systems.
Our findings also indicate that in the absence of a
carboxylic acid, nucleophilic addition of an isonitrile to a
carbonyl is promoted by appropriately positioned electron-
donating groups. Such donor groups may capture the
incipient nitrilium ion normally intercepted by the carboxyl
nucleophile in standard Passerini condensations.
Acknowledgment. Partial support of this work was
provided by the National Institutes of Health (GM 35712).
Qian Xia was the recipient of the S. M. Tsang Fellowship
at Cornell. Support of the Cornell NMR Facility has been
provided by NSF and NIH.
Supporting Information Available: Representative ex-
perimental procedures for the synthesis of new isonitrile
adducts described in Tables 1 and 2, as well as supporting
spectroscopic data. This material is available free of charge
via the Internet at http://pubs.acs.org.
Substituted 5-alkoxyoxazoles are known to be useful in
the synthesis of oxazolines,11 pyrrolopyridines,9 and other
(9) (a) Sun, X.; Janvier, P.; Zhao, G.; Bienayme´, H.; Zhu, J. Org. Lett.
2001, 3, 877-880. (b) Janvier, P.; Sun, X.; Bienayme´, H.; Zhu, J. J. Am.
Chem. Soc. 2002, 124, 2560-2567.
(10) (a) Tadesse, S.; Balan, C.; Jones, W.; Viswanadhan, V.; Hulme, C.
Abstracts of Papers, 223rd National Meeting of the American Chemical
Society, Orlando, FL, April 7-11, 2002; American Chemical Society:
Washington, DC, 2002; ORGN-241. (b) Banfi, L.; Guanti, G.; Riva, R.
Chem. Commun. 2000, 985-986.
OL025877C
(11) Suga, H.; Ikai, K.; Ibata, T. Tetrahedron Lett. 1998, 39, 869-872.
(12) (a) Ibata, T.; Isogami, Y.; Tamura, H. Chem. Lett. 1988, 1551-
1554. (b) Suga, H.; Ibata, T. Chem. Lett. 1991, 1221-1224.
(13) Evans, D. A.; Janey, J. M.; Magomedov, N.; Tedrow, J. S. Angew.
Chem., Int. Ed. 2001, 40, 1884-1888.
1634
Org. Lett., Vol. 4, No. 9, 2002