Preparation of peptide-like bicyclic lactams via a sequential Ugi reaction - Olefin metathesis approach

Article abstract of DOI:10.1016/j.tetlet.2004.06.052

Bicyclic lactams, suitable for incorporation into conformationally restricted peptide mimics, can be synthesized by using olefinic starting materials for the Ugi multicomponent reaction, setting up an olefin metathesis reaction, that is easily carried out with the Grubbs catalyst. The influence of the different starting materials is evaluated. In addition, the utilization of chiral, nonracemic amines is described.

Full text of DOI:10.1016/j.tetlet.2004.06.052

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5987–5990
Preparation of peptide-like bicyclic lactams via a sequential
Ugi reaction––olefin metathesis approach
q
Ralf Krelaus and Bernhard Westermann*
Leibniz-Institute of Plant Biochemistry, Department of Bioorganic Chemistry, Weinberg 3, 06120 Halle (Saale), Germany
Received 27 April 2004; revised 27 May 2004; accepted 11 June 2004
Abstract—Bicyclic lactams, suitable for incorporation into conformationally restricted peptide mimics, can be synthesized by using
olefinic starting materials for the Ugi multicomponent reaction, setting up an olefin metathesis reaction, that is easily carried out
with the Grubbs catalyst. The influence of the different starting materials is evaluated. In addition, the utilization of chiral, non-
racemic amines is described.
Ó 2004 Elsevier Ltd. All rights reserved.
The Ugi multicomponent reaction (MCR) is a very
versatile reaction for the preparation of peptides and
peptide mimics due to its high product diversity. This is
O
O
PGHN
PGHN
N
N
2
olefin
metathesis
of particular interest for the synthesis of compound
ensembles required for the discovery of new lead struc-
tures. This intriguing and simple reaction is remarkably
flexible. For suitable starting materials, a high degree of
functionalization in the amine, the carboxylic acid, the
isocyanide, and the carbonyl compound are tolerated.
However, the Ugi reaction has rarely been used to
synthesize lactams. In such cases, the classical Ugi 4-
component-4-center reaction would have to be reduced
to a 3-component-4-center reaction (Scheme 1).
CO2Et
CO2Et
Ugi-
MCR
CN
CO2Et
+
3
H2N
+
O
O
O
NPG
HO
In the preceding paper we have presented our approach
to the synthesis of bicyclic lactams, materials, that can
be used as peptide mimics. We are thus also interested
CO2Et
NHPG
4
Scheme 1.
in new pathways to obtain these products in a more
straightforward and short synthesis that would also al-
low diversification. A retrosynthetic scheme is presented
in Scheme 1. Key steps for the synthesis of bicyclic
lactams are a ring opening reaction of pyroglutamic
acid, an Ugi multi component reaction (MCR) and a
ring closing olefin metathesis. Recently, we have re-
ported the use of olefin metathesis on suitable lactams
O
N
( )
O
Grubbs
catalyst
N
m = 1,2
( )
(
)
n
n
= 1-4
n
( )
m
EtO2C
m
EtO2C
5
with the first generation Grubbs catalyst (Scheme 2).
Scheme 2.
To utilize the olefin metathesis for cyclization to form
the bicyclic heterocycle, olefinic starting materials have
to be used in order to do a Ugi MCR reaction. Very
recent reports utilized olefinic starting materials in the
Keywords: Ugi reaction; Olefin metathesis; Isocyanide; Peptide mimic.
q
See Ref. 1.
*
Corresponding author. Tel.: +49-345-5582-1340; fax: +49-345-5582-
309; e-mail: bwesterm@ipb-halle.de
1
0
040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.06.052

5988
R. Krelaus, B. Westermann / Tetrahedron Letters 45 (2004) 5987–5990
Table 1. Results of the Ugi MCR reaction of 1 with isocyanides 8, 10,
and amines 3, 4, 6 as starting materials
O
O
CO2H
Me
CO2H
Me
3
4
6
1
2
O
O
O
O
O
Me
NH2
NH2 HO
NH2
NH2
i-Bu
N
N
N
Ph
3
4
5
Me
Me
NH
Me
Me
OH
8
O
NH
n-Bu
NH
n-Bu
NH2
n-Bu
14
15
7 Me
13
6
81 %
86 %
92 %
O
O
O
O
Me
NC
NC
EtO2C
NC
i-Bu
8
9
10
N
N
N
Ph
Me
Me
NH
Me
Me
NC
1
0
NC
O
1
NH
O
NH
CO2Et
CO2Et
2
CO Et
11
12
16
17
18
2 %
51 %
7 %
O
N
R1
(
)
n
50 products
possible
Me
1,2 + 3-7 + 8-12
Table 2. Results of the Ugi MCR reaction of 2 with isocyanides 8, 10,
and amines 3, 4, 6 as starting materials
n = 1,2
O
NH
R
2
3
4
6
O
O
O
Me
Scheme 3.
i-Bu
N
N
N
Ph
Me
Me
O
Me
8
1
N
N
N
H n-Bu
H n-Bu
H n-Bu
6
19
20
21
O
O
Ugi reaction. Due to the limited information available,
we were interested in studying the influence of these
functional groups in the Ugi reaction. Here, we present
our results using olefinic starting compounds in an Ugi
5
9 %
65 %
46 %
O
O
O
Me
i-Bu
N
N
N
Ph
Me
Me
Me
3
-component-4-center reaction. In addition, we dem-
0
N
N
H
N
onstrate the use of chiral, nonracemic starting com-
pounds.
H
H
O
CO2Et
O
CO2Et
O
CO2Et
22
23
24
4
%
13 %
12 %
In preliminary experiments we applied the Ugi reaction
to lactams 13 (Scheme 3). For this, ketocarboxylic acids
1
1
and 2 in the presence of amines 3–7 and isocyanides 8–
2 were used. Levulinic acid 1 and its homologue 2 were
diastereomeric excess was observed; in both cases (15,
18) a 1:1 ratio is obtained. The same pattern of reactivity
and selectivity is achieved using d-keto carboxylic acid 2,
leading to products 19–24 (Table 2) in lower yields,
compared to the products formed by using 1.
chosen, because the methyl group in these substrates
allows easy characterization of the products. In race-
mates, it appears as a singlet in the H NMR spectra.
1
While employing chiral, nonracemic starting materials,
the diastereomeric ratio of the Ugi products can be
determined by integration of the two singlets. For the
formation of diastereomers, amines 6 and 7, and isocy-
anide 12 were used as chiral, nonracemic starting
materials. With these, the influence of stereocenters can
be studied. In the case of amines, allyl amine 4 was used
to introduce olefins.
The postulated mechanism for the Ugi reaction includes
1
the Mum-rearrangement. It can be assumed that the
transition state during this rearrangement, which will be
six-membered for levulinic acid 1, is favorable to the
seven-membered transition state for 2. This might
explain the higher yields for the formation of the c-
lactams.
In Table 1 the results of the Ugi reaction using levulinic
7
acid 1 are presented. It can be seen that the yield of the
Ugi reaction is highly dependent on the isocyanide. With
the very lipophilic isocyano butane 8, the yields (13–15)
are very high (81–92%), whereas ethyl 2-isocyano ace-
tate leads to 16–18 in low to moderate yields (7–51%).
To examine the influence of the isocyano substrate fur-
ther, isocyanides 9, 11, and 12 in the presence of the
chiral amines 6 and 7 were used. The corresponding
lactams 25–28 were also obtained in high yields, indi-
cating that isocyanides with a high CH-acidity at the a-
8
center of the isocyanides are not favorable.
Furthermore, the olefinic moieties of lactams 14 and 17
are not affected under the conditions of the Ugi reaction.
Therefore, olefinic building blocks such as allyl amine 4
can be employed. While using the chiral amine 6, no
It is interesting to note that even by employing two
chiral substrates, no diastereoselectivity can be
observed. Lactam 28 is obtained as a 1:1 mixture of
diastereomers.

R. Krelaus, B. Westermann / Tetrahedron Letters 45 (2004) 5987–5990
5989
O
O
Me
O
Me
O
O
O
a
b
c
3
5
N
Ph
N
Ph
N
Me
Me
4,10
+
c
CO2H
NH
t-Bu
O
NH
c-Hex
NH
3
6
CO2Et
2
5
26
9
3 %
91 %
O
(
Cy3P)2Cl2Ru=CHPh
mol%
O
O
N
5
N
O
O
Me
O
Me
+
NH
3
4 %
Ph
OH
NH
Ph
OH
NH
N
N
CO2Et
Me
Me
O
N
H
CO2Et
1:1
3
7
38
O
c-Hex
Ph
Me
Scheme 6.
2
7
28
9
1 %
76 %
from pyroglutamic acid requires only five steps to this
peptide mimics making it a very short synthesis of this
highly complex molecule.
For the synthesis of bicyclic lactams an easy approach
toward unsaturated keto carboxylic acids is required.
Protection of pyroglutamic ester 29 affords lactam 30.
Ring opening nucleophilic attack by Grignard reagents
derived from allyl bromide and 3-butenyl bromide,
respectively, achieves the synthesis of the two-centered,
olefinic amino acids. Subsequently, the Ugi reaction
employing 32 in the presence of allyl amine leads to
highly functionalized lactam 33 bearing two olefinic
moieties (Scheme 4).
An even shorter route can be envisioned by utilizing an
Ugi reaction to the tetraolefinic amide 36, setting up a
double RCM directly (Scheme 6).
Carbonyl compound 35 can be obtained by oxidation of
the commercially available alcohol. For the carboxylic
acid we chose crotonic acid, anticipating that a distinc-
tion in the RCM reaction should exist if one of the
double bonds is disubstituted. Two modes of cyclization
are possible: cyclization of a/c and b/c leading to 37 and
cyclization of a/b and c/c leading to 38.
In agreement with our previous results, a 1:1 ratio of
diastereomers was obtained. In the next step, a ring
closing olefin metathesis (RCM) using the Grubbs cat-
alyst was carried out (Scheme 5). In the presence of
As expected, the Ugi reaction led to the tetraolefinic
intermediate 36, which was cyclized without further
purification. After the RCM reaction, 37 and 38 were
isolated in a 1:1 ratio. This formation of analogous
[
3
Ru(PCy
3 2 2
) Cl ]@CHPh (5 mol %) the bicyclic product
4 is formed in 81% yield. The whole sequence starting
9
products was observed very recently by Ma and Ni.
Br
O
O
(
Mg
)n
They found that the outcome of the RCM is highly
dependent on whether a quaternary center or a tertiary
stereogenic center is formed. In the case of formation of
the quaternary center, both products are formed to the
same extent.
Boc2O
DMAP
O
Boc a:
NH
N
2
CO H
7
5 %
( )n
b: NaOH
CO2Me
CO2Me
NHBoc
29
30
31 n = 1, 78 %
3
2 n = 2, 82 %
In conclusion, we have shown that bicyclic lactams can
be synthesized by a sequence of Ugi reaction and ring
closing olefin metathesis. The 3-component-4-center Ugi
reaction seems to be advantageous for the preparation
of highly olefin-functionalized lactams. It can be selec-
tively bicyclized with the Grubbs catalyst. The Ugi
reaction toward a tetraolefinic intermediate followed by
a RCM should be by far the shortest route toward such
products, although the selectivity in the RCM will
require further studies.
Scheme 4.
O
BocHN
NH2
3
2
N
+
5
7 %
BuNC
Bu
Bu
O
O
N
H
33
(
Cy3P)2Cl2Ru=CHPh BocHN
mol%
N
5
Acknowledgements
8
1 %
O
N
Financial support was provided by the Fonds der
Chemischen Industrie and the Graduiertenstiftung
NRW (grant to R.K.), which is gratefully acknowl-
edged.
H
34
Scheme 5.

5990
R. Krelaus, B. Westermann / Tetrahedron Letters 45 (2004) 5987–5990
References and notes
. The work has been carried out mainly at the University of
7. General procedure for the synthesis of lactam 13: levu-
linic acid 1 (1.00 g, 8.6 mmol) was dissolved in dry
methanol (43 mL) and isobutylamine (0.79 g, 10.8
3
1
mmol) was added. The mixture was stirred for 30 min at
room temperature to allow formation of the imine.
Paderborn, Germany. It has been presented, in part, at the
Chemiedozententagung, abstract A33, Cologne, Germany,
1
-Isocyanobutane
8
(0.71 g, 8.6 mmol) was added
2
002.
. D o€ mling, A.; Ugi, I. Angew. Chem., Int. Ed. 2000, 39, 3168–
210.
. (a) Harriman, G. C. B. Tetrahedron Lett. 1997, 38, 5591–
and the mixture is stirred for two days at room tempera-
ture.
2
3
3
The solvent was removed and the remaining oil was
dissolved in dichloromethane and washed successively with
diluted hydrochloric acid and sodium carbonate solution.
The organic phase was dried over sodium sulfate
and filtered. Solvent was removed in vacuum to result
5594; (b) Short, K. M.; Mjalli, M. M. Tetrahedron Lett.
1997, 38, 359–362; (c) Hanusch-Kompa, C.; Ugi, I. Tetra-
hedron Lett. 1998, 39, 2725–2728; (d) Marcaccini, S.;
Pepino, R.; Torroba, T.; Miguel, D.; Garcia-Valverde, M.
Tetrahedron Lett. 2002, 43, 8591–8593; (e) Zimmer, R.;
Ziemer, A.; Gruner, M.; Br u€ dgam, I.; Hartl, H.; Reissig, H.
U. Synthesis 2001, 1649–1658.
. Westermann, B.; Diedrichs, N.; Krelaus, R.; Walter, A.;
Gedrath, I. Tetrahedron Lett., see preceding paper,
doi:10.1016/j.tetlet.2004.06.051.
. (a) Diedrichs, N.; Westermann, B. Synlett 1999,
1127–1129; (b) Diedrichs, N.; Krelaus, R.; Gedrath, I.;
Westermann, B. Can. J. Chem. 2002, 80, 686–691.
. (a) Lee, D.; Sello, J. K.; Schreiber, S. L. Org. Lett. 2000, 2,
709–712; (b) Hebach, C.; Kazmaier, U. Chem. Commun.
2
carboxamide 13 (yield 81%) as a yellow oil, which was
-(R,S)-N-butyl-1-isobutyl-2-methyl-5-oxo-2-pyrrolidine-
characterized without further purification.
1
H NMR (CDCl
3
) 0.93 (d, J ¼ 6:6 Hz, 6H), 0.93 (t,
4
5
6
J ¼ 6:9 Hz), 1.22–1.55 (m, 4H), 1.56 (s, 3H), 1.83–2.10
m, 2H), 2.26–2.47 (m, 2H), 2.72 (dd,
¼ 6:9 Hz,
¼ 13:8 Hz, 1H), 3.20–3.42 (m, 4H), 5.90 (br s, 1H).
(
J
1
J
2
3
1
3
C NMR (CDCl ) 14.1, 20.5, 20.7, 21.1, 24.0, 28.6, 30.0,
2.0, 33.5, 40.0, 49.3, 68.3, 173.8, 176.7. HRMS calcd
3
for
þ
C
14
H
26
N
2
O
2
[M]
m=e 254.19943, found m=e
2
. Orru, R., personal communication.
. (a) Ma, S.; Ni, B. Org. Lett. 2002, 4, 639–641; (b) Ma, S.;
Ni, B. Chem. Eur. J. 2004, online preprint.
54.20132.
8
9
2003, 596–597; (c) Beck, B.; Larbig, G.; Magnin-Lachaux,
M.; Picard, A.; Herdtweck, E.; D o€ mling, A. Org. Lett.
003, 5, 1047–1050.
2