Non Lewis acid catalysed epoxide ring opening with amino acid esters

Article abstract of DOI:10.1039/b902081k

The ring opening of epoxides by various amino acid esters is described in refluxing trifluoroethanol without any catalyst. Under these simple conditions the corresponding β-amino alcohols are obtained in good to excellent yields in relatively short reacti

Full text of DOI:10.1039/b902081k

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Non Lewis acid catalysed epoxide ring opening with amino acid esters†
Christine Philippe, Thierry Milcent, Benoit Crousse* and Danie`le Bonnet-Delpon
Received 3rd February 2009, Accepted 19th March 2009
First published as an Advance Article on the web 26th March 2009
DOI: 10.1039/b902081k
Table 1 Ring opening of mono and disubstituted epoxides with L-H-Val-
OMe in protic solvents
The ring opening of epoxides by various amino acid esters is
described in refluxing trifluoroethanol without any catalyst.
Under these simple conditions the corresponding b-amino
alcohols are obtained in good to excellent yields in rela-
tively short reaction times compared to previously reported
methods.
The direct ring opening of epoxides with amino acids should
be the most straightforward synthetic route to functionalised
peptidomimetics, in particular to hydroxyethylamine peptide
isosteres which are widely exploited in the design of potent
protease inhibitors.1 However, this reaction is not very efficient due
to the low nucleophilicity of amino acids. According to the scale
of nucleophilicity reported by Mayr et al.,2 the reactivity of amino
acids towards epoxides is expected to be comparable to that of
aromatic amines and thus to require an activator. Indeed, without
any catalyst, b-amino alcohols were reported to be obtained in
very low yields (8–14%).3 To improve the reaction, the use of
Lewis acids (Al2O3,4 Yb(OTf)3 or LiClO4 ), excess of reactants,
or other activation such as microwave irradiation is required.7
Despite these successful examples, drawbacks such as prolonged
reaction times and moderate yields still remain, depending on
the substrate. This indicates that there are no efficient general
conditions for the success of the reaction. To our knowledge,
only one preparative systematic study involving substoichiometric
quantities of Ca(OTf)2 in hot acetonitrile has been reported with
terminal oxiranes.8
Recently, efficient ring opening of oxiranes with amines has been
reported in protic solvents without any catalyst. For example,
aliphatic amines react easily in water9 while aromatic amines
have been proved to be more reactive with oxiranes in fluoro
alcohols as solvents.10 In connection with this, we now report the
successful opening of epoxides with a variety of amino acids in
trifluoroethanol.
First of all, the evaluation of the effect of various protic solvents
has been performed with the phenoxymethyl oxirane 1a and the
L-valine methyl ester.‡ The reaction was conducted at reflux
with 2 eq. of valine and prolonged until disappearance of the
epoxide. Performing the reaction in water did not result in the
formation of the b-amino alcohol 2a (Table 1, entry 1). The
same reaction was then carried out in ethanol and methanol: the
desired compound was obtained in good yields after around 2 h
as a 1:1 mixture of two diastereoisomers (Table 1, entries 2 and
3). Our efforts were then focused on fluoro alcohols. As strong
Entry Epoxide 1
Solvent
H2O
Time
—
Product (yielda [%])
1
2a (0)
2
3
4
5
6
7
8
EtOH
MeOH
HFIP
TFE
TFE (r.t.)
TFE
2 h
2.5 h
10 min (77)
10 min (92)
18 h
2 hb
1 h
(97)
(89)
(87)
(73)
2b (82)
TFE
5
6
9
10
EtOH
TFE
5.5 h
3 h
(85)
2c (96)
11
12
EtOH
TFE
24 h
18 h
(88)
2d (64)
13
14
15
EtOH
HFIP
iPrOH
48 h
18 h
48 h
(36)
(60)
(14)
a Isolated yields, 1:1 mixture of 2 diastereoisomers. b Reaction carried out
with 1 eq. of amino acid.
H-bond donors, the latter can be activators of oxirane opening
with weak nucleophiles10 while with strong nucleophiles, they form
complexes,11 thus inhibiting the reaction. The ring opening of 1a
in refluxing hexafluoroisopropanol (HFIP, b.p. = 58 ◦C) resulted
in the formation of the b-amino alcohol 2a isolated in good
yield (77%) after only 10 min (Table 1, entry 4). A similar result
was obtained in refluxing trifluoroethanol (TFE, b.p. = 78 ◦C),
with a higher yield (92%). The reaction time was considerably
increased when the reaction was performed at room temperature
(Table 1, entry 6). The reaction could also be achieved at reflux
with only 1 eq. of amino acid (Table 1, entry 7) but with a lower
isolated yield (73%). In all solvents, a complete regioselectivity
was observed, leading to the b-amino alcohol corresponding
to the nucleophilic attack at the less hindered carbon of the
epoxide.
Laboratoire BioCIS-CNRS, Faculte´ de Pharmacie, Univ. Paris-Sud, rue J. B.
Cle´ment, F-92296 Chaˆtenay-Malabry, France. E-mail: benoit.crousse@
u-psud.fr; Fax: +33(0)146835740; Tel: +33(0)146835739
† Electronic supplementary information (ESI) available: Experimental
procedure and characterization data (1H, 13C NMR and MS) of the
products 2a–12b. See DOI: 10.1039/b902081k
From these results, clearly amino acids are not so reactive as
aliphatic amines since the reaction failed in water. Nevertheless
amino acids exhibit a better reactivity than aromatic amines that
2026 | Org. Biomol. Chem., 2009, 7, 2026–2028
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
usually hardly react with oxiranes at refluxing EtOH and TFE.10a
The latter appeared to be the best solvent for the reaction, in
combined terms of reaction time and yield. Furthermore, due
to its poor nucleophilicity, TFE is not prone to be involved
in a transesterification reaction. This has been demonstrated
by performing the reaction with the L-valine benzyl ester: no
transesterification occurred in refluxing TFE whereas 10% of
transesterification was observed after 3 h in ethanol at reflux
(Scheme 1).
Table 2 Reactions of epoxides with various amino acids at refluxing TFE
Product
Entry Epoxide 1
Amino acid
Time
(yield [%])a
1
L-H-Val-OMe
10 min 2a (92)
2
3
4
5
6
7
L-H-Val-OBn
L-H-Tyr(Bn)-OMe
L-H-Leu-OtBu
L-H-Phe-OMe
H-Gly-OEt
1.5 h
1.5 h
3a (92)
4a (96)
10 min 5a (92)
45 min 6a (85)
24 hb
1 h
7a (54)
2b (82)
L-H-Val-OMe
Scheme 1 Ring opening with L-H-Val-OBn.
The efficiency of these conditions (2 eq. of amino acid in reflux-
ing TFE) was then evaluated with different epoxides: the phthal-
imide N-protected epoxide 1b for an access to peptidomimetics,
the decyl epoxide 1c as a lipophilic substrate and the cyclopentyl
epoxide 1d for an example of 1,2-disubstituted epoxide. With 1b
and 1c, corresponding b-amino alcohols 2b and 2c were obtained
in high yield after 1 h and 3 h respectively (Table 1, entries 8
and 10). With the disubstituted epoxide 1d, the reaction was more
difficult and required 18 h to yield 64% of 2d (entry 12). Taking into
account our previous results on facile disubstituted epoxides ring
opening with anilines in HFIP,10a,12 the reaction was carried out in
refluxing HFIP (Table 1, entry 14). Surprisingly, no improvement
was observed.
In all cases, the beneficial effect of TFE compared to EtOH was
again observed13 and is particularly crucial for the disubstituted
epoxide 1d (only 36% yield in EtOH after 48 h). To our knowledge
the ring opening of 1,2-disubstituted oxiranes with amino acids
had not been previously reported except when an activated
substituent was present.6a
8
9
10
11
12
H-Gly-OEt
L-H-Phe-OBn
L-H-Ile-OBn
L-H-Asp(OMe)-OMe 3.5 h
L-H-Val-OMe
1 h
2 h
1.5 h
7b (52)
8b (72)
9b (83)
10b (68)
2c (96)
3 h
13
14
15
L-H-Val-OBn
L-H-Leu-OtBu
L-H-Val-OMe
4 h
2.5 h
18 h
3c (78)
5c (90)
2d (64)
16
17
18
L-H-Val-OBn
L-H-Ile-OBn
L-H-Val-OMe
18 h
18 h
3d (54)
9d (79)
45 min 2e (80)c
a Isolated yields, 1:1 mixture of 2 diastereoisomers. b Carried out at room
temperature. c Only one diastereoisomer.
The reaction was then extended to various amino acids by
choosing TFE as solvent. In all cases, 2 eq. of enantiopure
C-protected amino acids were used. The results are summarized
in Table 2.
With 1a–c, the reaction was successful whatever the side chain
and the protection of amino acids were, except with glycine.
In most cases, reactions were complete within less than 2 h.
Functionalised amino acids such as protected tyrosine or aspartic
acid were also reactive under these uncatalysed conditions (entries
3 and 11). Interestingly, the reaction also gave rise to corresponding
b-amino alcohols from the disubstituted epoxide 1d albeit in a
lower yield. When long reaction times were needed, no transester-
ification was observed using the benzyl, t-butyl, methyl or ethyl
ester of the amino acids. No racemization occurred by treating the
enantiopure epoxide 1e with a chiral amino acid, since in this case
only one product was formed (Table 2, entry 18).
As an example for an access to more elaborated peptidomimet-
ics, the ring opening of 1a and 1b was also investigated with
the dipeptide 11 in refluxing TFE, and led to the corresponding
b-peptidyl alcohol derivatives 12a,b in good yields and within
reasonable reaction times (Scheme 2).
Scheme 2 Reaction of epoxides 1a and 1b with the dipeptide 11.
We have demonstrated that trifluoroethanol is a highly efficient
solvent for the direct epoxide ring opening with amino acids,
leading to the corresponding b-amino alcohols in high yields
and total regioselectivity. The reaction could be generalised to
various amino acids and structural patterns of epoxides, including
1,2-disubstituted ones. Our results clearly show the advantageous
effect of TFE. Compared with previously reported methods, sig-
nificant improvements were achieved: simple reaction conditions,
no need for catalyst or Lewis acid, reasonable reaction times,
increased yields and no by-product.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2026–2028 | 2027
©

View Article Online
These conditions are very attractive for the synthesis of hydrox-
yethylamine scaffolds which are widely used as transition-state
protease inhibitors.
Howes, I. Hussain, P. Jeffrey, G. Maile, R. Matico, J. Mosley, A. Naylor,
A. Brien, S. Redshaw, P. Rowland, V. Soleil, K. J. Smith, S. Sweitzer, P.
Theobald, D. Vesey, D. S. Walter and G. Wayne, J. Med. Chem., 2008,
51, 3313; (i) J. N. Freskos, Y. M. Fobian, T. E. Benson, M. J. Bienkowski,
D. L. Brown, T. L. Emmons, R. Heintz, A. Laborde, J. J. McDonald,
B. V. Mischke, J. M. Molyneaux, J. B. Moon, P. B. Mullins, D. B.
Prince, D. J. Paddock, A. G. Tomasselli and G. Winterrowd, Bioorg.
Med. Chem. Lett., 2007, 17, 73; (j) J. N. Freskos, Y. M. Fobian, T. E.
Benson, J. B. Moon, M. J. Bienkowski, D. L. Brown, T. L. Emmons, R.
Heintz, A. Laborde, J. J. McDonald, B. V. Mischke, J. M. Molyneaux,
P. B. Mullins, D. B. Prince, D. J. Paddock, A. G. Tomasselli and G.
Winterrowd, Bioorg. Med. Chem. Lett., 2007, 17, 78; (k) For elastase
inhibitors see: J.-P. Be´gue´, D. Bonnet-Delpon, N. Fischer-Durand, A.
Amour and M. Reboud-Ravaux, Tetrahedron: Asymmetry, 1994, 5,
1099.
2 (a) F. Brotzel, Y. C. Chu and H. Mayr, J. Org. Chem., 2007, 72, 3679;
(b) F. Brotzel and H. Mayr, Org. Biomol. Chem., 2007, 5, 3814.
3 P. W. Erhardt, C. M. Woo, R. J. Gorczynski and W. G. Anderson,
J. Med. Chem., 1982, 25, 1402.
4 N. Vaiana, L. Rizzi and S. Romeo, Chem. Lett., 2007, 36,
648.
5 O. Monasson, M. Ginisty, G. Bertho, C. Gravier-Pelletier and Y. Le
Merrer, Tetrahedron Lett., 2007, 48, 8149.
6 (a) K. C. Nicolaou, M. Zak, B. S. Safina, S. H. Lee and A. A. Estrada,
Angew. Chem. Int. Ed., 2004, 43, 5092; (b) S. Rodriguez-Escrich, D.
Popa, C. Jimeno, A. Vidal-Ferran and M. A. Pericas, Org. Lett., 2005,
7, 3829; (c) S. Higashibayashi, T. Mori, K. Shinko, K. Hashimoto and
M. Nakata, Heterocycles, 2002, 57, 111.
7 (a) B. O. Elemine, R. a. Besbes and M. R. Ennigrou, Synth. Commun.,
2007, 37, 3989; (b) H. L. Sham, D. A. Betebenner, W. Rosenbrook, T.
Herrin, A. Saldivar, S. Vasavanonda, J. J. Plattner and D. W. Norbeck,
Bioorg. Med. Chem. Lett., 2004, 14, 2643; (c) T. Tashiro, S. Fushiya and
S. Nozoe, Chem. Pharm. Bull., 1988, 36, 893.
8 (a) M. Sova, A. Babic, S. Pecar and S. Gobec, Tetrahedron, 2007, 63,
141; (b) A. Babic, M. Sova, S. Gobec and S. Pecar, Tetrahedron Lett.,
2006, 47, 1733.
Acknowledgements
We thank the French MENRT for awarding a research fellowship
to C. Philippe, Central Glass Co., Ltd. for the generous gift
of HFIP and A. Solgadi for performing mass spectra analysis
(SAMM platform, Chaˆtenay-Malabry).
Notes and references
‡ A typical procedure is as follows (Table 1, entry 5): L-H-Val-OMe·HCl
salt (1.5 mmol, 0.251 g) and K2CO3 (2.5 mmol, 0.345 g) were suspended
in water (3 mL). The free amino acid was extracted with diethyl ether
(3 ¥ 15 mL). The organic phase was then dried with MgSO4 and
concentrated under reduced pressure at ambient temperature. The free
amino acid (0.76 mmol, 0.099 g) was immediately diluted in 1.25 mL of
trifluoroethanol. Then, epoxide 1a (0.38 mmol, 0.058 g) was added. The
reaction mixture was stirred at reflux until the disappearance of the epoxide
(monitored by TLC). After 10 min of heating, the reaction medium was
concentrated under reduced pressure and the resulting oil was then purified
by chromatography on silica gel (cyclohexane/AcOEt : 8/2). The product
2a (0.098 g, 92%) was obtained as a colourless oil, as a mixture of two
diastereoisomers in a 1:1 ratio.
1 (a) For HIV-1 protease inhibitors see: J. Dohnalek, J. Hasek, J. Duskova,
H. Petrokova, M. Hradilek, M. Soucek, J. Konvalinka, J. Brynda, J.
Sedlacek and M. Fabry, J. Med. Chem., 2002, 45, 1432; (b) T. J. Tucker,
W. C. Lumma, L. S. Payne, J. M. Wai, S. J. Desolms, E. A. Giuliani, P. L.
Darke, J. C. Heimbach, J. A. Zugay, W. A. Schleif, J. C. Quintero, E. A.
Emini, J. R. Huff and P. S. Anderson, J. Med. Chem., 1992, 35, 2525;
(c) J. A. Martin, Antiviral Res., 1992, 17, 265; (d) For metalloprotease
inhibitors see: C. S. Thorsten Oost, M. Brewer, M. Goodnough, W.
Tepp, E. A. Johnson and D. H. Rich, Peptide Science, 2003, 71, 602;
(e) For plasmepsine inhibitors see: D. Muthas, D. Noteberg, Y. A.
Sabnis, E. Hamelink, L. Vrang, B. Samuelsson, A. Karlen and A.
Hallberg, Bioorg. Med. Chem., 2005, 13, 5371; (f) D. Noteberg, E.
Hamelink, J. Hulten, M. Wahlgren, L. Vrang, B. Samuelsson and A.
Hallberg, J. Med. Chem., 2003, 46, 734; (g) For cathepsin D inhibitors
see: R. M. McConnell, A. W. Green, C. J. Trana, M. S. McConnell,
J. F. Lindley, K. Sayyar, W. E. Godwin and S. E. Hatfield, Med. Chem.,
2006, 2, 27; (h) For b-secretase inhibitors see: N. Charrier, B. Clarke, L.
Cutler, E. Demont, C. Dingwall, R. Dunsdon, P. East, J. Hawkins, C.
9 N. Azizi and M. R. Saidi, Org. Lett., 2005, 7, 3649.
10 (a) U. Das, B. Crousse, V. Kesavan, D. Bonnet-Delpon and J. P. Begue,
´
´
J. Org. Chem., 2000, 65, 6749; (b) J. P. Begue, D. Bonnet-Delpon and
´
´
B. Crousse, Synlett, 2004, 18.
11 J. F. Berrien, M. Ourevitch, G. Morgant, N. E. Ghermani, B. Crousse
´
and D. Bonnet-Delpon, J. Fluorine Chem., 2007, 128, 839.
12 Reaction of 1.1 eq. of aniline with cyclopentene oxide in refluxing HFIP
afforded after 9 hours the corresponding b-amino alcohol in 88% yield.
13 For another comparison between EtOH and TFE, the reaction was
performed in refluxing EtOH and was quenched after the time required
for completion in refluxing TFE. Product, 2a was isolated in 34% after
10 min; product 2b in 56% after 1 hour; product 2c in 39% after 3 hours.
2028 | Org. Biomol. Chem., 2009, 7, 2026–2028
This journal is
The Royal Society of Chemistry 2009
©