Novel α-substituted β-amino diesters from acylnitroso-derived hetero-Diels-Alder cycloadducts

Article abstract of DOI:10.1016/S0040-4039(01)02353-X

α-Methyl β-amino diesters were synthesized in a simple three-step procedure starting from acylnitroso-derived Diels-Alder adducts of cyclopentadiene. Treatment of the cycloadducts with copper catalyst-modified methylmagnesium bromide gave anti-1,2-cyclope

Full text of DOI:10.1016/S0040-4039(01)02353-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 1131–1134
Novel a-substituted b-amino diesters from acylnitroso-derived
hetero-Diels–Alder cycloadducts
Matthew D. Surman, Mark J. Mulvihill and Marvin J. Miller*
Department of Chemistry and Biochemistry, University of Notre Dame Notre Dame, IN 46556, USA
Received 27 November 2001; revised 11 December 2001; accepted 12 December 2001
Abstract—a-Methyl b-amino diesters were synthesized in a simple three-step procedure starting from acylnitroso-derived
Diels–Alder adducts of cyclopentadiene. Treatment of the cycloadducts with copper catalyst-modified methylmagnesium bromide
gave anti-1,2-cyclopentenyl hydroxamic acids. Reduction of the hydroxamate NꢀO bond with titanium(III) chloride followed by
ozonolytic cleavage of the cyclopentenyl olefin provided the desired a-methyl b-amino diesters. © 2002 Elsevier Science Ltd. All
rights reserved.
1. Introduction
ties. Compound 8 was synthesized from b-amino diester
1 (R1=R2=Et, R3=R4=H).8 While b-amino diacids
and diesters have enjoyed a broad range of synthetic
applications as demonstrated above, they have been
used most extensively as precursors to b-lactams 9.9
Herein, we report a novel synthetic strategy for the
construction of b-amino diesters.
b-Amino diacids (1, R1=R2=H) and diesters (1, R1,
R2=alkyl) have served as valuable intermediates in the
synthesis of
a variety of biologically interesting
molecules (Fig. 1). For example, analogs of the C-ter-
minal tetrapeptide of gastrin (2) were synthesized from
b-homo aspartic acid (1, R1=t-Bu, R2=H, R3=H,
R4=Cbz) and showed potent gastrin antagonist activ-
ity.1 Harada and co-workers synthesized interesting
polypeptides 3 from 3-aminoglutaric acid (1, R1=R2=
R3=R4=H).2,3 Both enantiomers of 3,4-diamino-
butyric acid 4 were constructed from b-amino diester 1
(R1=t-Bu, R2=Me, R3=H, R4=Cbz) and served as
synthons for heterocyclic GABA-receptor agonists.4
Hultin and Jones showed that b-amino diesters 5 were
allosteric activators of trypsin catalysis.5 Amide-linked
triacridines 6 were synthesized as potential antitumor
agents from Boc-3-aminoglutaric acid (1, R1=R2=
R3=H, R4=Boc).6 Diethyl b-pyrrolidinylgluconate [1,
R1=R2=Et, R3, R4=CH2(CH2)2CH2] was used to con-
struct phenol 7, a potential intermediate in the synthesis
of natural products containing linear aromatic sys-
Carbapenems 9 makes up an important class of b-lac-
tam antibiotics with a wide spectrum of action against
Gram-positive and Gram-negative bacteria. They also
have potent in vitro and in vivo activity against the
clinically important Pseudomonas aeruginosa.10 The
potent activity of carbapenems has been attributed to
good diffusion through the bacterial outer membrane,
high affinity for penicillin-binding proteins, and high
stability and inhibitory activity against b-lactamases.
Extensive SAR studies have shown that 1b-methyl car-
bapenems (9, R5=CH3) are among the most potent in
this class of antibiotics.11
Even after decades of research, carbapenems continue
to capture the attention of synthetic chemists.9 The
densely functionalized bicyclic core of carbapenems
offers a number of synthetic challenges, including the
ability to obtain the required b-stereochemistry at the
C(1) position. Opening the synthetically versatile
acylnitroso Diels–Alder cycloadduct 10 with various
Grignard reagents was envisioned to provide anti-1,2-
cyclopentenes 11 that could be converted into a-substi-
tuted b-amino diesters 12. These diesters (12) possess
the required relative stereochemistry to serve as precur-
sors to 1b-substituted carbapenems 13 (Scheme 1).
tems.7
Substituted-9-azabicyclo[3.3.1]nonan-3-ones,
such as 3,7-dione derivative 8, often possess a wide
spectrum of biological activities, including anticholiner-
gic,
antidepressant,
neuroleptic,
cardiovascular,
antiparkinsonian, antidiabetic, and antitussive activi-
Keywords: acylnitroso; b-amino diesters; 1b-methyl carbapenems;
Grignard; hydroxamic acid.
* Corresponding author. Tel.: (219) 631-7571; fax: (219) 631-6652;
e-mail: marvin.j.miller.2@nd.edu
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)02353-X

1132
M. D. Surman et al. / Tetrahedron Letters 43 (2002) 1131–1134
Figure 1.
mately 70% yield of the desired and separable
anti-1,2-product 16. Similarly, treatment of N-Cbz
cycloadduct 15 with methylmagnesium bromide in the
presence of a catalytic amount of copper(II) chloride
provided a 98% yield of a 2.2:1 mixture of anti-1,2-
(17):anti-1,4-products, corresponding to a 67% yield of
the desired and separable product 17.16b,17 Attempts to
further optimize the reaction by changing the solvent
from tetrahydrofuran to diethyl ether were unsuccess-
ful. When copper was used in ether, the reaction with
cycloadduct 14 gave a 71% yield of a 13.5:4:1 ratio of
anti-1,2-(16):anti-1,4-:syn-1,4-products, and when cop-
per was omitted, only a 27% yield was obtained.
Scheme 1.
The synthesis was first continued with the benzyl pro-
tection of N-Boc hydroxamic acid 16, which proceeded
smoothly in 92% yield (Scheme 3). The next step
involved the oxidative cleavage of the cyclopentenyl
olefin. Initial attempts using ruthenium tetraoxide fol-
lowed by treatment with diazomethane gave rather low
2. Results and discussion
The synthesis of a-substituted b-amino diester 12 (R1=
Me) began with the opening of N-Boc cycloadduct 14
and N-Cbz cycloadduct 15 with methylmagnesium bro-
mide (Scheme 2).12 An initial attempt at the ring open-
ing of cycloadduct 14 in the absence of any copper
catalyst gave an unexpectedly low yield (12%) of a
1.6:1.4:1 ratio of anti-1,2-(16):anti-1,4-:syn-1,4-prod-
ucts. Inclusion of a catalytic amount of copper(II)
increased both the yield and selectivity of the reaction.
Use of ten mole percent of copper(II) chloride gave a
90% yield of a 27:8:1 ratio of anti-1,2-(16):anti-1,4-:syn-
1,4-products.14,16a This corresponded to an approxi-
Scheme 2.

M. D. Surman et al. / Tetrahedron Letters 43 (2002) 1131–1134
1133
methylmagnesium bromide gave anti-1,2-cyclopentenyl
hydroxamic acids 16 and 17. Reduction of the hydroxa-
mate NꢀO bond with titanium(III) chloride followed by
ozonolytic cleavage of the cyclopentenyl olefin provided
b-amino diesters 22 and 23. These a-methyl b-amino
diesters possess the required relative stereochemistry to
serve as important precursors to 1b-methyl
carbapenems.
Acknowledgements
Scheme 3.
We gratefully acknowledge the NIH for partial finan-
cial support of this research, the University of Notre
Dame for support in the form of a J. Peter Grace
fellowship, the Lizzadro Magnetic Resonance Research
Center at Notre Dame for NMR facilities, Dr. W.
Boggess and N. Sevova for mass spectrometry facilities,
and Maureen Metcalf for assistance with the
manuscript.
yields (up to 37%). An attempt was made to use
potassium permanganate as the oxidant, however the
yield was still quite low (17%).
The possibility of hydroxamate 18 binding to the metal
oxidants was thought to contribute to the inefficiency
of these oxidative cleavage reactions. In order to allevi-
ate this potential problem, a somewhat different
approach to the b-amino diester synthesis was taken.
The modified synthesis involved reducing the hydrox-
amic acid NꢀO bond prior to the oxidative cleavage of
the olefin. Hydroxamic acids 16 and 17 were treated
with titanium(III) chloride in a pH 7 potassium acetate
buffer to give amines 20 and 21 in good yields (82 and
93%, respectively) (Scheme 4).16c,d,19 Attempts to cleave
the cyclopentenyl olefin of amine 20 with either potas-
sium permanganate or ruthenium tetraoxide were
unsuccessful, giving complex mixtures of products. Suc-
cessful olefin cleavage was realized through the use of
ozonolysis conditions. Treatment of amines 20 and 21
with ozone in basic methanol provided direct access to
a-methyl b-amino dimethyl esters 22 and 23 in 63 and
49% yield, respectively.16e,f,20,21 Interestingly, even under
the basic reaction conditions, no evidence of epimeriza-
tion at the a-methyl position was observed.
References
1. Rodriguez, M.; Fulcrand, P.; Laur, J.; Aumelas, A.; Bali,
J. P.; Martinez, J. J. Med. Chem. 1989, 32, 522.
2. Munegumi, T.; Meng, Y.-Q.; Harada, K. Chem. Lett.
1988, 1643.
3. Munegumi, T.; Meng, Y.-Q.; Harada, K. Bull. Chem.
Soc. Jpn. 1989, 62, 2748.
4. Misiti, D.; Santaniello, M.; Zappia, G. Synth. Commun.
1992, 22, 883.
5. Hultin, P. G.; Jones, J. B. Bioorg. Chem. 1992, 20, 30.
6. Atwell, G. J.; Baguley, B. C.; Wilmanska, D.; Denny, W.
A. J. Med. Chem. 1986, 29, 69.
7. Singh, S.; Singh, I. Indian J. Chem. 1985, 24B, 586.
8. Rao, J.; Saxena, A. K. Indian J. Chem. 1989, 28B, 620.
9. Berks, A. H. Tetrahedron 1996, 52, 331.
10. Sumita, Y.; Yamaga, H.; Sunagawa, M. J. Antibiot. 1995,
48, 89.
11. Shih, D. H.; Baker, F.; Cama, L.; Christensen, B. G.
Heterocycles 1984, 21, 29.
3. Conclusions
b-Amino diacids and diesters are valuable synthetic
intermediates, particularly in the construction of b-lac-
tams. a-Methyl b-amino diesters 22 and 23 were readily
12. For the synthesis of cycloadducts 14 and 15, see: Mulvi-
hill, M. J.; Gage, J. L.; Miller, M. J. J. Org. Chem. 1998,
63, 4874. While the cycloadducts used in this synthesis
were racemic, the use of enantiomerically pure
cycloadducts13 should provide optically active a-substi-
tuted b-amino diesters.
accessed in
a simple three-step procedure from
acylnitroso Diels–Alder cycloadducts 14 and 15. Treat-
ment of 14 and 15 with copper catalyst-modified
13. Vogt, P. F.; Miller, M. J. Tetrahedron 1988, 54, 1317.
14. The opening of acylnitroso-derived Diels–Alder cycload-
ducts with other Grignard reagents and copper catalyst-
modified
Grignard
reagents
in
addition
to
methylmagnesium bromide has been explored and proved
to be successful.15
15. Surman, M. D.; Mulvihill, M. J.; Miller, M. J.,
manuscript submitted.
16. (a) Characterization data for 16: white solid; mp 56–
58°C; IR (TF) 3223, 2977, 2930, 1691, 1456, 1395, 1368,
1347, 1253, 1167, 1109, 907, 864, 757, 712 cm−1; 1H NMR
(300 MHz, CDCl3): l 1.08 (d, J=7.2, 3H), 1.47 (s, 9H),
2.47 (m, 1H), 2.65 (m, 1H), 2.99 (m, 1H), 4.26 (dt, J=7.2,
8.7 Hz, 1H), 5.55 (m, 1H), 5.62 (m, 1H), 7.38 (s, 1H); 13C
Scheme 4.

1134
M. D. Surman et al. / Tetrahedron Letters 43 (2002) 1131–1134
NMR (75 MHz, CDCl3): l 19.01, 28.30, 34.32, 41.61,
C14H18NO2 (M+H)+ 232.1338, found 232.1348.
66.14, 81.73, 127.61, 135.20, 157.14; HRMS (FAB) calcd
for C11H20NO3 (M+H)+ 214.1443, found 214.1448.
(b) Characterization data for 17: white solid; mp 71–
73°C; IR (TF) 3256, 2957, 2926, 2875, 1697, 1456, 1415,
(e) Characterization data for 22: clear oil; IR (TF) 3372,
2980, 1738, 1726, 1530, 1502, 1440, 1367, 1248, 1168
1
cm−1; H NMR (300 MHz, CDCl3): l 1.18 (d, J=7.5 Hz,
3H), 1.41 (s, 9H), 2.53 (m, 2H), 2.78 (m, 1H), 3.67 (s,
6H), 4.15 (m, 1H), 5.20 (d, J=9.0 Hz, 1H); 13C NMR (75
MHz, CDCl3): l 13.88, 28.28, 36.52, 43.16, 49.56, 51.72,
51.78, 79.49, 155.15, 171.75, 174.58; HRMS (FAB) calcd
for C13H24NO6 (M+H)+ 290.1604, found 290.1613.
(f) Characterization data for 23: clear oil; IR (TF) 3351,
1
1329, 1271, 1212, 1101, 697 cm−1; H NMR (300 MHz,
DMSO-d6): l 0.99 (d, J=7.2, 3H), 2.43 (m, 2H), 2.84 (m,
1H), 4.30 (q, J=7.5 Hz, 1H), 5.10 (s, 2H), 5.54 (m, 1H),
1
5.59 (m, 1H), 7.34 (m, 5H), 9.32 (s, 1H); H NMR (300
MHz, CDCl3): l 1.07 (d, J=7.2, 3H), 2.47 (m, 1H), 2.68
(m, 1H), 3.05 (m, 1H), 4.38 (overlapping ddd, J=7.2, 7.2,
8.7 Hz, 1H), 5.18 (d, J=3.0 Hz, 2H), 5.57 (m, 1H), 5.63
(m, 1H), 7.35 (m, 5H), 7.98 (bs, 1H); 13C NMR (75 MHz,
DMSO-d6): l 18.89, 34.02, 41.17, 65.55, 66.35, 127.69,
127.75, 127.90, 128.36, 135.04, 136.69, 156.51; 13C NMR
(75 MHz, CDCl3): l 18.79, 34.27, 41.46, 66.12, 67.85,
127.49, 127.87, 128.17, 128.45, 135.07, 135.89, 157.53;
HRMS (FAB) calcd for C14H18NO3 (M+H)+ 248.1287,
found 248.1292.
2953, 1734, 1522, 1457, 1437, 1238, 1205, 1044, 740 cm−1
;
1H NMR (500 MHz, CDCl3): l 1.21 (d, J=7.5 Hz, 3H),
2.58 (m, 2H), 2.83 (m, J=7.5 Hz, 1H), 3.66 (s, 3H), 3.68
(s, 3H), 4.23 (m, 1H), 5.09 (s, 2H), 5.47 (d, J=9.5 Hz,
1H), 7.34 (m, 5H); 13C NMR (75 MHz, CDCl3): l 14.01,
36.32, 43.11, 50.24, 51.77, 52.60, 66.85, 128.03, 128.11,
128.50, 136.50, 155.78, 171.65, 174.46; HRMS (FAB)
calcd for C16H22NO6 (M+H)+ 324.1447, found 324.1453.
17. The anti-1,2-stereochemical assignment of the major
product 17 was proven through the synthesis of known
trans-2-methyl-cyclopentyl-benzamide from hydroxamic
acid 17. anti-1,2-Hydroxamic acid 17 was exposed to
hydrogenation conditions to remove the Cbz group and
reduce the NꢀO bond and cyclopentenyl olefin. The
resulting cyclopentyl amine i was treated with benzoyl
chloride to give the benzamide derivative ii. The melting
point of ii (114–115°C) was consistent with the literature
value (116°C).18 The melting point of cis-2-methyl-
cyclopentyl-benzamide is 85°C.
(c) Characterization data for 20: white solid; mp=40–
43°C; IR (TF) 3343, 2977, 2930, 2870, 1692, 1530, 1502,
1
1366, 1280, 1248, 1172, 1076, 1042, 1011 cm−1; H NMR
(300 MHz, CDCl3): l 1.06 (d, J=6.9 Hz, 3H), 1.43 (s,
9H), 2.08 (dd, J=4.8, 16.5 Hz, 1H), 2.47 (m, 1H), 2.75
(dd, J=7.2, 16.5 Hz, 1H), 3.77 (bs, 1H), 4.67 (bs, 1H),
5.59 (s, 2H); 13C NMR (75 MHz, CDCl3): l 18.60, 28.38,
39.46, 47.52, 57.75, 78.99, 127.56, 135.14, 155.70; HRMS
(FAB) calcd for C11H20NO2 (M+H)+ 198.1494, found
198.1509.
(d) Characterization data for 21: white solid; mp=59–
60°C; IR (TF) 3315, 3032, 2956, 2870, 1685, 1541, 1455,
1279, 1246, 1080, 1034, 974, 760, 709, 694 cm−1; 1H NMR
(300 MHz, CDCl3): l 1.09 (d, J=6.9 Hz, 3H), 2.13 (ddd,
J=1.5, 4.8, 16.8 Hz, 1H), 2.52 (m, 1H), 2.81 (dd, J=7.5,
16.8 Hz, 1H), 3.87 (m, 1H), 4.84 (bs, 1H), 5.10 (s, 2H),
5.62 (s, 2H), 7.34 (m, 5H); 13C NMR (75 MHz, CDCl3):
l 18.58, 39.55, 47.62, 58.44, 66.63, 120.31, 127.55, 128.07,
128.52, 135.16, 136.73, 156.17; HRMS (FAB) calcd for
18. Hu¨ckel, W.; Kupka, R. Chem. Ber. 1956, 89, 1694.
19. Mattingly, P. G.; Miller, M. J. J. Org. Chem. 1980, 45,
410.
20. Marshall, J. A.; Garofalo, A. W. J. Org. Chem. 1993, 58,
3675.
21. Treatment of O-benzyl hydroxamate 18 under these
ozonolysis conditions resulted in only a modest yield
(31%) of the corresponding diester 19.