Selective synthesis of dehydroamino acids from threonines

Article abstract of DOI:10.1016/j.tetlet.2004.08.047

Dehydropeptides containing dehydroamino acid (ΔAA) are frequently found in natural resources with important biological activity. Herein, we report the selective synthesis of Z- and E-ΔAbu from L- and L-allo-threonine as starting materials through selenati

Full text of DOI:10.1016/j.tetlet.2004.08.047

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 7221–7224
Selective synthesis of dehydroamino acids from threonines
Kazuhiko Nakamura,a,* Tetsuya Isaka,a,b Hiroaki Toshimab and Masato Kodakaa
aNational Institute of Advanced Industrial Science and Technology (AIST), Central 6-9, Higashi 1-1-1,
Tsukuba, Ibaraki 305-8566, Japan
bDepartment of Bioresource Science, College of Agriculture, Ibaraki University, Ami-machi, Inashiki-gun,
Ibaraki 300-0393, Japan
Received 9 July 2004; revised 4 August 2004; accepted 6 August 2004
Available online 24 August 2004
Abstract—Dehydropeptides containing dehydroamino acid (DAA) are frequently found in natural resources with important bio-
logical activity. Herein, we report the selective synthesis of Z- and E-DAbu from L- and L-allo-threonine as starting materials
through selenation and oxidative elimination. The detailed reaction mechanism of phosphine-assisted selenoether formation is also
discussed.
Ó 2004 Elsevier Ltd. All rights reserved.
Dehydropeptides containing a,b-unsaturated amino
acid (or dehydroamino acid, DAA) residues, are fre-
quently found in natural resources with important bio-
logical activity.1,2 Their structures are rigid in both the
backbones and the side chains of the peptides because
of the presence of a double bond conjugated with a pep-
tide linkage. In addition, dehydropeptides are known as
fairly reactive Michael acceptors that react readily with
ÔsoftÕ nucleophiles, such as thiols or amines of biological
molecules. This reactivity is thought to be one of the
molecular mechanisms underlying the biological activi-
ties of dehydropeptides. A number of molecules from
this class were isolated from natural organisms, of which
AM-toxins,3,4 Sch-20561 (microorganisms)5 and kahala-
lide-F (marine organism)6 are known as typical exam-
ples. Biosynthetically, DAAs are derived from
proteinogenic amino acids, such as DAla from serine
and cystein, and a-dehydro-a-aminobutylic (DAbu) acid
from threonine, by activation of the b-hydroxy or thiol
function and posterior b-elimination. Chemical synthe-
ses of such DAAs with eliminative formation of a double
bond have been reported in numerous methods; for
example, activation of b-hydroxy group by tosylation
or other methods, followed by b-elimination;7 permethyl-
ation of a b-amino group, followed by Hofmann
elimination;8 N-chlorination of an a-amino group by
tert-butyl hypochlorite, followed by dehydrochlorina-
tion and isomerization under basic conditions.9 How-
ever, selective synthesis of an E- or Z- unsaturated
bond is rather difficult, though both E- and Z-DAbu
have been found in natural products as dehydroamino
acid components in cyclic depsipeptides.10
Recently, we developed new selenyl linkers for solid-
phase synthesis of dehydropeptides.11,12 Using these
linkers, we achieved the solid-phase synthesis of DAla-
containing dehydropeptides, AM-toxin,11 and RGD-
conjugate peptide12 using a protected serine as a starting
material. The novel linker (1) was designed for side-chain
anchoring, which enabled amino acid elongation to both
the N- and C-terminals of the starting amino acid by
orthogonal protection. The construction of an unstable
double bond could be realized simultaneously with
oxidative cleavage by syn-elimination from the solid
phase in the final stage of synthesis. We reported that
the synthesis of selenoether derived from serine was read-
ily carried out; however, threonine-derived selenoethers
were much more troublesome. In particular, b-seleno-
ether derived from L-threonine (2S,3R) could not be
obtained under the described conditions.12,13
In the present study, selective synthesis of Z- and E-
DAbu was attempted using L- and L-allo-threonine as
starting materials. We carefully re-examined the reac-
tion conditions for selenoether formation and estab-
lished a sufficient condition using pyridine as a solvent
under a cautious temperature control. We found that
it was important to have a low temperature (ꢀ40°C)
Keywords: Dehydroamino acid; Selenyl linker; Solid-phase synthesis;
Reaction mechanism.
*
Corresponding author. Tel.: +81 29 861 6186; fax: +81 29 861
6123; e-mail: kazuhiko.nakamura@aist.go.jp
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.08.047

7222
K. Nakamura et al. / Tetrahedron Letters 45 (2004) 7221–7224
COOtBu
H
H3C
OH
COOtBu
1
FmocNH
COO-Allyl
H
SeCN
Bu3P
Fmoc-L-Thr-OAllyl (2a)
H
H
+ dehydro product
H3C
Se
(4a)
pyridine
H
H3C
OH
FmocNH
COO-Allyl
FmocNH
COO-Allyl
H
3a,b (stereo isomers)
Fmoc-L-allo-Thr-OAllyl (2b)
product / yield*
starting material
condition
byproduct / yield*
-40˚ C / 40 min then r.t. / 90 min
-40˚ C / 20 min then r.t. / 70 min
2a
2b
3a / 29 % (38 %)
3b / 73 % (86 %)
4a / 47 % (62 %)
4a / 12 % (14 %)
* yields in parentheses calculated based on consumed starting material.
Figure 1. Synthesis of selenoether.
for the induction of an active intermediate and that a
much higher temperature (room temperature) was re-
quired to complete the selenium SN2 substitution; that
is, a bi-level temperature control was essential to obtain
the desired products (3a,b).14 In contrast, keeping the
reactants at the lower temperature (ꢀ40°C) or a higher
temperature (0°C or room temperature) led, respec-
tively, to no reaction or to the product of only an un-
desired product, a dehydro derivative. Interestingly,
the same dehydro product (4a) was obtained by b-elim-
ination from two stereochemical isomers, L (2a) and L-
allo (2b) (Fig. 1).
chemistry of precursor selenoether (and selenoxide)
could be also determined, as 5a (erythro) led to Z olefin
4a and 5b (threo) led to E olefin 4b, respectively. In addi-
tion, the undesired products at the selenation reaction
could also be determined to have Z geometry, by com-
parison with both dehydroamino acids.16
Two reaction mechanisms can be assumed for
phosphine-assisted selenoether formation: pentavalent
intramolecular selenation (A) or a bi-molecular mecha-
nism (B), as illustrated in Figure 3. In the former case,
the starting chiral alcohol and the resultant selenoether
should possess the same stereochemistry according to
the SNi reaction mechanism, while in the latter case
Walden inversion should be observed. According to
our experimental results, (threo-)L-threonine derivatives
gave erythro-selenoether, and (erythro-)L-allo-threonine
gave threo-selenoether. These results mean that the
selenation reaction underwent the stereo-inversion at
the b-position. Therefore, we can conclude that the
selenation reaction takes place via the phosphonium
intermediate (mechanism B). It can be understood that
bi-level temperature control is required for induction
of the phosphonium intermediate at a lower temperature
and the following nucleophilic substitution at an
Subsequent oxidative elimination readily proceeded
by treatment with hydrogen peroxide. The two desired
DAbuÕs (4a and 4b)14 with different geometries were
selectively obtained from the respective isomers 3a and
3b through the corresponding selenoxides as solo prod-
ucts. Considering the well-known reaction mechanism,
the selenoxide underwent precise syn-elimination by a
[2,3]-sigmatropic rearrangement.15 Therefore, the geom-
etry of the double bond reflects the relative stereochem-
istry of the corresponding selenoxide. The geometries of
the double bonds were determined by NOE experiments,
depicted in Figure 2. Consequently, the relative stereo-
NOE
COOtBu
H3C
(Z)
O
H H
H2O2 aq. / THF
0 ˚C 90 min then r.t. 3.5 h
COO-Allyl
O
H
N
H
H3C
Se
3a
O
54 %
4a
FmocNH
COO-Allyl
H
5a
COOtBu
NOE
H
CH3
H2O2 aq. / THF
0 ˚C 60 min
H
(E)
3b
H3C
Se
O
H
71 %
FmocNH
O
H
H
O
FmocNH
COO-Allyl
H
4b
5b
Figure 2. Synthesis of dehydroamino acids.

K. Nakamura et al. / Tetrahedron Letters 45 (2004) 7221–7224
7223
ArSe
PBu3
H
H3C
O
COOtBu
FmocNH
COO-Allyl
H
H
pentavalent intramolecular mechanism (A)
H3C
Se
2a
3a
(threo)
(erythro)
OR
H
SN2
FmocNH
COO-Allyl
H
+
H3C
O
P+Bu3
ArSe-
H3C
O
FmocNH
H
4a
elimination
OAllyl
FmocNH
COO-Allyl
phosphonium bi-molecular mechanism (B)
Figure 3. Reaction mechanism of selenoether formation.
9. Shimohigashi, Y.; Stammer, G. H. J. Chem. Soc., Perkin
Trans. 1 1983, 803–808.
elevated temperature. In addition, the dehydroamino
acid as an undesired product can be produced through
its elimination from the six-membered ring intermediate
with the assistance of strong phosphorus–oxygen inter-
action. b-Elimination reaction would proceed through
trans-relationship between the a-proton and the b-oxy-
phosphonium by formation of six-membered ring. The
geometry of the double bond (anti relationship between
the methyl and the ester) was clearly explained by this
reaction model regardless of the relative stereochemistry
at the a and b positions.
10. For example, kahalalide
F contains Z-DAbu and
Sch20561 contains E-DAbu, respectively.
11. Horikawa, E.; Kodaka, M.; Nakahara, Y.; Okuno, H.;
Nakamura, K. Tetrahedron Lett. 2001, 42, 8337–8339.
12. Nakamura, K.; Ohnishi, Y.; Horikawa, E.; Konakahara,
T.; Kodaka, M.; Okuno, H. Tetrahedron Lett. 2003, 44,
5445–5448.
13. Indirect syntheses of b-selenoether of amino acids from
cyclic compounds were reported: Higashibayashi, S.;
Kohno, M.; Goto, T.; Suzuki, K.; Mori, T.; Hashimoto,
K.; Nakata, M. Tetrahedron Lett. 2004, 45, 3707–3712;
Also see: Zhou, H.; vander Donk, W. A. Org. Lett. 2002,
4, 1335–1338; Zhu, Y.; Gieselman, M. D.; Zhou, H.;
Averin, O.; vander Donk, W. A. Org. Biomol. Chem. 2003,
1, 3304–3315.
In summary, we succeeded in the selective synthesis of
Z- and E-DAbu by selecting L- and L-allo-threonine as
starting materials and using selenation and oxidative
elimination processes with a selenyl linker. Since the use-
fulness of this linker for solid-phase synthesis of
dehydropeptides has been demonstrated,12,13 it should
be possible to synthesize natural dehydropeptides
including both E- and Z-DAbu by a solid-phase method.
14. Compound 3a: HRMS calcd for C33H35NO678Se m/z
25
619.1635, found m/z 619.1657; ½aꢁD +3.6 (c 1.0, CHCl3);
IR (film) 1716, 1508, 1303, 1216, 757cmꢀ1 1H NMR
;
(CDCl3) d 1.48 (3H, d, J = 6.8Hz), 1.58 (9H, s), 3.72 (1H,
m), 4.23 (1H, t, J = 7.5Hz), 4.38 (2H, d, J = 7.5Hz), 4.63
(2H, d, J = 5.9Hz, overlapped), 4.67 (1H, m, overlapped),
5.27 (1H, br d, J = 10.5Hz), 5.34 (1H, br d, J = 18.2Hz),
5.53 (1H, br d, J = 8.7Hz, NH), 5.90(1H, m), 7.32
(1H+2H, arom), 7.40(2H, arom), 7.59 (2H, arom), 7.76
(1H+2H, arom), 7.91 (1H, arom), 8.22 (1H, arom).
Compound 3b: HRMS calcd for C33H35NO678Se m/z
Acknowledgements
We would like to thank Professor Tadashi Nakata of
Tokyo University of Science for many helpful
discussions.
24
619.1635, found m/z 619.1641; ½aꢁD +44.3 (c 1.2 CHCl3);
IR (film) 1718, 1508, 1301, 1216, 758cmꢀ1 1H NMR
;
(CDCl3) d 1.54 (3H, d, J = 6.8Hz), 1.59 (9H, s), 3.91 (1H,
m), 4.25 (1H, t, J = 4.7Hz), 4.40(3H, overlapped), 4.68
(1H, dd, J = 3.2, 9.1Hz), 5.18 (1H, br d, J = 10.5Hz), 5.23
(1H, br d, J = 16.9Hz), 5.66 (1H, d, J = 9.1Hz, NH), 5.72
(1H, m), 7.33 (1H+2H, arom), 7.41 (2H, arom), 7.61 (2H,
arom), 7.74 (1H, arom), 7.78 (2H, arom), 7.91 (1H, arom),
8.19 (1H, arom). Compound 4a: colorless needles mp 73–
74°C (hexane–EtOAc); HRMS calcd for C22H21NO4 m/z
363.1468, found m/z 363.1458; IR (film) 1716, 1503, 1272,
References and notes
1. Humphrey, J. M.; Chamberlin, R. Chem. Rev. 1997, 97,
2243–2266.
2. Schmidt, U.; Lieberknecht, A.; Wild, J. Synthesis 1988,
159–172.
3. Okuno, T.; Ishita, Y.; Sawai, K.; Matsumoto, T. Chem.
Lett. 1974, 635–638.
1216, 757cmꢀ1
;
1H NMR (CDCl3) d 1.80(3H, d,
4. Ueno, T.; Nakashima, T.; Hayashi, Y.; Fukami, H. Agric.
Biol. Chem. 1975, 39, 1115–1122.
5. Afonso, A.; Hon, F.; Brambilla, R. J. Antibiot. 1999, 52,
J = 7.2Hz), 4.26 (1H, t, J = 6.9Hz), 4.45 (2H, d,
J = 6.9Hz), 4.68 (2H, d, J = 5.9Hz), 5.26 (1H, dd,
J = 1.4, 10.5Hz), 5.34 (1H, dd, J = 1.4, 16.9Hz), 5.94
(1H, m), 6.30(1H, br s, NH), 6.83 (1H, q, J = 7.2Hz), 7.32
(2H, arom), 7.41 (2H, arom), 7.67 (2H, arom), 7.77 (2H,
arom). Compound 4b: colorless needles mp 113–115°C
(hexane–EtOAc); HRMS calcd for C22H21NO4 m/z
363.1468, found m/z 363.1465; IR (KBr) 1726, 1705,
383–397.
`
´
`
´
6. Lopez-Macia, A.; Jimenez, J. C.; Royo, M.; Giralt, E.;
Albericio, F. J. Am. Chem. Soc. 2001, 123, 11398–11401.
7. Kolasa, T.; Gross, E. Int. J. Pept. Protein Res. 1982, 20,
259–266.
1
8. Nomoto, S.; Sano, A.; Shiba, T. Tetrahedron Lett. 1979,
20, 1049–1052.
1529, 1260, 1191cmꢀ1; H NMR (CDCl3) d 2.10(3H, d,
J = 7.3Hz), 4.25 (1H, t, J = 7.2Hz), 4.42 (2H, d,

7224
K. Nakamura et al. / Tetrahedron Letters 45 (2004) 7221–7224
J = 7.2Hz), 4.76 (2H, d, J = 5.5Hz), 5.30(1H, d, 15. Grieco, P. A.; Gilman, S.; Nishizawa, M. J. Org. Chem.
J = 10.5Hz), 5.38 (1H, d, J = 17.4Hz), 5.98 (1H, m),
6.86 (1H, br s, NH), 6.91 (1H, br m), 7.32 (2H, arom), 7.41
(2H, arom), 7.61(2H, arom), 7.77 (2H, arom).
1976, 41, 1485–1487.
16. E-DAbu isomerized partially into Z-DAbu at room
temperature within a few days.