Enantioselective synthesis of α-alkyl,α-vinyl amino acids via [2,3]-sigmatropic rearrangement of selenimides

Article abstract of DOI:10.1021/ol1030926

Chiral α-alkyl,α-vinyl amino acids (quaternary vinyl glycine derivatives) are prepared with high levels of enantiomeric purity by [2,3]-sigmatropic rearrangement of allylic selenimides. The required trisubstituted allylic selenides are prepared by an orga

Full text of DOI:10.1021/ol1030926

ORGANIC
LETTERS
2011
Vol. 13, No. 5
1040–1043
Enantioselective Synthesis of r-Alkyl,
r-Vinyl Amino Acids via [2,3]-Sigmatropic
Rearrangement of Selenimides
Alan Armstrong* and Daniel P. G. Emmerson
Department of Chemistry, Imperial College London, South Kensington,
London, SW7 2AZ, U.K.
A.Armstrong@imperial.ac.uk
Received December 21, 2010
ABSTRACT
Chiral R-alkyl,R-vinyl amino acids (quaternary vinyl glycine derivatives) are prepared with high levels of enantiomeric purity by [2,3]-sigmatropic
rearrangement of allylic selenimides. The required trisubstituted allylic selenides are prepared by an organocatalytic R-selenenylation of
aldehydes followed by Horner-Wadsworth-Emmons (HWE) olefination. Both (E)-and (Z)-geometrical isomers are available giving access to both
enantiomers of the desired products.
The asymmetric synthesis of quaternary amino acids has
attracted the interest of organic chemists due to the
biological importance of these structures. Although there
is now a wide range of methodology for the synthesis of
these compounds,1 few of these methods are useful for the
asymmetric synthesis of R-alkyl,R-vinyl amino acids.
R-Vinyl amino acids have shown potential as mechanism-
based inhibitors of PLP (pyridoxal phosphate) enzymes.2
Possible mechanisms of this irreversible inhibition
include R-deprotonation, resulting in conjugation of the
double bond giving a Michael acceptor, which reacts with
nucleophilic residues at the enzyme active site, irreversibly
inactivating it. R-Alkyl,R-vinyl amino acids lacking an
R-proton are also interesting targets for PLP-dependent
decarboxylase enzymes,3 in which an analogous ‘vinyl-
trigger’ mechanism can operate; here, an R-decarboxylation
(1) For reviews on the asymmetric synthesis of R,R-disubstituted R-
€
amino acids, see: (a) Vogt, H.; Brase, S. Org. Biomol. Chem. 2007, 5, 406–
430. (b) Cativiela, C.; Diaz-de-Villegas, M. D. Tetrahedron: Asymmetry
1998, 9, 3517–3599. (c) Cativiela, C.; Diaz-de-Villegas, M. D. Tetrahe-
dron: Asymmetry 2000, 11, 645–732. (d) Cativiela, C.; Diaz-de-Villegas,
M. D. Tetrahedron: Asymmetry 2007, 18, 569–623. (e) Cativiela, C.;
Diaz-de-Villegas, M. D. Tetrahedron: Asymmetry 2009, 20, 1–63. (f)
Mosey, R. A.; Fisk, J. S.; Tepe, J. J. Tetrahedron: Asymmetry 2008, 19,
2755–2762. For selected, more recent examples, see:(g) Branca, M.;
Pena, S.; Guillot, R.; Gori, D.; Alezra, V.; Kouklovsky, C. J. Am. Chem.
Soc. 2009, 131, 10711–10718. (h) Green, J. E.; Bender, D. M.; Jackson,
S.; O’Donnell, M. J.; McCarthy, J. R. Org. Lett. 2009, 11, 807–810. (i)
Fu, P.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2008, 130,
5530–5541. (j) Mitani, M.; Tanaka, Y.; Sawada, A.; Misu, A.; Matsumoto,
ꢀ
Y. Eur. J. Org. Chem. 2008, 1383–1391. (k) Cabrera, S.; Reyes, E.; Aleman,
J.; Milelli, A.; Kobbelgaard, S.; Jørgensen, K. A. J. Am. Chem. Soc. 2008,
130, 12031–12037. (l) Huang, X. L.; He, L.; Shao, P. L.; Ye, S. Angew.
Chem., Int. Ed. 2009, 48, 192–195. (m) Uraguchi, D.; Asai, Y.; Ooi, T.
Angew. Chem., Int. Ed. 2009, 48, 733–737. (n) Hayashi, Y.; Obi, K.; Ohta,
Y.; Okamura, D.; Ishikawa, H. Chem.;Asian J. 2009, 4, 246–249. (o)
Dickstein, J. S.; Fennie, M. W.; Norman, A. L.; Paulose, B. J.; Kozlowski,
M. C. J. Am. Chem. Soc. 2008, 130, 15794–15795. (p) Fesko, K.; Uhl, M.;
Steinreiber, J.; Gruber, K.; Griengl, H. Angew. Chem., Int. Ed. 2010,49,121–
124. (q) Konishi, H.; Lam, T. Y.; Malerich, J. P.; Rawal, V. H. Org. Lett.
2010, 12, 2028–2031.
(2) For a review covering synthesis and biochemical aspects of R-
vinyl amino acids, see: Berkowitz, D. B.; Charette, B. D.; Karukurichi,
K. R.; McFadden, J. M. Tetrahedron: Asymmetry 2006, 17, 869–882.
ꢀ
(3) (a) Ribereau-Gayon, G.; Danzin, C.; Palfreyman, M. G.; Aubry,
M.; Wagner, J.; Metcalf, B. W.; Jung, M. J. Biochem. Pharmacol. 1979,
28, 1331–1335. (b) Berkowitz, D. B.; Jahng, W.-J.; Pedersen, M. L.
Bioorg. Med. Chem. Lett. 1996, 6, 2151–2156. (c) Danzin, C.; Casara, P.;
Claverie, N.; Metcalf, B. W. J. Med. Chem. 1981, 24, 16–20.
r
10.1021/ol1030926
Published on Web 01/28/2011
2011 American Chemical Society

It is interesting to note the scarcity of methods relying on
the formation of the C-N bond as the key step.11 Moti-
vated by our longstanding interest in asymmetric carbon-
heteroatom bond formation, we wished to develop an
asymmetric synthesis of R-alkyl,R-vinyl amino acids.
As part of an ongoing research program on sulfur
amination/[2,3]-sigmatropic rearrangement of allylic and
propargylic sulfimides in our laboratatory,12 we recently
reported the synthesis of vinyl glycine derivatives, e.g. 4a
via amination/rearrangement of allylic sulfides (Scheme 1).
This method relies on an asymmetric organocatalytic R-
sulfenylation of aldehydes,13 followed by in situ Horner-
Wadsworth-Emmons (HWE) olefination giving disubsti-
tuted allylic sulfides, e.g. 3a.14 S-Amination is carried out
by treating allylic sulfides 3 with oxaziridine D,12a where-
upon [2,3]-sigmatropic rearrangement of the resulting
sulfimide occurs spontaneously; in situ treatment with
triethyl phosphite results in N-S bond cleavage giving
R-vinyl amino acids, e.g. 4a. However attempts to use this
method for the synthesis of R-alkyl,R-vinyl amino acid 4b
met with difficulties: olefination to give the required
trisubstituted allylic sulfide 3b needed higher temperatures
for full conversion such that partial racemization of the
R-sulfenyl aldehyde 2 was unavoidable (Scheme 1).
Attracted by recent reports of an organocatalytic R-
selenenylation of aldehydes,15 we sought to use selenium in
place of sulfur.16 This has a number of potential advan-
tages over the sulfur system: (a) the relative stability of
R-selenenyl aldehydes with respect to racemization has
been noted,15a and use of these aldehydes might prevent
significant racemization under the conditions of the HWE
reaction at the higher temperature required for full con-
version; (b) the synthesis of R-selenenyl aldehydes can be
conducted using a stable commercially available reagent
N-(phenylseleno)phthalimide (NPSP), avoiding the use of
triazole-derived sulfenylation reagent B which must be
Scheme 1. Vinyl Glycine Synthesis via Organocatalytic
Sulfenylation/HWE Olefination Followed by Amination/
Rearrangement12c
rather than an R-deprotonation brings the double bond into
conjugation.
The choice of currently available methods for the asym-
metric synthesis of R-alkyl,R-vinyl amino acids is domi-
nated by chiral auxiliary-directed methods. For instance,
Marsden and co-workers4 and Berkowitz and co-workers5
have reported decongugative alkylation of dienolates de-
rived from dehydro amino acids. Several other methods
rely on chiral auxiliary-directed stereoselective formation
of the carbon;vinyl bond through addition of organo-
metallic vinyl synthons.6 An alkylation of an oxazoline
derived from L-vinyl glycine is notable in that no chiral
auxiliary is required, chiral information being derived
instead from the starting material.5,7 Other methods typi-
cally rely on asymmetric alkylation using chiral auxiliaries
in combination with multistep installation of a vinyl
group.8 Catalytic methods are much rarer,9 the outstand-
ing example being Shibasaki’s bifunctional gadolinium-
derived catalyst for the Strecker reaction of vinyl
ketimines.10
(11) Palladium-catalyzed [3,3]-sigmatropic rearrangements of tri-
fluoro- or trichloro-acetimidates give protected β-amino alcohols which
may be converted to the corresponding vinyl amino acids by oxidation.
For an example giving racemic quaternary vinyl amino acids, see: (a)
Berkowitz, D. B.; Wu, B.; Li, H. Org. Lett. 2006, 8, 971–974. For an
example giving enantiomerically enriched vinyl amino acids, see:(b)
^
Mehmandoust, M; Petit, Y; Larcheveque, M. Tetrahedron Lett. 1992,
33, 4313–4316.
~
(4) Jones, M. C.; Marsden, S. P.; Munos Subtil, D. M. Org. Lett.
2006, 8, 5509–5512.
(5) Berkowitz, D. B.; Chisowa, E.; McFadden, J. M. Tetrahedron
2001, 57, 6329–6343.
(12) (a) Armstrong, A.; Cooke, R. S. Chem. Commun. 2002, 904–905.
(b) Armstrong, A.; Challinor, L.; Cooke, R. S.; Moir, J. H.; Treweeke,
N. R. J. Org. Chem. 2006, 71, 4028–4030. (c) Armstrong, A.; Challinor,
L.; Moir, J. H. Angew. Chem., Int. Ed. 2007, 46, 5369–5372. (d)
Armstrong, A.; Emmerson, D. P. G. Org. Lett. 2009, 11, 1547–1550.
(13) Marigo, M.; Wabnitz, T. C.; Fielenbach, D.; Jørgensen, K. A.
Angew. Chem., Int. Ed. 2005, 44, 794–797.
(14) An alternative method of preparing disubstituted allylic sulfides
in high ee has recently been reported: Sun, J.; Fu, G. C. J. Am. Chem.
Soc. 2010, 132, 4568–4569.
(6) (a) Crucianelli, M.; De Angelis, F.; Lazzaro, F.; Malpezzi, L.;
Volonterio, A.; Zanda, M. J. Fluorine Chem. 2004, 125, 573–577. (b)
Wipf, P.; Stephenson, C. R. J. Org. Lett. 2003, 5, 2449–2452. (c) Murga,
ꢀ
J.; Portoles, R.; Falomir, E.; Carda, M.; Marco, J. A. Tetrahedron:
ꢀ
Asymmetry 2005, 16, 1807–1816. (d) Portoles, R.; Murga, J.; Falomir,
E.; Carda, M.; Uriel, S.; Marco, J. A. Synlett 2002, 711–714.
(7) Berkowitz, D. B.; McFadden, J. M.; Chisowa, E.; Semerad, C. L.
J. Am. Chem. Soc. 2000, 122, 11031–11032. Pedersen, M. L.; Berkowitz,
D. B. J. Org. Chem. 1993, 58, 6965–6975.
(8) (a) Di Giacomo, M.; Vinci, V.; Serra, M.; Colombo, L. Tetra-
hedron: Asymmetry 2008, 19, 247–257. (b) Ma, D.; Zhu, W. J. Org.
Chem. 2000, 66, 348–350. (c) Avenoza, A.; Cativiela, C.; Corzana, F.;
Peregrina, J. M.; Zurbano, M. M. J. Org. Chem. 1999, 64, 8220–8225. (d)
Colson, P.-J.; Hegedus, L. S. J. Org. Chem. 1993, 58, 5918–5924.
(9) For a general review on organocatalytic formation of quaternary
stereocenters including R-amination of R-branched carbonyl com-
pounds, see: Bella, M.; Gasperi, T. Synthesis 2009, 1583–1614.
(10) Masumoto, S.; Usuda, H.; Suzuki, M.; Kanai, M.; Shibasaki, M.
J. Am. Chem. Soc. 2003, 125, 5634–5635.
(15) (a) Tiecco, M.; Carlone, A.; Sternativo, S.; Marini, F.; Bartoli,
G.; Melchiorre, P. Angew. Chem., Int. Ed. 2007, 46, 6882–6885. (b)
ꢀ
ꢀ
Sunden, H.; Rios, R.; Cordova, A. Tetrahedron Lett. 2007, 48, 7865–
7869.
(16) During the preparation of this manuscript two reports of a
similar strategy to synthesize R-hydroxy and R-chloro-(E)-β,γ-unsatu-
rated esters have appeared. These authors report an alternative method
for the in situ olefination giving disubstituted allylic sulfides: (a) Hess,
L. C.; Posner, G. H. Org. Lett. 2010, 12, 2120–2122. The R-chloro-(E)-
β,γ-unsaturated esters may be converted into vinylogogous R-amino
acids:(b) Genna, D. T.; Hencken, C. P.; Siegler, M. A.; Posner, G. H.
Org. Lett. 2010, 12, 4694–4697.
Org. Lett., Vol. 13, No. 5, 2011
1041

with very limited success: high conversion could not be
achieved even when a 3-fold excess of phosphonate anion
was used. Finally, we found that good yields of the allyl
selenides could be isolated when the organocatalytic reac-
tion was simply filtered prior to the HWE reaction, pre-
sumably removing phthalimide, the byproduct of the
selenenylation. We were pleased to find generally moder-
ate-to-good E/Z selectivities (Table 1, entries 1-5) evenfor
trisubstituted alkenes (R0 = Me). Furthermore, the geo-
metric isomers could easily be separated by column chro-
matography in all cases to give moderate-to-good yields
(38-75%, over two steps) of (E)-6. A more sterically
encumbered substrate proved the most challenging; longer
reaction times and a greater excess of phosphonate anion
were required in the synthesis of (E)-6c, which bears a tert-
butyl side chain (entry 3). Keen to expand the scope of
the asymmetric organocatalytic R-selenenylation to alde-
hydes bearing β-heteroatoms, previously unused in the
R-selenenylation,19 we were pleased to find that the alde-
hyde TBSO(CH2)2CHO was a substrate for the reaction,
albeit requiring further concentration and a prolonged
reaction time (entry 4).
Table 1. Organocatalytic R-Selenenation of Aldehydes/HWE
Olefinationa
entry
R
R0
E/Zb
yield/%c ee/%d
1
(E)-6ae Et
H
>20:1
4.7:1
12:1
75
69
38
54
52
63
41
44
70
60
50
96
95
91
88
94
95
ndh
ndh
97
95
88
2
(E)-6b
(E)-6cf
Et
Me
Me
H
3
t-Bu
4
(E)-6dg TBSOCH2
>20:1
4.6:1
5
(E)-6e
(Z)-6e
(Z)-6f
(Z)-6 g
Allyl
Allyl
i-Pr
Me
Me
Me
Et
6
1:>20
7
1:>20
1:>20
1:18
8
i-Pr
9
(Z)-6 h Bn
Me
Bn
10
11
(Z)-6i
(Z)-6j
Bn
1:15
TBSOCH2 Me
1:>20
Our attempts to prepare trisubstituted alkenes substi-
tuted with alkyl groups other than methyl by this method
were less successful, resulting in low E/Z selectivity and
poor conversion.20 Ando21 reported (Z)-selective HWE
olefinations of R-substituted ethyl (diphenylphosphono)-
acetates, and we hoped that this type of reagent might
allow access to a wider range of trisubstituted allylic sele-
nides (Z)-6.22 Furthermore, we have previously noted12c
that (Z)-geometric isomers of allylic sulfides undergo ami-
nation/rearrangement to give the enantiomer of the vinyl
glycine derivative produced from the (E)-isomer thus mak-
ing either enantiomer of 7 readily available without recourse
to the organocatalyst derived from the un-natural enantio-
mer of proline.
We therefore investigated the synthesis of the (Z)-
alkenes using the Ando-modified phosphonates (PhO)2-
P(O)CHR0CO2Et. Pleasingly, this was successful, giving
the allylic selenides (Z)-6e-j in acceptable yields and,
crucially, high ee’s and Z/E ratios (Table 1, entries
6-11). In contrast to the conventional HWE reagents,
a For full experimental details see Supporting Information. Alde-
hyde (0.4 mmol, 1.3 M), NPSP (0.48 mmol), A (5 mol %), O2NC6-
H4CO2H (5 mol %), 0 °C, 16 h; then phosphonate (0.8 mmol), n-BuLi
(0.76 mmol), CH2Cl2, -78 °C f 0 °C, 3 h. b By 1H NMR analysis of
crude reaction mixture. c Yield over two steps of the purified major
geometric isomer after column chromatography. d Ee determined by
chiral HPLC. e Methyl ester prepared. f 4 equiv of phosphonate anion
added in 3 stages. g Sulfenylation at a concentration of 2 M for 40 h.
h Separation of isomers by chiral hplc was not possible.
prepared immediately before use; (c) the amination/rear-
rangement can be carried out using commercially available
reagents according to the method of Hopkins17 (NCS and
an amide or carbamate: RNH2), potentially giving access
to a choice of nitrogen protecting group; (d) this method
requires no additional step to cleave the N-Se bond: this
weak bond is broken under the reaction conditions.
For our purposes it was crucial that there was no
unreacted aldehyde 1 remaining after the R-selenenylation
reaction as we wished to perform the HWE reaction
in situ.18 Since the established method15a for organocatalytic
R-selenenylation requires an excess of aldehyde, our first
task was to find conditions for this reaction under which
full consumption of aldehyde was possible, while main-
taining high enantioselectivity. Fortunately, we were able
to achieve this goal using 1.2 equiv of NPSP and a higher
reaction concentration (see Supporting Information for
further details).
(19) For a nonselective organocatalytic R-selenenylation of an R,β-
oxygen-substituted ketone, see: Wang, J.; Li, H.; Mei, Y. J.; Lou, B. S.;
Xu, D. G.; Xie, D. Q.; Guo, H.; Wang, W. J. Org. Chem. 2005, 70, 5678–
5687.
(20) For instance, in the HWE olefination to form (E)-6 (R = Bn,
R0 = Et) no conversion of 5 was observed under the conditions used in
the other examples. When 5 (R = Bn) was isolated and subjected to the
HWE conditions, 60% conversion and a 2.6:1 E/Z ratio were observed.
For further details, see Supporting Information.
Next, we needed to develop an in situ olefination which
would not result in racemization. Our initial attempts met
(21) Ando, K. J. Org. Chem. 1998, 63, 8411–8416.
(22) For an example of high (Z)-selectivity in Ando modified olefina-
tions on R-heteroatom-substituted aldehydes to afford trisubstituted
alkenes, see: (a) Wascholowski, V.; Giannis, A. Angew. Chem., Int. Ed.
2006, 45, 827–830. For examples of improved (Z)-selectivity compared
to Still-Gennari modified olefinations on such systems, see:(b) Dias,
L. C.; Meira, P. R. R. J. Org. Chem. 2005, 70, 4762–4773. (c) Vicario,
J. L.; Job, A.; Wolberg, M.; Muller, M.; Enders, D. Org. Lett. 2002, 4,
1023–1026. (d) Enders, D.; Vicario, J. L.; Job, A.; Wolberg, M.; Muller,
M. Chem.;Eur. J. 2002, 8, 4272–4284. (e) Yoshimura, T.; Yakushiji,
F.; Kondo, S.; Wu, X.; Shindo, M.; Shishido, K. Org. Lett. 2006, 8,
475–478.
(17) Shea, R. G.; Fitzner, J. N.; Fankhauser, J. E.; Spaltenstein, A.;
Carpino, P. A.; Peevey, R. M.; Pratt, D. V.; Tenge, B. J.; Hopkins, P. B.
J. Org. Chem. 1986, 51, 5243–5252. These authors include a single
example of an R-vinyl,R-alkyl amino acid, but the enantiomeric purity
was limited by the lack of methods to prepare allylic selenides in high ee
at that time.
(18) R-Selenenyl aldehydes undergo partial racemization during
purification by flash chromatography so a purification-free method is
required.
1042
Org. Lett., Vol. 13, No. 5, 2011

co-workers in the 1980s.17 The reaction has been shown
to proceed via chlorination of selenium followed by dis-
placement of chloride by the nitrogen reagent giving
the allylic selenimide which undergoes [2,3]-sigmatropic
rearrangement.17
Table 2. Amination/Rearrangementa
We initially chose benzyl carbamate as the nitrogen
source, since this would giveversatile Cbz-protectedamino
acid derivatives. Pleasingly, this method afforded R-alkyl,
R-vinyl amino acids 7 in moderate-to-good yields (Table 2).
A key advantage of this method is that the nitrogen source
in this reaction may be varied, giving a choice of protecting
groups. For example, we have used tert-butyl carbamate,
giving Boc-protected amino acids (Table 2, entries 9 and
11) and benzamide, giving a benzoyl-protected amino acid
(entry 3). In the latter case a lower yield (31%) could be
accounted for by observation of the unrearranged seleni-
mide intermediate, which was isolated from the reaction as
a single diastereomer. This supports a hypothesis in which
one diastereomer of the selenimide undergoes rearrange-
ment faster than the other due to a less sterically encum-
bered transition state, in line with a similar observation
by us in a related reaction.11c In all cases, the ee of the
products was at least 90% and was conserved during the
rearrangement, consistent with a concerted mechanism
(Table 2).
Particularly pleasing was the rearrangement of (E)-6d,
which occurred with loss of the TBS group in 57% yield
(Table 2, entry 5). Unfortunately, the rearrangement of
(Z)-6j, the corresponding (Z)-trisubstituted allylic sele-
nide, was unsuccessful (Table 2, entry 13).
In conclusion, we have developed a new, highly enantio-
selective synthesis of R-alkyl,R-vinyl amino acid de-
rivatives via [2,3]-sigmatropic rearrangement of allylic
selenimides. The required trisubstituted allylic selenides
were prepared via an asymmetric, organocatalytic R-selen-
enylation of aldehydes followed by HWE olefination.
This strategy is the first example of a general, asymmetric
synthesis of R-alkyl,R-vinyl amino acid derivatives in
which the C-N bond is formed in the key step. This
versatile method facilitates the introduction of a range of
protecting groups on nitrogen in the resulting quaternary
amino acid derivatives. The use of selenium in this report
shows several key advantages over the corresponding
sulfur-based method.
a For full experimental details see Supporting Information. b After
column chromatography. c Ee determined by chiral hplc. d After recrys-
tallization from Et2O/hexane. e Separation of isomers by chiral HPLC
was not possible. f Determined by chiral HPLC of Cbz protected product
(ent)-7f, prepared by deprotection/Cbz protection of the protuct (ent)-7 g.
g Complex mixture produced; none of the desired product isolated.
the R-ethyl- and R-benzyl-substituted Ando-modified
phosphonates underwent highly selective olefination af-
fording allylic selenides (Z)-6 g,i (Z/E ratios of >20:1 and
15:1 respectively, entries 8 and 10) with little or no race-
mization of the selenides. The aldehyde TBSO(CH2)2CHO
was once again used as a substrate, giving the R-oxygen
substituted selenide (Z)-6j.
Acknowledgment. We thank the EPSRC for support.
Supporting Information Available. Experimental de-
tails, copies of 1H and 13C NMR spectra and HPLC traces
for 6a-j and 7a-l. This material is available free of
charge via the Internet at http://pubs.acs.org.
We now investigated the NCS-mediated amination/
rearrangement using conditions reported by Hopkins and
Org. Lett., Vol. 13, No. 5, 2011
1043