Role of the countercation in diastereoselective alkylations of pyramidalized bicyclic serine enolates. An easy approach to α-benzylserine

Article abstract of DOI:10.1021/jo070656b

(Chemical Equation Presented) The use of a chiral serine equivalent as an excellent chiral building block has been demonstrated in the synthesis of α-benzylserine through a diastereoselective lithium enolate alkylation reaction and subsequent acid hydroly

Full text of DOI:10.1021/jo070656b

SCHEME 1. Retrosynthesis of r-Alkylserines from Bicyclic
Chiral Serine Equivalent 1
Role of the Countercation in Diastereoselective
Alkylations of Pyramidalized Bicyclic Serine
Enolates. An Easy Approach to r-Benzylserine†
Gonzalo Jime´nez-Ose´s, Carlos Aydillo, Jesu´s H. Busto,
Mar´ıa M. Zurbano,* Jesu´s M. Peregrina, and
Alberto Avenoza*
lective alkylation of this system oriented toward the large-scale
preparation of R-alkylserines (Scheme 1).
Departamento de Qu´ımica, UniVersidad de La Rioja, Grupo de
S´ıntesis Qu´ımica de La Rioja, UA-CSIC, 26006 Logron˜o, Spain
As a result of our theoretical investigations into the high
diastereoselectivity obtained experimentally (dr >20:1), we
proposed a highly pyramidalized ester enolate as the true source
of this stereodifferentiation. In fact, the inversion barrier of this
enolate was calculated to be greater than any other process
occurring under the reaction conditions.3 This proposal was
made on the basis of a “naked” enolate, because a base with a
noncoordinating cationslike potassium hexamethyldisilazane
(KHMDS)swas used in the experiments. Clearly, it was
necessary to test the validity of this hypothesis under other
reaction conditions. Moreover, we needed to minimize a
competitive retro-O-Michael reaction (often referred to as a
â-elimination), which reduced the yields of C-alkylation by
producing variable amounts of acrylate 3 with KHMDS as the
base (Table 1 graphic). The notable influence that the base
countercation has on alkylation processes is well-known, leading
to dramatic changes in regio- and stereoselectivities.4 The more
coordinating character of metalic ions like Li+ favors their
tighter association with the enolate moiety (mainly with oxygen
atoms), leading to complex structures like oligomers,5 chelates,6
and highly solvated adducts. In fact, the determination of the
major species in these reactants7 in solution is a challenging
task to rationalize the observed results in this kind of reaction.
Bearing these facts in mind, and with the aim of further
exploring the scope of the reaction, we decided to test the use
alberto.aVenoza@unirioja.es
ReceiVed April 4, 2007
The use of a chiral serine equivalent as an excellent chiral
building block has been demonstrated in the synthesis of
R-benzylserine through a diastereoselective lithium enolate
alkylation reaction and subsequent acid hydrolysis. The role
of a coordinating countercation (lithium) in the alkylation
reaction has been investigated. Theoretical studies have been
performed in order to elucidate the stereochemical outcome
of the alkylation process, which occurs with total retention
of configuration.
The current interest in R,R-dialkyl-R-amino acids1 relates to
the important effects they have on the biological activity of
peptides that incorporate these quaternary amino acids, because
they introduce alterations in the conformations of the backbone.
Among these compounds, chiral R-alkylserines have been
extensively studied owing to their important roles in synthetic
and biological chemistry.1
In the course of our studies aimed at finding synthetic routes
for R,R-disubstituted R-amino acids based on the use of five-
membered cyclic N,O-acetals2 (Garner aldehyde derivatives),
we became interested in diastereoselective alkylations of chiral
serine equivalents. In this context, we envisioned the design of
other, more stable chiral five-membered cyclic N,O-acetal serine
derivatives by incorporating a new ring in the structure.
Consequently, we recently reported the synthesis and reactivity
of a novel class of bicyclic N,O-acetal (1) derived from serine
(Scheme 1).3 More specifically, we assessed the diastereose-
(2) (a) Avenoza, A.; Cativiela, C.; Corzana, F.; Peregrina, J. M.; Zurbano,
M. M. J. Org. Chem. 1999, 64, 8220-8225. (b) Avenoza, A.; Cativiela,
C.; Peregrina, J. M.; Sucunza, D.; Zurbano, M. M. Tetrahedron: Asymmetry
1999, 10, 4653-4661. (c) Avenoza, A.; Cativiela, C.; Corzana, F.; Peregrina,
J. M.; Zurbano, M. M. Tetrahedron: Asymmetry 2000, 11, 2195-2204.
(d) Avenoza, A.; Cativiela, C.; Corzana, F.; Peregrina, J. M.; Sucunza, D.;
Zurbano, M. M. Tetrahedron: Asymmetry 2001, 12, 949-957. (e) Avenoza,
A.; Busto, J. H.; Cativiela, C.; Peregrina, J. M.; Sucunza, D.; Zurbano, M.
M. Tetrahedron: Asymmetry 2003, 14, 399-405. (f) Avenoza, A.; Busto,
J. H.; Corzana, F.; Peregrina, J. M.; Sucunza, D.; Zurbano, M. M.
Tetrahedron: Asymmetry 2004, 15, 719-724. (g) Avenoza, A.; Busto, J.
H.; Corzana, F.; Peregrina, J. M.; Sucunza, D.; Zurbano, M. M. Synthesis
2005, 575-578.
(3) Aydillo, C.; Jime´nez-Ose´s, G.; Busto, J. H.; Peregrina, J. M.; Zurbano,
M. M.; Avenoza, A. Chem. Eur. J. 2007, 13, 4840-4848.
(4) (a) Hu, Y.; Bishop, R. L.; Luxenburger, A.; Dong, S.; Paquette, L.
A. Org. Lett. 2006, 8, 2735-2737. (b) Boeckman, R. K., Jr.; Boehmler, D.
J.; Musselman, R. A. Org. Lett. 2001, 3, 3777-3780. (c) Alvarez-Ibarra,
C.; Csaky¨, A. G.; Maroto, R.; Quiroga, M. L. J. Org. Chem. 1995, 60,
7934-7940.
(5) (a) Williard, P. G.; Carpenter, G. B. J. Am. Chem. Soc. 1986, 108,
462-468. (b) Williard, P. G.; Carpenter, G. B. J. Am. Chem. Soc. 1985,
107, 3345-3346. (c) Seebach, D.; Amstutz, R.; Dunitz, J. D. HelV. Chim.
Acta 1981, 64, 2622-2626.
(6) (a) Williard, P. G.; Hintze, M. J. J. Am. Chem. Soc. 1990, 112, 8602-
8604. (b) Williard, P. G.; Hintze, M. J. J. Am. Chem. Soc. 1987, 109, 5539-
5541. (c) Sun, C.; Williard, P. G. J. Am. Chem. Soc. 2000, 122, 7829-
7830. (d) Collum, D. B. Acc. Chem. Res. 1992, 25, 448-454.
(7) (a) Abbotto, A.; Leung, S. S.-W.; Streitwieser, A.; Kilway, K. V. J.
Am. Chem. Soc. 1998, 120, 10807-10813. (b) Abbotto, A.; Streitwieser,
A.; Schleyer, P. v. R. J. Am. Chem. Soc. 1997, 119, 11255-11268.
* To whom correspondence should be addressed. Fax: +34 941 299655.
† Dedicated to Prof. Vicente Gotor on the occasion of his 60th birthday.
(1) (a) Wirth, T. Angew. Chem., Int. Ed. Engl. 1997, 36, 225-227. (b)
Cativiela, C.; Diaz-de-Villegas, M. D. Tetrahedron: Asymmetry 1998, 9,
3517-3599. (c) Cativiela, C.; Diaz-de-Villegas, M. D. Tetrahedron:
Asymmetry 2000, 11, 645-732. (d) Vogt, H.; Bra¨se, S. Org. Biomol. Chem.
2007, 5, 406-430. (e) Hruby, V. J. Acc. Chem. Res. 2001, 34, 389-397.
(f) Kotha, S. Acc. Chem. Res. 2003, 36, 342-351. (g) Cativiela, C.; Diaz-
de-Villegas, M. D. Tetrahedron: Asymmetry 2007, 18, 569-623.
10.1021/jo070656b CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/08/2007
J. Org. Chem. 2007, 72, 5399-5402
5399

TABLE 1. Diastereoselective Alkylation of Building Block 1 with
Li-Derived Bases
SCHEME 2. Synthesis of 4_ep from Building Block 1_ep
“Li”
base
product,
entry
RX
MeI
MeI
MeI
MeI
MeI
solvent
additive
2a-e/3/4a yieldb (%)
SCHEME 3. Synthesis of (S)-r-Benzylserine 5c
1
2
3
4
5
6
7
8
9
LDA
THF none
-/-/100
100/-/-
-/2/98
4,c 46
2a,d 90
4, 43
4, 56
4, 38
2a, 96
2a, 54
4, 54
LHDMS THF HMPA
LHDMS THF HMPT
LHDMS THF TMEDA
-/3/97
LHDMS THF 12-crown-4 -/2/98
MeOTf LHDMS THF none
MeOTf LHDMS Et2O none
EtOTf LHDMS THF none
EtOTf LHDMS THF HMPA
100/-/-
62/-/38
40/-/60
100/-/-
53/-/47
93/7/-
52/-/48
100/-/-
-/37/63
-/9/91
2b,e 94
2c,d 45
2c, 74
2d,e 40
2d, 92
4, 51
3, 4, 5, 11, 14, and 15, with the formation of bis-adduct 4 being
the alternative competitive pathway. This kind of “dimerization”
has been previously reported9 and is produced by the quick in
situ alkylation of the enolate of 1 with acceptor 3 in a
C-Michael-type reaction. This process proceeds with a high
10 BnBr
11 BnI
LHDMS THF HMPA
LHDMS THF HMPA
12 allylBr LHDMS THF HMPA
13 allylI LHDMS THF HMPA
14 n-HexI LHDMS THF HMPA
15 none LHDMS THF none
4, 87
1
diastereoselectivity (dr >20:1 as demonstrated by H NMR),
a Ratio 2a-e/3/4 determined by 1H NMR analysis of the reaction mixture.
b Yield of the major product isolated after column chromatography.
c Stereochemistry determined by X-ray analysis (this work). d Stereochem-
istry predicted by NOE (ref 3). e Stereochemistry determined by X-ray
analysis (ref 3).
and the absolute configuration of 4, which contains six chiral
carbons, was unambiguously determined by X-ray diffraction
analysis10 of monocrystals obtained by slow evaporation of a
solution in a mixture of hexane/ethyl acetate. Starting from the
enantiomer of bicycle 1 (1_ep), which could be obtained from
commercially available (R)-N-Boc-serine methyl ester,3 this
protocol allows the preparation of “dimer” 4_ep (Scheme 2),
which is an enantiomer of 4, as demonstrated by NMR, X-ray
diffraction analysis,10 and comparison of the optical activities
of the two compounds (see the Supporting Information).
In an effort to expand this methodology to the preparation
of R-alkylserines, we carried out the synthesis of (S)-R-
benzylserine 5c from the precursor 2c. Hydrolysis of the
alkylation substrate 2c in an acidic medium gave the required
amino acid 5c as an hydrochloride derivative with a 97% yield
(Scheme 3). An aliquot of this compound was treated with
propylene oxide in order to obtain the free amino acid 5c with
a 73% yield. The structural and optical properties of this
compound were identical to those reported in the literature
([R]25D ) +16.8).11 It is important to note that although several
synthetic methods have been reported for other chiral R-alkyl-
serines, there are only four reports on the asymmetric synthesis
of enantiomerically pure (S)- or (R)-R-benzylserine, also com-
monly named (S)- or (R)-R-hydroxymethylphenylalanine.11 The
interest in R-benzylserine has been illustrated in peptide
chemistry12 since several analogs of bioactive peptides (e.g.,
of lithium diisopropylamide (LDA) and lithium hexamethyld-
isilazane (LHDMS) as bases in the alkylation of building block
1. The influence of the leaving group was also studied by using
different commercially available alkyl halides and trifluo-
romethanesulfonates (ROTf). For less reactive alkyl groups or
poor leaving groups, several additives like hexamethylphos-
phoramide (HMPA), tris(dimethylamino)phosphine (HMPT),
N,N,N′,N′-tetramethylenediamine (TMEDA), and 12-crown-4
were used. Bearing in mind that different ethereal solvents are
often said to have different coordinating properties, causing a
profound effect on aggregation and therefore on the reactivity
of lithium enolates,8 two ethereal solvents, tetrahydrofuran
(THF) and diethyl ether (Et2O), were also tested. The results
obtained are summarized in Table 1. Alkylation reaction did
not occur using LDA as a base (entry 1). The same feature was
observed when the base LHMDS was combined with additives
different from HMPA (entries 2-5). From the observation of
entries 6 and 7, THF was selected as solvent instead Et2O.
As can be seen, the best results in terms of alkylation were
obtained with LHMDS as a base and MeOTf (without additives)
and EtOTf (in the presence of HMPA) as alkylating agents
(entries 6 and 9). These new conditions allowed us to improve
on the yield previously obtained in ref 3 with the total
suppression of secondary reactions. The alkylation products were
obtained with the same absolute configuration and diastereo-
meric purity (dr >20:1) achieved in ref 3, as demonstrated by
comparison of the NMR data and optical activity with the
previously reported values. Therefore, the change in the base
countercation from K+ to Li+ had a negligible effect on the
diastereoselectivity of the alkylation reaction.
(9) (a) Ley, S. V.; Dixon, D. J.; Guy, R. T.; Palomero, M. A.; Polara,
A.; Rodr´ıguez, F.; Sheppard, T. D. Org. Biomol. Chem. 2004, 2, 3618-
3627. (b) Ley, S. V.; Baeschlin, D. K.; Dixon, D. J.; Foster, A. C.; Ince, S.
J.; Priepke, H. W. M.; Reynolds, D. J. Chem. ReV. 2001, 101, 53-80.
(10) Sheldrick, G. M. SHELXL97. Program for the refinement of crystal
structures. University of Go¨ttingen, Germany, 1997.
(11) (a) Horikawa, M.; Nakajima, T.; Ohfune, Y. Synlett 1997, 253-
254. (b) Davis, F. A.; Zhang, Y.; Rao, A.; Zhang, Z. Tetrahedron 2001,
57, 6345-6352. (c) Jew, S.-s; Lee, Y.-J.; Lee, J.; Kang, M. J.; Jeong, B.-
S.; Lee, J.-H.; Yoo, M.-S.; Kim, M.-J.; Choi, S.-h.; Ku, J.-M.; Park, H.-g.
Angew. Chem., Int. Ed. 2004, 43, 2382-2385. (d) Vassiliou, S.; Yiotakis,
A.; Magriotis, P. A. Tetrahedron Lett. 2006, 47, 7339-7341.
(12) (a) Lempicka, E.; Slaninova, J.; Olma, A.; Lammek, B. J. Peptide
Res. 1999, 53, 554-559. (b) Olma, A.; Tourwe´, D. Lett. Pept. Sci. 2000,
7, 93-96. (c) Olma, A.; Chung, N. N.; Schiller, P. W.; Zabrocki, J. Acta
Biochim. Pol. 2001, 48, 1121-1124. (d) Olma, A.; Lachwa, M.; Lipkowski,
A. W. J. Peptide Res. 2003, 62, 45-52.
Interestingly, the retro-O-Michael product 3 was not observed
under almost any conditions, except as a minor product in entries
(8) Streitwieser, A.; Juaristi, E.; Kim, Y.-J.; Pugh, J. K. Org. Lett. 2000,
2, 3739-3741.
5400 J. Org. Chem., Vol. 72, No. 14, 2007

FIGURE 2. Transition structures for the reaction pathways evaluated
from enolate 1′, calculated at the B3LYP/6-31+G(d) level. Newman
projections from C2 (front) to C1 (back) and related torsion angles are
also displayed. Distances are given in Å, torsion and dihedral angles R
FIGURE 1. (a) Resonance forms of lithium enolate 1′ showing the
higher contribution of the enol form. (b) Minimum energy structure
together with HOMO of enolate 1′, calculated at the B3LYP/6-31+G-
(d) level. Distances are given in angstroms and angles R and θ in
degrees.
and θ in degrees, and relative activation energies in kcal mol-1
.
deprotonation, probably due to stereoelectronic factors inherent
in the structure of the bicyclic substrate.
In line with our previous observations,3 one of the most
remarkable geometrical features found in the calculated structure
of enolate 1′ is the high nonplanar character of the bridgehead
carbamate N-atom, measured in terms of the out-of-plane angle
(signed) between the bond vector N3-C1 and the plane N3-
C5-C6 (denoted as θ). This high level of pyramidalization is
most likely the result of the conformational restraints imposed
by the bicyclic structure and makes N3 formally chiral,14
showing S-configuration. Therefore, and in order to minimize
the stereoelectronic repulsions between the N3 lone pair and
the π bond between C1 and C4 in 1′, the pyramidalization of
the enolate (R) would be a consequence of this high level of
pyramidalization observed in N3 (θ).
The sources of stereoselectivity were then analyzed by
calculating the transition structures corresponding to approach
at the Re and Si faces (TS1r and TS1s, respectively), which
were properly located and characterized. The most remarkable
geometrical features are shown in Figure 2 along with the
relative energies of the calculated geometries. As can be seen
from these plots, the incipient C1-C7 and breaking C7-Br
bonds, together with the approach angles of the electrophile,
are very similar in both TS.
µ-selective opioid peptide agonist DALDA, endomorphin-2,
arginine vasopressin, deltorphin I) containing this quaternary
amino acid have been synthesized and the biological conse-
quences of replacing natural amino acids with (S)- or (R)-R-
benzylserine have been studied, with many examples showing
an increase in activity.
The aforementioned experimental findings encouraged us to
carry out a theoretical study of the C-alkylation process starting
from the enolate of bicycle 1 (see the Supporting Information).
Bromomethane (MeBr) was chosen as the alkylating agent for
the calculations, and discrete molecules of dimethyl ether
(Me2O) were used to explicitly solvate the lithium atom attached
to the enolate moiety. In addition, bulk solvent effects were
included in the estimation of energies in order to better
reproduce the experimental conditions (see the Supporting
Information).
After the exhaustive evaluation of a large number of
coordination and solvation possibilities (see the Supporting
Information for further details), the structure of the most stable
lithium enolate was found to be the seven-membered chelate 1′
by far (Figure 1a). One significant feature found in these
preliminary studies was the slightly nonplanar character of this
lithium enolate. According to Seebach et al.,13 this deviation
from planarity (pyramidalization) was measured in terms of the
out-of-plane angle (signed) between the bond vector C1-C4
and the plane C1-C2-N3 (denoted as R), which ranges from
0° in a pure sp2 center to 54.7° in a typical sp3 center.
Subsequently, the calculated value of R (around 6°), together
with the slightly asymmetric character at the nodal plane
observed for the HOMO along the C1-C4 bond (Figure 1b),
point toward retention of configuration of the stereocenter after
The energy barrier of the enolate alkylation by the Re face
(TS1r) is ca. 5 kcal mol-1 lower than by the Si face (TS1s),
which implies total diastereoselectivity (>99:1) in the reaction.
This situation is in total agreement with the experimentally
observed retention of configuration.
(14) (a) Calcagni, A.; Gavuzzo, E.; Lucente, G.; Mazza, F.; Morera, E.;
Paglialunga Paradisi, M.; Rossi, D. Biopolymers 2000, 54, 379-387. (b)
Kakuda, H.; Suzuki, T.; Takeuchi, Y.; Shiro, M. Chem. Commun. 1997,
85-86. (c) Ohwada, T.; Achiwa, T.; Okamoto, I.; Shudo, K.; Yamaguchi,
K. Tetrahedron Lett. 1998, 39, 865-868. (d) Ohwada, T.; Okamoto, I.;
Shudo, K.; Yamaguchi, K. Tetrahedron Lett. 1998, 39, 7877-7880. (e)
Srivastava, A.; Srivastava, V.; Verma, S. M. J. Org. Chem. 1994, 59, 3560-
3563.
(13) Seebach, D.; Maetzke, T.; Petter, W.; Klo¨tzer, B.; Plattner, D. A.
J. Am. Chem. Soc. 1991, 113, 1781-1786.
J. Org. Chem, Vol. 72, No. 14, 2007 5401

was stirred at -78 °C, and benzyl iodide (5.49 g, 25.20 mmol)
and a 1 M solution of LHMDS in THF (16.80 mL, 16.80 mmol)
were added by syringe (LHMDS was added slowly). After being
stirred for 5 min, the reaction was quenched with saturated NH4Cl
solution (100 mL). The mixture was warmed to rt and vigorously
stirred. The mixture was diluted with diethyl ether, and the aqueous
phase was separated and extracted with diethyl ether. The organic
layers were combined, washed with brine, and dried over anhydrous
Na2SO4. The solvent was evaporated, and the crude product was
purified by silica gel column chromatography to give 2.08 g of
compound 2c, 74% yield. Analytical and spectroscopic data for
this compound agree with those reported in the literature.3
Compound 2c (483 mg, 1.44 mmol) was charged into a round-
bottomed flask with 6 N aqueous HCl (10 mL). The mixture was
stirred and heated under reflux overnight. The solvent was removed
in vacuo and the residue was partitioned between water and ethyl
acetate. The aqueous phase was evaporated to give 323 mg of the
corresponding hydrochloride derivative of (S)-R-benzylserine, 97%
yield. An aliquot of this material (102 mg) was treated with (1:1)
ethanol/propylene oxide (3 mL) under reflux for 2 h to give (S)-
R-benzylserine 5c (62 mg, 0.32 mmol) as a white solid, 73%
Once the calculated energy barriers had reproduced the
experimental selectivity, we located the factors responsible for
this high level of stereocontrol in the C-alkylation. The
rehybridization of C1 to sp3 in TS1s and TS1r diminishes the
high pyramidalization of N3 in enolate 1′ in order to maintain
the seven-membered chelate (Figures 1 and 2). This planariza-
tion (lower values of θ) and the subsequent strain on the bicyclic
system are greater (by ca. 6°) in TS1s. Additionally, TS1r has
a more staggered arrangement (and consequently lower torsional
strain) than TS1s according to the Newman projections of the
C1-C2 bonds (Figure 2). These distinct torsional effects
occurring on each TS are in agreement with the stereoselectivity
observed for this reaction.
In summary, a short and general strategy for the preparation
of (S)-R-benzylserine 5c from chiral serine-derived building
block 1 has been developed. The synthetic route involves
alkylation of 1 followed by acid hydrolysis. We first investigated
the role of the countercation (lithium) in the diastereoselective
alkylation of this building block, which occurs with retention
of configurationsas it does when a base with a noncoordinating
cation (potassium) is used.3 Therefore, the change of the base
countercation from K+ to Li+ has a negligible effect on the
diastereoselectivity of the alkylation reaction, but this change
increases the yield of the reaction, minimizing the formation
of side-products. Additionally, the mechanism involved in this
alkylation process has been studied from a theoretical point of
view, including solvation effects.
yield: [R]25 +16.0 (c 1.00, H2O). NMR data agree with the
D
published data11 (see the Supporting Information). Anal. Calcd for
C10H13NO3: C, 61.53; H, 6.71; N, 7.18. Found: C, 61.45; H, 6.73;
N, 7.23.
Acknowledgment. We thank the Ministerio de Educacio´n
y Ciencia and FEDER (project CTQ2006-05825/BQU and
Ramo´n y Cajal contract to J.H.B.).
Supporting Information Available: Experimental details,
spectroscopic characterization of all new compounds, crystal
structure data, and computational details. This material is available
free of charge via the Internet at http://pubs.acs.org.
Experimental Section
(S)-R-Benzylserine (5c). Compound 1 (2.06 g, 8.40 mmol) was
dissolved in dry THF (100 mL) under an argon atmosphere, and
HMPA (6.02 g, 33.60 mmol) was added by syringe. The solution
JO070656B
5402 J. Org. Chem., Vol. 72, No. 14, 2007