A short stereoselective synthesis of disubstituted cyclic amino acids

Article abstract of DOI:10.1016/S0040-4039(02)01267-4

A new synthetic route to enantiomerically pure disubstituted derivatives of cyclic amino acids is reported. Key step of this synthesis is an oxidative cleavage of azabicycloalkene precursors that are synthesized in enantiomerically pure form via aza-Diels

Full text of DOI:10.1016/S0040-4039(02)01267-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 5957–5960
A short stereoselective synthesis of disubstituted cyclic
amino acids
Wolfgang Maisona,* and Gunadi Adiwidjajab
aInstitut fu¨r Organische Chemie der Universita¨t Hamburg, Martin-Luther-King-Platz 6, 20146 Hamburg, Germany
bMineralogisch-Petrographisches Institut der Universita¨t Hamburg, Grindelallee 48, 20146 Hamburg, Germany
Received 4 June 2002; revised 25 June 2002; accepted 27 June 2002
Abstract—A new synthetic route to enantiomerically pure disubstituted derivatives of cyclic amino acids is reported. Key step of
this synthesis is an oxidative cleavage of azabicycloalkene precursors that are synthesized in enantiomerically pure form via
aza-Diels–Alder reaction. A range of different disubstituted pipecolic acid derivatives has been synthesized and their structure has
been evaluated by X-ray analysis. © 2002 Elsevier Science Ltd. All rights reserved.
a-Amino acids with a rigid backbone structure are
synthetically interesting targets because they can be
used as building blocks for the preparation of peptides
or peptidomimetic structures with biological activity.1
In particular, incorporation of cyclic amino acids based
on proline or pipecolic acid with well-defined structural
properties into peptides leads to useful model com-
pounds for studying peptide conformation and protein
folding.2 In addition, these substitutions often go along
with improved pharmacological profiles in pharmaco-
logically relevant peptides.3 Derivatives of proline and
pipecolic acid have been used as constrained analogues
of proteinogenic amino acids4 and as rigid modules to
constrain peptide conformation.5 In particular, 5-sub-
stituted prolines6 as well as 3,6-substituted pipecolic
acids7 have been demonstrated to serve as rigid ana-
logues of the cis- and trans-prolyl amide portion in
natural peptides and have been used as educts for the
synthesis of constrained dipeptide analogues.8 In addi-
tion, these compounds serve as valuable intermediates
in the synthesis of a range of different pyrollidine and
piperidine alkaloids for example of the indolizidine and
quinolizidine type.9
All these facts have contributed to the growing interest
in finding short and efficient stereoselective synthetic
routes to substituted cyclic amino acids. Efforts in this
direction have met with some success in the past and a
number of stereoselective routes to substituted
prolines10 and pipecolic acids11 have been developed.
However, most routes are multistep procedures and
general synthetic schemes suitable for 3,5-substituted
prolines and 3,6-substituted pipecolic acids are still
missing.
R*
N
R*
+
N
CO2R
n
n
I
CO2R
n = 0, 1
R* = chiral auxiliary
H
N
In the course of a project directed to the development
of rigid bioactive peptidomimetics, that should serve as
modular ligands for cancer specific receptors, we
became interested in highly functionalized derivatives of
proline and pipecolic acid of the general structure II
(Fig. 1) in enantiomerically pure form. Although some
multistep protocols to proline derivatives II (n=0) are
known,12 routes to the corresponding pipecolic acids II
(n=1) are extremely rare.11b
CO2H
R
R
n
II
Figure 1. Retrosynthetic analysis of cyclic amino acids II.
Keywords: proline; pipecolic acid; aza-Diels–Alder; peptidomimetic;
alkaloid.
A general route to cyclic amino acids of type II would
be especially valuable if both residues R were easy to
* Corresponding author. Tel.: +49-(0)40-428382813; fax: +49-(0)40-
428382495; e-mail: maison@chemie.uni-hamburg.de
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01267-4

5958
W. Maison, G. Adiwidjaja / Tetrahedron Letters 43 (2002) 5957–5960
Me
CO2Et
Ph
convert into a range of different natural amino acid
side chains or other functionalities, thus offering a
route to constrained amino acid analogues and versatile
intermediates for natural product synthesis.
N
1. O3, MeOH, -80 °C
2. DMS, rt
O
6
3
O
exo-1
6 %
2
We envisioned azabicycloalkenes I as ideal precursors
for cyclic amino acids of type II. As depicted in Fig. 1
these bicyclic precursors I in turn should be easily
accessible via stereoselective aza-Diels–Alder reaction
of a cyclic dien and a chiral imine. This kind of
stereoselective aza-Diels–Alder reaction has been stud-
ied extensively in the past and was shown to yield
various nitrogen heterocycles in a highly stereoselective
way if chiral amines are used as the stereochemical
source of information.13 Oxidative cleavage of the dou-
ble bond in resulting azabicycloalkenes I should then
give the desired disubstituted amino acid derivatives II
with suitable functionalities right in place for further
manipulation.
Me
Ph
1. O3, MeOH, -80 °C
2. DMS, rt
3. NaBH4
N
CO2Et
OH
HO
3
6
exo-1
78 %
3
Me
Ph
N
3
CO2Et
R
R
i, ii
6
3
4a, R = Br, i, 89 %
4b, R = Mes, ii, 56 %
Scheme 2. Synthesis of disubstituted pipecolic acid deriva-
tives. (i) Br2, PPh3, CH2Cl2, pyridine, 0°C; (ii) MesCl,
DIPEA, CH2Cl2, 0°C (compound 4b was isolated as its
hydrochloride salt).
Adopting a known protocol,14 we started with a one-
pot synthesis of azabicyclooctenes 1 according to
Scheme 1 from 1,3-cyclohexadiene and a chiral imine,
which was derived from (R)-phenylethylamine and
ethylglyoxylate. A mixture of exo-1 and endo-1 strongly
favoring the exo product was obtained in reasonable
yield. It should be noted that the reaction may be
performed on a large scale, thus offering a convenient
access to precursors of the general type I in Fig. 1. Both
isomers exo-1 and endo-1 are easily separated by
column chromatography on silica gel.
procedure yielded pipecolic acid derivative 3 in good
yield after chromatographic workup. The relative stere-
ochemistry of substituents attached to the piperidine
ring was established by NOESY NMR and was found
to be not altered by the oxidation reaction. Finally, the
absolute stereochemistry was determined by X-ray crys-
tal structure analysis15 of dibromide 4a (Fig. 2) which
was prepared according to Scheme 2 from diol 3 with
triphenylphosphine bromide in good yield.
With azabicyclooctenes
1 in hand, we tried an
ozonolytic cleavage of the double bond under standard
conditions to the bisaldehyde 2 according to Scheme 2.
Unfortunately we were not able to isolate bisaldehyde 2
using a standard flash chromatographic workup proce-
dure although TLC and NMR of the crude reaction
mixture indicated that 2 was formed. However, a rapid
workup yielded 2 in disappointingly low yield of 18%
after flash filtration over a silica gel plug. Suspecting
that the chemical instability of amino aldehyde 2 under
workup conditions did not allow its isolation in better
yields, we tried the ozonolysis of exo-1 with subsequent
reduction of the intermediate bisaldehyde 2 in one pot
with sodium borohydride next. As anticipated, this
Diol 3, dibromide 4a and mesylate 4b may serve as
versatile precursors for a number of different 3,6-sym-
metrically substituted pipecolic acid derivatives by stan-
dard functional group transformations. The intrinsic
Ph
TFA, CH2Cl2
BF3 * Et2O
N
Me
+
2 h at -80 °C,
72 h at rt
CO2Et
Ph
Ph
Me
N
+
N
Me
CO2Et
CO2Et
endo-1, 8 %
exo-1, 65 %
Figure 2. ORTEP plot of dibromide 4a (heteroatoms are
marked with element symbols and hydrogen atoms are omit-
ted for clarity).
Scheme 1. Aza-Diels–Alder protocol to azabicyclooctenes
exo-1 and endo-1.

W. Maison, G. Adiwidjaja / Tetrahedron Letters 43 (2002) 5957–5960
5959
Acknowledgements
basicity of compound 4b, due to the internal tertiary
amine, caused initial problems with its work up
because of elimination of mesylate. However, these
problems were solved by isolating 4b as its hydro-
chloride salt.
We gratefully acknowledge material support as well
as helpful discussions of Professor Dr. Chris Meier
and are grateful for a gift of Pd/C from Degussa AG.
Taking advantage of the vicinal aminoalcohol moiety
in diol 3, it can furthermore be derivatized selectively
at substituents in 3- or 6-position enlarging its syn-
thetic potential. Compound 3 was therefore debenzyl-
ated with hydrogen (balloon) and 5% palladium on
activated charcoal in ethanol for 24 h according to
Scheme 3 to give N-terminally deprotected pipecolic
acid 5 in quantitative yield. When treated with benz-
aldehyde in toluene (Dean–Stark conditions for 12 h),
selective protection of the 1,2-aminoalcohol moiety as
a benzylidene aminal in 616 was achieved in good
yield.
References
1. (a) Gante, J. Angew. Chem. 1994, 106, 1780–1802; Angew.
Chem., Int. Ed. Engl. 1994, 33, 1699–1720; (b) Giannis,
A.; Kolter, T. Angew. Chem. 1993, 105, 1303–1326;
Angew. Chem., Int. Ed. Engl. 1993, 32, 1244–1267; (c)
Hruby, V. J.; Li, G.; Haskell-Luevano, C.; Shenderovich,
M. Biopolymers 1997, 43, 219–266.
2. (a) Hanessian, S.; McNaughton-Smith, G.; Lombart, H.-
G.; Lubell, W. D. Tetrahedron 1997, 38, 12789–12854; (b)
MacArthur, M. W.; Thornton, J. M. J. Mol. Biol. 1991,
218, 397–412; (c) Maison, W.; Kennedy, R.; Kemp, D. S.
Tetrahedron Lett. 2001, 42, 4975–4977; (d) Maison, W.;
Kennedy, B.; Kemp, D. S. Angew. Chem. 2001, 113,
3936–3938; Angew. Chem., Int. Ed. Engl. 2001, 40, 3819–
3821.
Protected pipecolic acid
6
was obtained in
diastereomerically pure form and the stereochemistry
at the aminal C assigned to be (R)-configuration
based on NOESY NMR experiments. Compound 6
may serve as a precursor for selective functionaliza-
tion at C3 and C6 of the piperidine ring, giving thus
access to a number of enantiomerically pure con-
strained amino acid derivatives.
3. For example see: Smith, A. B., III; Benowitz, A. B.;
Sprengeler, P. A.; Barbosa, J.; Guzman, M. C.;
Hirschmann, R.; Schweiger, E. J.; Bolin, D. R.; Nagy, Z.;
Campbell, R. M.; Cox, D. C.; Olson, G. L. J. Am. Chem.
Soc. 1999, 121, 9286–9298.
4. Gibson, S. E.; Guillo, N.; Tozer, M. J. Tetrahedron 1999,
55, 585–615.
5. (a) Blanco, M.-J.; Penide, M. R.; Sardina, F. J. J. Org.
Chem. 1999, 64, 8786–8793; (b) Dumy, P.; Keller, M.;
Ryan, D. E.; Rohwedder, B.; Wo¨hr, T.; Mutter, M. J.
Am. Chem. Soc. 1997, 119, 918–925.
6. (a) Beausoleil, E.; Lubell, W. D. J. Am. Chem. Soc. 1996,
118, 12902–12908; (b) Halab, L.; Lubell, W. D. J. Org.
Chem. 1999, 64, 3312–3321.
7. (a) Maison, W.; Lu¨tzen, A.; Kosten, M.; Schlemminger,
I.; Westerhoff, O.; Saak, W.; Martens, J. J. Chem. Soc.,
Perkin Trans. 1 2000, 1867–1871; (b) Swarbrick, M. E.;
Gosselin, F.; Lubell, W. D. J. Org. Chem. 1999, 64,
1993–2002.
8. (a) Maison, W.; Acre, E.; Renold, P.; Kennedy, R.;
Kemp, D. S. J. Am. Chem. Soc. 2001, 123, 10245–10254;
(b) Belvisi, L.; Colombo, L.; Colombo, M.; Di Giacomo,
M.; Manzoni, L.; Vodopivec, B.; Scolastico, C. Tetra-
hedron 2001, 57, 6463–6473; (c) Tong, Y.; Fobian, Y. M.;
Wu, M.; Boyd, N. D.; Moeller, K. D. J. Org. Chem.
2000, 65, 2484–2493.
In summary, we have developed an efficient protocol
for the stereoselective synthesis of polyfunctionalized
derivatives of pipecolic acid. Key steps of this synthe-
sis are an aza-Diels–Alder reaction of a chiral imine
and subsequent oxidative cleavage of the resulting
azabicycloalkene. It should be noted that this proto-
col may easily be expanded to the synthesis of proline
derivatives by using cyclopentadien instead of cyclo-
hexadien in the aza-Diels–Alder reaction. Further-
more, both enantiomers of the chiral educt, (R)- and
(S)-phenylethylamine, are commercially available at
low cost thus giving access to both enantiomers of
cyclic amino acid derivatives like 3 in only two steps.
We are currently using these highly functionalized
amino acids as building blocks for the synthesis of
conformationally constrained dipeptide mimics (results
of this work will be reported elsewhere), but they
may also serve as intermediates for natural product
synthesis.
9. (a) Michel, P.; Rassat, A.; Daly, J. W.; Spande, T. F. J.
Org. Chem. 2000, 65, 8908–8918; (b) Schro¨der, F.;
Francke, W. Tetrahedron 1998, 54, 5259–5264; (c)
Ahman, J.; Somfai, P. Tetrahedron 1995, 51, 9747–9756.
10. (a) Noguchi, Y.; Uchiro, H.; Yamada, T.; Kobayashi, S.
Tetrahedron Lett. 2001, 42, 5253–5256; (b) Halab, L.;
Be´lec, L.; Lubell, W. D. Tetrahedron 2001, 57, 6439–
6446; (c) Davis, F. A.; Zhang, H.; Lee, S. H. Org. Lett.
2001, 3, 759–762; (d) Mitchinson, A.; Nadin, A. J. Chem.
Soc., Perkin Trans. 1 2000, 2862–2892; (e) Wang, Q.;
Sasaki, N. A.; Riche, C.; Potier, P. J. Org. Chem. 1999,
64, 8602–8607; (f) Wang, Q.; Tran Huu Dau, M.-E.;
Sasaki, N. A.; Potier, P. Tetrahedron 2001, 57, 6455–
6462.
H
N
CO2Et
OH
H2, Pd / C
78 %
HO
3
5
Ph
O
benzaldehyde,
toluene, NH4Cl
N
CO2Et
OH
5
84 %
6
Scheme 3. Synthesis of pipecolic acid 6.

5960
W. Maison, G. Adiwidjaja / Tetrahedron Letters 43 (2002) 5957–5960
11. (a) Naoki, N.; Tanaka, K.; Momose, T.; Daly, J. W.;
Garraffo, H. M. Tetrahedron 1997, 53, 9553–9574; (b)
Agami, C.; Comesse, S.; Kadouri-Puchot, C. J. Org.
Chem. 2000, 65, 4435–4439; (c) Di Nardo, C.; Varela, O.
J. Org. Chem. 1999, 64, 6119–6125; (d) Cheˆnevert, R.;
Morin, M.-P. J. Org. Chem. 1999, 64, 3178–3180; (e)
Haddad, M.; Larcheveˆque, M. Tetrahedron: Asymmetry
1999, 4231–4237; (f) Keenan, T. P.; Yaeger, D.; Holt, D.
A. Tetrahedron: Asymmetry 1999, 4331–4341; (g) Rutjes,
F. P. J. T.; Veerman, J. J. N.; Meester, W. J. N.;
Hiemstra, H.; Schoemaker, H. E. Eur. J. Org. Chem.
1999, 1127–1135; (h) Tjen, K. C. M. F.; Kinderman, S.
S.; Schoemaker, H. E.; Hiemstra, H.; Rutjes, F. P. J. T.
Chem. Commun. 2000, 699–700.
12. (a) Mulzer, J.; Meier, A.; Buschmann, J.; Luger, P.
Synthesis 1996, 123–132; (b) Lorthiois, E.; Marek, I.;
Normant, J. F. J. Org. Chem. 1998, 63, 2442–2450.
13. Abraham, H.; Stella, L. Tetrahedron 1992, 48, 9707–9718.
14. Alonso, D. A.; Bertilsson, S. K.; Johnsson, S. Y.; Nordin,
S. J. M.; So¨dergren, M. J.; Anderson, P. G. J. Org.
Chem. 1999, 64, 2276–2280.
318.1705, found 318.1702. 3 1H NMR (500 MHz, CDCl3)
l/ppm 7.41 (d, 2H, J=7.3 Hz), 7.36 (m, 2H), 7.26 (t, 1H,
J=7.2 Hz), 4.50 (q, 1H, J=7.0 Hz), 4.17 (m, 2H), 3.88
(dd, 1H, J=3.4 Hz, 11.0 Hz), 3.84 (m, 1H), 3.76 (dd, 1H,
J=3.8 Hz, 10.7 Hz), 3.54 (dd, 1H, J=5.3 Hz, 10.7 Hz),
3.53 (m, 1H), 3.52 (d, 1H, J=2.6 Hz), 1.92 (m, 1H), 1.82
(m, 1H), 1.76 (m, 1H), 1.57 (m, 1H), 1.50 (m, 1H), 1.33
(d, 3H, J=7.0 Hz), 1.29 (t, 3H, J=7.3 Hz); HRMS
(FAB) calcd for C18H27NO4 (MH+) 322.2018, found
322.2014. 4a 1H NMR (500 MHz, CDCl3) l/ppm 7.47
(m, 2H), 7.34 (m, 2H), 7.26 (m, 1H), 4.48 (q, 1H, J=6.9
Hz), 4.08–4.15 (m, 3H), 3.78 (dd, 1H, J=3.6 Hz, 11.4
Hz), 3.55 (d, 1H, J=9.5 Hz), 3.50 (d, 1H, J=11.4 Hz),
3.36 (dd, 1H, J=6.9 Hz, 9.5 Hz), 3.30 (s, 1H), 2.03 (m,
1H), 1.78–1.88 (m, 3H), 1.49 (m, 1H), 1.30 (d, 3H, J=6.9
Hz), 1.29 (t, 3H, J=6.9 Hz); HRMS (FAB) calcd for
C18H25Br2NO2 (MH+) 446.0330, found 446.0332. 4b 1H
NMR (500 MHz, CDCl3) l/ppm 7.36 (m, 4H), 7.25 (m,
1H), 4.40 (q, 1H, J=7.0 Hz), 4.32 (dd, 1H, J=4.0 Hz,
10.4 Hz), 4.12–4.24 (m, 6H), 3.37 (s, 1H), 2.98 (s, 3H),
2.82 (s, 3H), 2.20 (m, 1H), 1.83 (m, 1H), 1.66 (m, 3H),
1.33 (d, 3H, J=7.0 Hz), 1.30 (t, 3H, J=7.1 Hz); HRMS
(FAB) calcd for C20H31NO8S2 (MH+) 478.1569, found
15. Crystals of 4a were grown from dichloromethane. Crys-
tallographic data (excluding structure factors) for the
structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplemen-
tary publication numbers CCDC 185159. Copies of the
data can be obtained, free of charge, on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK [fax:
+44 (0) 1223-336033 or e-mail: deposit@ccdc.cam.ac.uk].
Ortep 3 for Windows: Farrugia, L. J. J. Appl. Crystallogr.
1997, 30, 565.
1
478.1571. 5 H NMR (500 MHz, CDCl3) l/ppm 4.22 (m,
2H), 3.87 (dd, 1H, J=8.2 Hz, 10.7 Hz), 3.84 (d, 1H,
J=2.5 Hz), 3.70 (dd, 1H, J=5.4 Hz, 10.7 Hz), 3.64 (br,
3H), 3.57 (dd, 1H, J=3.4 Hz, 11.1 Hz), 3.42 (dd, 1H,
J=7.9 Hz, 11.1 Hz), 2.99 (m, 1H), 2.35 (m, 1H), 1.62 (m,
2H), 1.37 (m, 2H), 1.29 (t, 3H, J=7.3 Hz); HRMS (FAB)
calcd for C10H19NO4 (MH+) 218.1392, found 218.1391. 6
1H NMR (500 MHz, CDCl3) l/ppm 7.35–7.39 (m, 5H),
5.38 (s, 1H), 4.12–4.20 (m, 3H), 3.80 (m, 1H), 3.68 (m,
2H), 3.61 (dd, 1H, J=6.4 Hz, 6.6 Hz), 3.39 (d, 1H,
J=1.3 Hz), 2.19 (m, 1H), 1.93 (m, 1H), 1.77 (m, 3H),
1.28 (t, 3H, J=7.3 Hz); HRMS (FAB) calcd for
C17H23NO4 (MH+) 306.1705, found 306.1708.
1
16. Selected analytical data: 2 H NMR (500 MHz, CDCl3):
l/ppm 9.50 (d, 1H, J=1.3 Hz), 8.97 (d, 1H, J=4.7 Hz),
7.33 (m, 5H), 4.19–4.27 (m, 2H), 4.14 (d, 2H, J=2.6 Hz),
4.06–4.11 (m, 2H), 2.62 (m, 1H), 2.21 (m, 1H), 1.72 (m,
1H), 1.5 (m, 2H), 1.37 (d, 3H, J=6.9 Hz), 1.32 (t, 3H,
J=7.3 Hz); HRMS (FAB) calcd for C18H23NO4 (MH+)