Theoretical evidence for pyramidalized bicyclic serine enolates in highly diastereoselective alkylations

Article abstract of DOI:10.1002/chem.200601746

A new chiral serine equivalent and its enantiomer have been synthesized from (S)- and (R)-N-Bocserine methyl esters (Boc: tert-butyloxy-carbonyl). The use of these compounds as chiral building blocks has been demonstrated in the synthesis of α-alkyl α-amino acids by diastereoselective po tassium enolate alkylation reactions and subsequent acid hydrolyses. Theoretical studies were performed to eluci date the stereochemical outcome of both the formation of five-membered cyclic N,O-acetals and the subsequent alkylation process, which occurs with total retention of configuration. This feature could be explained in terms of the high degree of pyramidalization of enolate intermediates.

Full text of DOI:10.1002/chem.200601746

DOI: 10.1002/chem.200601746
Theoretical Evidence for Pyramidalized Bicyclic Serine Enolates in Highly
Diastereoselective Alkylations
Carlos Aydillo, Gonzalo JimØnez-OsØs, Jesffls H. Busto, JesfflsM. Peregrina,
María M. Zurbano,* and Alberto Avenoza*[a]
Abstract: A new chiral serine equiva-
lent and its enantiomer have been syn-
thesized from (S)- and (R)-N-Boc-
serine methyl esters (Boc: tert-butyloxy-
carbonyl). The use of these compounds
as chiral building blocks has been dem-
onstrated in the synthesis of a-alkyl a-
amino acids by diastereoselective po-
tassium enolate alkylation reactions
and subsequent acid hydrolyses. Theo-
retical studies were performed to eluci-
date the stereochemical outcome of
both the formation of five-membered
cyclic N,O-acetals and the subsequent
alkylation process, which occurs with
total retention of configuration. This
feature could be explained in terms of
the high degree of pyramidalization of
enolate intermediates.
Keywords: ab initio calculations ·
alkylation · amino acids · asymmet-
ric synthesis · carbanions
Introduction
synthesis.[2] Diastereoselective methylation of an N,O-acetal
serine derivative incorporating a new stereogenic center at
the acetal carbon atom has recently been described.[2g] How-
ever, due to the propensity of these materials to polymerize,
racemize, and form hydrates, the compounds must be freshly
prepared, and their storage is not recommended. An alter-
native approach to prepare stable and readily available de-
rivatives is therefore needed to overcome these problems,
particularly in terms of scale-up. With this aim in mind, we
envisioned the design of other, more stable, chiral, five-
membered cyclic N,O-acetal serine derivatives by incorpo-
rating a new ring in the structure; it was envisioned that this
ring could also favor diastereoselectivity in subsequent alky-
lations by means of increased rigidity.
In the course of our studies towards synthetic routes for
a,a-disubstituted a-amino acids based on the use of five-
membered cyclic N,O-acetals,[1] we became interested in dia-
stereoselective alkylations of chiral serine equivalents
(Scheme 1).
Scheme 1. Five-membered cyclic N,O-acetals as chiral building blocks.
Results and Discussion
In this context, five-membered cyclic N,O-acetals derived
from l-serine or l-serinal (i.e., Garner aldehyde) have been
extensively used as three-carbon building blocks in organic
Synthesis of bicyclic N,O-acetal serine derivatives as chiral
building blocks: We first treated 2,3-dimethoxybutadiene in
the presence of a catalytic amount of triphenylphosphine hy-
drobromide[3] with commercially available N-Boc-l-serine
methyl ester 1 at room temperature. Building block 2 was
obtained in 11% yield as the only product from a complex
reaction mixture (Scheme 2). In an effort to increase this
yield, we investigated the reaction of 2,2,3,3-tetramethoxy-
butane[4] with protected serine 1 in the presence of p-tolue-
nesulfonic acid in refluxing toluene. This again gave chiral
bicyclic N,O-acetal 2, but in this case the yield, on a multi-
gram scale, was 75% under the same conditions. This com-
[a] C. Aydillo, G. JimØnez-OsØs, Dr. J. H. Busto, Dr. J.M. Peregrina,
Dr. M. M. Zurbano, Dr. A. Avenoza
Departamento de Química
Universidad de La Rioja, UA-CSIC
26006 LogroÇo (Spain)
Fax: ( +34)941-299-655
E-mail: marimar.zurbano@unirioja.es
alberto.avenoza@unirioja.es
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
4840
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4840 – 4848

FULL PAPER
Scheme 2. Synthesis of chiral bicyclic serine equivalents 2 and 4. Re-
agents and conditions: a) 2,2,3,3-Tetramethoxybutane, p-TsOH·H2O, tolu-
ene, reflux.
pound was obtained as a single diastereomer, as determined
by H NMR spectroscopy (d.r.>20:1).[5] This bicyclic struc-
ture was assessed by complete NMR analysis, including
NOE experiments.
1
Scheme 3. Proposed mechanism for the formation of building block 2.
Starting from N-Boc-d-serine methyl ester 3, this protocol
allows the preparation of bicyclic N,O-acetal 4, which is the
enantiomer of 2, as demonstrated by NMR spectroscopy
and comparison of the optical activities of the two com-
pounds (see the Supporting Information). This finding is not
trivial and implies complete control of stereoselectivity
throughout the whole multistep process. It is noteworthy
that the application of a similar synthetic methodology to
other alkoxycarbonylamino alcohols[2f,6] proceeds via the
formation of six-membered cyclic butane-2,3-diacetals.
ble participation of the Boc group can be regarded as the
driving force for the global process. A significant piece of
experimental evidence that supports this mechanism is that
when the same reaction was carried out with benzyloxycar-
bonyl (Cbz)-protected methyl serinate instead of Boc-pro-
tected methyl serinate, bicyclic N,O-acetals were not isolat-
ed from the rather complexmitxure obtained under the
same conditions.
Generation of 2 and 4 with three stereogenic centers as
single diastereomers necessarily involves a high level of dia-
stereocontrol in all reaction steps. We observed that the con-
figuration of the starting material governs the configuration
of the final product. In this sense, when the enantiomer of
the reacting serinate 1 (serinate 3) was employed, the abso-
lute configuration of all stereocenters was inverted in the
final product 4, as mentioned above.
Mechanism of formation of building block 2: The proposed
mechanism for this reaction is depicted in Scheme 3 and in-
volves the typical acid-catalyzed formation of N,O-acetals A
starting from protected amino alcohols and acetalic com-
pounds[2] as the first steps.
The final steps of the reaction are promoted by the well-
known nucleophilic character of the tert-butyloxycarbonyl
(Boc) group,[7,1c,e,f] which is able to trap the intermediate
species generated by expelling the tert-butyl fragment, lead-
ing to stable bicyclic 2. Considering the large number of
equilibria involved in this reaction, we believe that irreversi-
To gain insight into this synthetic route, we carried out
theoretical calculations employing DFT methods (see Com-
putational Details). Due to the greater complexity of model-
ing the initial reaction steps, we centered our study on dia-
stereoselective ring closure to give the five-membered N,O-
acetalic species, promoted by the action of the Boc group.
The reactant in this final stage was considered to be
mainly A on account of its greater thermodynamic stability
compared to that of its epimer at the N,O-acetalic carbon
atom (A_ep, Figure 1). All calculated conformers of A are
about 1 kcalmolÀ1 lower in Gibbs free energy, which is in ac-
cordance with the diastereomeric ratio previously reported
for the reaction of 1 with pivalaldehyde (3:1).[8] This situa-
tion is probably due to the more stable trans disposition of
bulky groups in the five-membered ring.
Several approaches to the intramolecular attack of the
Boc group on the remaining acetalic region were investigat-
ed, including the pure carbocationic form and others with
pseudo-concerted loss of methanol. Unfortunately, none of
the transition structures (TS) involved in these pathways
could be located by the theoretical methods used in this
work, and consequently “kinetic” information for this pro-
Abstract in Spanish: Se ha sintetizado un nuevo equivalente
de serina quiral y su enantiómero a partir de los Østeres metí-
licos de (S)-o ( R)-N-Boc-serina. Se ha demostrado su uso
como excelentes building blocks quirales mediante la síntesis
de varios a-alquil-a-aminoµcidos, empleando reacciones de
alquilación diastereoselectivas con enolatos de potasio y pos-
terior hidrólisis µcida. Se han realizado estudios teóricos para
elucidar el comportamiento estereoquímico tanto de la for-
mación de los N,O-acetales cíclicos de cinco miembros como
del posterior proceso de alquilación, transcurriendo Øste con
total retención de configuración. Esta característica podría
explicarse atendiendo a una importante piramidalización en
los enolatos intermedios.
Chem. Eur. J. 2007, 13, 4840 – 4848
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4841

A. Avenoza, M. M. Zurbano et al.
pounds under standard conditions. Treatment of 2 with
methyl triflate (MeOTf) at À788C in THF as solvent, fol-
lowed by the addition of potassium hexamethyldisilazide
(KHMDS), produced the methylated derivative as a single
diastereomer (d.r.>20:1).[5]
The reaction proceeds with high yield in a short time
(5 min). For other less reactive alkyl groups or worse leaving
groups (Table 1), the addition of four equivalents of hexa-
Table 1. Diastereoselective alkylation of building block 2.
Entry RX
HMPA
(equiv)
Conversion[a] 5a–d/ Product
6
(yield [%])[b]
1
2
3
4
5
6
MeOTf
MeI
EtOTf
BnI
AllylI
none
0
4
4
4
4
0
100
100
80
100
100
100
57/10 5a[c] (85)
10/10 5a[c] (50)
70/10 5b[d] (70)
4.3/10 5c[c] (30)
23/10 5d[d] (70)
6[c] (95)
Figure 1. Calculated [B3LYP/6-31+G(d)] minimum-energy geometries of
intermediate A, its epimer at the N,O-acetalic carbon atom A_ep, bicycle
2 and its epimer at the O,O-acetalic carbon atom 2_ep. Relative energies
are given in kcalmolÀ1 and distances in Š.
–
[a] Determined by integration of the CH2 signal in the 1H NMR spectra.
[b] Yield after column chromatography. [c] Stereochemistry predicted by
NOE. [d] Stereochemistry determined by X-ray analysis.
cess could not be obtained. This problem is due to the ex-
treme flatness of the potential-energy surface (PES) in the
neighbourhood of the TS, which precludes convergence of
the desired structures. Therefore, only the thermodynamic
stabilities of the products could be evaluated. Several con-
formers of the bicycle 2 and its epimer 2_ep were calculated
(see the Supporting Information and Figure 1). The struc-
tures with the experimentally observed distribution of ste-
reocenters (Re attack) were found to be significantly more
stable (by ca. 3–4 kcalmolÀ1) than their epimers at the O,O-
acetalic carbon atom, and this led to the exclusive formation
of compound 2. This feature is mainly due to the lower
steric congestion in the trans disposition of the acetalic me-
thoxyl group and the N,O-acetalic oxygen atom in 2 with re-
spect to its epimer 2_ep. Therefore, in the absence of more
helpful experimental information, we propose thermody-
namic control along the whole
C
reaction profile, which leads to
the most stable compound at
the end of the global process.
Diastereoselective alkylation
of building block 2: Given the
easy access to chiral serine-de-
rived building blocks 2 and 4,
and with the aim of assessing
their use as precursors for the
asymmetric synthesis of a-alkyl
a-amino acids, we attempted
the alkylation of these com-
Figure 2. X-ray diffraction structures of the alkylation products 5b and 5d.
4842
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4840 – 4848

Pyramidalized Bicyclic Serine Enolates
FULL PAPER
alkylation reaction involves a process with retention of con-
figuration. To gain insight into the origin of this feature, we
carried out a thorough theoretical study of all reaction chan-
nels accessible from the enolate of bicycle 2 (see Computa-
tional Details). The initial pathways to be evaluated were C-
alkylation (5a), O-alkylation (although experimentally this
was not detected) and retro-O-Michael reaction (6). Bromo-
methane was chosen as the alkylating agent in calculating
the aforementioned pathways. The choice of this electro-
phile allowed us to study thoroughly the complexreaction
scenario using a good level of theory while maintaining a
good approximation to the real system. Additionally, the
charged nature of most of the analyzed species forced us to
include solvent effects in the estimation of energies in order
to better reproduce the experimental conditions (see Com-
putational Details).
The acceptability of excluding the cation in all calcula-
tions was demonstrated by comparing the structures of
“naked” and potassium enolates (2’ and 2’·K, respectively).
In accordance with the well-known weakly coordinating
character of potassium, the geometries displayed in Figure 3
both with and without the cation are very similar, and there-
fore the absence of potassium, which in turn simplifies the
calculations, is justified.
strong evidence pointing to retention of configuration of the
stereocenter after deprotonation. This phenomenon is seen
to a greater extent among other carbanionic species,[13] but
very little theoretical evidence for pyramidalized enolates
has been reported to date.[14] Moreover, to the best of our
knowledge, all reported calculations concerning the stereo-
selective alkylation of enolates refer to cyclic or bicyclic lac-
tams (diazepine-2-ones,[15] pyrrolidinones,[16] oxazolopiperi-
dones[17]) or lactones,[18] in which the enolate moiety is endo-
cyclic; in our case the enolate is exocyclic.
This theoretically found pyramidalization is supported by
previous X-ray crystal structures of two imidazolidinone-de-
rived silyl enol ethers.[12] These cyclic pyramidalized struc-
tures observed in the solid state retain the configuration of
the starting imidazolidinones, which could be correlated
with the high stereoselectivity achieved in the alkylation re-
actions.[12]
Unfortunately, the existence of the aforementioned com-
petitive retro-O-Michael reaction, even at very short reac-
tion times, made it impossible to experimentally measure
the stereoselectivity of enolate protonation, at least in an ac-
ceptable way. This could have been a powerful tool to evalu-
ate the energy barrier of pyramidal inversion.
It was possible, however, to isolate a small amount of
starting bicycle 2 when the retro-O-Michael reaction was
not fully complete (1 min). The properties of this compound
(NMR, optical activity) were found to be identical to those
previously measured prior to the reaction. Moreover, on
quenching this test reaction with saturated [D4]ammonium
chloride, the a-deuterated analogue of 2 could be isolated as
a single diastereomer (d.r.>20:1)[5] with the same starting
configuration at all stereocenters, as demonstrated by NMR
spectroscopy in an analogous manner to 2. These experi-
mental probes demonstrated the retention of chirality
during the deprotonation/protonation (deuteration) process.
With these findings in hand, it was reasonable to believe
that the highly pyramidalized enolate must invert before
being alkylated on the other face. We therefore investigated
pyramidal inversion as a competitive pathway which could
be the true source of the high degree of stereodifferentiation
observed. All of the reaction profiles calculated from eno-
late 2’ are shown in Figure 4 and the energy barriers and in-
termediates (only the minimum-energy conformers) are
summarized in Table 2.
As can be seen from Table 2, the inclusion of solvation
dramatically alters the distribution of relative energies and
makes some reaction channels more accessible in solution.
The reacting species whose negative charge is partially or
completely cancelled (i.e., in alkylation processes) are
poorly solvated compared to those whose negative charge
remains unaltered (i.e., in enolate inversion and retro-O-Mi-
chael reactions). As a consequence, diastereoselective C-al-
kylation (the exclusive reaction in the gas phase) and retro-
O-Michael reaction become competitive pathways in solu-
tion, and this leads to a 94:5 ratio of products 5a and 6
based on the Boltzmann distribution obtained from Gibbs
free energies of each TS under Curtin–Hammett conditions.
Figure 3. a) Structure of enolate 2’ showing the greater contribution of
the carbanionic resonance form. b) Minimum-energy structures and
HOMO of “naked” (2’) and potassium (2’·K) enolates, calculated at the
B3LYP/6-31+G(d) level of theory. Distances are given in Š and angle a
in degrees.
A key feature found in these preliminary studies was the
abnormally pronounced nonplanar character of both the
“naked” and potassium enolates. Seebach et al.[12] measured
this deviation from planarity (pyramidalization) in terms of
À
the out-of-plane angle a between the C1 C4 bond vector
and the C1-C2-N3 plane, which ranges from 08 in a pure sp2
center to 54.78 in a typical sp3 center. Thus, the large calcu-
lated values of a (ca. 298), together with the non-symmetric
character (different shape and size of each lobe) of the
À
HOMO along the C1 C4 bond (Figure 3), provide the first
Chem. Eur. J. 2007, 13, 4840 – 4848
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4843

A. Avenoza, M. M. Zurbano et al.
The greater stability of 2’
with respect to 2’_ep (by
9.8 kcalmolÀ1) is mainly caused
by the strong stereoelectronic
repulsions between the N3
lone pair and the p bond be-
tween C1 and C4 in 2’_ep,
where the most important
charge contribution is located
at the pz orbital of C1. This
atom bears most of the nega-
tive charge of that bond and
consequently repels the N3
lone pair, as we confirmed
through NBO calculations (see
Computational Details and
Supporting Information). More
interestingly, the barrier for
pyramidal inversion of enolate
2’ (TS1) is about 2 kcalmolÀ1
higher in energy than TS2
(leading to less than 1% of
epimer 2’_ep), a situation that
implies complete retention of
chirality in the C-alkylation of
this substrate. Moreover, once
the enolate has inverted, it
must overcome a second acti-
vation barrier to complete the
disfavored C-alkylation, which
represents an overall barrier of
about 15 kcalmolÀ1. Applying
the equations of the Eyring
transition-state theory to the
Figure 4. Reaction channels including minimum-energy paths (MEP) explored in the computational study
starting from enolate 2’: pyramidal inversion, C-alkylation, O-alkylation and retro-O-Michael reaction.
calculated energy barriers, we
could estimate half-lives (t1/2)
of 59.2 and 0.3 ms for enolate 2’ in inversion and C-alkyla-
tion processes at À788C, respectively. These results are in
very good agreement with the experimental observations
discussed before.
Table 2. Calculated [B3LYP/6-31+G(d)] activation barriers and relative
energies[a] [kcalmolÀ1
(TS).[b]
] of the intermediates and transition structures
Gas phase (e=1.00)
DE0 DG
THF (e=7.58)
[c]
Structure
DDGsolv
DE0
DG
0.00
10.30
9.80
3.20
8.31
À40.37
14.00
14.69
À41.05
4.77
16.78
4.49
16.57
À6.59
9.48
2’
TS1
2’_ep
preTS2
TS2
5a·Br
preTS2_ep
TS2_ep
5a_ep·Br
preTS3
TS3
preTS3_ep
TS3_ep
7a·Br
TS4
0.00
12.58
10.71
À13.08
À5.42
À49.20
À2.69
0.35
0.00
0.00
À2.74
À1.17
10.33
6.79
1.56
10.47
7.39
0.67
10.33
6.15
9.34
6.42
0.00
9.84
9.54
The geometries of the minimum energy transition struc-
tures (TS) for each process are shown in Figure 5 (see Sup-
porting Information for complete characterization of all con-
formers located for each stationary point, including pre- and
post-transition structures). The pyramidalization of C1 in all
species, as revealed by their a values, is noteworthy.
Interestingly, TS1 maintains some pyramidalization away
from the ideal planar transition structure due to its late
character, which makes it very similar in geometry and
energy (by less than 1 kcalmolÀ1 in solution) to 2’_ep. The
even more pyramidal character of 2’_ep with respect to 2’
(ca. 88) is related to the lower energy barrier calculated
from 2’_ep to TS2_ep (4.89 kcalmolÀ1 in solution) with re-
spect to that from 2’ to TS2 (8.31 kcalmolÀ1 in solution), in
accordance with Hammondꢁs postulate.
The HOMO in TS1 has a more symmetrical character
than those in 2’ and 2’_ep. The pyramidal inversion from
13.04
10.97
À7.13
1.52
À2.75
1.37
À41.93
À47.64
7.78
7.74
À48.39
À0.12
10.54
À0.89
10.23
À13.86
9.59
3.53
7.30
À49.06
À10.45
4.39
À41.72
À5.56
10.63
À4.85
10.15
À8.29
6.73
À10.23
3.81
À15.56
6.84
1.70
2.75
TS4_ep
6’
13.24
À9.10
13.81
À9.43
0.00
1.55
13.24
À7.55
13.81
À7.88
[a] DE0 values include zero-point energy (ZPE) corrections at the same
level of theory. [b] The initial enolate and methyl bromide were arbitrari-
ly chosen as the zero level in the relative-energy calculations. [c] Calcu-
lated at the IEF-PCM/B3LYP/6-31+G(d) level.
4844
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4840 – 4848

Pyramidalized Bicyclic Serine Enolates
FULL PAPER
goal, and as a representative example, we transformed 5a
and 5b into the corresponding (S)-a-methylserine (8a) and
(S)-a-ethylserine (8b).
This was achieved by simple acid hydrolysis with 6n aque-
ous HCl under reflux(Scheme 4). The epxerimental data
Scheme 4. Synthesis of a-alkyl serines 8a, 8b, and 10a.
obtained for these compounds agree with those previously
reported in the literature.[1d,19] Considering the importance
and the extended use of non-natural a-methylated a-amino
acids,[20] especially (S)-a-methylserine (8a) as an inducer of
Figure 5. Minimum-energy transition structures for all reaction pathways
evaluated from enolate 2’ including intermediate 2’_ep, calculated at the
B3LYP/6-31+G(d) level of theory. Distances are given in Š and angles
a in degrees. HOMOs are displayed for TS1 and 2’_ep.
[21]
a-helixconformations when incorporated into peptides,
and the small number of suitable methods to obtain them
on a large scale,[1d,22] we also present here a new synthetic
strategy to obtain the R isomer. Thus, alkylation of chiral
building block 4 with methyl triflate followed by acid hy-
drolysis gave (R)-a-methylserine (10a) in good yield in
three steps from N-Boc-d-serine methyl ester 3 (Scheme 4).
enolate 2’ to 2’_ep is rather complexand is associated with a
conformational change at the oxazolidine ring. This is the
reason why IRC calculations failed when following the reac-
tion coordinate to the most stable enolate 2’, even when
using a broad range of step sizes (0.1–1.0 amu1/2 Bohr). For
the remainder of the structures, this lack of planarity is
more noticeable in C-alkylation and, to a lesser extent, in
the retro-O-Michael TS. However, O-alkylation TS have a
more planar character (moreover, C1 has inverted in
TS3_ep), as expected on account of the delocalization of
electronic density along the incipient ketene acetalic system.
In summary, the stereoselectivities achieved with these
substrates along with the theoretical study in this work rep-
resent another case of the proposal enunciated by See-
bach,[12] in which the stereoselectivity of a reaction can be
correlated with pyramidalization.
Conclusion
In summary, a short and general strategy for the multigram-
scale preparation of chiral serine-derived building blocks
has been developed. We first investigated their outstanding
diastereoselective alkylation, which occurs with efficient re-
tention of configuration. The mechanism involved in this al-
kylation process has been studied from a theoretical point
of view, with inclusion of solvation effects, and the retention
of the configuration at the defined stereocenter derives
from the high degree of pyramidalization of the ester eno-
late. As a synthetic application, we also synthesized a varie-
ty of enantiomerically pure quaternary a-alkyl a-amino
acids with the serine skeleton. The extensive use of these
chiral serine-derived building blocks as important starting
materials in stereocontrolled organic synthesis will continue
to be explored in the future.
Synthesis of quaternary a-alkyl a-amino acids: Given the re-
sults on the asymmetric alkylation of chiral building block 2,
we decided to extend this methodology to the synthesis of
some quaternary a-alkyl a-amino acids.[20] To achieve this
Chem. Eur. J. 2007, 13, 4840 – 4848
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4845

A. Avenoza, M. M. Zurbano et al.
8.8 Hz, 1H; CH2), 3.87 (d, J=8.8 Hz, 1H; CH2), 3.81 (s, 3H; CO2CH3),
3.46 (s, 3H; OCH3), 1.81 (s, 3H; CH3), 1.54 (s, 3H; CH3), 1.34 ppm (s,
3H; CH3); 13C NMR (100 MHz, CDCl3): d=173.2 (CO2CH3), 155.0
(NCO2), 106.8 (CCH3OCH3), 102.6 (CNCH3OCH2), 76.5 (CH2), 66.2
(CCO2CH3), 53.0 (CO2CH3), 51.3 (OCH3), 20.5 (CH3), 17.8 (CH3),
16.3 ppm (CH3); MS (ESI+): m/z: 260.1; elemental analysis calcd (%) for
C11H17NO6: C 50.96, H 6.61, N 5.40; found: C 50.66, H 6.70, N 5.38.
Experimental Section
General procedures: Solvents were purified according to standard proce-
dures. Analytical TLC was performed on Polychrom SI F254 plates.
Column chromatography was performed with silica gel 60 (230–
400 mesh). 1H and 13C NMR spectra were recorded on Bruker ARX 300
1
and Bruker Avance 400 spectrometers. H and 13C NMR spectra were re-
(3S,7R,7aS)-Tetrahydro-3-ethyl-7-methoxy-7,7a-dimethyl-5-oxo-2H-
corded in CDCl3 with TMS as internal standard and in D2O with TMS as
external standard in a coaxial microtube. Melting points were determined
on a Büchi B-545 melting point apparatus and are uncorrected. Optical
rotations were measured on a Perkin-Elmer 341 polarimeter. Microanaly-
ses were carried out on a CE Instruments EA-1110 analyser and are in
good agreement with the calculated values.
oxazolo[3,2-c]oxazole-3-carboxylic acid methyl ester (5b): Compound 5b
R
(64 mg, 0.24 mmol) was obtained from compound 2 (46 mg, 0.17 mmol),
after purification (hexane/ethyl acetate 1/1), as a colorless solid; yield:
70%. m.p. 71–738C; [a]D25 =À39.1 (c=1.02 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=4.64 (d, J=8.9 Hz, 1H; CH2), 3.91 (d, J=8.9 Hz,
1H; CH2), 3.81 (s, 3H; CO2CH3), 3.45 (s, 3H; OCH3), 2.44–2.53 (m, 1H;
CH2CH3), 2.05–2.14 (m, 1H; CH2CH3), 1.54 (s, 3H; CH3), 1.32 (s, 3H;
CH3), 0.95 ppm (t, J=7.4, 3H; CH2CH3); 13C NMR (100 MHz, CDCl3):
d=172.5 (CO2CH3), 154.6 (NCO2), 106.7 (CCH3OCH3), 102.4
(CNCH3OCH2), 73.9 (CH2), 70.0 (CCO2CH3), 52.8 (CO2CH3), 51.2
(OCH3), 26.0 (CH2CH3), 18.1 (CH3), 16.5 (CH3), 9.2 ppm (CH2CH3); MS
(ESI+): m/z: 274.1; elemental analysis calcd (%) for C12H19NO6: C 52.74,
H 7.01, N 5.13; found: C 52.70, H 7.03, N 5.13.
(3S,7R,7aS)-Tetrahydro-7-methoxy-7,7a-dimethyl-5-oxo-2H-oxazolo[3,2-
G
c]oxazole-3-carboxylic acid methyl ester (2): Method A: Triphenylphos-
phine hydrobromide (20.3 mg, 0.06 mmol) was added to a solution of (S)-
N-Boc-serine methyl ester (100 mg, 0.46 mmol) and 2,3-dimethoxy-1,3-
butadiene (63 mg, 0.55 mmol) in dry CH2Cl2 (5 mL) at room tempera-
ture. After stirring for 24 h, the mixture was diluted with CH2Cl2
(10 mL). The organic phase was washed with saturated NaHCO3 and
dried over anhydrous Na2SO4. The solvent was evaporated and the resi-
due was purified by column chromatography on silica gel (hexane/ethyl
acetate 8/2) to give 2 as a yellow oil (12.8 mg, 0.05 mmol, 11% yield).
Method B: A round-bottomed flask was charged with (S)-N-Boc-serine
methyl ester (11.18 g, 45.58 mmol), 2,2,3,3-tetramethoxybutane (16.26 g,
91.19 mmol), toluene (300 mL) and TsOH·H2O (0.86 g, 4.57 mmol). The
solution was heated under refluxand stirred for 3 h. The reaction mixture
was cooled to room temperature, diluted with diethyl ether (100 mL) and
then quenched with saturated NaHCO3 (100 mL). The aqueous phase
was extracted with diethyl ether (2”80 mL). The organic layers were
combined, washed with brine and dried over anhydrous Na2SO4. The sol-
vent was evaporated and the crude product purified by column chroma-
tography (hexane/ethyl acetate 9/1) to give 2 as a yellow oil (9.37 g,
31.91 mmol, 75% yield). [a]2D5 =À118.1 (c=1.20 in CHCl3); 1H NMR
(400 MHz, CDCl3): d=4.77 (dd, J=5.9 Hz, J=9.0 Hz, 1H; CH), 4.30 (t,
J=8.9 Hz, 1H; CH2), 4.15 (dd, J=5.9 Hz, J=8.8 Hz, 1H; CH2), 3.82 (s,
3H; CO2CH3), 3.47 (s, 3H; OCH3), 1.58 (s, 3H; CH3), 1.36 ppm (s, 3H;
CH3); 13C NMR (100 MHz, CDCl3): d=170.5 (CO2CH3), 160.6 (NCO2),
107.1 (CCH3OCH3), 101.5 (CNCH3OCH2), 66.7 (CH2), 59.9 (CH), 52.9
(CO2CH3), 51.0 (OCH3), 16.2 (CH3), 15.5 ppm (CH3); MS (ESI+): m/z:
246.1; elemental analysis calcd (%) for C10H15NO6: C 48.98, H 6.17, N
5.71; found: C 48.88, H 6.22, N 5.68.
(3S,7R,7aS)-Tetrahydro-3-benzyl-7-methoxy-7,7a-dimethyl-5-oxo-2H-
oxazolo[3,2-c]oxazole-3-carboxylic acid methyl ester (5c): Compound 5c
N
(80 mg, 0.24 mmol) was obtained from 2 (196 mg, 0.80 mmol), after pu-
rification (hexane/ethyl acetate 8/2), as a white solid; yield: 30%. m.p.
111–1138C; [a]2D5 =À117.4 (c=1.09 in CHCl3); 1H NMR (400 MHz,
CDCl3): d=7.32–7.10 (m, 5H; Ph), 4.56 (d, J=9.1 Hz, 1H; CH2), 4.18 (d,
J=9.1 Hz, 1H; CH2), 4.00 (d, J=13.8 Hz, 1H; CH2Ph), 3.74 (s, 3H;
CO2CH3), 3.45 (s, 3H; OCH3), 3.23 (d, J=13.8 Hz, 1H; CH2Ph), 1.41 (s,
3H; CH3), 1.32 ppm (s, 3H; CH3); 13C NMR (100 MHz, CDCl3): d=
171.4 (CO2CH3), 154.6 (NCO2), 134.4, 130.0, 128.7, 127.6 (Ph), 107.3
(CCH3OCH3), 102.2 (CNCH3OCH2), 73.5 (CH2), 69.6 (CCO2CH3), 52.9
(CO2CH3), 51.5 (OCH3), 38.2 (CH2Ph), 18.5 (CH3), 16.4 ppm (CH3); MS
(ESI+): m/z: 336.1; elemental analysis calcd (%) for C17H21NO6: C 60.89,
H 6.31, N 4.18; found: C 60.81, H 6.32, N 4.21.
(3S,7R,7aS)-Tetrahydro-3-allyl-7-methoxy-7,7a-dimethyl-5-oxo-2H-
oxazolo[3,2-c]oxazole-3-carboxylic acid methyl ester (5d): Compound 5d
G
(28 mg, 0.10 mmol) was obtained from compound 2 (35 mg, 0.14 mmol),
after purification (hexane/ethyl acetate 8/2), as a white solid; yield: 70%.
m.p. 47–498C; [a]2D5 =À74.5 (c=1.04 in CHCl3); 1H NMR (400 MHz,
CDCl3): d=5.82–5.64 (m, 1H; CH2CH=CH2), 5.15–5.24 (m, 2H;
CH2CH=CH2), 4.58 (d, J=9.1 Hz, 1H; CH2), 4.04 (d, J=9.1 Hz, 1H;
CH2), 3.79 (s, 3H; CO2CH3), 3.45 (s, 3H; OCH3), 3.12 (dd, J=14.1 Hz,
J=6.6 Hz, 1H; CH2CH=CH2), 2.86 (dd, J=14.1 Hz, J=7.8 Hz, 1H;
CH2CH=CH2), 1.53 (s, 3H; CH3), 1.33 ppm (s, 3H; CH3); 13C NMR
(100 MHz, CDCl3): d=172.1 (CO2CH3), 154.6 (NCO2), 131.2 (CH2CH=
CH2), 120.9 (CH2CH=CH2), 107.1 (CCH3OCH3), 102.5 (CNCH3OCH2),
73.9 (CH2), 68.6 (CCO2CH3), 53.1 (CO2CH3), 51.5 (OCH3), 37.3
(CH2CH=CH2), 18.0 (CH3), 16.7 ppm (CH3); MS (ESI+): m/z=286.1; el-
emental analysis calcd (%) for C13H19NO6: C 54.73, H 6.71, N 4.91;
found: C 54.82, H 6.68, N 4.89.
(3R,7S,7aR)-Tetrahydro-7-methoxy-7,7a-dimethyl-5-oxo-2H-oxazolo[3,2-
G
c]oxazole-3-carboxylic acid methyl ester (4): As described for its enantio-
mer 2, compound 4 (1.73 g, 7.03 mmol) was obtained by the same meth-
odology, but by using (R)-N-Boc-serine methyl ester (2.06 g, 9.39 mmol)
as starting material; yield: 75%. [a]2D5 =+120.2 (c=1.27 in CHCl3). The
spectroscopic data are identical to those described for its enantiomer. El-
emental analysis calcd (%) for C10H15NO6: C 48.98, H 6.17, N 5.71;
found: C 48.91, H 6.15, N 5.69.
General procedure for alkylation reactions: Compound
2 or 4
(0.90 mmol) was dissolved in dry THF (15 mL) under argon atmosphere
and, if necessary, HMPA (3.60 mmol) was added with a syringe, and then
the solution was stirred at À788C. Alkylating agent (2.70 mmol) and a
0.91m solution of KHMDS in THF (1.35 mmol) were added with a sy-
ringe (KHMDS was added slowly). After the reaction mixture was stirred
for 5 min, the reaction was quenched with saturated NH4Cl solution
(15 mL). This mixture was warmed to room temperature and vigorously
stirred. The crude product was diluted with diethyl ether and the aqueous
phase was separated and extracted again with more diethyl ether. The or-
ganic layers were combined, washed with brine and dried over anhydrous
Na2SO4. The solvent was evaporated, and the crude purified by column
chromatography on silica gel to give 5a–d (from 2) or 9a (from 4).
AHCTREUNG
AHCTREUNG
solved in dry THF (5 mL) under argon atmosphere. The solution was
cooled to À788C and stirred. A 0.91m solution of KHMDS in THF
(0.18 mL, 0.17 mmol) was added slowly with a syringe. After 10 min, the
reaction was quenched with saturated NH4Cl solution (5 mL). The mix-
ture was heated to room temperature in a warm water bath and vigorous-
ly stirred. The crude product was diluted with diethyl ether and the aque-
ous phase was extracted with more diethyl ether. The organic layers were
combined, washed with brine, and dried over anhydrous Na2SO4. The sol-
vent was evaporated and the crude product was purified by column chro-
matography on silica gel (hexane/ethyl acetate 6/4) to give compound 6
as a colourless oil (26 mg, 0.10 mmol); yield: 95%. [a]2D5 =À122.7 (c=
1.10 in CHCl3); 1H NMR (400 MHz, CDCl3): d=6.57 (s, 1H; C=CH2),
5.99 (s, 1H; C=CH2), 4.71 (s, 1H; OH), 3.87 (s, 3H; CO2CH3), 3.44 (s,
3H; OCH3), 1.62 (s, 3H; CH3), 1.36 ppm (s, 3H; CH3); 13C NMR
(100 MHz, CDCl3): d=166.1 (CO2CH3), 154.9 (NCO2), 131.8 (C=CH2),
(3S,7R,7aS)-Tetrahydro-7-methoxy-3,7,7a-trimethyl-5-oxo-2H-oxazolo-
ACHTREUNG
cation (hexane/ethyl acetate 1/1), as a colourless oil; yield: 85%. [a]D25
=
À95.4 (c=1.05 in CHCl3); 1H NMR (400 MHz, CDCl3): d=4.59 (d, J=
4846
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4840 – 4848

Pyramidalized Bicyclic Serine Enolates
FULL PAPER
129.1 (C=CH2), 108.7 (CCH3OCH3), 90.8 (CNCH3OH), 53.2 (CO2CH3),
50.7 (OCH3), 19.6 (CH3), 14.6 ppm (CH3); MS (ESI+): m/z: 246.1; ele-
mental analysis calcd (%) for C10H15NO6: C 48.98, H 6.17, N 5.71; found:
C 49.03, H 6.15, N 5.77.
Depending on the individual case, the internally stored parameters for
toluene and tetrahydrofuran were used to calculate solvation energies
(DGsolv). Zero-point corrected (DE0) and Gibbs free energies (DG) were
used for the discussion on the relative stabilities of the considered struc-
tures. Second-order perturbation energies (E2),[28] as well as pairwise
steric exchange energies for disjoint interactions (ES),[29] were calculated
by natural bond orbital/natural localized molecular orbital (NBO/
NLMO) analysis with the NBO 5.G program[30] and upgraded Gaussi-
an03 as interface. These attractive and repulsive interactions were esti-
mated by means of the localized s, p, n, s* and p* valence orbitals.
(S)-a-Methylserine (8a): A round-bottomed flask was charged with com-
pound 5a (221 mg, 0.85 mmol) and aqueous 6n HCl solution (10 mL).
The mixture was stirred overnight under reflux. The solvent was removed
in vacuo, and the residue was partitioned between water and ethyl ace-
tate. The aqueous phase was evaporated to give the hydrochloride deriv-
ative of compound 8a (130 mg, 98%). An aliquot of this material
(72 mg) was treated with ethanol/propylene oxide (1/1, 2 mL) under
refluxfor 2 h to give compound 8a (23 mg, 0.20 mmol) as a white solid;
yield 41%. [a]2D5 =+5.3 (c=1.16 in H2O). The spectroscopic data were
identical to those reported in the literature.[1d] MS (ESI+): m/z: 120.0; el-
emental analysis calcd (%) for C4H9NO3: C 40.33, H 7.62, N 11.76;
found: C 40.24, H 7.63, N 11.72.
Cartesian coordinates, electronic energies, entropies, enthalpies, Gibbs
free energies, and lowest frequencies of the different conformations of all
structures considered are available as Supporting Information.
X-ray diffraction analysis: Crystal data for 5b: C12H19NO6, Mr =273.28,
colourless prism, 0.5”0.3”0.2 mm, T=173 K, orthorhombic, space group
P212121, Z=4, a=5.72920(10), b=8.1401(2), c=29.1363(8) Š, V=
1358.81(6) Š3, 1calcd =1.336 gcmÀ3
,
F(000)=584, l=0.71073 Š (MoKa),
A
(S)-a-Ethylserine (8b): Starting from compound 5b (64 mg, 0.24 mmol)
and following the protocol described above for 8a, compound 8b (41 mg,
98%) was obtained as its hydrochloride derivative. The spectroscopic
data were identical to those reported in the literature.[19b,c] An analytical
sample of this (S)-a-ethylserine hydrochloride (20 mg) was treated with
ethanol/propylene oxide (1/1, 2 mL) under reflux to give compound 8b
(9.5 mg, 0.07 mmol) as a white solid; yield 60%. [a]2D5 =À1.5 (c=0.95 in
m=0.107 mmÀ1, Nonius kappa CCD diffractometer, q range 0.21–27.888,
6939 collected reflections, 3156 unique reflections, full-matrixleast-
squares refinement (SHELXL97),[31] R1 =0.0373, wR2 =0.0895, [R1 =
0.0523, wR2 =0.0970 (all data)], GOF=1.036, residual electron density
between 0.227 and À0.139 eŠÀ3. Hydrogen atoms were located from
electron-density maps. Crystal data for 5d: C13H19NO6, Mr =285.29, color-
less prism, 0.4”0.2”0.1 mm, T=173 K, orthorhombic, space group
P212121, Z=4, a=5.83440(10), b=14.2883(2), c=17.2023(3) Š, V=
1
H2O). H NMR (300 MHz, D2O): d=3.94 (d, J=11.9 Hz, 1H; CH2), 3.68
(d, J=11.9 Hz, 1H; CH2), 1.62–1.94 (m, 2H; CH2CH3), 0.94 ppm (dd, J=
10.1 Hz, J=5.0 Hz, 3H; CH2CH3); 13C NMR (75 MHz, D2O): d=177.6
(CO2H), 69.6 (CCH2OH), 67.2 (CCH2OH), 28.4 (CH2CH3), 10.0 ppm
(CH2CH3); MS (ESI+): m/z: 134.1; elemental analysis calcd (%) for
C5H11NO3: C 45.10, H 8.33, N 10.52; found: C 45.20, H 8.30, N 10.46.
1434.05(4) Š3, 1calcd =1.322 gcmÀ3
,
F(000)=608, l=0.71073 Š (MoKa),
A
m=0.105 mmÀ1, Nonius kappa CCD diffractometer, q range 0.21–27.888,
22960 collected reflections, 3420 unique reflections, full-matrixleast-
squares refinement (SHELXL97),[31] R1 =0.0395, wR2 =0.0930, [R1 =
0.0529, wR2 =0.0991 (all data)], GOF=1.069, residual electron density
between 0.192 and À0.166 eŠÀ3. Hydrogen atoms were located by mixed
methods (electron-density maps and theoretical positions).
(3R,7S,7aR)-Tetrahydro-7-methoxy-3,7,7a-trimethyl-5-oxo-2H-oxazolo-
A
enantiomer 5a, compound 9a (204 mg, 0.78 mmol) was obtained from
compound 4 (200 mg, 0.82 mmol); yield: 95%. [a]2D5 =+92.1 (c=1.05 in
CHCl3). The spectroscopic data were identical to those described for its
enantiomer; elemental analysis calcd (%) for C11H17NO6: C 50.96, H
6.61, N 5.40; found: C 50.95, H 6.70, N 5.42.
Acknowledgements
(R)-a-Methylserine (10a): As described for its enantiomer 8a, the hydro-
chloride derivative of compound 10a (109 mg, 98%) was obtained from
compound 9a (212 mg, 0.82 mmol). Similarly, an analytical sample
(53 mg) was treated with ethanol/propylene oxide (1/1, 2 mL) under
refluxto give compound 10a (29 mg, 0.24 mmol) as a white solid; yield
70%. [a]2D5 =À6.3 (c=0.98 in H2O). The spectroscopic data were identical
to those described for its enantiomer; elemental analysis calcd (%) for
C4H9NO3: C 40.33, H 7.62, N 11.76; found: C 40.24, H 7.63, N 11.72.
We thank the Ministerio de Educación y Ciencia and FEDER (project
CTQ2006-05825/BQU and Ramón y Cajal contract J.H.B.), the Universi-
dad de La Rioja (project API-05/B01 and grant of G.J.-O.), and the Go-
bierno de La Rioja (ANGI-2004/03 and ANGI-2005/01 projects). We
also thank CESGA for computer support.
2D NMR experiments: NMR experiments were carried out on a Bruker
Avance 400 spectrometer at 298 K. Magnitude-mode ge-2D COSY spec-
tra were recorded with gradients by using the cosygpqf pulse program
with 908 pulse width. Phase-sensitive ge-2D HSQC spectra were recorded
with a z-filter and selection before t1 with removal of the decoupling
during acquisition by using the invigpndph pulse program with CNST2
(JHC)=145. 2D NOESY experiments were carried out by using phase-
sensitive ge-2D NOESY for spectra recorded in CDCl3.
[1] a) A. Avenoza, C. Cativiela, F. Corzana, J. M. Peregrina, M. M. Zur-
bano, J. Org. Chem. 1999, 64, 8220–8225; b) A. Avenoza, C. Cativie-
la, J. M. Peregrina, D. Sucunza, M. M. Zurbano, Tetrahedron: Asym-
metry 1999, 10, 4653–4661; c) A. Avenoza, C. Cativiela, F. Corzana,
J. M. Peregrina, M. M. Zurbano, Tetrahedron: Asymmetry 2000, 11,
2195–2204; d) A. Avenoza, C. Cativiela, F. Corzana, J. M. Peregrina,
D. Sucunza, M. M. Zurbano, Tetrahedron: Asymmetry 2001, 12,
949–957; e) A. Avenoza, J. H. Busto, C. Cativiela, J. M. Peregrina,
D. Sucunza, M. M. Zurbano, Tetrahedron: Asymmetry 2003, 14,
399–405; f) A. Avenoza, J. H. Busto, F. Corzana, J. M. Peregrina, D.
Sucunza, M. M. Zurbano, Tetrahedron: Asymmetry 2004, 15, 719–
724; g) A. Avenoza, J. H. Busto, F. Corzana, J. M. Peregrina, D. Su-
cunza, M. M. Zurbano, Synthesis 2005, 575–578.
[2] a) P. Garner, J. M. Park, J. Org. Chem. 1987, 52, 2361–2364; b) P.
Garner, J. M. Park, Org. Synth. 1992, 70, 18–27; c) D. Seebach, J. D.
Aebi, M. Gauder-Coquoz, R. Naef, Helv. Chim. Acta 1987, 70,
1194–1216; d) A. Avenoza, C. Cativiela, F. Corzana, J. M. Peregrina,
M. M. Zurbano, Synthesis 1997, 1146–1150; e) X. Liang, J. An-
dersch, M. Bols, J. Chem. Soc. Perkin Trans. 1 2001, 2136–2157;
f) D. J. Dixon, C. I. Harding, S. V. Ley, D. M. G. Tilbrook, Chem.
Commun. 2003, 468–469; g) M. Brunner, T. Straub, P. Saarenketo,
K. Rissanen, A. M. P. Koskinen, Lett. Org. Chem. 2004, 1, 268–270;
h) M. Brunner, A. M. P. Koskinen, Tetrahedron Lett. 2004, 45, 3063–
3065; i) A. McKillop, R. J. K. Taylor, R. J. Watson, N. Lewis, Synthe-
Computational details: All calculations were carried out with the B3LYP
hybrid functional[23] and 6-31+G(d) basis set. Full geometry optimiza-
tions and transition structure (TS) searches were carried out with the
Gaussian03 package.[24] The possibility of different conformations was
taken into account for all structures. BSSE corrections were not consid-
ered in this work. Frequency analyses were carried out at the same level
used in the geometry optimizations, and the nature of the stationary
points was determined in each case according to the appropriate number
of negative eigenvalues of the Hessian matrix. Scaled frequencies were
not considered since significant errors in the calculated thermodynamic
properties are not found at this theoretical level.[25] Where necessary,
mass-weighted intrinsic reaction coordinate (IRC) calculations were car-
ried out by using the Gonzalez and Schlegel scheme[26] in order to ensure
that the TSs indeed connected the appropriate reactants and products.
Solvent effects were taken into account through the polarized continuum
model (IEF-PCM)[27] using UAHF radii, as implemented in Gaussian03.
Chem. Eur. J. 2007, 13, 4840 – 4848
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4847

A. Avenoza, M. M. Zurbano et al.
sis 1994, 31–32; j) L. Williams, Z. Zhang, F. Shao, P. J. Carroll,
M. M. JoulliØ, Tetrahedron 1996, 52, 11673–11694.
[3] a) S. V. Ley, E. Díez, D. J. Dixon, R. T. Guy, P. Michel, G. L. Nat-
trass, T. D. Sheppard, Org. Biomol. Chem. 2004, 2, 3608–3617;
b) N. L. Douglas, S. V. Ley, H. M. L. Osborn, D. R. Owen, H. W. M.
Priepke, S. L. Warriner, Synlett 1996, 793–795.
[21] a) A. L. Morris, M. W. MacArthur, E. G. Hutchinson, J. M. Thorn-
ton, Protein Struct. Funct. Genet. 1992, 12, 345–364; b) E. Altmann,
K.-H. Altmann, K. Nebel, M. Mutter, Int. J. Pept. Protein Res. 1988,
32, 344–349; c) V. Pavone, B. Di Blasio, A. Lombardi, O. Maglio, C.
Isernia, C. Pedone, E. Benedetti, E. Altmann, M. Mutter, Int. J.
Pept. Protein Res. 1993, 41, 15–20; d) H. Mickos, K. Sundberg, B.
Luning, Acta Chem. Scand. 1992, 46, 989–993; e) D. Obrecht, M.
Altorfer, C. Lehmann, P. Schçnholzer, K. Müller, J. Org. Chem.
1996, 61, 4080–4086.
[22] a) N. D. Smith, M. Goodman, Org. Lett. 2003, 5, 1035–1037; b) N. D.
Smith, A. M. Wohlrab, M. Goodman, Org. Lett. 2005, 7, 255–258.
[23] a) C. Lee, W. Yang, R. Parr, Phys. Rev. B 1988, 37, 785–789;
b) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[4] J.-L. Montchamp, F. Tian, M. E. Hart, J. W. Frost, J. Org. Chem.
1996, 61, 3897–3899.
[5] We have expressed the corresponding diastereomeric ratio as fol-
lows: d.r.>20:1, because a single stereoisomer of each of the final
1
compounds was detected by H NMR.
[6] C. I. Harding, D. J. Dixon, S. V. Ley, Tetrahedron 2004, 60, 7679–
7692.
[7] C. Agami, F. Couty, Tetrahedron 2002, 58, 2701–2724.
[8] M. Brunner, P. Saarenketo, T. Straub, K. Rissanen, A. M. P. Koski-
nen, Eur. J. Org. Chem. 2004, 3879–3883.
[9] CCDC-629276 (5b) and CCDC-629275 (5d) contain the supplemen-
tary crystallographic data for this paper. These data can be obtained
free of charge from the Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
[10] D. Seebach, B. Lamatsch, R. Amstutz, A. K. Beck, M. Dobler, M.
Egli, R. Fitzi, M. Gautschi, B. Herradon, P. C. Hidber, J. J. Irwin, R.
Locher, M. Maestro, T. Maetzke, A. Mourino, E. Pfammatter, D. A.
Plattner, C. Schickli, W. B. Schweizer, P. Seiler, G. Stucky, W. Petter,
J. Escalante, E. Juaristi, D. Quintana, C. Miravitles, E. Molins, Helv.
Chim. Acta 1992, 75, 913–934.
[11] a) H. Park, J. Lee, M. J. Kang, Y.-J. Lee, B.-S. Jeong, J.-H. Lee, M.-S.
Yoo, M.-J. Kim, S. Choi, S. Jew, Tetrahedron 2004, 60, 4243–4249;
b) P. J. Reider, R. S. Eichen, E. Conn, P. Davis, V. J. Granda, A. J.
Zambito, E. J. J. Gravowski, J. Org. Chem. 1987, 52, 3326–3334.
[12] D. Seebach, T. Maetzke, W. Petter, B. Klçtzer, D. A. Plattner, J.
Am. Chem. Soc. 1991, 113, 1781–1786.
[13] a) E. Buncel, J. M. Dust, Carbanion Chemistry: Structures and Mech-
anisms, American Chemical Society, Washington, DC, 2003; b) H.-
U. Reissig, I. Bohm, J. Am. Chem. Soc. 1982, 104, 1735–1737.
[14] a) S. Hoz, C. Azran, A. Sella, J. Am. Chem. Soc. 1996, 118, 5456–
5461; b) Y. Ikuta, S. Tomoda, Tetrahedron Lett. 2003, 44, 5931–5934.
[15] a) H. Zhao, D. C. Hsu, P. R. Carlier, Synthesis 2005, 1–16; b) P. R.
Carlier, P. C.-H. Lam, J. C. DeGuzman, H. Zhao, Tetrahedron:
Asymmetry 2005, 16, 2998–3002.
[24] Gaussian03, RevisionC.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomer-
y, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyen-
gar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y.
Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O.
Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Fores-
man, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski,
B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Na-
nayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford
CT, 2004.
[25] C. W. Bauschlicher, Jr., Chem. Phys. Lett. 1995, 246, 40–44.
[26] a) C. Gonzalez, H. B. Schlegel, J. Chem. Phys. 1989, 90, 2154–2161;
b) C. Gonzalez, H. B. Schlegel, J. Chem. Phys. 1990, 94, 5523–5527.
[27] a) M. Cossi, V. Barone, B. Mennucci, J. Tomasi, Chem. Phys. Lett.
1998, 286, 253–260; b) J. Tomasi, B. Mennucci, E. Cancꢂs, J. Mol.
Struct. 1999, 464, 211–226; c) M. Cossi, G. Scalmani, N. Rega, V.
Barone, J. Chem. Phys. 2002, 117, 43–54, and references therein;
d) J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105, 2999–
3093.
[16] K. Ando, N. S. Green, Y. Lee, K. N. Houk, J. Am. Chem. Soc. 1999,
121, 5334–5335.
[17] I. Soteras, O. Lozano, A. Gómez-EsquØ, C. Escolano, M. Orozco,
M. Amat, J. Bosch, F. J. Luque, J. Am. Chem. Soc. 2006, 128, 6581–
6588.
[28] a) A. E. Reed, L. A. Curtiss, F. Weinhold, Chem. Rev. 1988, 88, 899–
926; b) F. Weinhold, Natural Bond Orbital Methods in Encyclopedia
of Computational Chemistry, Vol. 3, (Ed.: P. von R. Schleyer), 1998,
pp. 1792–1811.
[29] a) J. K. Badenhoop, F. Weinhold, J. Chem. Phys. 1997, 107, 5406–
5421; b) J. K. Badenhoop, F. Weinhold, J. Chem. Phys. 1997, 107,
5422–5432; c) J. K. Badenhoop, F. Weinhold, Int. J. Quantum Chem.
1999, 72, 269–280.
[30] NBO 5.0, E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Car-
penter, J. A. Bohmann, C. M. Morales, F. Weinhold, Theoretical
Chemistry Institute, University of Wisconsin, Madison, 2001.
[31] G. M. Sheldrick, SHELXL97, Program for the refinement of crystal
structures, University of Gçttingen, Germany, 1997.
[18] a) K. Ando, J. Am. Chem. Soc. 2005, 127, 3964–3972; b) Y. Ikuta, S.
Tomoda, Org. Lett. 2004, 6, 189–192.
[19] a) D. Seebach, J. D. Aebi, M. Gander-Coquoz, R. Naef, Helv. Chim.
Acta 1987, 70, 1194–1216; b) Y. Ito, M. Sawamura, E. Shirakawa, K.
Hayashizaki, T. Hayashi, Tetrahedron Lett. 1988, 29, 235–238; c) N.
Takamura, S. Terashima, K. Arhiwa, S. Yamada, Chem. Pharm.
Bull. 1967, 15, 1776–1784.
[20] a) S. Mzengeza, T. K. Venkatachalam, M. Diksic, Amino Acids 2000,
18, 81–88; b) C. Cativiela, M. D. Díaz-de-Villegas, Tetrahedron:
Asymmetry 1998, 9, 3517–3599; c) Y. Ohfune , T. Shinada, Eur. J.
Org. Chem. 2005, 5127–5143.
Received: December 5, 2006
Published online: March 15, 2007
4848
www.chemeurj.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2007, 13, 4840 – 4848