Asymmetric synthesis employing a chiral 5-methoxy-1,4-oxazin-2-one derivative: Preparation of enantiomerically pure α-quaternary α-amino acids

Article abstract of DOI:10.1002/(SICI)1099-0690(199908)1999:8<1967::AID-EJOC1967>3.0.CO;2-G

A new asymmetric synthesis of disubstituted α-amino acids is presented. This synthesis is based on the chiral 5-methoxy-1,4-oxazin-2-one derivative 5 relying on the α-hydroxy acid 1 as a chiral auxiliary. Alkylation reactions of the glycine equivalent 5 are performed by deprotonation with secbutyllithium and subsequent reaction with alkyl halides, yielding the monoalkylated compounds 13 and 14. A second alkylation step of the lithium enolates of 13 and 14 leads to the α,α-disubstituted compounds 17. Both steps proceed with good yields and excellent stereoselectivities (up to 99% de). From the major diastereomers 17c-d the corresponding α-amino acids 19c- d are obtained enantiomerically pure upon hydrolytic cleavage with aqueous sodium hydroxide. Alkylation of the enolate ion of 5 with epichlorohydrines as bifunctional electrophiles provides the cyclopropyl derivatives 20a-b. Direct hydrolysis or oxidation of 20a-b, followed by reductive amination and hydrolysis leads to the substituted 1-aminocyclopropanecarboxylic acids 21a- b and 24a-b.

Full text of DOI:10.1002/(SICI)1099-0690(199908)1999:8<1967::AID-EJOC1967>3.0.CO;2-G

FULL PAPER
Asymmetric Synthesis Employing a Chiral 5-Methoxy-1,4-oxazin-2-one
Derivative: Preparation of Enantiomerically Pure ␣-Quaternary
␣-Amino Acids
Oliver Achatz,[a] Andrea Grandl,[a] and Klaus Theodor Wanner*[a]
Dedicated to Prof. Dr. Dr. h.c. mult. H. Nöth on the occasion of his 70th birthday
Keywords: 1,4-Oxazines / Asymmetric synthesis / Amino acids / Glycine derivatives / Aminocyclopropanecarboxylic acid
derivatives
A new asymmetric synthesis of disubstituted α-amino acids
is presented. This synthesis is based on the chiral 5-methoxy-
1,4-oxazin-2-one derivative 5 relying on the α-hydroxy acid
From the major diastereomers 17c–d the corresponding α-
amino acids 19c–d are obtained enantiomerically pure upon
hydrolytic cleavage with aqueous sodium hydroxide.
1 as a chiral auxiliary. Alkylation reactions of the glycine Alkylation of the enolate ion of 5 with epichlorohydrines as
equivalent 5 are performed by deprotonation with sec- bifunctional electrophiles provides the cyclopropyl
butyllithium and subsequent reaction with alkyl halides, derivatives 20a–b. Direct hydrolysis or oxidation of 20a–b,
yielding the monoalkylated compounds 13 and 14. A second followed by reductive amination and hydrolysis leads to the
alkylation step of the lithium enolates of 13 and 14 leads to
the α,α-disubstituted compounds 17. Both steps proceed with
good yields and excellent stereoselectivities (up to 99% de).
substituted 1-aminocyclopropanecarboxylic acids 21a–b and
24a–b.
Introduction
Nonproteinogenic unnatural α-amino acids play an im-
portant role in the field of peptide chemistry and for the
design of enzyme inhibitors. In this context α-quaternary α-
amino acids are of special interest because of their rigidity
resulting from the quaternary center at the α-carbon atom.
Furthermore, such compounds are valuable chiral building
blocks in the asymmetric synthesis of more complex mol-
ecules. Although there are numerous methods for the asym-
metric synthesis of monosubstituted α-amino acids in the
literature,[1] there are only a few for the preparation of α,α-
disubstituted compounds.[2]
Scheme 1. Synthesis of α-quaterny α-amino acids employing oxazi-
nedione 2 and oxazinone 5
In a recent paper[3] we described the synthesis of chiral
α-substituted pipecolic acid derivatives 3 employing 1,4-
oxazine-2,5-dione 2 as a building block with the chiral in-
formation derived from the enantiomerically pure α-
hydroxy carboxylic acid 1. In this report we describe our
efforts to develop the imidate 5 derived from the 1,4-oxa-
zine-2,5-dione 4 as a versatile substrate for the asymmetric
construction of α,α-disubstituted α-amino acids 6. The im-
idate 5 was selected for this purpose as it appeared to be
more suitable for alkylation reactions than the amide pre-
cursor 4.
Results and Discussion
Synthesis of the Glycine Equivalent 5
We developed two alternative routes (Scheme 2) for the
preparation of 5, both starting from 1 as the chiral auxili-
ary, which is easily accessible by two synthetic steps.[4]
In the case of route 1, in the first step compound 1 was
transformed into the corresponding phenyl ester 7a in order
to modestly activate the carboxylic acid function. This
transformation could be effected in a yield of 81% by treat-
ing 1 with N,N-carbonyldiimidazole (CDI) and phenol. In
order to set up the glycine moiety, 7a was treated with an
excess of chloroacetyl chloride to give the diester 8a (96%),
which upon heating with 2 equiv. of sodium azide in aceto-
nitrile provided the azide 9a (93%). In a one-pot reaction
[a]
Institut für Pharmazie Ϫ Zentrum für Pharmaforschung,
Butenandtstraße 5Ϫ13, Haus C, D-81377 München, Germany
Fax: (internat.) ϩ 49(0)89/2180-7247
E-mail: klaus.wanner@cup.uni-muenchen.de
Eur. J. Org. Chem. 1999, 1967Ϫ1978
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
1434Ϫ193X/99/0808Ϫ1967v $ 17.50ϩ.50/0
1967

O. Achatz, A. Grandl, K. Th. Wanner
FULL PAPER
by hydrogenation in the presence of Pd/C and subsequent
Finally, a more direct and efficient strategy for the prep-
heating of the crude reaction mixture to reflux, azide 9a was aration of 4 was found. Treatment of the carboxylic acid 1
then transformed into 4 (83%). Though a clean cyclization with cbz-protected glycine that had been activated with
occurred, it took seven days to obtain a complete reaction. CDI (in THF, room temperature, 30 min.) yielded the ester
Therefore, the more reactive pentafluorophenyl ester 9b, 10 (91%). Upon removal of the cbz group by catalytic
prepared in analogy to 9a, was also tested as an alternative hydrogenation (Pd/C, H2) the amino acid 11 (88%) was ob-
precursor to 4. In this case the ring closure of the amine, tained. This acid underwent a clean cyclization upon heat-
generated from 9b (by hydrogenation in the presence of Pd/ ing when treated with MukaiyamaЈs reagent[6] (2-chloro-1-
C), occurred already at room temperature and was complete methylpyridinium iodide) in the presence of ethyldiisopro-
after stirring the reaction mixture for about 12 h (85%).
pylamine (8 h) to give 4 in a yield of 89%.
Scheme 2. Syntheses of the chiral glycine equivalent 5
1
The appearance of the H-NMR spectrum of the oxazi-
The reaction of 4 with 2 equiv. of trimethyloxonium
nedione 4 was strongly dependent on the amount of water tetrafluoroborate[7] afforded finally the oxazinone 5 in a
present in the solvent used (CDCl3). In the absence of water yield of 91%.
(in anhydrous CDCl3) the two signals for the diastereotopic
methylene protons at C-8 appeared as doublets at δ ϭ 4.14
and δ ϭ 4.24 (J ϭ 18.8 Hz), which were accompanied by
two further, but very weak signals that overlapped with the
Enolate Alkylations with Alkyl Halides
former [δ ϭ 4.15 (dd, J ϭ 18.8/4.1 Hz), 4.25 (d, J ϭ
The results of the alkylation reactions obtained for the
18.8 Hz)]. The two major signals, however, disappeared enolate of 5 were strongly dependent on the reaction con-
completely after a small amount of water (one drop) had ditions employed. The first experiments were performed
been added to the NMR sample, whereas the two minor with THF as solvent generating the enolate 12 by treating
signals increased to the expected size. This phenomenon 5 with strong bases like sec-butyllithium or potassium hexa-
may be the result of hydrogen bonds involving the lactam methyldisilylamide at low temperatures (Ϫ50 and Ϫ78°C,
function which lead to dimers or even oligomers of 4.[5] 15Ϫ20 min). Though the addition of alkyl halides
Such aggregates could dominate in the absence of water and (PhCH2Br, allyl bromide) led to the diastereomeric alky-
be diminished in favor of a monomeric species when water lation products 13 and 14 with good to excellent diastereo-
is added. But it may also be a simple hydration of 4 (present selectivities (> 90% de), the yields of these reactions were,
as a monomer) that changes the nature and thus the NMR however, generally very low. Surprisingly, the crude reaction
spectrum of this compound.
products contained large amounts of the starting material
From the two methods for the preparation of 4 presented besides significant amounts of dialkylation products 15.
above the one that was based on a phenyl ester gave a total This indicated that the poor yields obtained for the monoal-
yield of 60% (1b Ǟ 4) but was less convenient because of kylation products 13/14 were not simply due to an insuf-
the duration of this cyclization reaction (7 d). In contrast, ficient deprotonation of 5 for whatever reason. Extensive
the time required for the cyclization of the pentafluoro- studies performed for the benzylation revealed that the sol-
phenyl ester was markedly less, but in this case the overall vent was crucial for the product distribution (13ϩ14 versus
yield was somewhat lower (1b Ǟ 4; 44%).
15 and 5) of these alkylations. In Table 1 some representa-
1968
Eur. J. Org. Chem. 1999, 1967Ϫ1978

Preparation of Enantiomerically Pure α-Quaternary α-Amino Acids
FULL PAPER
tive results obtained for alkylations of enolate 12, formed benzyl) and with excellent diastereoselectivities (d.s. > 98:2
by deprotonation of 5 with sec-butyllithium [except for en- to > 99:1).
try 5 in Table 1 where KN(SiMe3)2 was used] are shown.
For the second alkylation step the above-mentioned pro-
cedure developed for the monoalkylation reactions was em-
ployed. This afforded the dialkylation products 17 in good
to high yields and with excellent diastereoselectivities (see
Table 2). Similar results were obtained when the reactions
were performed in THF (at Ϫ78°C, sBuLi), indicating that
the second alkylation step is quite insensitive to the reaction
conditions. In each case both diastereomers (17a/c and 17b/
d) were available as major compounds on changing the se-
quence for the introduction of the different substituents.
The diastereoselectivities followed from the 1H-NMR spec-
tra of the crude products. Though both diastereomers (17)
were available as reference compounds, in no case could
the corresponding minor diastereomer (17c,d for 17a,b and
17a,b for 17c,d) be detected. This indicated that the dia-
stereoselectivities should be at least d.s. > 99:1. In the case
of the benzylation of 13b/14b (see entry 2 Table 2) minute
amounts of a side product were present that partially inter-
Scheme 3. Alkylation of the oxazinone 5 via the enolate 12
Table 1. Enolate alkylation reactions of 5 with alkyl halides
Entry
R1ϪHal
Reaction conditions[a]
Product
13,14
Yield (%)
d.s.[b]
13/14
Conversion rate[c]
13؉14
5[d]
13 ϩ 14
15
5[d]
1
2
3
4
5
6
7
8
9
MeI
Ϫ55°C, DME
a
b
b
c
c
c
c
c
c
67
61
74
57
25
51
55
82
76
> 99:1[e]
99.2:0.8
92.4:7.6
Ϫ
93
94
Ϫ
60
25
73
94
92.5
95
3
4
AllBr
Ϫ55°C, DME
6
0
AllBr
Ϫ55°C Ǟ room temp., DME
Ϫ80°C, THF
Ϫ
3
20
8.5
4.5
3.5
1
Ϫ
37
55
18.5
1.5
4
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
PhCH2Br
35
40
8
Ϫ78°C, THF, inverse[f]
Ϫ80°C Ǟ room temp., THF
Ϫ80°C, THF/DMPU 2:1
Ϫ55°C Ǟ room temp., DME
Ϫ55°C, DME
Ϫ
96:4
94:6
95:5
98.4:1.6
4
[a]
[b]
Deprotonation with 1 equiv. of sBuLi, except entry 5. Ϫ For entries 2, 3 and 9 determined by analytical HPLC, in all other cases
from the integrals of the 1H-NMR spectra of the crude products. Ϫ [c] Calculated from the integrals of the 1H-NMR spectra of the crude
[d]
[e]
1
products. Ϫ
Recovered starting material. Ϫ For entry 1 in the H-NMR spectrum only a trace amount of 14a was detectable (ca.
[f]
0.2%). Ϫ Addition of 1.1 equiv. of KN(SiMe3)2 to a mixture of 1 equiv. of 5 and 1.2 equiv. of PhCH2Br.
For the benzylation in THF severe amounts of the double fered with the areas where the signals for the minor dia-
alkylation product 15 and unchanged starting material 5 stereomer were expected. Therefore, in this case a conserva-
were found, when the reaction temperature was kept tive diastereoselectivity of only d.s. > 98:2 is claimed. The
atϪ80°C (for conversion rates see Table 1, entry 4) or raised results are summarized in Table 2.
to room temperature (see Table 1, entry 6). With an inverse
alkylation protocol the product distribution became even
The 1H-NMR spectra of the mono- and disubstituted
worse (see Table 1, entry 5). According to Williams et al.[8] benzyl derivatives 13c, 14c, 15c and 17a؊d showed a signifi-
such side reactions may be due to aggregates formed from cant behavior with respect to the chemical shift for one or
the enolate ions. In a further alkylation reaction we added even both of the two methyl groups at C-1 (of the chiral
DMPU as a cosolvent (see Table 1, entry 7) in order to auxiliary). In the case of the monosubstituted compounds
counteract this phenomenon that might apply here, too. In- the stereoisomer 13c with the benzyl group being on the
deed this led to a significantly improved conversion rate opposite side of the two methyl substituents at C-1 the sig-
with respect to the diastereomeric monoalkylation products nal of the most shielded methyl group was detected at δ ϭ
13 ϩ 14. But the yield of isolated product (13/14) was still 0.85. This is close to the range of δ ϭ 0.90 to δ ϭ 1.02
moderate (55%), possibly as a result of the aqueous workup where the respective signals of the methyl- and allyl-substi-
to remove the cosolvent DMPU. Finally, we found that in tuted compounds 13a؊b, 14a؊b and 15a؊b appear. For
dimethoxyethane[9] (DME) the reactions proceeded in the the products 14c and 15c where a benzyl group is at the
desired manner, the double alkylation reactions being al- same side as the methyl groups at C-1 the signal of one
most completely suppressed. Thus, as outlined in Table 1 (14c: δ ϭ 0.69) or both (15c: δ ϭ 0.01 and δ ϭ 0.28) of
(entry 1, 2 and 9) the monoalkylation products were iso- those methyl substituents were shifted significantly to
lated in good yields (61% for R1 ϭ allyl to 76% for R1 ϭ higher field.
Eur. J. Org. Chem. 1999, 1967Ϫ1978
1969

O. Achatz, A. Grandl, K. Th. Wanner
FULL PAPER
In the final step the alkylation products 17 had to be
hydrolyzed to give the free amino acids 19. This could be
best performed under basic conditions with 40% aqueous
sodium hydroxide, whereas a two-step procedure that had
successfully been applied for the cyclopropyl compounds
23a؊b described below gave only poor results. For the hy-
drolysis the diastereomers 17c؊d each representing one iso-
mer of the two pairs of diastereomers 17 were selected. The
reaction was performed by stirring the compounds at room
temperature (with 40% aqueous sodium hydroxide) for
eight (17c) and twelve days (17d), respectively. Workup in-
cluding purification by ion exchange chromatography pro-
vided the amino acids 19c[10a,10c] and 19d[10b] in a yield of
52% and 48%, respectively. For the hydrolysis of 17c, in
addition to the amino acid 19c the amide 18c (35%) was
isolated, indicating that the reaction time applied had been
insufficient. With a prolonged reaction time (twelve days)
as applied for the cleavage of 17d no such side product
was found.
Scheme 4. Alkylation of the monoalkylation products 13/14 via the
enolate 16
From the optical rotations found for the amino acids
19c؊d, which are known in the literature,[10] it became evi-
dent that these compounds 19c؊d have (S) configuration.
According to these results the precursors 17c؊d must have
(S) configuration (at C-8) as well and as a consequence this
stereocenter (at C-8) must have (R) configuration in the dia-
stereomers 17a؊b.
Table 2. Enolate alkyation of the monoalkylation products 13/14
with alkyl halides[a]
Entry
R1
R2ϪHal[b]
Yield (%)
17
d.s.[c]
1
2
3
4
Me
PhCH2Br
PhCH2Br
MeI
55
85
78
88
> 99:1
> 98:2
> 99:1
> 99:1
All
PhCH2
PhCH2
The chiral center in 5 will keep its integrity even under
strong basic conditions. For this reason compound 5 is a
useful tool for the synthesis of α,α-dialkylated α-amino ac-
ids.
AllBr
[a] A diastereomeric mixture of 13 ϩ 14 was used, as obtained from
[b]
various alkylations. Ϫ Reaction conditions: 1) 1.5Ϫ2.0 equiv. of
sBuLi, Ϫ55°C, DME; 2) 2.0Ϫ3.0 equiv. of R2ϪHal, Ϫ55°C, DME.
From this point of view it was only of minor interest to
employ 5 also in the synthesis of α-monoalkylated α-amino
acids. Therefore and as the hydrolysis of the dialkylation
products 17 proceeded only under strong alkaline reaction
conditions no attempts for the hydrolysis of 13/14 to give
the respective α-tertiary amino acids were made.
[c]
1
Ϫ
Determined from the H-NMR spectra of the crude reaction
products; the diastereomers were inseparable or only insufficiently
separable by analytical HPLC.
For the 1H-NMR spectra of the disubstituted com-
pounds 17 a related phenomenon was found. For the com-
pounds 17a and 17b with the benzyl groups on the ”bottom
side” the signals of the methyl groups were detected at δ ϭ
0.83 and δ ϭ 0.80, respectively. And again as it had been
observed for 15c in the case of 17c and 17d with the benzyl
group up the signals of both methyl groups (at C-1) exhib-
ited a marked upfield shift (17c: δ ϭ 0.26, 0.53; 17d: δ ϭ
0.16, 0.46).
It is obvious that this phenomenon is indicative of the
stereochemistry at C-8 and thus may be used for the assign-
ment of the absolute configuration of the newly created
stereocenter.
Broad applicability and high diastereoselectivities for
monoalkylation reactions are main features of the bislactim
ether method developed by Schöllkopf et al. as well as the
mild acidic conditions sufficient to liberate the amino acids.
However, the suitability of the bislactim ethers for the syn-
thesis of α-quaternary α-amino acids is strongly limited.
The second alkylation step includes the risk of deprotonat-
ing the stereocenter of the chiral auxiliary. Therefore, usu-
ally only monoalkylations are performed with bislactim
ethers, whereas the preparations of dialkylation products
are limited to some more or less rare examples with special
structural features.[11]
In contrast to the former reagents compounds like BOC-
BMI (N-BOC-tert-butyl-methyl-imidazolidinone) devel-
oped by Seebach et al. (”self-reproduction of chirality”) al-
low the synthesis of mono- as well as of α,α-disubstituted
derivatives almost equally well. Yields and diastereoselec-
tivities for the various steps are generally very high. For the
hydrolysis of the α,α-disubstituted glycine derivatives to the
free α-amino acids, however, mostly severe reaction con-
ditions, or multistep sequences are necessary.[12]
Scheme 5. Hydrolysis of the dialkylation products 17cϪd
1970
Eur. J. Org. Chem. 1999, 1967Ϫ1978

Preparation of Enantiomerically Pure α-Quaternary α-Amino Acids
Thus, with respect to the synthesis of α-quaternary α-
FULL PAPER
Further efforts were directed towards the development of
amino acids compound 5 may be a valuable alternative to a synthetic sequence for the preparation of amino deriva-
the aforementioned methods.
tives like 24a and 24b from the alcohols 20a and 20b. It
turned out that this could easily be accomplished by oxi-
dation of 20a and 20b according to Swern to give the alde-
hydes 22a and 22b, respectively, which upon reductive amin-
ation Ϫ performed with aniline and NaBH3CN (in THF at
room temperature) Ϫ provided 23a and 23b. The total
yields for the two steps amounted to 86% (23a) and 90%
(23b), and there was no loss of diastereomeric purity to be
observed. Finally, the hydrolysis of 23a and 23b, performed
by the two-step procedure described above, afforded the two
diastereomeric amino acids 24a and 24b in reasonable
yields of 62% and 59%, respectively.
Alkylation with Epichlorohydrines as Bifunctional
Electrophiles for the Synthesis of
Aminocyclopropanecarboxylic Acids
In the context of a project directed toward the develop-
ment of new CNS-active compounds we were interested in
a synthetic access to 1-aminocyclopropanecarboxylic acids
like 21 and 24. For the construction of these compounds we
performed double alkylations employing enantiomerically
pure epichlorohydrines. Upon treatment of 5 with 2.1
equivalents of NaN(SiMe3)2 (in THF at Ϫ10°C) followed
by 4 equivalents of (S)-(ϩ)-epichlorohydrine the cyclopro-
pane derivative 20a was formed in a yield of 69%. This pro-
cedure was also effective with (R)-(Ϫ)-epichlorohydrine
providing the diastereomer 20b in a yield of 66%. In both
cases no other stereoisomers could be detected, neither by
1H-NMR spectroscopy nor by HPLC.
The hydrolysis of 20a and 20b to give the free amino ac-
ids could be best performed by a two-step procedure using
first hydrochloric acid (c ϭ 0.2 mol/L) to cleave the imidate
function and then 40% aqueous sodium hydroxide to finally
hydrolyze the remaining ester. This gave the two dia-
stereomeric 1-amino-2-(hydroxymethyl)-cyclopropanecar-
boxylic acids 21a and 21b in overall yields of 40% and 43%,
respectively. A comparison of the spectroscopic data and
the optical rotations of these compounds with those of
authentic materials [(E)- and (Z)-1-amino-2-(hydroxymeth-
yl)cyclopropanecarboxylic acids] reported in the litera-
ture[13] revealed that their structure corresponds to 21a
and 21b.
Model for the Asymmetric Induction
As is evident from the data reported above the alkylation
reactions of the enolates 12 and 16 proceeded consistently
with re selectivity. As a rationale for this stereoselection the
model in Scheme 7 representing the putative structure of
the enolates 12 and 16 is proposed. It is reasonable to as-
sume that in the enolate ions of 12 and 16 O-6, C-7, C-8,
N-9 and C-10 are substantially coplanar, while C-5 lies
above this plane defined by the aforementioned atoms.[16]
The stereochemical outcome of these reactions is also in
accord with results observed for related reactions,[14][15] in-
dicating that the anion of 5 has attacked the halide-bearing
carbon atom of the epichlorohydrins.
Scheme 7. Preferential attack in the alkylation of oxazinone 5 and
of monoalkylation products 13/14
In this conformation the geminal dimethyl group at C-1
is placed in a favorable pseudoaxial orientation in which
the allylic strain of that group with the methoxy group at
C-10 in the neighborhood is avoided. As a result of the
differences in steric hindrance of the top versus the bottom
face of the enolates 12 and 16, caused by the disposition
of this group in an axial orientation, the approach of the
electrophile should proceed preferentially from below. This
approach would give rise to a re stereoselectivity and it is
this stereoselectivity (re) that has been, indeed, observed for
all alkylation reactions.
Conclusion
In summary, we have prepared compound 5 as a new
chiral glycine equivalent. For the alkylation reactions DME
appeared to be advantageous over THF as a solvent. In the
former solvent (DME) good yields and diastereoselectivities
for the monoalkylation products were obtained, whereas (in
contrast to THF) only insignificant amounts of double al-
Scheme 6. Synthesis of 1-aminocyclopropanecarboxylic acids
21aϪb and 24aϪb
Eur. J. Org. Chem. 1999, 1967Ϫ1978
1971

O. Achatz, A. Grandl, K. Th. Wanner
FULL PAPER
Rf ϭ 0.25 (n-heptane/ethyl acetate, 25:75; detected with cerium
molybdate solution). Ϫ 1H NMR (CDCl3): δ ϭ 1.02 (s, 3 H, CH3),
1.04 (s, 3 H, CH3), 1.05 (s, 3 H, CH3), 1.47Ϫ1.58 (m, 1 H,
CH2CH2), 1.94Ϫ2.04 (m, 2 H, CH2CH2), 2.77Ϫ2.86 (m, 1 H,
CH2CH2), 3.25 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H,
CH2OCH3), 3.40 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 4.14 (d, J ϭ
18.8 Hz, 1 H, COCH2NH), 4.24 (d, J ϭ 18.8 Hz, 1 H, COCH2NH),
6.32 (s, br., 1 H, CONHCH2). Ϫ 1H NMR (CDCl3/H2O); after
addition of a small amount of H2O to the 1H-NMR sample, we
observed different chemical shifts and a different coupling pattern
kylation products were formed. The second alkylation step
proceeded in comparison to the first with improved yields
(55Ϫ88%) and with almost complete asymmetric induction
as never any minor isomer could be detected. Hydrolysis
led to the corresponding α-quaternary amino acids 19. In
addition the alkylation method could be successfully ex-
tended to the stereoselective synthesis of the substituted
aminocyclopropanecarboxylic acids 21a؊b and 24a؊b. The
absolute configuration of the alkylation products could be
delineated by comparing the optical rotation of the amino for the two protons of the COCH2N group: δ ϭ 4.15 (dd, J ϭ 18.8/
4.1 Hz, 1 H, COCH2NH), 4.25 (d, J ϭ 18.8 Hz, 1 H, COCH2NH).
Ϫ IR: ν˜ ϭ 3433 cmϪ1, 1757, 1686. Ϫ MS (CH4; CI); m/z (%): 256
(100) [M ϩ Hϩ]. Ϫ C13H21NO4 (255.3): calcd. C 61.16, H 8.29, N
5.49; found C 61.20, H 8.33, N 5.40.
acids (19c؊d, 21a؊b) with literature data.
Experimental Section
General Remarks: Standard vacuum techniques were used in the
handling of air-sensitive materials. Solvents were dried and kept
under N2 and freshly distilled before use. All reagents were used as
commercially available. Phenol was distilled in vacuo and stored
under nitrogen. Ϫ M.p. (uncorrected values): Melting point appa-
ratus according to Dr. Tottoli (Büchi no. 510). Ϫ Optical rotations:
Polarimeter 241 MC (PerkinϪElmer). Ϫ 1H NMR: JNMR-GX
400 (Jeol), 400 MHz, chemical shifts (δ) are reported in ppm, TMS
as internal standard. Ϫ IR: FT-IR spectrometer 1600 and FT-IR
spectrometer Paragon 1000 (PerkinϪElmer), liquids were run as
(2R,5S)-10-Methoxy-2-methoxymethyl-1,1,2-trimethyl-6-oxa-9-aza-
spiro[4.5]dec-9-en-7-one (5): To a solution of 3.78 g (14.8 mmol) of
4 in 150 mL of CH2Cl2 4.52 g (30.7 mmol, 2.1 equiv.) of Me3OϩBF
Ϫ was added and the resulting mixture was stirred for 16 h at room
4
temp. Then the mixture was cooled with ice, and treated under
stirring with 30 mL of phosphate buffer (pH 7, c ϭ 1.0 mol/L).
After saturation with solid NaCl, the organic layer was separated
and the aqueous layer was extracted with CH2Cl2 (3 ϫ 10 mL).
The combined organic layers were washed with brine, dried
(MgSO4), and concentrated in vacuo. Purification by CC (n-hep-
films, solids as KBr pellets. Ϫ MS: 5989 Mass spectrometer with tane/ethyl acetate, 70:30) afforded 3.62 g (91%) of 5 as colorless
20
59980 B particle beam LC/MS interface (Hewlett Packard). Ϫ
Combustion analysis: CHN Rapid (Heraeus). Ϫ Column chroma-
crystals, m.p. 45Ϫ47°C. Ϫ [α]D ϭ ϩ160.0 (c ϭ 1.45, CHCl3). Ϫ
TLC: Rf ϭ 0.20 (n-heptane/ethyl acetate, 70:30; detected with ce-
tograhy (CC): Flash chromatography[17] on silica gel (Merck 60 rium molybdate solution). Ϫ HPLC: tR ϭ 11.0 min (n-heptane/
1
F-254, 0.040Ϫ0.063 mm). Ϫ TLC: TLC plates 60 F-254 (Merck),
detection with UV (λ ϭ 254 nm) or with ammonium cerium(IV)
ethyl acetate, 70:30; 1.0 mL/min). Ϫ H NMR (CDCl3): δ ϭ 0.90
(s, 3 H, CH3), 0.95 (s, 3 H, CH3), 0.99 (s, 3 H, CH3), 1.47Ϫ1.53
heptamolybdate. Ϫ Analytical HPLC: L-6000 pump, L-4000 UV/ (m, 1 H, CH2CH2), 1.90Ϫ2.05 (m, 2 H, CH2CH2), 2.60Ϫ2.67 (m,
Vis detector (λ ϭ 254 nm), D-7500 and D-2500 Chromato Inte-
grator (Merck-Hitachi), column: LiChroCart with LiChrospher
1 H, CH2CH2), 3.17 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H,
CH2OCH3), 3.45 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.70 (s, 3 H,
Si 60 cartridge (5 µm, 250 ϫ 4 mm with precolumn 4 ϫ 4 mm), NϭCOCH3), 4.18 (d, J ϭ 21.6 Hz, 1 H, COCH2N), 4.32 (d, J ϭ
(Merck). Ϫ Preparative HPLC: L-6000 pump, L-4000 UV/Vis (λ ϭ 21.6 Hz, 1 H, COCH2N). Ϫ IR: ν˜ ϭ 1755 cmϪ1, 1687. Ϫ MS
254 nm), D-2000 Chromato Integrator(Merck-Hitachi), column:
(70 eV); m/z (%): 269 (28) [Mϩ], 113 (100). Ϫ C14H23NO4 (269.3):
Hibar RT LiChrosorb Si 60 (7 µm, 250 ϫ 25 mm) (Merck).
calcd. C 62.43, H 8.61, N 5.20; found C 62.26, H 8.59, N 5.40.
Preparation of the Chiral Glycine Equivalent 5 from 1
Phenyl (1S,3R)-1-Hydroxy-3-methoxymethyl-2,2,3-trimethylcyclo-
pentane-1-carboxylate (7a): To a solution of 3.63 g (16.8 mmol) of
1 in 150 mL of THF 2.99 g (18.5 mmol, 1.1 equiv.) of CDI, dis-
solved in 25 mL of THF, was added. After stirring for 16 h at room
(2R,5S)-2-Methoxymethyl-1,1,2-trimethyl-6-oxa-9-azaspiro-
[4.5]decane-7,10-dione (4): A) To a solution of 1.75 g (4.66 mmol)
of 9a in 470 mL of dioxane 150 mg of Pd/C (10% Pd) was added
and the resulting mixture was stirred at room temp. under H2 for temp., the solution was treated with 1.67 g (17.7 mmol, 1.05 equiv.)
5 h. After H2 had been substituted by N2 the mixture was heated
to reflux for 7 d. Finally, the solvent was evaporated in vacuo and
the crude product, still contaminated with the catalyst, was purified
by CC (n-heptane/ethyl acetate, 25:75; Rf ϭ 0.25) to give 0.987 g
of phenol, dissolved in 10 mL of THF, and heated to reflux for 7 h.
The residue obtained after concentration in vacuo was dissolved in
50 mL of Et2O and washed with 0.5  NaOH (4 ϫ 15 mL). The
aqueous layers were reextracted with Et2O (2 ϫ 10 mL), the com-
(83%) of 4 as colorless oil, which crystallized upon standing at bined organic extracts were dried (MgSO4) and concentrated in
room temp., affording colorless crystals. Ϫ B) To a solution of vacuo. CC (cyclohexane/ethyl acetate/AcOH, 85:13:2) yielded
1.50 g (3.23 mmol) of 9b in 500 mL of THF 170 mg of Pd/C (10% 4.00 g (81%) of 7a as colorless oil. Ϫ [α]D ϭ Ϫ8.4 (c ϭ 0.27,
20
Pd) was added and the resulting mixture was stirred at room temp.
under H2 for 15 h. After the solvent was evaporated in vacuo, the
crude product, still contaminated with the catalyst, was purified by
CHCl3). Ϫ TLC: Rf ϭ 0.47 (cyclohexane/ethyl acetate/AcOH,
64:34:2; detected with cerium molybdate solution) Ϫ 1H NMR
(CDCl3): δ ϭ 0.94 (s, 3 H, CH3), 0.99 (s, 3 H, CH3), 1.14 (s, 3 H,
CC (petroleum ether/ethyl acetate, 2:8; Rf ϭ 0.39) to give 0.326 g CH3), 1.64 (ddd, J ϭ 13.3/11.1/4.9 Hz, 1 H, CH2CH2), 1.89 (ddd,
(85%) of 4 as colorless crystals. Ϫ C) A suspension of 0.252 g J ϭ 14.3/9.4/4.9 Hz, 1 H, CH2CH2), 2.04 (ddd, J ϭ 13.3/9.4/7.3 Hz,
(0.92 mmol) of 11 in 20 mL of CH2Cl2 was treated with 0.280 g 1 H, CH2CH2), 2.79 (ddd, J ϭ 14.3/11.1/7.3 Hz, 1 H, CH2CH2),
(1.1 mmol, 1.2 equiv.) of 2-chloro-1-methylpyridinium iodide. After 3.12 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.40 (d, J ϭ 9.0 Hz, 1 H,
addition of 0.5 mL of diisopropylethylamine, the suspension turned CH2OCH3), 3.41 (s, 3 H, CH2OCH3), 5.36 (s, br., 1 H, OH), 7.12
into a clear solution, which was refluxed for 8 h before CH2Cl2
was removed in vacuo. Purification by CC (petroleum ether/ethyl
(dd, J ϭ 8.8/1.1 Hz, 2 H, aromatic H), 7.21Ϫ7.25 (m, 1 H, aromatic
H), 7.36Ϫ7.41 (m, 2 H, aromatic H). Ϫ IR: ν˜ ϭ 3354 cmϪ1, 1753.
acetate, 2:8, Rf ϭ 0.39) gave 0.209 g of 4 (89%) as colorless crystals. Ϫ MS (CH4; CI); m/z (%): 293 (16) [MϩHϩ], 171 (100). Ϫ
20
Ϫ M.p. 92Ϫ95°C. Ϫ [α]D ϭ ϩ122.0 (c ϭ 0.10, CHCl3). Ϫ TLC:
C17H24O4 (292.4): calcd. C 69.84, H 8.27; found C 69.79, H 8.32.
1972
Eur. J. Org. Chem. 1999, 1967Ϫ1978

Preparation of Enantiomerically Pure α-Quaternary α-Amino Acids
FULL PAPER
Pentafluorophenyl (1S,3R)-1-Hydroxy-3-methoxymethyl-2,2,3-tri-
methyl-cyclopentane-1-carboxylate (7b): To a solution of 0.395 g
(1.83 mmol) of 1 in 15 mL of THF 0.334 g (2.0 mmol, 1.1 equiv.)
(2.27 mmol) of 8a in 25 mL of MeCN 0.298 g (4.59 mmol) of NaN3
was added. The resulting mixture was heated for 7 h (70Ϫ75°C)
and then concentrated in vacuo. The resulting oil was dissolved in
of CDI, dissolved in 8 mL of THF, was added. After stirring the 20 mL of Et2O, washed with a 20% NaHCO3 solution (3 ϫ 3 mL)
mixture for 22 h at room temp., a solution of 0.336 g (1.83 mmol)
of pentafluorophenol in 5 mL of THF was added. After refluxing
the mixture for 15 h, THF was removed in vacuo, and the remain-
and with brine (2 ϫ 5 mL). The aqueous layers were reextracted
with Et2O (2 ϫ 5 mL) and the combined organic layers were dried
(MgSO4) and concentrated in vacuo. Purification by CC (cyclohex-
ing oil was dissolved in 5 mL of Et2O. The ethereal solution was ane/ethyl acetate, 80:20) yielded 0.791 g (93%) of 9a as a pale yellow
washed with 0.5  NaOH (5 ϫ 2 mL) and brine (1 ϫ 2 mL). The
aqueous phases obtained were reextracted with Et2O (2 ϫ 5 mL),
the combined organic layers were dried (MgSO4) and concentrated
in vacuo. Purification by CC (petroleum ether/ethyl acetate, 7:3;
oil. Ϫ [α]D20 ϭ ϩ31.6 (c ϭ 0.36, CHCl3). Ϫ TLC: Rf ϭ 0.45 (cyclo-
hexane/ethyl acetate, 80:20; detetermined by UV). Ϫ 1H NMR
(CDCl3): δ ϭ 0.95 (s, 3 H, CH3), 0.98 (s, 3 H, CH3), 1.10 (s, 3 H,
CH3), 1.45Ϫ1.54 (m,
1 H, CH2CH2), 1.92Ϫ2.02 (m, 2 H,
Rf ϭ 0.35) gave 0.475 g of 7b (68%) as colorless crystals, m.p. 59°C. CH2CH2), 2.98Ϫ3.06 (m, 1 H, CH2CH2), 3.16 (d, J ϭ 9.0 Hz, 1
20
1
Ϫ [α]D ϭ ϩ34.8 (c ϭ 0.48, CHCl3). Ϫ H NMR (CDCl3): δ ϭ
0.93 (s, 3 H, CH3), 0.96 (s, 3 H, CH3), 1.15 (s, 3 H, CH3), 1.69 CH2OCH3), 3.86 (d, J ϭ 16.9 Hz, 1 H, COCH2N3), 3.92 (d, J ϭ
(ddd, J ϭ 13.6/11.3/4.1 Hz, 1 H, CH2CH2), 1.93 (ddd, J ϭ 14.2/ 16.9 Hz, 1 H, COCH2N3), 6.99Ϫ7.01 (t, J ϭ 7.7 Hz, 2 H, aromatic
H, CH2OCH3), 3.29 (s, 3 H, CH2OCH3), 3.40 (d, J ϭ 9.0 Hz, 1 H,
9.6/4.1 Hz, 1 H, CH2CH2), 2.06 (ddd, J ϭ 13.6/9.6/7.7 Hz, 1 H, H), 7.15Ϫ7.18 (t, J ϭ 7.7 Hz, 1 H, aromatic p-H), 7.29Ϫ7.34 (t,
CH2CH2), 2.74 (ddd, J ϭ 14.2/11.3/7.7 Hz, 1 H, CH2CH2), 3.10
(d, J ϭ 9.4 Hz, 1 H, CH2OCH3), 3.36 (d, J ϭ 9.4 Hz, 1 H,
J ϭ 7.7 Hz, 2 H, aromatic H). Ϫ IR: ν˜ ϭ 2108 cmϪ1, 1760, 1748.
Ϫ MS (CH4; CI); m/z (%): 376 (7) [M ϩ Hϩ], 275 (100). Ϫ
CH2OCH3), 3.43 (s, 3 H, OCH3), 6.08 (s, br., 1 H, OH). Ϫ IR: ν˜ ϭ C19H25N3O5 (375.4): calcd. C 60.79, H 6.71, N 11.19; found C
3324 cmϪ1, 1779, 1520. Ϫ MS (CH4; CI); m/z (%): 383 (3) [M ϩ
Hϩ], 355 (100). Ϫ C17H19F5O4 (382.3): calcd. C 53.41, H 5.01;
found C 53.44, H 4.96.
60.72, H 6.79, N 11.17.
Pentafluorophenyl (1S,3R)-1-Azidoacetoxy-3-methoxymethyl-2,2,3-
trimethylcyclopentane-1-carboxylate (9b): A solution of 0.138 g
(0.30 mmol) of 8b in 5 mL of MeCN was treated with 0.042 g
(0.63 mmol) of NaN3 and stirred for 7 h at 70°C. MeCN was re-
moved in vacuo. The residue was taken up in 10 mL of Et2O and
washed with 5% NaHCO3 solution (3 ϫ 5 mL) and brine (2 ϫ
5 mL). The aqueous layers were reextracted with Et2O (2 ϫ 5 mL),
Phenyl (1S,3R)-1-Chloracetoxy-3-methoxymethyl-2,2,3-trimethyl-
cyclopentane-1-carboxylate (8a): A solution of 2.91 g (9.95 mmol)
of 7a in 8.0 mL (100 mmol, 10 equiv.) of chloroacetyl chloride was
stirred for 24 h at room temp. The chloroacetyl chloride was re-
moved in vacuo. The oily residue was dissolved in 50 mL of Et2O,
washed with sat. Na2CO3 solution (3 ϫ 5 mL) and with brine (2
ϫ 10 mL). The aqueous layers were reextracted with Et2O (2 ϫ
10 mL). The combined organic layers were dried (MgSO4) and con-
centrated to give 3.52 g (96%) of 8a as colorless crystals, m.p.
84Ϫ88°C. Ϫ [α]D20 ϭ ϩ46.3 (c ϭ 0.40, CHCl3). Ϫ TLC: Rf ϭ 0.38
(cyclohexane/ethyl acetate, 80:20; determined by UV). Ϫ 1H NMR
(CDCl3): δ ϭ 1.01 (s, 3 H, CH3), 1.05 (s, 3 H, CH3), 1.17 (s, 3 H,
the combined organic layers were dried (MgSO4) and concentrated
in vacuo to give 0.122 g (87%) of 9b as colorless oil. Ϫ [α]D
20
ϭ
1
ϩ30.8 (c ϭ 0.45, CHCl3). Ϫ H NMR (CDCl3): δ ϭ 0.93 (s, 3 H,
CH3), 0.98 (s, 3 H, CH3), 1.09 (s, 3 H, CH3), 1.51 (ddd, J ϭ 16.4/
5.6/5.2 Hz, 1 H, CH2CH2), 1.92Ϫ2.06 (m, 2 H, CH2CH2), 2.96
(ddd, J ϭ 20.4/5.5/5.2 Hz, 1 H, CH2CH2), 3.17 (d, J ϭ 9.2 Hz, 1
H, CH2OCH3), 3.29 (s, 3 H, OCH3), 3.46 (d, J ϭ 9.2 Hz, 1 H,
CH2OCH3), 4.04 (d, J ϭ 14.5 Hz, 1 H, COCH2N3), 4.09 (d, J ϭ
14.5 Hz, 1 H, COCH2N3). Ϫ IR: ν˜ ϭ 2109 cmϪ1, 1789, 1753, 1522.
Ϫ MS (CH4; CI); m/z (%): 466 (2) [M ϩ Hϩ], 181 (100). Ϫ
C19H20F5N3O5 (465.4): calcd. C 49.04, H 4.33, N 9.03; found C
49.32, H 4.24, N 8.76.
CH3), 1.57Ϫ1.63 (m,
1 H, CH2CH2), 1.97Ϫ2.05 (m, 2 H,
CH2CH2), 3.04Ϫ3.10 (m, 1 H, CH2CH2), 3.24 (d, J ϭ 9.0 Hz, 1
H, CH2OCH3), 3.36 (s, 3 H, CH2OCH3), 3.53 (d, J ϭ 9.0 Hz, 1 H,
CH2OCH3), 4.10 (d, J ϭ 14.5 Hz, 1 H, COCH2Cl), 4.14 (d, J ϭ
14.5 Hz, 1 H, COCH2Cl), 7.07 (d, J ϭ 7.7 Hz, 2 H, aromatic H),
7.24 (t, J ϭ 7.7 Hz, 1 H, aromatic H), 7.38 (t, J ϭ 7.7 Hz, 2 H,
aromatic H). Ϫ IR: ν˜ ϭ 1755 cmϪ1, 1756. Ϫ MS (CH4; CI); m/z
(%): 369 (7) [M ϩ Hϩ], 275 (100). Ϫ C19H25ClO5 (368.9): calcd. C
61.87, H 6.83; found C 61.85, H 6.85.
(1S,3R)-1-(N-Benzyloxycarbonylaminoacetoxy)-3-methoxy-
methyl-2,2,3-trimethylcyclopentane-1-carboxylic Acid (10): To a
solution of 0.548 g (2.62 mmol) of cbz-glycine in 5 mL of THF
0.424 g (2.62 mmol) of CDI, dissolved in 5 mL of THF, was added.
The resulting mixture was stirred for 30 min at room temp. Then a
solution of 0.283 g (1.31 mmol) of 1 in 5 mL of THF was added.
After stirring the mixture for 15 h at room temp., THF was re-
moved in vacuo. Purification by CC (petroleum ether/ethyl acetate/
AcOH, 10:3:1; Rf ϭ 0.22) gave 0.51 g (96%) of 10 as colorless oil.
Pentafluorophenyl
2,2,3-trimethylcyclopentane-1-carboxylate (8b):
(1S,3R)-1-Chloroacetoxy-3-methoxymethyl-
solution of
A
0.217 g (0.57 mmol) of 7b in 0.5 mL of chloroacetyl chloride was
stirred for 24 h at room temp. Chloroacetyl chloride was removed
in vacuo. The remaining oil was taken up in 10 mL of Et2O and
washed with a 5% NaHCO3 solution (5 ϫ 5 mL). The ethereal
phase was dried (MgSO4) and concentrated in vacuo to give 0.218 g
20
1
Ϫ [α]D ϭ ϩ89.4 (c ϭ 0.85, CHCl3). Ϫ H NMR (CDCl3): δ ϭ
0.88 (s, 3 H, CH3), 0.98 (s, 3 H, CH3), 1.00 (s, 3 H, CH3), 1.45Ϫ1.50
(m, 1 H, CH2CH2), 1.87Ϫ1.95 (m, 2 H, CH2CH2), 2.86Ϫ2.91 (m,
1 H, CH2CH2), 3.12 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.32 (s, 3 H,
OCH3), 3.41 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.97 (dd, J ϭ 18.4/
5.6 Hz, 1 H, COCH2NH), 4.03 (dd, J ϭ 18.4/5.6 Hz, 1 H,
COCH2NH), 5.11 (s, 2 H, OCH2C6H5), 5.45 (t, J ϭ 5.6 Hz, 1 H,
NH), 7.28Ϫ7.35 (m, 5 H, C6H5). Ϫ IR: ν˜ ϭ 3342 cmϪ1, 2970, 1723,
1530. Ϫ MS (CH4; CI); m/z (%): 408 (72) [M ϩ Hϩ], 280 (100). Ϫ
C21H29NO7 (407.5): calcd. C 61.90, H 7.17, N 3.44; found C 61.79,
H 7.27, N 3.45.
20
of 8b (84%) as colorless oil. Ϫ [α]D ϭ ϩ41.7 (c ϭ 0.41, CHCl3).
1
Ϫ H NMR (CDCl3): δ ϭ 1.00 (s, 3 H, CH3), 1.05 (s, 3 H, CH3),
1.16 (s, 3 H, CH3), 1.60 (ddd, J ϭ 10.6/5.6/5.3 Hz, 1 H, CH2CH2),
1.99Ϫ2.12 (m, 2 H, CH2CH2), 3.02 (ddd, J ϭ 14.8/6.0/5.3 Hz, 1 H,
CH2CH2), 3.24 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.36 (s, 3 H,
OCH3), 3.54 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 4.12 (d, J ϭ 14.5 Hz,
1 H, COCH2Cl), 4.16 (d, J ϭ 14.5 Hz, 1 H, COCH2Cl). Ϫ IR: ν˜ ϭ
1787 cmϪ1, 1743, 1523. Ϫ MS (CH4; CI); m/z (%): 460 (8) [M ϩ
Hϩ], 153 (100). Ϫ C19H20ClF5O5 (458.8): calcd. C 49.74, H 4.39;
found C 49.66, H 4.47.
Phenyl (1S,3R)-1-Azidoacetoxy-3-methoxymethyl-2,2,3-trimethyl-
(1S,3R)-1-Aminoacetoxy-3-methoxymethyl-2,2,3-trimethyl-
cyclopentane-1-carboxylate (9a): To
a
solution of 0.837 g cyclopentanecarboxylic Acid (11): To
a
solution of 0.490 g
Eur. J. Org. Chem. 1999, 1967Ϫ1978
1973

O. Achatz, A. Grandl, K. Th. Wanner
(1.20 mmol) of 10 in 20 mL of ethanol 0.024 g of Pd/C (10% Pd) (14) [Mϩ], 268 (18), 255 (60), 83 (100). Ϫ C15H25NO4 (283.4): calcd.
FULL PAPER
was added and the resulting heterogenous mixture was stirred for
15 h under H2 at room temp. The solid formed was dissolved by
addition of 5 mL of H2O and the solution was filtered to remove
Pd/C. The filtrate was concentrated in vacuo to give 0.287 g (88%)
of 11 as colorless crystals, m.p. 129°C. Ϫ [α]D20 ϭ ϩ81.8 (c ϭ 0.50,
H2O). Ϫ 1H NMR (D2O): δ ϭ 0.92 (s, 3 H, CH3), 0.97 (s, 3 H,
CH3), 1.02 (s, 3 H, CH3), 1.46Ϫ1.48 (m, 1 H, CH2-CH2), 1.83Ϫ1.94
(m, 2 H, CH2-CH2), 2.87Ϫ2.94 (m, 1 H, CH2CH2), 3.16 (d, J ϭ
9.0 Hz, 1 H, CH2OCH3), 3.38 (s, 3 H, OCH3), 3.45 (d, J ϭ 9.0 Hz,
1 H, CH2OCH3), 3.68 (d, J ϭ 17.0 Hz, 1 H, COCH2NH2), 3.82 (d,
J ϭ 17.0 Hz, 1 H, COCH2NH2). Ϫ IR: ν˜ ϭ 3423 cmϪ1, 2968, 1731,
1650. Ϫ MS (CH4; CI); m/z (%): 274 (2) [M ϩ Hϩ], 256 (100). Ϫ
C13H23NO5 (273.3): calcd. C 57.14, H 8.48, N 5.14; found C 57.28,
H 8.69, N 5.18.
C 63.58, H 8.89, N 4.94; found C 63.51, H 8.98, N 4.92.
14a: A small amount (11.0 mg, 0.038 mmol) of 13a in 1 mL of
DME was treated with 30 µL (0.039 mmol, 1 equiv.) of sBuLi at
Ϫ55°C. After stirring for 20 min, quenching at Ϫ55°C, workup
and purification by CC, as described for the alkylation reaction,
10.0 mg (91%) of 13a and 14a were obtained as a diastereomeric
mixture (d.s. ϭ 48:52). In the 1H-NMR spectrum the following
1
signals for 14a were observed. Ϫ H NMR (CDCl3): δ ϭ 0.90 (s,
3 H, CH3), 0.96 (s, 3 H, CH3), 1.00 (s, 3 H, CH3), 1.39Ϫ1.50 (m,
1 H, CH2CH2), 1.56 (d, J ϭ 7.6 Hz, 3 H, CH3CH), 1.88Ϫ2.04 (m,
2 H, CH2CH2), 2.54Ϫ2.68 (m, 1 H, CH2CH2), 3.17 (d, J ϭ 9.0 Hz,
1 H, CH2OCH3), 3.32 (s, 3 H, CH2OCH3), 3.42 (d, J ϭ 9.0 Hz, 1
H, CH2OCH3), 3.67 (s, 3 H, CH3OCϭN), 4.37 (q, J ϭ 7.6 Hz, 1
H, CH3CH).
Alkylation Reactions with Monofunctional Alkyl Halides
15a: TLC: Rf ϭ 0.51 (n-heptane/ethyl acetate, 70:30; detected with
cerium molybdate solution). Ϫ H NMR (CDCl3): δ ϭ 0.90 (s, 3
1
General Procedure 1 (GP 1). ؊ Alkylation Reactions of 5: A solu-
tion of 1.0 equiv. of 5 in DME was cooled to Ϫ55°C and treated
with 1.0 equiv. of sBuLi (c ϭ 1.3 mol/l, n-heptane/cyclohexane,
90:10). After stirring for 15 min at Ϫ55°C, a solution of 2.0Ϫ3.0
equiv. of alkyl halide in DME was added. The mixture was stirred
for 16 h at Ϫ55°C and then either quenched at this temperature
(and warmed to room temp.) or after it had been slowly warmed
to room temp. The quenching was performed by addition of 1.0 mL
of phospate buffer (pH ϭ 7, c ϭ 1.0 mol/L). Then the reaction
mixture was concentrated in vacuo and the residue was dissolved
in Et2O (10Ϫ20 mL). The organic layer was washed with brine. The
aqueous layers were extracted with Et2O (2 ϫ 5Ϫ10 mL), and the
combined organic layers were washed with brine (2 ϫ 3Ϫ5 mL),
dried (MgSO4), and concentrated. The resulting residue was puri-
fied by CC.
H, CH3), 0.91 (s, 3 H, CH3), 1.00 (s, 3 H, CH3), 1.40Ϫ1.48 (m, 1
H, CH2CH2), 1.48 (s, 3 H, CH3), 1.49 (s, 3 H, CH3), 1.89Ϫ2.02 (m,
2 H, CH2CH2), 2.52Ϫ2.62 (m, 1 H, CH2CH2), 3.17 (d, J ϭ 9.0 Hz,
1 H, CH2OCH3), 3.33 (s, 3 H, CH2OCH3), 3.44 (d, J ϭ 9.0 Hz, 1
H, CH2OCH3), 3.66 (s, 3 H, CH3OCϭN). Ϫ IR: ν˜ ϭ 1741 cmϪ1
1688. Ϫ MS (CH4; CI); m/z (%): 298 (100) [M ϩ Hϩ], 270 (80).
,
(2R,5S,8R)-8-Allyl-10-methoxy-2-methoxymethyl-1,1,2-tri-
methyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (13b), (2R,5S,8S)-8-Allyl-
10-methoxy-2-methoxymethyl-1,1,2-trimethyl-6-oxa-9-azaspiro-
[4.5]dec-9-en-7-one (14b) and (2R,5S)-8,8-Diallyl-10-methoxy-2-
methoxymethyl-1,1,2-trimethyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one
(15b): According to GP 1 from 69.3 mg (0.257 mmol) of 5 in 7 mL
of DME with 200 µL (0.260 mmol, 1.0 equiv.) of sBuLi and 63 mg
(45 µL, 0.52 mmol, 2.0 equiv.) of allyl bromide; quenching at
1
Ϫ55°C. The H-NMR spectrum of the crude product showed that
General Procedure 2 (GP 2). ؊ Alkylation Reactions of 13 and 14:
The reactions were performed in analogy to the alkylation reactions
of 5 employing a mixture of 13 and 14 instead of 5.
the mixture consisted of 94% of 13b and 14b and 6% of 15b. Purifi-
cation by CC (n-heptane/ethyl acetate, 85:15) yielded 48.2 mg (61%)
of 13b and 14b (d.s. ϭ 99.2:0.8) and 5.0 mg (5%) of 15b. The mix-
ture of the diastereomeric compounds 13b and 14b was separated
by prep. HPLC (n-heptane/ethyl acetate, 85:15; 13.5 mL/min). The
diastereoselectivity was determined by analytical HPLC (n-hep-
tane/ethyl acetate, 85:15; 1.0 mL/min) from the crude product:
(2R,5S,8R)-10-Methoxy-2-methoxymethyl-1,1,2,8-tetramethyl-6-
oxa-9-azaspiro[4.5]dec-9-en-7-one (13a), (2R,5S,8S)-10-Methoxy-2-
methoxymethyl-1,1,2,8-tetramethyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-
one (14a), and (2R,5S)-10-Methoxy-2-methoxymethyl-1,1,2,8,8-
pentamethyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (15a): According
to GP 1 from 133 mg (0.494 mmol) of 5 in 8 mL of DME, with
0.38 mL (0.494 mmol, 1.0 equiv.) of sBuLi and 141 mg (62 µL,
0.992 mmol, 2.0 equiv.) of MeI; quenching at Ϫ55°C. The 1H-
NMR spectrum of the crude product showed that the mixture con-
sisted of 93% of 13a and 14a (d.s. > 99:1), 3% of 15a and 4% of
unchanged 5. Purification by CC (n-heptane/ethyl acetate, 80:20)
afforded 94.0 mg (67%) of 13a and 14a and 1.8 mg (3%) of 15a as
colorless oils. The major diastereomer 13a was purified by prepara-
tive HPLC (n-heptane/ethyl acetate, 85:15; 13.5 mL/min). Analyti-
cal HPLC gave only an incomplete resolution of the two dia-
stereomers (n-heptane/ethyl acetate, 85:15; 1.0 mL/min; major dia-
d.s. ϭ 99.2:0.8; 13b: tR ϭ 6.7 min; 14b: tR ϭ 8.7 min; 15b: tR
ϭ
5.7 min.
13b: Colorless oil. Ϫ [α]D20 ϭ ϩ150.7 (c ϭ 0.76 in CHCl3). Ϫ TLC:
Rf ϭ 0.13 (n-heptane/ethyl acetate, 85:15; detected with cerium
molybdate solution). Ϫ 1H NMR (CDCl3): δ ϭ 0.90 (s, 3 H, CH3),
0.94 (s, 3 H, CH3), 0.99 (s, 3 H, CH3), 1.47 (ddd, J ϭ 12.4/9.2/
6.9 Hz, 1 H, CH2CH2), 1.92 (ddd, J ϭ 14.0/8.2/6.9 Hz, 1 H,
CH2CH2), 2.01 (ddd, J ϭ 12.4/8.2/5.6 Hz, 1 H, CH2CH2), 2.61 (dtt,
J ϭ 14.0/7.0/1.2 Hz, 1 H, CH2CHϭCH2), 2.63 (ddd, J ϭ 14.0/9.2/
5.6 Hz, 1 H, CH2CH2), 2.77 (dddt, J ϭ 14.0/7.0/4.6/1.2 Hz, 1 H,
CH2CHϭCH2), 3.17 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3
H, CH2OCH3), 3.45 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.72 (s, 3 H,
CH3OCϭN), 4.14 (dd, J ϭ 7.0/4.6 Hz, 1 H, CHCH2CHϭCH2),
5.10 (ddt, J ϭ 10.1/2.2/1.2 Hz, 1 H, CH2CHϭCH2), 5.17 (ddt, J ϭ
17.1/2.2/1.2 Hz, 1 H, CH2CHϭCH2), 5.88 (ddt, J ϭ 17.1/10.1/
7.0 Hz, 1 H, CH2CHϭCH2). Ϫ IR: ν˜ ϭ 1743 cmϪ1, 1683. Ϫ MS
(CH4; CI); m/z (%): 310 (100) [M ϩ Hϩ]. Ϫ C17H27NO4 (309.4):
calcd. C 65.99, H 8.80, N 4.53; found C 65.87, H 8.70, N 4.39.
1
stereomer 13a: tR ϭ 10.7 min; minor diastereomer 14a: tR
ϭ
11.0 min).
20
13a: [α]D ϭ ϩ175.5 (c ϭ 0.92 in CHCl3). Ϫ TLC: Rf ϭ 0.32 (n-
heptane/ethyl acetate, 70:30; detected with cerium molybdate solu-
1
tion). Ϫ H NMR (CDCl3): δ ϭ 0.91 (s, 3 H, CH3), 0.94 (s, 3 H,
CH3), 0.98 (s, 3 H, CH3), 1.42Ϫ1.50 (m, 1 H, CH2CH2), 1.53 (d,
J ϭ 7.2 Hz, 3 H, CH3CH), 1.88Ϫ2.04 (m, 2 H, CH2CH2), 14b: Colorless oil. Ϫ H NMR (CDCl3): δ ϭ 0.91 (s, 6 H, CH3),
2.59Ϫ2.67 (m, 1 H, CH2CH2), 3.16 (d, J ϭ 9.0 Hz, 1 H, 1.01 (s, 3 H, CH3), 1.41Ϫ1.50 (m, 1 H, CH2CH2), 1.91Ϫ2.01 (m,
CH2OCH3), 3.32 (s, 3 H, CH2OCH3), 3.43 (d, J ϭ 9.0 Hz, 1 H, 2 H, CH2CH2), 2.39 (dddt, J ϭ 14.0/9.3/7.0/1.2 Hz, 1 H, CH2CHϭ
CH2OCH3), 3.70 (s, 3 H, CH3OCϭN), 4.12 (q, J ϭ 7.2 Hz, 1 H, CH2), 2.54Ϫ2.63 (m, 1 H, CH2CH2), 2.79 (dddt, J ϭ 14.0/7.0/5.4/
CH3CH). Ϫ IR: ν˜ ϭ 1746 cmϪ1, 1683. Ϫ MS (70 eV); m/z (%): 283 1.2 Hz, 1 H, CH2CHϭCH2), 3.16 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3),
1974
Eur. J. Org. Chem. 1999, 1967Ϫ1978

Preparation of Enantiomerically Pure α-Quaternary α-Amino Acids
FULL PAPER
3.34 (s, 3 H, CH2OCH3), 3.42 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), δ ϭ 0.01 (s, 3 H, CH3), 0.28 (s, 3 H, CH3), 0.44Ϫ0.51 (m, 1 H,
3.70 (s, 3 H, CH3OCϭN), 4.32 (dd, J ϭ 9.3/5.4 Hz, 1 H, CH2CH2), 0.77 (s, 3 H, CH3), 1.01Ϫ1.08 (m, 1 H, CH2CH2),
CHCH2CHϭCH2), 5.13Ϫ5.22 (m, 2 H, CH2CHϭCH2), 6.00 (ddt, 1.40Ϫ1.55 (m, 2 H, CH2CH2), 2.87 (d, J ϭ 9.0 Hz, 1 H,
J ϭ 17.0/10.1/7.0 Hz, 1 H, CH2CHϭCH2). Ϫ IR: ν˜ ϭ 1745 cmϪ1
1686. Ϫ MS (CH4; CI); m/z (%): 310 (100) [M ϩ Hϩ].
,
CH2OCH3), 2.96 (d, J ϭ 12.5 Hz, 1 H, CH2Ph), 3.07 (d, J ϭ
12.9 Hz, 1 H, CH2Ph), 3.08 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.20
(s, 3 H, CH2OCH3), 3.36 (d, J ϭ 12.9 Hz, 1 H, CH2Ph), 3.58 (d,
J ϭ 12.5 Hz, 1 H, CH2Ph), 3.77 (s, 3 H, CH3OCϭN), 7.08Ϫ7.26
(m, 10 H, aromatic H). Ϫ IR: ν˜ ϭ 1734 cmϪ1, 1691. Ϫ MS (70 eV);
m/z (%): 449 (15) [Mϩ], 358 (100). Ϫ C28H35NO4 (449.6): calcd. C
74.80, H 7.85, N 3.12; found C 74.74, H 7.90, N 3.13.
15b: Colorless oil. Ϫ TLC: Rf ϭ 0.51 (n-heptane/ethyl acetate,
70:30; detected with cerium molybdate solution). Ϫ 1H NMR
(CDCl3): δ ϭ 0.90 (s, 3 H, CH3), 0.93 (s, 3 H, CH3), 1.02 (s, 3 H,
CH3), 1.40Ϫ1.48 (m,
1 H, CH2CH2), 1.83Ϫ1.98 (m, 2 H,
CH2CH2), 2.37Ϫ2.66 (m, 1 H, CH2CH2, 4 H, 2 ϫ CH2CHϭCH2),
3.16 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.32 (s, 3 H, CH2OCH3), (2R,5S,8R)-8-Benzyl-10-methoxy-2-methoxymethyl-1,1,2,8-tetra-
3.39 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.68 (s, 3 H, CH3OCϭN), methyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (17a): According to GP
5.03Ϫ5.18 (m, 4 H, 2 ϫ CH2CHϭCH2), 5.56 (ddt, 1 H, J ϭ 16.0/ 2 from 46.4 mg (0.164 mmol) of 13a and 14a in 5 mL of DME,
11.5/7.3 Hz, CH2CHϭCH2), 5.85 (ddt, 1 H, J ϭ 16.9/10.4/7.3 Hz,
with 190 µL (0.247 mmol, 1.5 equiv.) of sBuLi and 56 mg (40 µL,
CH2CHϭCH2). Ϫ IR: ν˜ ϭ 1740 cmϪ1, 1691. Ϫ MS (CH4; CI); m/z 0.33 mmol, 2.0 equiv.) of PhCH2Br; quench at Ϫ55°C. Purification
(%): 298 (100) [Mϩ].
by CC (n-heptane/ethyl acetate, 85:15) yielded 33.5 mg (55%) of
1
17a (d.s. > 99:1, 17a/17c). In the H-NMR spectrum of the crude
(2R,5S,8R)-8-Benzyl-10-methoxy-2-methoxymethyl-1,1,2-tri-
methyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (13c), (2R,5S,8S)-8-Benzyl-
10-methoxy-2-methoxymethyl-1,1,2-trimethyl-6-oxa-9-azaspiro-
[4.5]dec-9-en-7-one (14c) and (2R,5S)-8,8-Dibenzyl-10-methoxy-2-
methoxymethyl-1,1,2-trimethyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one
(15c): According to GP 1 from 64.5 mg (0.240 mmol) of 5 in 6 mL
of DME, with 180 µL (0.234 mmol, 1.0 equiv.) of sBuLi and 84 mg
product no signals of the minor diasteromer 17c could be detected.
17a: Colorless oil. Ϫ [α]D20 ϭ ϩ63.0 (c ϭ 0.135 in CHCl3). Ϫ TLC:
Rf ϭ 0.35 (n-heptane/ethyl acetate, 85:15; detected by UV, 254 nm,
and with cerium molybdate solution). Ϫ HPLC: tR ϭ 13.4 min (n-
1
heptane/ethyl acetate, 90:10; 0.75 mL/min). Ϫ H NMR (CDCl3):
δ ϭ 0.83 (s, 3 H, CH3), 0.84 (s, 3 H, CH3), 0.93 (s, 3 H, CH3), 1.04
(60 µL, 0.49 mmol, 2.0 equiv.) of PhCH2Br; quenching at room (ddd, J ϭ 14.3/8.1/7.0 Hz, 1 H, CH2CH2), 1.27 (ddd, J ϭ 13.0/8.6/
temp. The 1H-NMR spectrum of the crude product showed that
the mixture consisted of 95% of 13c and 14c, 1% of 15c and 4% of
7.0 Hz, 1 H, CH2CH2), 1.62 (s, 3 H, CH3CCH2Ph), 1.75 (ddd, J ϭ
13.0/8.1/7.0 Hz, 1 H, CH2CH2), 1.96 (ddd, J ϭ 14.3/8.6/7.0 Hz, 1
unchanged 5. Purification by CC (n-heptane/ethyl acetate, 75:25) H, CH2CH2), 2.89 (d, J ϭ 12.7 Hz, 1 H, CH2Ph), 3.11 (d, J ϭ
afforded 59.6 mg (76%) of 13c and 14c (d.s. ϭ 98.4:1.6) and 0.2 mg 9.0 Hz, 1 H, CH2OCH3), 3.26 (d, J ϭ 12.7 Hz, 1 H, CH2Ph), 3.28
(2%) of 15c. The diastereomeric mixture containing 13c and 14c (s, 3 H, CH2OCH3), 3.31 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.70 (s,
was separated by prep. HPLC (n-heptane/ethyl acetate, 85:15;
3 H, CH3OCϭN), 7.08Ϫ7.13 (m, 2 H, aromatic H), 7.15Ϫ7.25 (m,
13.5 mL/min). The diastereoselectivity was determined by analyti-
3 H, aromatic H). Ϫ IR: ν˜ ϭ 1738 cmϪ1, 1693, 701. Ϫ MS (CH4;
cal HPLC (n-heptane/ethyl acetate, 90:10; 1.0 mL/min) from the CI); m/z (%) ϭ 374 (100) [M ϩ Hϩ], 346 (72). Ϫ C22H31NO4
crude product: d.s. ϭ 98.4:1.6; 13c: tR ϭ 13.6 min; 14c: tR
20.0 min; 15c: tR ϭ 8.0 min.
ϭ
(373.5): calcd. C 70.75, H 8.37, N 3.75; found C 70.70, H 8.58,
N 3.59.
20
13c: Colorless oil. Ϫ [α]D ϭ ϩ92.2 (c ϭ 1.01 in CHCl3). Ϫ TLC:
(2R,5S,8R)-8-Allyl-8-benzyl-10-methoxy-2-methoxymethyl-1,1,2-
Rf ϭ 0.48 (cyclohexane/ethyl acetate, 70:30; detected by UV). Ϫ 1H trmethyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (17b): According to
NMR (CDCl3): δ ϭ 0.85 (s, 3 H, CH3), 0.91 (s, 3 H, CH3), 0.96 (s,
GP 2 from 39.3 mg (0.127 mmol) of 13b and 14b in 2 mL of DME,
3 H, CH3), 1.43 (ddd, J ϭ 13.1/9.2/7.2 Hz, 1 H, CH2CH2), 1.76 with 150 µL (0.195, 1.5 equiv.) of sBuLi and 70 mg (50 µL,
(ddd, J ϭ 14.5/8.5/7.2 Hz, 1 H, CH2CH2), 1.95 (ddd, J ϭ 13.1/
8.5/6.2 Hz, 1 H, CH2CH2), 2.50 (ddd, J ϭ 14.5/9.2/6.2 Hz, 1 H,
0.40 mmol, 3.0 equiv.) of PhCH2Br; quenching at Ϫ55°C. Purifi-
cation by CC (n-heptane/ethyl acetate, 90:10) yielded 42.9 mg (85%)
CH2CH2), 3.11 (dd, J ϭ 13.5/7.5 Hz, 1 H, CH2Ph), 3.16 (d, J ϭ of 17b (d.s. > 98:2, 17b/17d). In the 1H-NMR spectrum of the crude
9.0 Hz, 1 H, CH2OCH3), 3.32 (s, 3 H, CH2OCH3), 3.37 (dd, J ϭ product no signals of the minor diastereomer 17d could be de-
13.5/4.3 Hz, 1 H, CH2Ph), 3.43 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3),
3.68 (s, 3 H, CH3OCϭN), 4.31 (dd, J ϭ 7.5/4.3 Hz, 1 H,
CHCH2Ph), 7.18Ϫ7.32 (m, 5 H, aromatic H). Ϫ IR: ν˜ ϭ 1745
cmϪ1, 1688, 701. Ϫ MS (CH4; CI); m/z (%): 360 (100) [M ϩ Hϩ].
Ϫ C21H29NO4 (359.5): calcd. C 70.17, H 8.13, N 3.90; found C
70.20, H 8.09, N 3.91.
tected.
20
17b: Colorless oil. Ϫ [α]D ϭ ϩ30.0 ( c ϭ 1.0 in CHCl3). Ϫ TLC:
Rf ϭ 0.39 (n-heptane/ethyl acetate, 70:30; deteced by UV, 254 nm,
and with cerium molybdate solution). Ϫ HPLC: tR ϭ 6.90 min (n-
heptane/ethyl acetate, 90:10; 1.0 mL/min). Ϫ 1H NMR (CDCl3):
δ ϭ 0.63Ϫ0.74 (m, 1 H, CH2CH2), 0.80 (s, 3 H, CH3), 0.84 (s, 3
H, CH3), 0.91 (s, 3 H, CH3), 1.12Ϫ1.21 (m, 1 H, CH2CH2),
1.58Ϫ1.71 (m, 2 H, CH2CH2), 2.62 (dd, J ϭ 13.4/7.4 Hz, 1 H,
CH2CHϭCH2), 2.82 (dd, J ϭ 13.4/7.3 Hz, 1 H, CH2CHϭCH2),
20
14c: Colorless oil. Ϫ [α]D ϭ ϩ55.1 (c ϭ 0.16 in CHCl3). Ϫ 1H
NMR (CDCl3): δ ϭ 0.69 (s, 3 H, CH3), 0.80 (s, 3 H, CH3), 0.98 (s,
3 H, CH3), 1.40Ϫ1.48 (m, 1 H, CH2CH2), 1.89Ϫ2.00 (m, 2 H,
CH2CH2), 2.50Ϫ2.59 (m, 1 H, CH2CH2), 2.98 (dd, J ϭ 13.5/ 2.88 (d, J ϭ 12.6 Hz, 1 H, CH2Ph), 3.05 (d, J ϭ 9.0 Hz, 1 H,
8.8 Hz, 1 H, CH2Ph), 3.12 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.32 CH2OCH3), 3.19 (d, J ϭ 12.6 Hz, 1 H, CH2Ph), 3.21 (d, J ϭ
(s, 3 H, CH2OCH3), 3.39 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.40 9.0 Hz, 1 H, CH2OCH3), 3.26 (s, 3 H, CH2OCH3), 3.71 (s, 3 H,
(dd, J ϭ 13.5/4.3 Hz, 1 H, CH2Ph), 3.66 (s, 3 H, CH3OCϭN), 4.52
CH3OCϭN), 5.12Ϫ5.22 (m, 2 H, CH2CHϭCH2), 5.83 (ddt, J ϭ
(dd, J ϭ 8.8/4.3 Hz, 1 H, CHCH2Ph), 7.20Ϫ7.33 (m, 5 H, aromatic 17.3/10.0/7.3 Hz, 1 H, CH2CHϭCH2), 7.04Ϫ7.08 (m, 2 H, aro-
H). Ϫ IR: ν˜ ϭ 1740 cmϪ1, 1685, 702. Ϫ MS (CH4; CI); m/z (%):
360 (100) [M ϩ Hϩ].
matic H), 7.14Ϫ7.24 (m, 3 H, aromatic H). Ϫ IR: ν˜ ϭ 1733 cmϪ1
,
1689, 702. Ϫ MS (CH4; CI); m/z (%) ϭ 400 (100) [M ϩ Hϩ], 372
(53). Ϫ C24H33NO4 (399.5): calcd. C 72.15, H 8.33, N 3.51; found
C 72.10, H 8.40, N 3.48.
15c: A sufficient quantity for characterization was obtained by
combining the samples of several experiments: Colorless oil. Ϫ
20
[α]D ϭ ϩ37.0 (c ϭ 0.57 in CHCl3). Ϫ TLC: Rf ϭ 0.55 (cyclohex-
(2R,5S,8S)-8-Benzyl-10-methoxy-2-methoxymethyl-1,1,2,8-tetra-
ane/ethyl acetate, 70:30; detected by UV). Ϫ 1H NMR (CDCl3): methyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (17c): According to GP
Eur. J. Org. Chem. 1999, 1967Ϫ1978
1975

O. Achatz, A. Grandl, K. Th. Wanner
FULL PAPER
2 from a solution of 59.0 mg (0.164 mmol) of 13c and 14c in 6 mL
of DME, with 250 µL (0.325 mmol, 2.0 equiv.) of sBuLi and 48 mg aromatic H) (the H-NMR-spectroscopic data are in good accord-
H, CH2), 7.12Ϫ7.16 (m, 2 H, aromatic H), 7.22Ϫ7.30 (m, 3 H,
1
(21 µL, 0.34 mmol, 2.0 equiv.) of MeI; quenching at Ϫ55°C. Purifi-
ance with the data given in ref.[10a,10c]; the chemical shifts, given in
cation by CC (n-heptane/ethyl acetate, 85:15) yielded 47.6 mg (78%) ref.[10b], however, showed a difference from the former for each sig-
of 17c (d.s. > 99:1, 17c/17a). In the 1H-NMR spectrum of the crude
product no signals of the minor diastereomer 17a could be de-
tected.
nal of 0.16 ppm). Ϫ IR: ν˜ ϭ 1651 cmϪ1. Ϫ MS (CH4; CI); m/z
(%) ϭ 180 (100) [M ϩ Hϩ]. Ϫ The above-mentioned CH2Cl2 ex-
tracts were combined, dried (MgSO4), and concentrated. The re-
sulting residue was purified by CC (n-heptane/ethyl acetate/AcOH,
65:25:10). This gave 15.0 mg (35%) of 18c as a colorless oil. Ϫ TLC:
Rf ϭ 0.55 (2-propanol/H2O, 95:5; detected by UV, 254 nm, and
20
17c: Colorless oil. Ϫ [α]D ϭ ϩ73.5 (c ϭ 1.51 in CHCl3). Ϫ TLC:
Rf ϭ 0.33 (n-heptane/ethyl acetate, 85:15; detected by UV, 254 nm,
and with cerium molybdate solution). Ϫ HPLC: tR ϭ 9.5 min (n-
heptane/ethyl acetate, 90:10; 1.0 mL/min). Ϫ 1H NMR (CDCl3):
δ ϭ 0.26 (s, 3 H, CH3), 0.53 (s, 3 H, CH3), 0.91 (s, 3 H, CH3),
1.35Ϫ1.42 (m, 1 H, CH2CH2), 1.49 (s, 3 H, CH3CCH2Ph),
1.83Ϫ1.94 (m, 2 H, CH2CH2), 2.83Ϫ2.47 (m, 1 H, CH2CH2), 2.97
(d, J ϭ 13.1 Hz, 1 H, CH2Ph), 3.03 (d, J ϭ 9.0 Hz, 1 H,
CH2OCH3), 3.28 (s, 3 H, CH2OCH3), 3.29 (d, J ϭ 13.1 Hz, 1 H,
CH2Ph), 3.33 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.70 (s, 3 H,
CH3OCϭN), 7.16Ϫ7.24 (m, 5 H, aromatic H). Ϫ IR: ν˜ ϭ 1741
cmϪ1, 1689. Ϫ MS (CH4; CI); m/z (%) ϭ 374 (87) [M ϩ Hϩ], 346
(100). Ϫ C22H31NO4 (373.5): calcd. C 70.75, H 8.37, N 3.75; found
C 70.75, H 8.31, N 3.80.
1
with cerium molybdate solution). Ϫ H NMR (CDCl3): δ ϭ 0.74
(s, 3 H, CH3), 0.85 (s, 6 H, CH3), 1.46 (s, 3 H, CH3CCH2Ph),
1.59Ϫ1.69 (m, 2 H, CH2CH2), 1.91Ϫ2.03 (m, 1 H, CH2CH2),
2.65Ϫ2.77 (m, 1 H, CH2CH2), 3.03 (d, J ϭ 9.5 Hz, 1 H,
CH2OCH3), 3.14 (d, J ϭ 13.8 Hz, 1 H, CH2Ph), 3.24 (d, J ϭ
9.5 Hz, 1 H, CH2OCH3), 3.39 (s, 3 H, CH2OCH3), 3.47 (d, J ϭ
13.8 Hz, 1 H, CH2Ph), 6.05 (s, 1 H, NH), 7.16Ϫ7.19 (m, 2 H, aro-
matic H), 7.23Ϫ7.33 (m, 3 H, aromatic H). Ϫ IR: ν˜ ϭ 1737 cmϪ1
,
1648, 1517. Ϫ MS (CH4; CI); m/z (%) ϭ 378 (100) [M ϩ Hϩ]. Ϫ
C21H31NO5 (377.5): calcd. C 66.82 H 8.28 N 3.71; found C 67.01
H 8.36 N 3.49.
(S)-2-Amino-2-benzyl-4-pentenoic Acid (19d): The preparation was
performed in analogy to the procedure described for 19c the reac-
tion time being increased to 12 d. From 33.7 mg of 17d 8.3 mg
(48%) of 19d was obtained as colorless crystals, m.p. 220°C (dec.).
(2R,5S,8S)-8-Allyl-8-benzyl-10-methoxy-2-methoxymethyl-1,1,2-tri-
methyl-6-oxa-9-azaspiro[4.5]dec-9-en-7-one (17d): According to GP
2 from 56.6 mg (0.158 mmol) of 13c and 14c in 6 mL of DME,
with 180 µL (0.234 mmol, 1.5 equiv.) of sBuLi and 39 mg (27 µL,
0.32 mmol, 2.0 equiv.) of allyl bromide. Purification by CC (n-hep-
tane/ethyl acetate, 80:20) yielded 55.4 mg (88%) of 17d (d.s. > 99:1,
17d/17b). In the 1H-NMR spectrum of the crude product no signals
of the minor diastereomer 17b could be detected.
20
20
Ϫ [α]D ϭ ϩ26 (c ϭ 0.11, H2O) {ref.[10b] [α]D ϭ ϩ27.3 (c ϭ 1,
H2O)}. Ϫ 1H NMR (D2O): δ ϭ 2.38 (dd, J ϭ 14.6/8.8 Hz, 1 H,
CH2CH), 2.70 (dd, J ϭ 14.6/6.3 Hz, 1 H, CH2CH), 2.89 (d, J ϭ
14.4 Hz, 1 H, CH2Ph), 3.22 (d, J ϭ 14.4 Hz, 1 H, CH2Ph),
5.12Ϫ5.18 (m, 2 H, CH2ϭCH), 5.65 (dddd, J ϭ 15.9/11.0/8.8/
6.3 Hz, 1 H, CH2CHϭCH2), 7.14Ϫ7.18 (m, 2 H, aromatic H),
7.22Ϫ7.31 (m, 3 H, aromatic H) (as described for 19c the 1H-NMR
signals given in ref.[10b] showed a difference in chemical shifts of
0.16 ppm from those reported above). Ϫ IR: ν˜ ϭ 1627 cmϪ1. Ϫ
MS (CH4; CI); m/z (%) ϭ 206 (100) [M ϩ Hϩ].
17d: Colorless crystals, m.p. 66Ϫ67°C. Ϫ [α]D20 ϭ ϩ59.8 (c ϭ 0.47
in CHCl3). Ϫ TLC: Rf ϭ 0.48 (n-heptane/ethyl acetate, 70:30; de-
tected by UV, 254 nm, and with cerium molybdate solution). Ϫ
HPLC: tR ϭ 7.04 min (n-heptane/ethyl acetate, 90:10; 1.0 mL/min).
1
Ϫ H NMR (CDCl3): δ ϭ 0.16 (s, 3 H, CH3), 0.46 (s, 3 H, CH3),
0.90 (s, 3 H, CH3), 1.31Ϫ1.39 (m, 1 H, CH2CH2), 1.77Ϫ1.90 (m,
2 H, CH2CH2), 2.32Ϫ2.40 (m, 1 H, CH2CH2), 2.43 (dd, J ϭ 13.0/
7.5 Hz, 1 H, CH2CHϭCH2), 2.73 (dd, J ϭ 13.0/7.5 Hz, 1 H,
Synthesis of Aminocyclopropanecarboxylic Acids
(1R,3R,6S,8R)-1-Hydroxymethyl-11-methoxy-8-methoxymethyl-7,7,8-
CH2CHϭCH2), 2.99 (d, J ϭ 13.0 Hz, 1 H, CH2Ph), 3.01 (d, J ϭ trimethyl-5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-4-one (20a): To a
9.0 Hz, 1 H, CH2OCH3), 3.27 (s, 3 H, CH2OCH3), 3.31 (d, J ϭ solution of 0.21 g (0.78 mmol) of 5 in 8 mL of THF, 1.64 mL
9.0 Hz, 1 H, CH2OCH3), 3.37 (d, J ϭ 13.0 Hz, 1 H, CH2Ph), 3.78 (1.64 mmol, 2.1 equiv.) of a solution of NaN(SiMe3)2 (c ϭ 1.0 mol/
(s, 3 H, CH3OCϭN), 5.06Ϫ5.13 (m, 2 H, CH2CHϭCH2), 5.59
(ddt, J ϭ 16.2/11.0/7.5 Hz, 1 H, CH2CHϭCH2), 7.16Ϫ7.26 (m, 5 drin was added at Ϫ10°C. The mixture was stirred at Ϫ10°C for
H, aromatic H). Ϫ IR: ν˜ ϭ 1726 cmϪ1, 1686, 700. Ϫ MS (CH4;
24 h. After addition of 5 mL of phosphate buffer (pH ϭ 7, c ϭ 1
CI); m/z (%) ϭ 400 (99) [M ϩ Hϩ], 372 (100). Ϫ C24H33NO4 mol/L), THF was removed in vacuo, and the remaining aqueous
L in THF) and 0.29 g (3.12 mmol, 4 equiv.) of (S)-(ϩ)-epichlorohy-
(399.5): calcd. C 72.15, H 8.33, N 3.51; found C 72.01, H 8.51,
N 3.54.
phase was extracted with CH2Cl2 (3 ϫ 5 mL). The combined or-
ganic layers were dried (MgSO4), and concentrated in vacuo. CC
(petroleum ether/ethyl acetate, 7:3, Rf ϭ 0.20) gave 0.17 g (69%) of
(S)-2-Amino-2-methyl-3-phenylpropanoic Acid (19c) and (2S)-2-N-
[(1S,3R)-1-Hydroxy-3-methoxymethyl-2,2,3-trimethylcyclopentyl-1-
carbonyl]amino-2-methyl-3-phenylpropanoic Acid (18c): To a solu-
tion of 42.8 mg of 17c in 0.1 mL of THF 0.3 mL of NaOH (40%)
was added. After stirring for 8 d, the mixture was diluted with
2 mL of H2O and washed with CH2Cl2 (2 ϫ 1 mL). The organic
layers were reextracted with H2O (2 ϫ 2 mL). The combined aque-
ous phases were acidified with 2  HCl and after washing with
CH2Cl2 (2 ϫ 1 mL) concentrated in vacuo. The residue was dis-
solved in a minimum amount of H2O and subjected to ion ex-
change chromatography (Dowex 50 W X 8 cation exchange resin,
Hϩ form, elution with H2O until the eluent was neutral and chlo-
20
20a as colorless crystals, m.p. 68°C. Ϫ [α]D ϭ Ϫ35.8 (c ϭ 0.65,
1
CHCl3). Ϫ H NMR (CDCl3): δ ϭ 0.89 (s, 3 H, CH3), 0.96 (s, 3
H, CH3), 0.99 (s, 3 H, CH3), 1.46Ϫ1.53 (m, 1 H, CH2CH2), 1.56
(dd, J ϭ 7.7/4.3 Hz, 1 H, CH2CH), 1.95 (dd, J ϭ 9.8/4.3 Hz, 1 H,
CH2CH), 1.99Ϫ2.08 (m, 3 H, CH2CH2 and CH2CH), 2.49 (s, 1 H,
OH), 2.61Ϫ2.68 (m, 1 H, CH2CH2), 3.16 (d, J ϭ 9.1 Hz, 1 H,
CH2OCH3), 3.34 (s, 3 H, CH2OCH3), 3.51 (d, J ϭ 9.1 Hz, 1 H,
CH2OCH3), 3.66 (s, 3 H, OCH3), 3.88 (dd, J ϭ 12.2/5.1 Hz, 1 H,
CH2OH), 4.00 (dd, J ϭ 12.2/2.6 Hz, 1 H, CH2OH). Ϫ IR: ν˜ ϭ
3450 cmϪ1, 1732, 1682. Ϫ MS (CH4; CI); m/z (%): 326 (100) [M ϩ
Hϩ]. Ϫ C17H27NO5 (325.4): calcd. C 62.75, H 8.36, N 4.30; found
C 62.73, H 8.42, N 4.35.
ride-free, then 10% NH3) to give 10.7 mg (52%) of 19c as colorless
20
crystals, m.p. > 300°C. Ϫ [α]D ϭ Ϫ20 (c ϭ 0.17, H2O) {ref.[10a]
,
(1S,3R,6S,8R)-1-Hydroxymethyl-11-methoxy-8-methoxymethyl-7,7,8-
trimethyl-5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-4-one (20b): To a
20
1
[α]D ϭ Ϫ21.5 (c ϭ 1, H2O)}. Ϫ H NMR (D2O): δ ϭ 1.40 (s, 3
H, CH3), 2.83 (d, J ϭ 14.2 Hz, 1 H, CH2), 3.15 (d, J ϭ 14.2 Hz, 1 solution of 0.13 g (0.48 mmol) of 5 in 5 mL of THF 1.01 mL
1976 Eur. J. Org. Chem. 1999, 1967Ϫ1978

Preparation of Enantiomerically Pure α-Quaternary α-Amino Acids
(1.01 mmol, 2.1 equiv.) of a solution of NaN(SiMe3)2 (c ϭ 1.0 mol/
FULL PAPER
2.61Ϫ2.69 (m, 1 H, CH2CH2), 2.83 (ddd, J ϭ 8.1/7.8/6.0 Hz, 1 H,
L in THF) and 0.18 g (1.92 mmol, 4 equiv.) of (R)-(Ϫ)-epichlorohy- CHCHO), 3.19 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H,
drine was added at Ϫ10°C. The mixture was stirred at Ϫ10°C for CH2OCH3), 3.46 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.64 (s, 3 H,
24 h. After addition of 3 mL of phosphate buffer (pH ϭ 7, c ϭ 1 OCH3), 9.42 (d, J ϭ 6.0 Hz, 1 H, CHO). Ϫ IR: ν˜ ϭ 1743 cmϪ1
,
mol/L), THF was removed in vacuo, and the remaining aqueous
phase was extracted with CH2Cl2 (3 ϫ 3 mL). The combined or-
ganic layers were dried (MgSO4), filtered and concentrated in vac-
uo. Purification by CC (petroleum ether/ethyl acetate, 7:3, Rf ϭ
0.20) gave 0.10 g (66%) of 20b as colorless crystals, m.p. 66°C. Ϫ
[α]D ϭ ϩ88.2 (c ϭ 0.70, CHCl3). Ϫ H NMR (CDCl3): δ ϭ 0.88
(s, 3 H, CH3), 0.97 (s, 3 H, CH3), 1.00 (s, 3 H, CH3), 1.44Ϫ1.52
(m, 1 H, CH2CH2), 1.50 (dd, J ϭ 7.3/4.7 Hz, 1 H, CH2CH), 1.71
(dd, J ϭ 9.6/4.7 Hz, 1 H, CH2CH), 1.95Ϫ2.06 (m, 2 H, CH2CH2),
2.25Ϫ2.30 (m, 1 H, CH2CH), 2.63Ϫ2.70 (m, 1 H, CH2CH2), 2.97
(s, 1 H, OH), 3.18 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H,
CH2OCH3), 3.48 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.66 (s, 3 H,
OCH3), 3.88 (dd, J ϭ 12.2/5.8 Hz, 1 H, CH2OH), 4.11 (dd, J ϭ
12.2/2.8 Hz, 1 H, CH2OH). Ϫ IR: ν˜ ϭ 3440 cmϪ1, 1734, 1682. Ϫ
MS (CH4; CI); m/z (%): 326 (100) [M ϩ Hϩ]. Ϫ C17H27NO5
(325.4): calcd. C 62.75, H 8.36, N 4.30; found C 62.67, H 8.58,
N 4.16.
1712, 1675. Ϫ MS (CH4; CI); m/z (%): 324 (100) [M ϩ Hϩ]. Ϫ
C17H25NO5 (323.4): calcd. C 63.14, H 7.79, N 4.33; found C 62.95,
H 7.99, N 4.15.
(1R,3R,6S,8R)-11-Methoxy-8-methoxymethyl-7,7,8-trimethyl-1-
phenylaminomethyl-5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-4-one
(23a): 0.048 g (0.15 mmol) of 22a was dissolved in 3 mL of THF
and treated with 0.042 g (0.45 mmol) of aniline and 0.009 g
(0.15 mmol) of NaBH3CN. The mixture was stirred at room temp.
for 1 h. THF was removed in vacuo, the residue was taken up in
8 mL of Et2O and washed with 5% NaHCO3 solution (3 ϫ 5 mL)
and brine (2 ϫ 5 mL). The aqueous phases were reextracted with
Et2O (2 ϫ 5 mL), the combined organic layers were dried (MgSO4)
20
1
and concentrated in vacuo. CC (petroleum ether/ethyl acetate, 9:1,
20
Rf ϭ 0.26) gave 0.052 g (86%) of 23a as colorless oil. Ϫ [α]D
ϭ
1
Ϫ89.0 (c ϭ 0.50, CHCl3) Ϫ H NMR (CDCl3): δ ϭ 0.80 (s, 3 H,
CH3), 0.87 (s, 3 H, CH3), 0.91 (s, 3 H, CH3), 1.29 (dd, J ϭ 7.7/
4.3 Hz, 1 H, CH2CH), 1.46Ϫ1.53 (m, 1 H, CH2CH2), 1.97 (dd, J ϭ
(1R,3R,6S,8R)-11-Methoxy-8-methoxymethyl-7,7,8-trimethyl-4-oxo-
5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-1-carbaldehyde (22a): To a
solution of 0.037 g, (26 µL, 0.29 mmol) of oxalyl chloride in 1 mL
of CH2Cl2, a solution of 0.050 g (46 µL, 0.64 mmol) of DMSO in
1 mL of CH2Cl2 was added slowly at Ϫ60°C. After stirring for
15 min at Ϫ60°C, a solution of 0.074 g (0.226 mmol) of 20a in
4 mL of CH2Cl2, after another 15 min at Ϫ60°C, 0.127 g (0.18 mL,
1.26 mmol) of NEt3 was added. The reaction was still kept for
10 min at Ϫ60°C and then for 5 min at room temp., before it was
quenched by addition of 4 mL of H2O. Then, after 10 min, the
organic layer was separated, and the remaining aqueous phase was
extracted with CH2Cl2 (3 ϫ 5 mL). The combined organic layers
were dried (MgSO4) and concentrated in vacuo. CC (petroleum
ether/ethyl acetate, 9:1, Rf ϭ 0.18) gave 0.066 g (90%) of 22a as
9.4/4.3 Hz,
1 H, CH2CH), 1.99Ϫ2.06 (m, 2 H, CH2CH2),
2.33Ϫ2.41 (m, 1 H, CH2CH), 2.61Ϫ2.68 (m, 1 H, CH2CH2), 3.14
(d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H, CH2OCH3), 3.38
(dd, J ϭ 13.3/5.6 Hz, 1 H, CH2NH), 3.44 (dd, J ϭ 13.3/6.8 Hz, 1
H, CH2NH), 3.50 (d, J ϭ 9.0 Hz, 1 H, CH2OCH3), 3.66 (s, 3 H,
OCH3), 3.84 (s, 1 H, NH), 6.61 (d, J ϭ 7.7 Hz, 2 H, aromatic H),
6.72 (t, J ϭ 7.5 Hz, 1 H, aromatic H), 7.18 (t, J ϭ 7.7 Hz, 2 H,
aromatic H). Ϫ IR: ν˜ ϭ 3380 cmϪ1, 1732, 1682, 1604. Ϫ MS (CH4;
CI); m/z (%): 401 (100) [M ϩ Hϩ]. Ϫ C23H32N2O4 (400.5): calcd.
C 68.97, H 8.05, N 6.99; found C 68.68, H 7.88, N 7.19.
(1S,3R,6S,8R)-11-Methoxy-8-methoxymethyl-7,7,8-trimethyl-1-phenyl-
aminomethyl-5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-4-one (23b):
0.060 g (0.18 mmol) of 22b was dissolved in 3 mL of THF and
treated with 0.050 g (0.54 mmol) of aniline and 0.011 g (0.18 mmol)
of NaBH3CN. The mixture was stirred at room temp. for 1 h. THF
was removed in vacuo, the residue was taken up in 8 mL of Et2O
and washed with 5% NaHCO3 solution (3 ϫ 5 mL) and brine (2
ϫ 5 mL). The aqueous phases were reextracted with Et2O (2 ϫ
5 mL), the combined organic layers were dried (MgSO4) and con-
centrated in vacuo. CC (petroleum ether/ethyl acetate, 9:1, Rf ϭ
20
colorless crystals, m.p. 44°C. Ϫ [α]D ϭ Ϫ41.9 (c ϭ 0.40, CHCl3).
1
Ϫ H NMR (CDCl3): δ ϭ 0.81 (s, 3 H, CH3), 0.92 (s, 3 H, CH3),
0.98 (s, 3 H, CH3), 1.46Ϫ1.52 (m, 1 H, CH2CH2), 1.95Ϫ2.08 (m,
2 H, CH2CH2), 2.16 (dd, J ϭ 7.1/4.9 Hz, 1 H, CH2CH), 2.37 (dd,
J ϭ 9.2/4.9 Hz, 1 H, CH2CH), 2.51 (ddd, J ϭ 9.2/7.1/6.8 Hz, 1 H,
CHCHO), 2.61Ϫ2.68 (m, 1 H, CH2CH2), 3.14 (d, J ϭ 9.0 Hz, 1
H, CH2OCH3), 3.33 (s, 3 H, CH2OCH3), 3.47 (d, J ϭ 9.0 Hz, 1 H,
CH2OCH3), 3.67 (s, 3 H, OCH3), 9.26 (d, J ϭ 6.8 Hz, 1 H, CHO).
Ϫ IR: ν˜ ϭ 1744 cmϪ1, 1719, 1676. Ϫ MS (CH4; CI); m/z (%): 324
(100) [M ϩ Hϩ]. Ϫ C17H25NO5 (323.4): calcd. C 63.14, H 7.79, N
4.33; found C 62.97, H 8.00, N 4.34 .
20
0.26) gave 0.064 g (88%) of 23b as colorless oil. Ϫ [α]D ϭ ϩ 98.5
(c ϭ 0.52, CHCl3). Ϫ H NMR (CDCl3): δ ϭ 0.88 (s, 3 H, CH3),
1
0.96 (s, 3 H, CH3), 1.00 (s, 3 H, CH3), 1.17 (dd, J ϭ 7.3/4.4 Hz, 1
H, CH2CH), 1.46Ϫ1.55 (m, 1 H, CH2CH2), 1.72 (dd, J ϭ 9.4/
4.4 Hz, 1 H, CH2CH), 1.93Ϫ2.08 (m, 2 H, CH2CH2), 2.34Ϫ2.41
(m, 1 H, CH2CH), 2.62Ϫ2.70 (m, 1 H, CH2CH2), 3.17 (d, J ϭ
8.8 Hz, 1 H, CH2OCH3), 3.33 (s, 3 H, CH2OCH3), 3.38 (dd, J ϭ
12.9/5.1 Hz, 1 H, CH2NH), 3.49 (d, J ϭ 8.8 Hz, 1 H, CH2OCH3),
3.54 (dd, J ϭ 12.9/7.8 Hz, 1 H, CH2NH), 3.64 (s, 3 H, OCH3), 3.89
(s, 1 H, NH), 6.64 (d, J ϭ 7.7 Hz, 2 H, aromatic H), 6.71 (t, J ϭ
7.3 Hz, 1 H, aromatic H), 7.18 (t, J ϭ 7.3 Hz, 2 H, aromatic H).
Ϫ IR: ν˜ ϭ 3385 cmϪ1, 1734, 1679, 1603. Ϫ MS (CH4; CI); m/z
(%): 401 (100) [M ϩ Hϩ]. Ϫ C23H32N2O4 (400.5): calcd. C 68.97,
H 8.05, N 6.99; found C 68.68, H 7.94, N 7.20.
(1S,3R,6S,8R)-11-Methoxy-8-methoxymethyl-7,7,8-trimethyl-4-oxo-
5-oxa-12-azadispiro[2.2.4.2]dodec-11-en-1-carbaldehyde (22b): To a
solution of 0.032 g (22 µL, 0.25 mmol) of oxalyl chloride in 1 mL
of CH2Cl2, a solution of 0.043 g (40 µL, 0.55 mmol) of DMSO in
1 mL of CH2Cl2 was added slowly at Ϫ60°C. After stirring for
15 min at Ϫ60°C, a solution of 0.064 g (0.196 mmol) of 20b in
4 mL of CH2Cl2, after another 15 min at Ϫ60°C, 0.109 g (0.16 mL,
1.08 mmol) of NEt3 was added. The reaction was still kept at
Ϫ60°C for 10 min, then for 5 min at room temp., before it was
quenched by addition of H2O. The organic layer was separated,
and the remaining aqueous phase was extracted with CH2Cl2 (3
ϫ 5 mL). The combined organic layers were dried (MgSO4) and
concentrated in vacuo. CC (petroleum ether/ethyl acetate, 9:1, Rf ϭ
0.18) gave 0.057 g (89%) of 22b as colorless crystals, m.p. 48°C. Ϫ
Hydrolysis of 20a؊b and 23a؊b. ؊ General Procedure 3 (GP 3): A
solution of 0.1 mmol of the respective compound in 1 mL of 0.2 
HCl was stirred at room temp. for 24 h and then concentrated in
vacuo. The residue was dissolved in 1 mL of 40% aqueous NaOH
[α]D20 ϭ ϩ 80.7 (c ϭ 0.49, CHCl3). Ϫ 1H NMR (CDCl3): δ ϭ 0.90 and the resulting solution was stirred for 24 h at room temp. Then
(s, 3 H, CH3), 0.97 (s, 3 H, CH3), 1.01 (s, 3 H, CH3), 1.48Ϫ1.54
it was cooled to 0°C and, after adjusting the pH to 1Ϫ2 with
(m, 1 H, CH2CH2), 1.95Ϫ2.07 (m, 4 H, CH2CH2 and CH2CH), concd. HCl, extracted with Et2O (3 ϫ 2 mL). The aqueous phase
Eur. J. Org. Chem. 1999, 1967Ϫ1978
1977

O. Achatz, A. Grandl, K. Th. Wanner
FULL PAPER
was concentrated in vacuo and the residue dissolved in a minimum
amount of H2O and subjected to ion exchange chromatography
(Dowex 50 W X 8 cation exchange resin, elution with H2O until
the eluent was neutral and chloride-free, then elution with 10%
aqueous NH3) to give the free amino acid.
Acknowledgments
Financial support of this work by the Deutsche Forschungsgemein-
schaft and the Fonds der Chemischen Industrie is gratefully
acknowledged.
(1R,2R)-1-Amino-2-(hydroxymethyl)cyclopropanecarboxylic Acid
(21a): According to GP 3 from 33 mg of 20a 8 mg (59%) of 21a
[1] [1a]
For a recent review see: T. Wirth, Angew. Chem. 1997, 109,
20
was obtained as colorless crystals, m.p. 198°C (dec). Ϫ [α]D
ϭ
235Ϫ237; Angew. Chem. Int. Ed. Engl. 1997, 36, 225Ϫ227. Ϫ
[1b] For a review see also: R. M. Williams, Synthesis of Optically
Active α-Amino Acids, Pergamon Press, Oxford, 1989.
24
Ϫ22.5 (c ϭ 0.70, H2O) {ref.[11]; (1S,2S) enantiomer: [α]D ϭ ϩ
23.3 (c ϭ 0.95, H2O)}. Ϫ 1H NMR (D2O): δ ϭ 1.41 (dd, J ϭ 10.2/
6.5 Hz, 1 H, CH2CH), 1.45 (t, J ϭ 6.5 Hz, 1 H, CH2CH),
1.75Ϫ1.85 (m, 1 H, CH2CH), 3.81 (dd, J ϭ 11.5/7.2 Hz, 1 H,
CH2OH), 3.84 (dd, J ϭ 11.5/6.7 Hz, 1 H, CH2OH). Ϫ IR: ν˜ ϭ
3250 cmϪ1, 1630, 1570. Ϫ MS (CH4; CI); m/z (%): 132 (1) [M ϩ
Hϩ], 101 (100). Ϫ C5H9NO3 (131.1): calcd. C 45.80, H 6.92, N
10.68; found C 45.73, H 6.77, N 10.96.
[2]
[2a]
For example see:
G. Porzi, S. Sandri, Tetrahedron: Asym-
[2b]
metry 1998, 9, 3411Ϫ3420. Ϫ
A. Carloni, G. Porzi, S. San-
[2c]
dri, Tetrahedron: Asymmetry 1998, 9, 2987Ϫ2998. Ϫ
K.
Hammer, K. Undheim, Tetrahedron 1997, 53, 5925Ϫ5936. Ϫ [2d]
K. Hammer, K. Undheim, Tetrahedron 1997, 53, 2309Ϫ2322.
[2e]
Ϫ
For a review on asymmetric syntheses of α-quaternary
aminocyclopropanecarboxylic acids see: K. Burgess, K.-K. Ho,
D. Moye-Sherman, Synlett 1994, 575Ϫ583.
K. Th. Wanner, S. Stamenitis, Liebigs Ann. Chem. 1993,
477Ϫ484.
K. Th. Wanner, A. Kärtner, Synthesis 1988, 968Ϫ970.
M. B. Robin, F. A. Bovey, H. Basch, in: The Chemistry of Am-
ides (Ed. J. Zabicky); in: The Chemistry of Functional Groups
(Sr. Ed.: C. Patai), John Wiley & Sons, New York, 1970, pp.
1Ϫ72.
T. Mukaiyama, E. Bald, K. Saigo, Chem. Lett. 1975,
1163Ϫ1167.
[3]
(1R,2S)-1-Amino-2-(hydroxymethyl)cyclopropanecarboxylic Acid
[4]
[5]
(21b): According to GP 3 from 33 mg of 20b 9 mg (66%) of 21b
20
was obtained as colorless crystals, m.p. 230°C (dec.) Ϫ [α]D
ϭ
25
ϩ72.8 (c ϭ 0.56, H2O) {ref.[18]; [α]D ϭ ϩ73.81 (c ϭ 0.48, H2O)}.
Ϫ H NMR (D2O): δ ϭ 1.18 (t, J ϭ 6.5 Hz, 1 H, CH2CH), 1.50
1
[6]
[7]
[8]
[9]
(dd, J ϭ 10.0/6.5 Hz, 1 H, CH2CH), 1.83Ϫ1.95 (m, 1 H, CH2CH),
3.77 (dd, J ϭ 12.0/6.8 Hz, 1 H, CH2OH), 3.98 (dd, J ϭ 12.0/5.2 Hz,
1 H, CH2OH). Ϫ IR: ν˜ ϭ 3235 cmϪ1, 1639, 1568. Ϫ MS (CH4;
CI); m/z (%): 132 (1) [M ϩ Hϩ], 101 (100) Ϫ C5H9NO3 (131.1):
calcd. C 45.80, H 6.92, N 10.68; found C 45.97, H 7.14, N 10.37.
U. Schöllkopf, U. Groth, C. Deng, Angew. Chem. 1981, 93,
793Ϫ795; Angew. Chem. Int. Ed. Engl. 1981, 20, 798Ϫ800.
R. M. Williams, M.-N. Im, J. Am. Chem. Soc. 1991, 113,
9276Ϫ9286.
J. F. Dellaria, B. D. Santarsiero, J. Org. Chem. 1989, 54,
3916Ϫ3926.
(1R,2R)-1-Amino-2-(phenylaminomethyl)cyclopropanecarboxylic
Acid (24a): According to GP 3 from 40 mg of 23a 13 mg (62%) of
24a was obtained as colorless crystals, m.p. 180°C (dec). Ϫ [α]D20 ϭ
Ϫ66.4 (c ϭ 0.75, H2O). Ϫ 1H NMR (D2O): δ ϭ 1.18 (t, J ϭ 6.2 Hz,
1 H, CH2CH), 1.32 (dd, J ϭ 9.8/6.2 Hz, 1 H, CH2CH), 1.62Ϫ1.73
(m, 1 H, CH2CH), 3.16 (dd, J ϭ 11.0/7.2 Hz, 1 H, CH2NH), 3.32
(dd, J ϭ 11.0/6.7 Hz, 1 H, CH2NH), 6.85Ϫ6.92 (m, 3 H, aromatic
[10] [10a]
C. Cativiela, M. D. Diaz-de-Villegas, J. A. Galvez, Tetra-
[10b]
hedron: Asymmetry 1994, 5, 261Ϫ268. Ϫ
T. M. Zydowsky,
D. Seebach, A. Fadel, Helv. Chim. Acta 1985, 68,
E. de Lara, S. G. Spanton, J. Org. Chem. 1990, 55, 5437Ϫ5439.
[10c]
Ϫ
1243Ϫ1250.
[11] [11a]
W. Hartwig, Methods Org. Chem. (Houben-Weyl) 1995,
vol. E21a, pp. 1041Ϫ1055. Ϫ [11b] U. Schöllkopf, W. Hartwig, U.
Groth, K.-O. Westphalen, Liebigs Ann. Chem. 1981, 696Ϫ708.
[12] [12a]
H), 7.30 (t, J ϭ 7.2 Hz, 2 H, aromatic H). Ϫ IR: ν˜ ϭ 3228 cmϪ1
,
H.-E. Högberg Methods Org. Chem. (Houben-Weyl) 1995,
1645, 1580. Ϫ MS (CH4; CI); m/z (%): 207(1) [M ϩ Hϩ], 106 (100).
Ϫ C11H14N2O2 (206.3): calcd. C 64.06, H 6.84, N 13.58; found C
64.14, H 6.68, N 13.77.
[12b]
vol. E21a, pp. 791Ϫ915. Ϫ
A. Studer, D. Seebach, Liebigs
[12c]
Ann. 1995, 217Ϫ222. Ϫ
1988, 44, 5277Ϫ5292.
R. Fitzi, D. Seebach, Tetrahedron
[13]
[14]
[15]
[16]
D. J. Aitken, J. Royer, H.-P. Husson, J. Org. Chem. 1990, 55,
2814Ϫ2820.
(1R,2S)-1-Amino-2-(phenylaminomethyl)cyclopropanecarboxylic
Acid (24b): According to GP 3 from 40 mg of 23b 12 mg (59%) of
24b was obtained as colorless crystals, m.p. 205°C (dec). Ϫ [α]D20 ϭ
ϩ90.5 (c ϭ 0.60, H2O) Ϫ 1H NMR (D2O): δ ϭ 1.05 (t, J ϭ 6.5 Hz,
1 H, CH2CH), 1.14 (dd, J ϭ 10.2/6.5 Hz, 1 H, CH2CH), 1.48Ϫ1.56
(m, 1 H, CH2CH), 3.02 (dd, J ϭ 11.5/7.7 Hz, 1 H, CH2NH), 3.19
(dd, J ϭ 11.5/5.9 Hz, 1 H, CH2NH), 6.70Ϫ6.80 (m, 3 H, aromatic
S. Shuto, S. Ono, Y. Hase, N. Kamiyama, H. Takada, K. Yama-
sihita, A. Matsuda, J. Org. Chem. 1996, 61, 915Ϫ923.
D. E. McClure, B. H. Arison, J. J. Baldwin, J. Am. Chem. Soc.
1979, 101, 3666Ϫ3668.
For a model where a related structure has been suggested, see:
G. Porzi, S. Sandri, P. Verrocchio, Tetrahedron: Asymmetry
1998, 9, 119Ϫ132.
[17]
[18]
W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43,
2923Ϫ2925.
H), 7.16 (t, J ϭ 7.9 Hz, 2 H, aromatic H). Ϫ IR: ν˜ ϭ 3195 cmϪ1
,
M. C. Pirrung, S. E. Dunlap, U. P. Trinks, Helv. Chim. Acta
1989, 72, 1301Ϫ1309.
1627, 1539. Ϫ MS (CH4; CI); m/z (%): 207(1) [M ϩ Hϩ], 106 (100).
Ϫ C11H14N2O2 (206.3): calcd. C 64.06, H 6.84, N 13.58; found C
63.84, H 7.10, N 13.84.
Received December 17, 1998
[O98567]
1978
Eur. J. Org. Chem. 1999, 1967Ϫ1978