Stereoselective synthesis of pyrrolidinyl glycines from nitrones: complementarity of nucleophilic addition and 1,3-dipolar cycloaddition

Article abstract of DOI:10.1016/j.tetlet.2006.05.114

Epimeric pyrrolidinyl glycines, a sort of conformationally constrained α,β-diaminoacids, were stereoselectively prepared using complementary approaches based on nitrone chemistry. Nucleophilic additions to pyrrolidinyl nitrones and 1,3-dipolar cycloadditi

Full text of DOI:10.1016/j.tetlet.2006.05.114

Tetrahedron Letters 47 (2006) 5013–5016
Stereoselective synthesis of pyrrolidinyl glycines from
nitrones: complementarity of nucleophilic addition and
1,3-dipolar cycloaddition
Pedro Merino,a, Petra Padar, Ignacio Delso,a Muniappan Thirumalaikumar,a
a,b
*
´ ´
a
b,
*
´
´
Tomas Tejero and Lajos Kovacs
aLaboratorio de Sintesis Asimetrica, Departamento de Qu´ımica Orga´nica, Instituto de Ciencia de Materiales de Aragon,
Universidad de Zaragoza, Consejo Superior de Investigaciones Cientificas, E-50009 Zaragoza, Aragon, Spain
bDepartment of Medicinal Chemistry, University of Szeged, H-6720 Szeged, Hungary
Received 10 April 2006; revised 14 May 2006; accepted 17 May 2006
Abstract—Epimeric pyrrolidinyl glycines, a sort of conformationally constrained a,b-diaminoacids, were stereoselectively prepared
using complementary approaches based on nitrone chemistry. Nucleophilic additions to pyrrolidinyl nitrones and 1,3-dipolar cyclo-
additions of L-serine derived nitrones to form the corresponding hydroxylamines and isoxazolidines, respectively, provided key
intermediates for the synthesis of the target compounds. Whereas the nucleophilic addition route afforded the syn adduct, the
1,3-dipolar cycloaddition approach furnished the precursor for the preparation of the corresponding anti compound.
Ó 2006 Elsevier Ltd. All rights reserved.
a,b-Diaminoacids 1 are interesting targets because of
their utility as synthetic intermediates as well as their
presence in several biologically active compounds.1 In
addition, much attention was focused during the last
years on the preparation of conformationally con-
strained amino acids.2 A recent surge of activity into
the synthetic chemistry of hetaryl glycines, in particular,
tetrahydrofuranyl derivatives is also indicative of the
importance of this area.3
As a part of our continuing research program on the syn-
thetic potential of nitrones, and in the preparation of
non-proteinogenic aminoacids,8 including a,b-diamino
acids,9 we now report on two new complementary routes
leading to epimers at the b-stereogenic center of pyrroli-
dinyl glycines.
The synthetic approaches are based on nitrone chemis-
try and they make use of the two different and well-
known reactivities of the nitrone functionality in 1,3-
dipolar cycloadditions and nucleophilic additions.10
Pyrrolidinyl glycines 2 can be considered as conforma-
tionally restricted a,b-diamino acids whose preparation
is still scarcely considered. To the best of our knowledge,
such compounds have only been synthesized in parti-
cular cases. Compounds 3 have been used as synthetic
intermediates for the preparation of bicyclic piperazine
2-carboxylic acids4 that have been further utilized in
asymmetric synthesis as either catalysts or building
blocks.5 Bicyclic pyrrolidinyl glycines 4 have also been
employed as intermediates in the synthesis of models
for the azinomycin6 and ficellomycin7 antibiotics
(Fig. 1).
*
Corresponding authors. Tel./fax: +34 976 762075 (P.M.); e-mail:
pmerino@unizar.es
Figure 1.
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.114

5014
P. Merino et al. / Tetrahedron Letters 47 (2006) 5013–5016
Scheme 1. Reagents and conditions: (i) Ac2O, Py, rt; (ii) Bu4NF, THF, rt; (iii) RuCl3, NaIO4, CH3CN–CCl4–H2O (3:2:2), rt; (iv) CH2N2, Et2O, rt;
(v) H2, 100 atm, Pd(OH)2–C, Boc2O, rt.
Our first approach started from the completely stereo-
selective addition of lithium trimethylsilyl acetylide to
the pyrrolidinyl nitrone 5 (Scheme 1). Only one isomer
could be detected in the reaction mixture and compound
6 ([a]D +5 (c 0.21, CHCl3)) was obtained in 93% chem-
ical yield.11 The stereochemical course of the reaction
can be explained on the basis of our previously reported
model for nucleophilic additions of Grignard reagents to
pyrrolidinyl nitrones.12 The absolute configuration of 6
was unambiguously assigned following our previously
reported empirical rule based on NMR measurements.13
After protection of the N-hydroxyamino group as an
acetyl derivative, compound 7 (100%; [a]D À27 (c 0.41,
CHCl3)) was obtained. Unfortunately, all attempts of
oxidizing the ethynyl group to a carboxylic acid either
with RuO2 or RuCl3 in the presence of an excess of
sodium periodate as a reoxidant, failed; only decomposi-
tion products were recovered from the reaction
mixtures. A similar behavior had already been observed
in our laboratory.14 Thus the silyl groups in 6 were
cleaved with TBAF in THF and after acetylation of
mixture revealed the presence of four isomers in 35:7:7:1
ratio, which was further confirmed by HPLC (XTerra
C18, 5 lm, MeOH–H2O, 3:2). After separation of the
adducts by OPLC (OPLC-50, 0.2 mm HTSorbTM 5 lm
silica gel layer, hexane/EtOAc 8:2, 50 bar, 500 lL/min)
the major adduct 12a was obtained in 56% isolated
yield.16 The stereochemical assignment of 12a–d was
determined by careful NMR analysis utilizing homonu-
clear decoupling, multiple-difference NOE and 2D
experiments including COSY, ROESY, and HMQC.
Moreover, the observed stereochemical induction is in
agreement with previous observations for a-alkoxy and
a-amino nitrones for which, in all cases, the diastereo-
facial induction showed to be anti with respect to the
heteroatom in a position to the nitrone moiety.17 Tenta-
tively, we suggest that this stereochemical outcome is in
agreement with the transition state model A illustrated
in Figure 2 corresponding to an endo approach of the
dipolarophile to the Re face of the nitrone.
the resulting crude product, compound
8 (89%;
[a]D À37 (c 0.38, CHCl3)) was obtained.
In this case, unmasking of the carboxyl moiety was car-
ried out by oxidation of the triple bond with the system
RuCl3–NaIO4 in good yield, thus demonstrating that
the replacement of the O-silyl protecting group by the
acetyl one and/or removal of the C-silyl group is crucial
for the success of the oxidation. After esterification of
the crude carboxylic acid with freshly prepared diazo-
methane, the N-(acetoxy) pyrrolidinyl glycine 9 (73%;
[a]D À39 (c 0.25, CHCl3)) was obtained after purifica-
tion by radial chromatography. Hydrogenation under
pressure (100 atm) in the presence of the Pearlman’s cat-
alyst and Boc2O afforded the protected pyrrolidinyl gly-
cine 10 in 57.4% overall yield (six steps from nitrone
5).15
Our second approach was based on the hitherto un-
known 1,3-dipolar cycloaddition between the L-serine
derived nitrone 11 and methyl acrylate (Scheme 2).
The reaction was conducted without solvent at 90 °C
in a sealed tube for 5 h. The NMR analysis of the crude
Scheme 2. Cycloaddition between 11 and methyl acrylate.

P. Merino et al. / Tetrahedron Letters 47 (2006) 5013–5016
5015
Figure 2. Proposed model for the 1,3-dipolar cycloaddition between 11
and methyl acrylate.
Catalytic hydrogenation of 12a using Pearlman’s cata-
lyst provided pyrrolidinone 13 ([a]D À33 (c 0.21,
CHCl3); mp 164–166 °C) in 95% yield (Scheme 3).
Compound 13 was treated with tert-butyldiphenyl silyl
chloride and Boc2O to afford the protected pyrrolidin-
2-one 14 ([a]D +34 (c 0.22, CHCl3)). However, acidic
hydrolysis of the oxazolidine moiety to liberate the pri-
mary hydroxyl group as a previous step for the oxida-
tion reaction afforded a 3:2 mixture of the expected
compound 15a ([a]D +25 (c 0.17, CHCl3)) and 15b
([a]D À6 (c 0.30, CHCl3)), the latter coming from an
unexpected silyl migration.
Scheme 4. Reagents and conditions: (i) PhCOCl, Py, CH2Cl2, 0 °C; (ii)
Boc2O, Et3N, DMAP, CH2Cl2; (iii) p-TsOH, MeOH, 45 °C, 6 h; (iv)
TEMPO, [bis(acetoxy)iodo]benzene, MeCN–H2O; (v) CH2N2, Et2O,
rt.
oxidized with the system TEMPO-BAIB18 to afford the
crude carboxylic acid. This compound was isolated and
fully characterized19 as the corresponding methyl ester
18 which was obtained by treatment of the acid with
an ethereal solution of freshly prepared diazomethane.
Compound 18 was obtained in 17% overall yield (seven
steps from nitrone 11).
In order to avoid the silyl migration, we then decided to
change the protecting group of the hydroxyl group.
After benzoylation and N-protection of 13, compound
16 (80%, [a]D +45 (c 0.25, CHCl3)) was obtained as
the immediate precursor of the target compound
(Scheme 4). Treatment of 16 with catalytic p-TsOH in
methanol gave rise to the free primary alcohol 17
(80%, [a]D +20 (c 0.27, CHCl3)) which was subsequently
We also attempted the reduction of the lactam moiety in
compound 18 to obtain the corresponding saturated
pyrrolidine, following the Garcia-Ruano’s procedure.20
Unfortunately, the reduction failed and only the starting
compound was obtained.21
In conclusion, two complementary routes leading to syn-
and anti-pyrrolidinyl glycines via a nucleophilic addition
to nitrone 5 and a cycloaddition reaction of 11 with
methyl acrylate, respectively, have been achieved. In
the first approach, an ethynyl group has been used as
a synthetic equivalent of the carboxyl unit. For the sec-
ond approach, the synthetic equivalence between the
oxazolidine ring and the glycine unit has been utilized.
Unusual conformationally constrained a,b-diamino-
acids containing both a saturated ring of pyrrolidine and
a pyrrolidin-2-one ring have been successfully prepared
in this work. The preparation of other pyrrolidinyl gly-
cines of interest through these methods, as well as chem-
ical modifications of the prepared compounds are now
underway in our laboratories.
Acknowledgements
We thank the Ministerio de Educacion y Ciencia (MEC,
Project CTQ2004-0421), the European Union (Project
TRIoH, LSHB-CT-2003-503480), and the Regional
Government of Aragon (DGA) for financial support.
I.D. thanks DGA for a pre-doctoral grant.
Scheme 3. Reagents and conditions: (i) H2, Pd(OH)2–C, 100 atm, rt;
(ii) BuPh2SiCl, imidazole, DMF, 70 °C; (iii) Boc2O, Et3N, DMAP,
CH2Cl2; (iv) p-TsOH, MeOH, 45 °C, 6 h.
t

5016
P. Merino et al. / Tetrahedron Letters 47 (2006) 5013–5016
1H), 3.15 (dd, J = 3.5, 12.0 Hz, 1H), 3.59 (s, 3H), 4.10 (m,
References and notes
1H), 4.70 (d, J = 8.1 Hz, 1H), 6.10 (br s, 1H). Anal. Calcd
for C21H36N2O6 C, 61.14; H, 8.80; N, 6.79. Found: C,
61.37; H, 8.61; N, 6.55.
1. Viso, A.; Fernandez de la Pradilla, R.; Garcia, A.; Flores,
A. Chem. Rev. 2005, 105, 3167–3196.
16. Data for 12a: white solid; mp 63–66 °C. [a]D À54 (c 0.11,
CHCl3); 1H NMR (400 MHz, 70 °C, DMSO-d6): d 1.44 (s,
3H), 1.45 (s, 12H), 2.56 (ddd, J = 2.4, 8.3, 12.8 Hz, 1H),
2.63 (ddd, J = 7.5, 8.1, 12.8 Hz, 1H), 3.45 (ddd, J = 2.4,
7.4, 7.5 Hz, 1H), 3.72 (s, 3H), 3.80–3.83 (m, 4H), 4.07 (d,
J = 13.6 Hz, 1H), 4.60 (dd, J = 8.1, 8.3 Hz, 1H), 7.26–7.30
(m, 5H). Anal. Calcd for C22H32N2O6 C, 62.84; H, 7.67;
N, 6.66. Found: C, 62.71; H, 7.53; N, 6.70.
17. (a) Chiacchio, U.; Borrello, L.; Iannazzo, D.; Merino, P.;
Piperno, A.; Rescifina, A.; Richichi, B.; Romeo, G.
Tetrahedron: Asymmetry 2003, 14, 2419–2425; (b) Merino,
P.; Franco, S.; Merchan, F. L.; Tejero, T. J. Org. Chem.
2000, 65, 5575–5589.
2. (a) Miller, J. A.; Nguyen, S. B. T. Mini-Rev. Org. Chem.
2005, 2, 39–45; (b) Galeazzi, R.; Mobbili, G.; Orena, M.
Curr. Org. Chem. 2004, 8, 1799–1829; (c) Several authors.
In Tetrahedron Symposia-In-Print; Hruby, V. J.; Solos-
honok, V. A., Eds., 2001; Vol. 57, pp 6329–6650.
3. Jirgensons, A.; Marinozzi, M.; Pelliciari, R. Tetrahedron
2005, 61, 373–377.
4. Hanessian, S.; Sharma, R. Heterocycles 2000, 52, 1231–
1232.
5. Foresti, E.; Palmieri, G.; Petrini, M.; Profeta, R. Org.
Bioorg. Chem. 2003, 1, 4275–4281.
6. Moran, E. J.; Tellew, J. E.; Zhao, Z. C.; Armstrong, R. W.
J. Org. Chem. 1993, 58, 7848–7859.
18. Epp, J. B.; Widlanski, T. S. J. Org. Chem. 1999, 64, 293–
295.
7. Paumier, D.; Garcia, M.; Shipman, M.; Muir, J. C. Synlett
2004, 2212–2214.
8. For an account, see: Merino, P.; Franco, S.; Merchan, F.
L.; Tejero, T. Synlett 2000, 442–454.
9. (a) Merino, P.; Lanaspa, A.; Merchan, F. L.; Tejero, T.
Tetrahedron Lett. 1997, 38, 1813–1816; (b) Merino, P.;
Lanaspa, A.; Merchan, F. L.; Tejero, T. Tetrahedron:
Asymmetry 1998, 9, 629–646.
10. Merino, P. Nitrones and their Cyclic Derivatives. In Science
of Synthesis; Padwa, A., Ed.; Thieme: Stuttgart, 2004, Vol.
27, pp 511–580.
19. Data for 18: oil, [a]D À2 (c 0.13, CHCl3). 1H NMR
(400 MHz, 70 °C, DMSO-d6): d 1.39 (s, 9H), 1.50 (s, 9H),
2.04 (ddd, J = 6.3, 6.7, 13.7 Hz, 1H), 2.52–2.54 (m, 1H),
3.70 (s, 3H), 4.54–4.56 (m, 1H), 5.03 (dd, J = 3.5, 9.1 Hz,
1H), 5.56 (dd, J = 6.7, 9.8 Hz, 1H), 7.01 (d, J = 9.1 Hz,
1H), 7.50–7.54 (m, 2H), 7.67–7.70 (m, 1H), 8.02–8.05 (m,
2H). Anal. Calcd for C24H32N2O9 C, 58.53; H, 6.55; N,
5.69. Found: C, 58.49; H, 6.60; N, 5.76.
´
20. Ezquerra, J.; Rubio, A.; Pedregal, C.; Sanz, G.; Rodrı-
guez, J. H.; Garcia-Ruano, J. L. Tetrahedron Lett. 1993,
34, 4952–4989.
11. Merino, P.; Franco, S.; Gascon, J. M.; Merchan, F. L.;
Tejero, T. Tetrahedron: Asymmetry 1999, 10, 1867–1871.
12. Merino, P.; Tejero, T. Tetrahedron 2001, 57, 8125–8128.
13. Merino, P.; Franco, S.; Gascon, J. M.; Merchan, F. L.;
Tejero, T. Tetrahedron: Asymmetry 1999, 10, 1861–1865.
14. Merino, P.; Franco, S.; Merchan, F. L.; Tejero, T. J. Org.
Chem. 1998, 63, 5627–5630.
21. Quite probably the reaction conditions are too mild for
performing the reaction, and other conditions could lack
of chemoselectivity with the different ester functionalities
present in the molecule. This reduction step should be
attempted at a earlier stage (i.e., with compound 13) in
order to have some guarantee of success. We are currently
studying this possibility and it will be reported in due
course.
15. Data for 10: oil, [a]D À54 (c 0.11, CHCl3). 1H NMR
(400 MHz, 100 °C, DMSO-d6): d 1.42 (s, 9H), 1.44 (s, 9H),
1.96 (s, 3H), 2.10–2.30 (m, 2H), 3.06 (dd, J = 3.8, 12.0 Hz,