Asymmetric synthesis with 6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one as a new chiral glycine equivalent: Preparation of enantiomerically pure α-tertiary and α-quaternary α-amino acids

FULL PAPER

Asymmetric Synthesis with 6-tert-Butyl-5-methoxy-6-methyl-3,6-dihydro-2H-

1,4-oxazin-2-one as a New Chiral Glycine Equivalent: Preparation of

Enantiomerically Pure α-Tertiary and α-Quaternary α-Amino Acids

[a]

ˇ

´

Claus-Jürgen Koch, Sona Simonyiova, Jörg Pabel, Annerose Kärtner, Kurt Polborn,

and Klaus Theodor Wanner*^[a]

Dedicated to Prof. M.-H. Zenk on the occasion of his 70th birthday

Keywords: Amino acids / Asymmetric synthesis / Carboxylic acids / Glycine equivalent / α-Hydroxy acids

The chiral oxazinone 2 has been developed as a new chiral

glycine equivalent for the asymmetric synthesis of mono- and

the mono- and the disubstituted derivatives of 2 the corres-

ponding α-amino acids 33a−c and 34a−d were obtained in

disubstituted α-amino acids. It is derived from the α-hydroxy- high enantiomeric purity by hydrolytic cleavage of the oxazi-

carboxylic acid 1, which serves as a chiral auxiliary, and is

easily accessible in enantiomerically pure form by optical

resolution of the racemic compound (RS)-1. For alkylation re-

none ring, accomplished either in two steps with aqueous

TFA and aqueous NaOH or in one with either aqueous

NaOH or 3

N HBr. Alkylation of the enolate ions of (S)-2 or

actions, 2 was deprotonated with sBuLi or phosphazenic (R)-2 with epichlorohydrins as bifunctional electrophiles pro-

base. Subsequent treatment with alkyl halides yielded the vided the hydroxymethylenecyclopropyl derivatives 21 and

monosubstituted compounds 13/14a−c, e, f, (ent)-13d, (ent)- 22. Hydrolysis of 21 and 22 afforded the free amino acids 35

14d, while a second alkylation step, via the corresponding

and (ent)-35. Reductive amination with aniline after oxida-

enolates, provided the disubstituted compounds 17/18a−d.

tion of 21 and 22 to the corresponding aldehydes 24 and 26

Both alkylation steps proceeded with good yields and excel- provided the compounds 25 and 27, whereas Mitsunobu

lent diastereoselectivities (up to 99% de) and even less react- treatment of 21 and 22 with 1-phenyl-3-(trifluoroacetyl)urea

ive electrophiles such as isopropyl iodide could be used. The

(28) afforded the urea derivatives 29 and 31. Hydrolysis of

results obtained in this reaction supported the assumption these compounds yielded the corresponding 1-aminocylopro-

that the enolate of 2, as well as those of the monosubstituted

derivatives of 2, have less tendency to form the aggregates

that hamper alkylation reactions with other systems with

higher oxygen content. From the major diastereomers of both

panecarboxylic acid derivatives 36/(ent)-36 and (ent)-37.

( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,

Germany, 2003)

Introduction

Though there exists a broad variety of methods for the

asymmetric construction of α-amino acids, the number of

processes for the asymmetric synthesis of α-quaternary α-

amino acids was formerly quite limited, but this field is

meanwhile also expanding very rapidly.^[3]

In recent years, optically active, nonproteinogenic, unnat-

ural α-amino acids have come to play an important role in

the field of peptide chemistry. α-Quaternary α-amino acids

especially are of major interest, as the introduction of con-

formational constraints into peptides may result in im-

proved properties or beneficial physiological effects.^[1]

Moreover, α-quaternary α-amino acids appear to be power-

ful enzyme inhibitors, and thus promising drug candidates

in the field of medicinal chemistry.^[2]In addition, enantiom-

erically pure α-quaternary α-amino acids may serve as valu-

able chiral building blocks for the construction of more

complex molecules.^[1]

We have previously reported the stereoselective prepara-

tion of α-monoalkylated and α,α-dialkylated α-amino acids

by making use of the chiral glycine equivalent 4, developed

by us.^[4]Both mono- and dialkylation reactions of the enol-

ate of 4 proceeded with high diastereoselectivities. However,

the reaction conditions, especially for the first alkylation,

had to be carefully optimized in order to avoid undesired

double alkylation reactions, resulting in a decrease in the

yield of the desired product. Large-scale syntheses with 4

are furthermore restricted by the fact that the preparation

of the enantiomerically pure α-hydroxycarboxylic acid 3,

which represents the chiral auxiliary in 4, is rather costly.

For the synthesis of the enantiomer of 3 this is even more

true, as the more expensive (1S,3R)-(Ϫ)-camphoric acid is

needed.^[5]

[a]

Department Pharmazie Ϫ Zentrum für Pharmaforschung Ϫ

LMU München,

Butenandtstr. 7, Haus C, 81377 München, Germany

Fax: (internat.) ϩ 49-(0)89/2180-7247

E-mail: klaus.wanner@cup.uni-muenchen.de

1244

 2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ193X/03/0407Ϫ1244 $ 20.00ϩ.50/0 Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

In this paper we report the results of our efforts to de- hydroxycarboxylic acid 1 in enantiomerically pure form,

velop the chiral glycine equivalent 2 as a new building block was investigated.

for the asymmetric construction of α-monosubstituted α-

amino acids,^[1a,6]and, which seems even more important, α-hydroxycarboxylic acid (RS)-1 was attempted. As known

for the synthesis of disubstituted derivatives.^[3,7]methods for the synthesis of (RS)-1 or an ester thereof^[9b,9c]

As a very promising approach, the resolution of racemic

In (S)-2, in comparison with 4, the α-hydroxycarboxylic did not prove satisfactory, the synthesis of (RS)-1 was ac-

acid 3 is replaced by the chiral auxiliary 1 (Scheme 1). This complished as follows. Upon oxidation of pinacolone (5)

was considered to be advantageous for different reasons: with KMnO₄in aqueous NaOH^[9b]and distillation, the bu-

the synthesis of 1 might be carried out in an easy and eco- tyric acid derivative 6 was obtained in a yield of 85%. The

nomic manner, and in the case of a racemic synthesis of 1 butyric acid derivative 6 was then treated with 2.2 equiv. of

an optical resolution might give direct access to both en- MeMgCl to give, after recrystallization from n-heptane, the

antiomers. In addition, the low molecular mass of 1 would desired (RS)-1 in 92% yield. After extensive experimenta-

result in a favorable mass balance for the overall sequence tion, phenylalaninol (7), which is commercially available in

of the amino acid synthesis. A further advantage may be both enantiomeric forms at a reasonable price,^[10]proved to

seen in the fact that the chiral auxiliary 1 contributes only be a well suited resolving agent for the resolution of the

1

singlets to the H NMR spectra of 2 and the subsequent racemic acid (RS)-1.

alkylation products, thus enabling simple characterization

of the products. The most important fact behind our de- equiv. of (R)-phenylalaninol [(R)-7] in iPrOH, a precipitate

cision to evaluate 1 as a chiral auxiliary, however, was that of salt containing (S)-1 was immediately formed

When a solution of (RS)-1 in iPr₂O was treated with 0.45

a

there are no extra functionalities in 1 besides the amino (Scheme 2). The highest resolution was observed (96% ee)

and the carboxy group. Williams et al. had observed side when the resolving agent was added slowly at an elevated

reactions during the alkylation of their chiral glycine temperature (ϩ 60 °C) and the reaction product was al-

equivalent,^[7c]which they put down to the formation of ag- lowed to equilibrate (36 h at ϩ60 °C, 36 h at room temp.).

gregates of the enolate generated for the alkylation reac- A further increase in the enantiomeric excess was achieved

tion,^[8]and we had had to deal with similar problems when when the salt resulting from the above procedure was sus-

we performed alkylation reactions with the enolate derived pended in a solvent mixture of iPr₂O and EtOH for several

from 4. Both glycine equivalents display additional func- days (36 h at ϩ60 °C, 36 h at room temp.). Subsequent re-

tional groups that might favor the formation of such ag- lease of (S)-1 from the resulting precipitate by use of

gregates. This gave rise to the assumption that the oxazine NaHSO₄gave a 73% yield [based on the 0.45 equiv. of (R)-

2 might be better suited than 4 as a chiral glycine equiva- 7] of the carboxylic acid (S)-1 with Ն 99.2% ee. Of course,

lent, as it is devoid of any additional functionalities that in full analogy to this procedure, (R)-1 could be obtained

might promote the formation of aggregates of the enolate.

Scheme 1. Delineation of the chiral glycine equivalent (S)-2 from 4

Results and Discussion

First an efficient route to the pure enantiomers of the α-

hydroxycarboxylic acid 1 was needed. As known methods

seemed to be hardly suitable,^[9]an alternative route, ex-

pected to allow an easy large-scale preparation of the α- (RS)-1

Scheme 2. Preparation of (S)-1 by optical resolution of racemic

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1245

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

simply by switching to (S)-phenylalaninol [(S)-7] as resolv- Enolate Alkylations with Alkyl Halides

ing agent.

Monoalkylation Reactions

The absolute configuration of (S)-1 [as well as of (R)-1]

was evident from the optical rotation found for this com-

pound in comparison to the data reported in the literatu-

re.^[9d,9f]

The enantiomeric purities of samples of the α-hydroxy-

carboxylic acid 1 were determined by chiral HPLC. A sta-

tionary phase with hydroxyproline for ligand-exchange

chromatography (with a 0.5 m CuSO₄solution as elu-

ent^[11]) was used, and separated the enantiomers of 1 very

nicely.

Because of the absence of additional functional groups

in (S)-2, we expected that alkylation reactions of the enolate

of (S)-2 should be far less problematic than those of the

enolate of 4,^[4]with respect to the influence of reaction con-

ditions on the outcome of the alkylation reactions.

When the reactions were performed under standard con-

ditions (THF, Ϫ78 °C, sBuLi) the results were indeed satis-

fying. In contrast to the results obtained with 4, the forma-

tion of the dialkylation products 15 while starting material

was still present was almost negligible in the case of (S)-2.

At low temperatures (Ϫ78 °C) only minute amounts, if any

at all, of the dialkylation products 15 were observed,

slightly increasing if the reaction temperature was raised to

0 °C (see Table 1). There was therefore no need for tedious

optimization of the reaction conditions, especially of the

solvent system (use of a co-solvent such as DMPU, or of

DME in place of THF) to avoid these undesired side reac-

tions. These results also indirectly support the assumption

put forward by Williams^[7c]that these undesired dial-

kylation reactions arise from aggregates that form from the

lithiated compounds.^[8]

Synthesis of the Chiral Glycine Equivalents (S)-2 and (R)-2

The chiral glycine equivalent (S)-2 was synthesized ana-

logously to the construction of the previously reported spir-

ocyclic oxazinone derivative 4.^[4]In the first step, the carb-

oxylic acid (S)-1 was treated with CDI-activated Cbz-gly-

cine, yielding the N-protected ester (S)-9 (86%). Removal of

the Cbz group by catalytic hydrogenolysis (Pd/C, H₂) gave

the free amino acid (S)-10 (96%), which underwent a clean

cyclization to give (S)-11 in a yield of 84% upon treatment

with Mukaiyama’s reagent (2-chloro-1-methyl-pyridinium

iodide)^[12]in the presence of diisopropylethylamine. Finally,

by treatment of (S)-11 with an excess of Meerwein’s salt

(trimethyloxonium tetrafluoroborate),^[13](S)-2 was ob-

tained in a yield of 89% (see Scheme 3). The synthesis of

(R)-2 was accomplished by the otherwise identical, but en-

antiomorphic, synthetic sequence (see Scheme 3).

Table 1. Alkylation reactions between (S)-2 and alkyl halides

Entry^[a]R¹ϪHal

Reaction temp.

ds^[b]

Yield (%)

13/14

13 ϩ 14 15

1

2

3

4

5

6

7

8

9

PhCH₂Br Ϫ78 °C

99.7:0.3 90

2

3

1

PhCH₂Br Ϫ78 °C Ǟ 0 °C 95.2:4.8 66

AllBr

MeI

Ϫ78 °C

Ϫ90 °C

99.2:0.8 76

99.7:0.3 94

Ϫ78 °C Ǟ 0 °C 95.8:4.2 76

4

Ϫ78 °C

Ϫ93 °C

Ϫ50 °C

Ϫ78 °C

96.8:3.2 72

98.7:1.3 62

97.0:3.0 72

92.7:7.3 63

MeI

nBuI

iPrI

5

[a]

All experiments were carried out in THF with 1.1 equiv. of

sBuLi, except for entry 8, in which DME was used, and for entry

9, in which 1.1 equiv. of tBuϪP₄were used. ^[b]All diastereoselectiv-

ities were determined by analytical HPLC.

The experiments could therefore be simply carried out in

THF, mostly at a temperature of Ϫ78 °C. For the depro-

tonation, 1.1 equiv. sBuLi were used and for the alkylation

3 equiv. of the respective alkyl halide. When isopropyl iod-

ide was employed, however, the reactivity of the lithium en-

olate (S)-12 appeared to be too low, so phosphazenic base

tBuϪP₄{tBuNϭP[NϭP(NMe₂)₃]₃} had to be used for the

deprotonation of (S)-2 in this case.

Treatment of (S)-12 at Ϫ78 °C with benzyl bromide, allyl

bromide, and methyl iodide provided the corresponding al-

kylation products 13/14aϪc with good to excellent yields

(from 72 to 90%) and diastereoselectivities (from 96.8:3.2

to 99.7:0.3 ds; Table 1, entries 1, 3, 6). Only in the case of

13/14aϪb was the product contaminated with some dial-

kylation product, but the amount was very small (15a: 2%,

15b: 1%). Additional reactions with allyl bromide and

Scheme 3. Synthesis of the chiral glycine equivalents (S)-2 and (R)-

2: i) Cbz-glycine, CDI, THF; ii) H₂, Pd/C, EtOH; iii) 2-chloro-1-

methylpyridinium iodide, iPr₂EtN, CH₂Cl₂; iv) Me₃O^ϩBF₄

,

Ϫ

CH₂Cl₂

1246

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

methyl iodide at still lower temperatures (Ϫ90 °C and be-

After minor modifications of the reaction conditions,

low) gave even higher diastereoselectivities (99.7:0.3 and reasonable results were also obtained for alkylation reac-

98.7:1.3) without any significant drop in yields (see Table 1, tions with the only moderately reactive electrophiles n-butyl

entries 4 and 7). Furthermore, a marked effect was seen iodide and isopropyl iodide. Treatment with n-butyl iodide

when a precooled (Ϫ78 °C) solution of [¹³C]MeI in THF was best performed at Ϫ50 °C in DME as solvent, whereas

was used for the alkylation reaction (performed in this case with isopropyl iodide the enolate of (S)-2 had to be gener-

with (R)-2 rather than (S)-2). Under these conditions we ated with the phosphazenic base tBuϪP₄. Both yields and

obtained the highest diastereoselectivity (99.7:0.3 ds, see diastereoselectivities were good (13/14e 72%, 97.0:3.0 ds; 13/

Scheme 4) as well as the highest yield for the methylation 14f 63%, 92.7:7.3 ds; see Table 1, entries 8 and 9), though

reactions. As the material (ent)-13/(ent)-14d had been in addition small amounts of the dialkylated products oc-

needed for the synthesis of radiolabeled alanine for a biolo- curred (15e 5%; 15f 5%, see Table 1, entries 8 and 9).

1

gical assay, the reaction had been run on a relatively large

scale. Probably because of this the loss of material during

workup was reduced, which might explain the higher yield

observed in this case.

The H NMR spectra of the monobenzyl derivatives of

(S)-2 showed a significant effect with respect to the chem-

ical shifts of the methyl group and the tert-butyl group at

C-6 of the chiral auxiliary, similar to that observed for the

glycine equivalent 4.^[4]In the case of 13a, in which the

benzyl group at C-3 is located on the same side of the oxazi-

none ring as the methyl group at C-6, the latter was de-

tected at δ ϭ 0.92 ppm. This represents a significant upfield

shift, because the C-6 methyl group in the absence of a

benzyl group is usually found around δ ϭ 1.50 ppm. In the

case of compound 14a, in which the benzyl group is now

positioned trans to the methyl but cis to the tert-butyl group

(at C-6 of the oxazinone ring), the tert-butyl group is now

shifted (from δ ϭ 1.00 ppm) to higher field, becoming loc-

ated at δ ϭ 0.82 ppm. This phenomenon may be useful for

the determination of the stereochemistry at C-3 for related

derivatives of 2. It should be mentioned that the above de-

termination of the stereochemistry was verified by the op-

tical rotation found for -phenylalanine (33a) obtained by

hydrolysis of 13a (see below).

Dialkylation Reactions

All the second alkylation reactions were performed with

the mixtures of diastereomers 13/14 obtained during the

first alkylation step, and they were carried out in THF at

Ϫ78 °C with 1.1 equiv. of sBuLi, except for alkylations with

n-butyl and isopropyl iodide (see below).

As already observed for 4,^[4]it was also found for 2 that

the sense of the asymmetric induction for the alkylation re-

actions remains unchanged. Both enantiomers of the final

products are therefore easily accessible simply by changing

the sequence of the introduction of the substituents at C-3

(of 2). In order to demonstrate this useful feature clearly,

the derivatives 13/14bϪc and 13/14eϪf, with allyl, methyl,

n-butyl, or isopropyl substituents, were alkylated with

benzyl bromide, whereas the monobenzyl derivatives 13/14a

were subjected to alkylation reactions with methyl iodide,

n-butyl iodide, allyl bromide, and isopropyl iodide (see

Scheme 5).

Scheme 4. Alkylation of the oxazinone (S)-2 and (R)-2 via the cor-

responding enolates

On the other hand, and consistently with the above re-

sults, the diastereoselectivities became significantly lower

when the mixtures were allowed to warm up to 0 °C during

the alkylation experiments (with benzyl bromide and allyl

bromide; see Table 1, entries 2 and 5). In addition, the

yields of 13/14a and 13/14b also dropped, whereas the

The results may be seen in Table 2. The reactions generally

proceeded smoothly and gave reasonable to good yields

(60Ϫ85%) and high diastereoselectivities (94.1:5.9Ϫ99.6:0.4

ds).

n-Butyl iodide was also sufficiently reactive for a smooth

amounts of the dialkylated oxazines rose slightly (15a 3%, alkylation reaction, but only when the reaction was per-

15b 4%; see Table 1, entries 2 and 5).

formed in DME at Ϫ50 °C (see Table 2, entry 3). For the

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1247

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

Scheme 5. Second alkylation of the monoalkylation products 13/14 a؊c, e, f.

Table 2. Alkylation of the monoalkylated oxazines 13aϪc, e, f/14aϪc, e, f

Entry^[a]

Starting material/R¹

R²ϪHal

Product

major ϩ minor diast.

Yield (%)

17 ϩ 18

ds^[b]

17/18

1

2

3

4

5

6

7

8

9

13/14a

PhCH₂

Me

All

nBu

iPr

MeI

AllBr

nBuI

iPrI

17a ϩ 18a

17b ϩ 18b

17c ϩ 18c

19^[c]

17d ϩ 18d

18a ϩ 17a

18b ϩ 17b

18c ϩ 17c

18d ϩ 17d

79

60

66

Ϫ

86

82

85

79

95.3:4.7

96.6:3.4

94.1:5.9

Ϫ

95.1:4.9

1.5:98.5

1.0:99.0

0.8:99.2

0.4:99.6

13/14a

13/14c

13/14b

13/14e

13/14f

iPrI

PhCH₂Br

[a]

All experiments were carried out in THF at Ϫ78 °C with 1.1 equiv. of sBuLi, except for entry 3, in which the reaction was run in

[b]

DME at Ϫ50 °C and for entry 5, in which 1.1 equiv. of phosphazenic base tBuϪP₄were employed.

determined by analytical HPLC. Instead of 17/18d, compound 19 was obtained in a yield of 83% when sBuLi was employed.

All diastereoselectivities were

[c]

reaction with isopropyl iodide, phosphazenic base tBuϪP₄a remarkable yield of 83% (see Table 2, entry 4). The forma-

again had to be used for the deprotonation of 13/14a to tion of 19 had to be a result of oxygen that had entered the

generate an enolate more reactive than the lithium enolate flask and which was finally trapped to give 19. That oxygen

16, which did not react properly.

was able to penetrate the rubber septum used to seal the

Interestingly, in the treatment of the lithium enolate 16 reaction flask may be blamed on the long period the reac-

with isopropyl iodide, which failed to give the alkylation tion was allowed to run. The structure of 19 was further

product 17/18, the hydroperoxy derivative 19 was found in verified by an independent synthesis in which a definite vol-

1248

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

ume of dry air was bubbled through a solution of the li-

thium enolate 16, which produced 19 in a yield of 81%.

Upon treatment of 19 with Ac₂O the ester 20 was formed,

which also supports the constitution assigned to 19. As re-

markable upfield shifts were observed for the tert-butyl

1

groups of 19 and 20 in the H NMR spectra (0.64 and δ ϭ

0.60 ppm respectively), it is also reasonable to assume that

19 and 20 have the stereochemistry at C-3 as shown (R).

1

The H NMR spectra of the dialkylated derivatives 17/

18 with benzylic side chains in the C-3 position in their

oxazine rings showed very similar behavior with respect to

the chemical shift of the methyl group and the tert-butyl

group at C-6 of the chiral auxiliary, as already described for

the monoalkylated derivatives 13/14a (see above). Hence,

the tert-butyl group at C-6 of the chiral auxiliary undergoes

a shift from a δ value of about 1.00 ppm to δ ϭ 0.55 ppm

Ϯ 0.05 ppm for compounds 17aϪd, while the methyl group

at C-6 is significantly shifted from δ ഠ 1.52 ppm towards

higher field in compounds 18aϪd (18a δ ϭ 0.52 ppm, 18b

δ ϭ 0.38 ppm, 18c δ ϭ 0.32 ppm, 18d δ ϭ 0.18 ppm). The

degree of chemical shifting of the 6-methyl group in 18aϪd

also very closely reflects the steric demand of the alkyl

group introduced into the 3-position of the oxazine ring

subsequently to the benzyl substituent. Thus, in the case of

the isopropyl-substituted compound 18d the benzyl group

(at C-3) is pushed to the greatest extent towards the methyl

group at C-6 (far more strongly than observed for 18aϪc

with sterically less demanding groups at C-3), resulting in

the most pronounced upfield shift for this substituent.

Scheme 6. Direct conversion of (S)-2 and (R)-2 into 21 and 22 by

use of epichlorohydrine

tution reaction of the halogen-bearing carbon atom. In ad-

dition, the subsequent ring-closure through C-2 (of the ep-

ichlorohydrin) had proceeded inconsistently, from the bot-

tom face of the oxazine ring for 21 and from the top face

for 22.

Cyclopropanation Reactions

In connection with our studies aimed towards the devel-

opment of new ligands acting at the glycine binding site of

the NMDA-receptor complex, we were interested in various

1-aminocyclopropanecarboxylic acids,^[4]especially in deriv-

atives with a substituent in the 2-position of the cyclopro-

pane skeleton, such as 35؊36 and (ent)-35-(ent)-37. The

synthesis of these compounds was ideally to be accomp-

lished with the chiral glycine equivalents (S)-2 and (R)-2, in

order to evaluate the utility of these new building blocks

further, while known strategies should be used for the con-

struction of the cyclopropane skeleton.

Alkylation reactions of (S)-2 and (R)-2 with (S)-epichlo-

rohydrin and (R)-epichlorohydrin appeared most promis-

ing, as these should give access to all four stereoisomers of

the desired amino acids.^[4,14]When (S)-2 was deprotonated

with NaN(SiMe₃)₂(2.2 equiv.) in THF at Ϫ20 °C in the

presence of (S)- or (R)-epichlorohydrin, identified as the

optimum reaction conditions, the cyclopropane derivatives

21 and 22 were obtained in 42% and 44% yields, respect-

ively (Scheme 6).

Figure 1. X-ray diffraction analysis of 21

At present, the reasons for this phenomenon are unclear.

Thus, because of the unforeseen stereochemical course of

these reactions, the hydrolysis of 21 and 22 afforded the

enantiomeric amino acids 35 and (ent)-35 (see below,

Scheme 11) and not a pair of diastereomers. The spectro-

scopic data found for the amino acids 35 and (ent)-35 were,

of course, in good accord with those reported in the literat-

ure.^[16]As is evident from these results the original strategy

designed for the synthesis of all four stereoisomers of the

above amino acids was no longer applicable. The enantio-

morphic sequence starting from (R)-2 would just give the

same stereoisomers of the desired amino acids^[17]as had

been obtained with (S)-2.

In addition, a Furukawa-style cyclopropanation reaction

(Et₂Zn, 2 equiv. CH₂I₂, Et₂O, room temperature) was also

performed with 23a/b^[18]and also provided the known ste-

reoisomers 21 and 22. As observed in the alkylation reac-

tions, the reagent had approached the oxazine ring prefer-

According to X-ray analyses performed for 21 and 22

(see Figure 1),^[15]these compounds did not posses the ex-

pected stereochemistry. Clearly the reaction sequence had

commenced with the addition of the enolate to the terminal

carbon of the epoxide ring of the respective epichlorohydrin entially from the ‘‘bottom face’’, although with the ratio

and not, as observed in many related syntheses, by a substi- amounting to about 4:6 (ϭ 21/22, according to the ¹H

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1249

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

NMR spectrum of the crude product) the reaction was al-

most devoid of any stereoselectivity (Scheme 7).

Further attempts to synthesize the amino derivatives of

21 and 22 remained unsuccessful. The aldehydes 24 and 26

could not be directly converted into the corresponding

primary amines by any of the reductive amination proced-

ures investigated (e.g., reductive amination with ammonia,

ammonium acetate or ammonium chloride in the presence

of sodium cyanoborohydride, or the use of hexamethyldisil-

azane to generate an imine for subsequent reduction).

Either no conversion occurred, or else the secondary and

tertiary amines were formed even though starting material

was still present. Substitution of the tosylates derived from

21 and 22 with NaN(SiMe₃)₂was also unsuccessful.

Finally it was discovered that the hydroxy functions in 21

and 22 could successfully be replaced by 1-phenyl-3-(tri-

fluoroacetyl)urea (28) under Mitsunobu conditions,^[19]pro-

viding 29 and 31 in good yields (63% and 65%). To the best

Scheme 7. Cyclopropanation of 23a/b by use of CH₂I₂/Et₂Zn ac-

cording to Furukawa

The starting material 23a/b had been obtained by means of our knowledge, this is the first time that 28 has been

of an aldol condensation between (S)-2 and glycolaldehyde employed in this type of reaction, in which the trifluoroace-

(8 equiv.) in the presence of 1.0 equiv. of KOH, carried out tyl moiety in 28 serves as an activating group for the urea

at 0 °C in EtOH and yielding 23a and 23b in 79% yield and moiety. This gives rise to a reasonable acidity, necessary if

as a 96:4 mixture. The stereochemistries of the double the Mitsunobu reaction is to be performed successfully,^[19e]

bonds of 23a and 23b were assigned on the basis of the and is easily removable after the reaction as well.

stereochemical outcome of the cyclopropanation reactions

of these compounds (see above).

Thus, when 29 and 31 were stirred in aqueous K₂CO₃to

remove the trifluoroacetyl group the desired compounds 30

and 32 were obtained in good yields (74 and 76%, see

Scheme 9).

In addition it should be mentioned that the required urea

28^[20]may be efficiently prepared in a new procedure by

stirring phenylurea with trifluoroacetic acid anhydride at

room temperature (yield 86%).

Although 1-aminocyclopropanecarboxylic acids with an

aminomethylene side chain were of interest for biological

studies, as were those with the original phenylurea group

still present, it was questionable whether the amino acids

latent in 30 and 32 could be released without destruction

of the urea side chain. Unsurprisingly, despite extensive ex-

perimentation with 30 and 32, all attempts to obtain the

corresponding amino acid with the urea side chain still pre-

sent failed and the free diamino acid (ent)-37, for example,

was obtained instead (see Scheme 11).^[21]

Transformations of the Hydroxymethylene Cyclopropane

Derivatives 21 and 22 into the Corresponding

Aminomethylene Cyclopropane Derivatives

With regard to the development of bioactive compounds,

various derivatives of 21 and 22 in which the hydroxy group

had been replaced by suitable nitrogen-derived functionali-

ties were of interest.

In order to replace the hydroxy group by an anilino sub-

stituent, the stereoisomers 21 and 22 were first oxidized

with ortho-iodoxybenzoic acid to give the corresponding al-

dehydes 24 and 26 (63% and 88%, respectively). Subsequent

reductive amination with aniline, acetic acid, and sodium

cyanoborohydride provided the desired compounds 25 and

27, both in good yields (67% and 73%), as can be seen in

Scheme 8. The same sequence had been utilized previously

for the chiral glycine equivalent 4.^[4]

Liberation of the Amino Acids

A two-step sequence was employed for the hydrolysis of

the monosubstituted and the cyclopropyl derivatives of (S)-

2 and (R)-2 to give the free amino acids. In the first step

the compounds were treated with TFA at 60 °C for 20 h.

This resulted in complete cleavage of the imidate func-

tion.^[22]Finally, the cleavage products thus obtained were

treated with NaOH in various solvents at either room temp.

(13aϪc, 20 h), 60 °C (21 and 22, 20 h), or 90 °C (25 and

27, 5d) to hydrolyze the remaining ester function. After

workup with 6  HCl and subsequent purification by ion-

exchange chromatography (Dowex 50 W X 8) the free

monoalkylated amino acids 33aϪc and the cyclopropyl de-

rivatives 35, (ent)-35, 36, and (ent)-36 were obtained. The

yields for the monosubstituted amino acids 33aϪc were low

to moderate (25%Ϫ69%, see Scheme 10), whereas the yields

for the cyclopropyl derivatives 35, (ent)-35, 36, and (ent)-36

Scheme 8. Oxidation and subsequent reductive amination of 21,

and 22 to the corresponding aniline derivatives. IBX ϭ ortho-

iodoxybenzoic acid

1250

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

Scheme 9. Preparation of the urea derivatives 30 and 32 by Mitsunobu reaction of 21 and 22 with 1-phenyl-3-(2,2,2-trifluoroacetyl)urea

(28) and subsequent removal of the trifluoroacetyl group by basic hydrolysis

were reasonable to very good (50Ϫ97%, see Scheme 11). In

the case of the disubstituted amino acids an enantiomeric

purity of Ͼ 99% ee may be assumed, as racemization can-

not occur because of the structure of these compounds and

as the precursors had been employed in diastereomerically

(Ͼ99.5% de) and enantiomerically pure form (Ͼ99.5% ee).

Scheme 11. Hydrolysis of cyclopropyl derivatives to the corres-

ponding free amino acids

were performed, different reaction conditions being tested

with regard to their suitability. Thus, in one case, 5 equiv.

of LiOH in a mixture of dioxane/H₂O (1:1) was employed

in place of an excess of 40% aqueous NaOH for the hydro-

lysis of the ester function of 13c after the compound had

been treated with 0.2  TFA. This was accompanied by

severe drops in both the chemical and the optical yields

(26%, 77% ee determined by chiral HPLC^[11]), however, and

so this kind of hydrolysis was not pursued any further.

Finally, in further hydrolysis experiments, 13c and (ent)-

Scheme 10. Hydrolysis of mono- and dialkylated derivatives of 2

13d were treated with 25 equiv. of 3  aqueous HBr in a

pressure tube at 130 °C for one day.^[23]When the resulting

hydrobromides of the amino acids were purified by ion-ex-

The monosubstituted amino acids 33aϪc were subjected change chromatography (Dowex 50 W X 8 resin) 85% of -

to chiral HPLC analysis, which revealed enantiomeric purit- alanine (33c) and 83% of the [¹³C]-labeled -alanine [(ent)-

ies of 87% ee (33a), 90% ee (33b), and 94% ee (33c), where 33d] were obtained. Gratifyingly, a loss of enantiomeric

a value of at least 99% ee had been expected. Accordingly, purity of only 1%Ϫ2% ee had occurred in this case [deter-

some significant racemization had occurred during the hy- mined by HPLC,^[11]33c and (ent)-33d 97Ϫ98% ee].

drolysis in this case.

As mentioned above, all attempts at selective hydrolysis

In order to find a method in which the extent of racemiz- of 30 and 32 with the urea side chain remaining untouched

ation would be reduced, additional hydrolysis experiments failed.^[24]Complete hydrolysis also appeared to be difficult,

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1251

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

though. It turned out that a two-step procedure based on

treatment with 0.1  TFA (60 °C overnight) followed by

aqueous NaOH at 90 °C (in DME) for the cleavage of the

lactim ether and the ester function provided the best results,

Conclusion

From the results presented in this paper, the chiral glycine

though these were still not fully satisfactory. Thus, when equivalent 2 compares well with other literature methods

applied as an example to 31, chosen for purposes of con- used for the asymmetric construction of α-amino acids. The

venience, only a 24% yield of the diamino acid (ent)-37 was syntheses of both enantiomers of 2 Ϫ (S)-2 and (R)-2 Ϫ are

obtained after ion-change chromatography (DOWEX easily and efficiently accomplishable starting from the pure

50Wx8). Though the yield was already low, the material was enantiomers of the chiral auxiliary 1. These are accessible

still contaminated by some inseparable by-product.

in a short and high-yielding process, starting from cheap

As there was no risk of racemization, the hydrolysis of and readily available compounds. The optical resolution of

the doubly alkylated compounds 17aϪd was effected in a racemic (RS)-1 appears to be highly efficient and may be

single step by use of 40% aqueous NaOH in methanol (4:1 performed even on large scale.

mixture), although 3 days at 60 °C were required for 17aϪc

The chiral glycine equivalents (R)-2 and (S)-2 appear to

to make the reaction go to completion and 2 days at reflux be especially useful for the construction of α,α-disubstituted

for the bulkier 17d. After the usual workup the free doubly amino acids, but may also be used for the synthesis of α-

alkylated amino acids 34aϪd were obtained in reasonable monosubstituted amino acids. In contrast, other systems

to good yields (51%Ϫ91%, see Scheme 10).

(Schöllkopf’s bislactim ether method, for example) are of-

ten limited to a greater or lesser extent to the synthesis of

α-monosubstituted amino acids. Most alkylation reactions

of 2 proceeded smoothly in pure THF in the absence of

additives such as HMPA or DMPU. Only for the alkylation

with butyl iodide was a different solvent (DME) used. In

general, as long as reactive alkyl halides were used, there

was also no need to employ particular bases for the depro-

tonation of 2 in order to perform alkylation reactions, as

only very small amounts Ϫ if any at all Ϫ of double al-

kylation products were formed under these conditions (in

the first alkylation step). This result nicely supports the as-

sumption that the presence of fewer oxygen functions in an

intermediate such as 12 (or 16) reduces the tendency to the

formation of aggregates, which would otherwise hamper al-

kylation reactions. Furthermore, alkylation reactions al-

most always proceeded in good yields and with excellent

stereoselectivities, even when less reactive alkyl halides were

used, this being true both for the first and for the second

alkylation step, whereas other systems are often limited if

less reactive electrophiles are employed. Hydrolysis of com-

pounds 13 was carried out in a two-step procedure (first by

using TFA for cleaving the lactim ether function, then by

using NaOH for cleaving the ester function), providing the

monosubstituted amino acids 33aϪc. Racemization oc-

curred to some extent. From the hydrolysis of the dial-

kylated compounds 17aϪd, performed in a single step with

aqueous NaOH, the disubstituted amino acids 34aϪd were

obtained. This method also proved useful for the prepara-

tion of the substituted cyclopropanecarboxyclic acids

35؊36 and (ent)-35-(ent)-37. The nature of the chiral auxili-

ary 1 also had additional benefits: the mass balance of the

synthetic sequence may be deemed to be quite favorable as

Model for the Asymmetric Induction

From the stereochemistry found for the alkylation prod-

ucts, the alkylation reactions of the enolates (S)-12 and (R)-

16 had proceeded consistently with re selectivity, and those

for the enantiomers of the former compounds [(R)-12 and

(R)-16] with, of course, si selectivity. To explain the ob-

served sense of the asymmetric induction, the model in

Scheme 12, representing the putative structure of the enol-

ates (S)-12 and (R)-16, is put forward. As previously pro-

posed for 4,^[4]it is reasonable to assume that the ring atoms

O-1, C-2, C-3, N-4, and C-5 are all essentially located in

one plane, whereas C-6 adopts a position above this plane

in order to reduce allylic strain.^[7a]

Scheme 12. Hypothetical model for the asymmetric induction in

the alkylation reactions of the enolates (S)-12 and (S)-16

Because of the steric encumbrance of the top face arising

from the pseudoaxial disposition of the tert-butyl group,

electrophiles should approach the enolate from the ‘‘bottom

face’’ of the oxazine ring. This would give rise to re stereo-

selectivity for (S)-12 and (R)-16 and to si stereoselectivity

for the enantiomeric enolates, as was indeed the case for all

alkylation reaction. Only the formation of the cyclopropane

derivative 22 by treatment of (S)-2 with (R)-epichlorohydrin

did not obey this rule.

1

the molecular mass of 1 is relatively low, and the H NMR

spectra of 2 and of subsequent products show only a few

signals Ϫ all singlets Ϫ resulting from the chiral auxiliary,

which is very helpful for the interpretation of these spectra.

In addition, the stereochemistry of the alkylation products

may be predicted with high reliability, as the same sense of

asymmetric induction was observed in all cases [except for

the reaction with (R)-epichlorohydrin], with the electrophile

1252

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

removal of the solvent produced a white precipitate. Recrystalliza-

tion from hot n-heptane (50 mL) and subsequent washing of the

precipitate with cold n-heptane (3 ϫ) yielded (RS)-1 (22.07 g, 92%)

as colorless crystals, m.p. 139Ϫ140 °C {ref.^[27]141Ϫ142 °C}. ¹H

NMR (CDCl₃): δ ϭ 1.03 [s, 9 H, C(CH₃)₃], 1.43 (s, 3 H, CH₃) ppm.

¹³C NMR (CDCl₃): δ ϭ 20.4 (CH₃), 25.1 [C(CH₃)₃], 36.6

preferentially approaching the enolates from the face op-

posite to the one defined by the tert-butyl group.

Experimental Section

[C(CH₃)₃], 78.3 (COH), 178.6 (CO₂H) ppm. IR: ν˜ ϭ 3453 cm^Ϫ1

,

General Remarks: All experiments were carried out in oven-dried

glassware under a dry N₂atmosphere. Standard vacuum techniques

were used for the handling of air-sensitive materials. Solvents were

dried and kept under N₂and freshly distilled before use. Reagents

were used as commercially available. Solvents used for HPLC were

degassed prior to use. M.p. (uncorrected values): Melting point ap-

paratus according to Dr. Tottoli (Büchi no. 510). Optical rotation:

2964, 1725, 1376, 1265. MS (CH₄; CI): m/z (%) ϭ 147 (18) [M ϩ

H^ϩ], 129 (89), 101 (100), 71(56). C₇H₁₄O₃(146.2): calcd. C 57.51,

H 9.65; found C 57.33, H 9.84.

(S)-(؊)-2-Hydroxy-2,3,3-trimethylbutanoic Acid [(S)-1]. By Optical

Resolution: A solution of enantiomerically pure (R)-phenylalaninol

[(R)-7, 6.81 g, 45.0 mmol] in iPrOH (75 mL) was added at 60 °C

over a period of 2.5 h to (RS)-1 (14.62 g, 100.0 mmol) in 370 mL

of iPr₂O. The resulting suspension was stirred for 36 h at 60 °C and

then allowed to cool to room temp., at which it was kept whilst

stirring for another 36 h. After filtration and washing with iPr₂O,

the salt obtained was dried. The diastereomeric salt was then sus-

pended in a mixture of iPr₂O/EtOH (3:1, 800 mL), and stirred for

36 h at 60 °C and then for 36 h at room temp. After filtration and

washing with iPr₂O, the diastereomeric salt was dried and then

treated with ice-cooled NaHSO₄(0.5 , 260 mL). The solution was

extracted with cold Et₂O (4 ϫ) and the combined organic layers

were washed with water and dried over CaCl₂. Filtration and re-

moval of the solvent in vacuo gave (S)-1 (5.31 g, 73%) as colorless

crystals, m.p. 99 °C. Chiral HPLC with a Nucleosil^Chiral-1 col-

umn^[11](H₂O, 0.5 mmol CuSO₄·5 H₂O/L; 1.0 mL/min), kept at

1

241 MC polarimeter (PerkinϪElmer). H NMR spectra: J NMR-

GX 400 (Jeol), 400 MHz, ¹³C NMR spectra: J NMR-GX 400

(Jeol), 100 MHz, chemical shifts (δ) are reported in ppm, TMS as

internal standard. IR: FT-IR spectrometer 1600 and FT-IR spec-

trometer Paragon 1000 (PerkinϪElmer), liquids were run as films,

solids as KBr pellets. MS: 5989 mass spectrometer with 59980 B

particle beam LC/MS interface (Hewlett Packard); API 2000 LC/

MS/MS System (Applied Biosystems). HRMS: MStation 700

(Jeol). Combustion analysis: CHN Rapid (Heraeus). Column chro-

matography (CC): Flash chromatography^[25]on silica gel (Merck

60 0.040Ϫ0.063 mm). TLC: TLC plates 60 F-254, detection with

UV (λ ϭ 254 nm) or with ammonium cerium() heptamolybdate.

Analytical HPLC: L-6000 pump and L-6200 intelligent pump, L-

4000 and L-7400 UV/Vis detectors, D-2500, D-7500, and D-7000

(PC, cartridge supported) Chromato integrators (MerckϪHitachi),

columns: LiChroCart^with LiChrospher^Si 60 cartridge

(5 µm, 250 ϫ 4 mm with precolumn 4 ϫ 4 mm) (Merck), ET250/4

Nucleosil^Chiral-1 (5 µm, 250 ϫ 4 mm) (MachereyϪNagel),

column oven: High temperature oven (Knauer). Preparative HPLC:

L-6000 pump, L-4000 UV/Vis, D-2000 Chromato Integrator

(MerckϪHitachi), column: Hibar RT LiChrosorb^Si 60 (7 µm,

250 ϫ 25 mm) (Merck).

55Ϫ60 °C, revealed 99.6:0.4 es; (S)-1: t_Rϭ 5.25 min; (R)-1: t_R

ϭ

11.87 min. [α]²_D⁰ϭ Ϫ1.1 (c ϭ 2.50, C₂H₅OH); {ref.^[9d,9e,9f]: [α]²_D⁰

ϭ

Ϫ1.4 (c ϭ 8.63, C₂H₅OH)}. [α]²_D⁰ϭ Ϫ11.7 (c ϭ 1.80, CHCl₃).

C₇H₁₄O₃(146.2): calcd. C 57.51, H 9.65; found C 57.21, H 9.95.

Recovery of (R)-Phenylalaninol [(R)-7]: The filtrates from the crys-

tallization and the recrystallization steps and the aqueous phase of

the extraction step were combined, treated with an excess of

NaOH, and subsequently extracted with CH₂Cl₂(4 ϫ). The com-

bined organic extracts were dried over Na₂SO₄(1 h), filtered, and

concentrated in vacuo, and the residue was recrystallized from n-

Preparation of the Chiral Glycine Equivalent 2 from Pinacolone (5):

3,3-Dimethyl-2-oxobutanoic Acid (6)^[9b,26](Procedure Modified from hexane/ethyl acetate (1:1) to give (R)-7 (5.41 g, 80%).^[10]

the Literature Method^[9b]): KMnO₄(61.60 g, 390.0 mmol) was ad-

(R)-(؉)-2-Hydroxy-2,3,3-trimethylbutanoic Acid (R)-1. By Optical

ded at 0 °C, in portions and with vigorous stirring, to a solution

Resolution: The same procedure as described for the preparation of

of NaOH (16.20 g, 405.0 mmol) and 5 (20.25 g, 202.0 mmol,

(S)-1 was employed, except that (S)-phenylalaninol [(S)-7] was used

for the resolution. Colorless crystals, m.p. 98 °C. Chiral HPLC [as

25.00 mL) in H₂O (750 mL). The mixture was stirred for 5 h at 0

°C and for another 12 h at room temp. The resulting suspension

specified for (S)-1]: 99.5:0.5 es; (S)-1: t_Rϭ 6.35 min; (R)-1: t_R

8.75 min. [α]²_D⁰ϭ ϩ1.2 (c ϭ 2.85, C₂H₅OH); {ref.^[9d,9e,9f]: [α]²_D⁰

ϭ

was filtered and the MnO₂was washed several times with H₂O.

The filtrate was then treated, with cooling, with HCl_conc.(50 mL)

and extracted with Et₂O (4 ϫ 250 mL). The combined organic

layers were washed with brine and dried over MgSO₄. Removal of

the solvent gave a pale yellow liquid, which upon distillation

(20 Torr/85 °C) yielded 6 (21.55 g, 85%) as a colorless oil. ¹H NMR

(CDCl₃): δ ϭ 1.33 (s, 9 H, CH₃) ppm. ¹³C NMR (CDCl₃): δ ϭ 25.6

(CH₃), 42.5 [C(CH₃)₃], 163.7 (CO₂H), 201.9 (CO) ppm. C₆H₁₀O₃

(130.1): calcd. C 55.37, H 7.74; found C 54.89, H 7.74.

ϩ1.5 (c ϭ 9.83, C₂H₅OH)}. [α]²_D⁰ϭ ϩ11.8 (c ϭ 2.10, CHCl₃).

C₇H₁₄O₃(146.2): calcd. C 57.51, H 9.65; found C 57.69, H 9.47.

Recovery of (S)-Phenylalaninol [(S)-7]: As described above for the

recovery of (R)-7.

(S)-2-(N-Benzyloxycarbonylaminoacetoxy)-2,3,3-trimethylbutanoic

Acid [(S)-9]:

(256 mg, 1.57 mmol) in THF (4 mL) was added to a solution of

(R,S)-2-Hydroxy-2,3,3-trimethylbutanoic Acid [(RS)-1]: Compound Cbz-glycine (330 mg, 1.57 mmol) in THF (4 mL), followed after

A solution of N,NЈ-carbonyldiimidazole (CDI)

6 (21.40 g, 164.5 mmol) was dissolved in THF (160 mL) and the

mixture was cooled to 0 °C. With vigorous stirring, a solution of

MeMgCl in THF (3.0 , 120 mL, 362.0 mmol, 2.2 equiv.) were

slowly added in such a way as to maintain the temperature below

20 °C. The reaction mixture was stirred for another 24 h at room

4 h of stirring at room temp by (S)-1 (115 mg, 0.790 mmol, 0.5

equiv.) in THF (4 mL). After the mixture had been stirred for an-

other 20 h at room temp. the solvent was removed in vacuo. CC

(petroleum ether/ethyl acetate/acetic acid, 10:3:1) yielded (S)-9

(229 mg, 86%) as colorless crystals, m.p. 58 °C. [α]²_D⁰ϭ ϩ12.4 (c ϭ

temp. and finally carefully hydrolyzed (with cooling!) with HCl (6 0.485, CHCl₃). TLC: R_fϭ 0.30 (petroleum ether/ethyl acetate/

, 100 mL). After removal of the organic solvent, the aqueous solu-

tion was extracted with Et₂O (4 ϫ 200 mL). Washing of the com-

bined organic layers with brine, drying over CaCl₂, filtration, and

acetic acid, 10:3:1). ¹H NMR (CDCl₃): δ ϭ 1.02 [s, 9 H, (CH₃)₃C],

1.64 (s, 3 H, CH₃), 3.95 (dd, J ϭ 18.1/4.9 Hz, 1 H, NHCH₂CO),

4.05 (dd, J ϭ 18.1/6.0 Hz, 1 H, NHCH₂CO), 5.11 (s, 2 H,

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1253

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

OCH₂Ph), 5.41 (s, broad, 1 H, NH), 7.30Ϫ7.35 ppm (m, 5 H, NCH₂CO), 4.39 (d, J ϭ 21.6 Hz, 1 H, NCH₂CO) ppm. IR: ν˜ ϭ

C₆H₅), CO₂H not located. IR: ν˜ ϭ 3348 cm^Ϫ1, 2977, 1720, 1528,

2975 cm^Ϫ1, 1744, 1689, 1472, 1374, 1336. MS (CH₄; CI): m/z (%) ϭ

1377, 1277. MS (CH₄; CI): m/z (%) ϭ 338 (76) [M ϩ H^ϩ], 210 200 (1) [M ϩ H^ϩ], 83 (100). C₁₀H₁₇NO₃(199.3): calcd. C 60.28, H

(100). C₁₇H₂₃NO₆(337.4): calcd. C 60.52, H 6.87, N 4.15; found C 8.60, N 7.03; found C 60.29, H 8.85, N 6.77.

60.55, H 7.13, N 3.86.

(R)-6-tert-Butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-

(R)-2-(N-Benzyloxycarbonylaminoacetoxy)-2,3,3-trimethylbutanoic

Acid [(R)-9]: This compound was obtained by the procedure de-

one [(R)-2]: This compound was obtained by the procedure de-

scribed for the synthesis of (S)-2, from (R)-11 (350 mg, 1.89 mmol).

scribed for the synthesis of (S)-9, from (R)-1 (500 mg, 3.42 mmol). Yield: 331 mg (88%) as colorless crystals, m.p. 81 °C. C₁₀H₁₇NO₃

Yield: 1.027 g (89%) as colorless crystals, m.p. 56 °C. C₁₇H₂₃NO₆

(337.4): calcd. C 60.52, H 6.87, N 4.15; found C 60.80, H 7.12,

N 3.93.

(199.3): calcd. C 60.28, H 8.60, N 7.03; found C 60.39, H 8.80,

N 6.99.

Alkylation Reactions with Monofunctional Alkyl Halides. General

(S)-2-Aminoacetoxy-2,3,3-trimethylbutanoic Acid [(S)-10]: Pd/C

Procedure (GP A) for the Monoalkylation of (S)-2 and (R)-2: Unless

(10% Pd, 14 mg) was added to (S)-9 (200 mg, 0.580 mmol) in eth- otherwise stated, a solution of (S)-2 or (R)-2 in THF (0.1 ) was

anol (8 mL), and the resulting slurry was stirred overnight under

H₂. The precipitate that had formed was redissolved by addition

of H₂O (6 mL). Filtration and removal of the solvent yielded (S)-

treated at Ϫ78 °C with sBuLi (1.3  solution in hexane, 1.1 equiv.).

After 30 min, the alkyl halide (3.0 equiv.) was added and the solu-

tion was stirred overnight at Ϫ78 °C. The mixture was then either

quenched at this temperature (and allowed to warm to room temp.)

10 (113 mg, 96%) as colorless crystals, which were used without

1

further purification. H NMR (D₂O): δ ϭ 1.01 [s, 9 H, (CH₃)₃C], or it was slowly allowed to warm to room temp. (ഠ 5 h). The hydro-

1.58 (s, 3 H, CH₃), 3.90 (d, J ϭ 17.5 Hz, 1 H, COCH₂NH₂), 3.96 lysis was performed by addition of phosphate buffer (pH ϭ 7, c ϭ

(d, J ϭ 17.5 Hz, 1 H, COCH₂NH₂) ppm. IR: ν˜ ϭ 3380 cm^Ϫ1, 2977, 1.0 mol/L). The organic layer was separated and the aqueous layer

1759, 1635, 1596, 1370, 1247, 1212. MS (CH₄; CI): m/z (%) ϭ 204 was extracted with Et₂O (3 ϫ 5Ϫ20 mL). The combined organic

(69) [M ϩ H^ϩ], 76 (100).

layers were washed with brine, dried over MgSO₄, filtered, and con-

centrated. The resulting residue was purified by CC.

(R)-2-Aminoacetoxy-2,3,3-trimethylbutanoic Acid [(R)-10]: This

compound was produced by the procedure described for the syn-

thesis of (S)-10, from (R)-9 (1.00 g, 2.96 mmol). Yield: 590 mg

(98%) as colorless crystals, which were used without further puri-

fication.

(3R,6S)-3-Benzyl-6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-

1,4-oxazin-2-one (13a), (3S,6S)-3-Benzyl-6-tert-butyl-5-methoxy-6-

methyl-3,6-dihydro-2H-1,4-oxazin-2-one (14a), and (S)-3,3-Di-

benzyl-6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-

2-one (15a). A: The compounds were produced by GP A, from

(S)-2 (100 mg, 0.50 mmol), sBuLi (423 µL, 0.55 mmol), and benzyl

bromide (257 mg, 178 µL, 1.5 mmol), with quenching at Ϫ78 °C.

Purification by CC (n-pentane/ethyl acetate, 96:4) without separa-

tion of the diastereomers yielded 13a/14a (130.5 mg, 90%). The dia-

stereoselectivity was determined by analytical HPLC (n-heptane/

ethyl acetate, 95:5, 1.75 mL/min) of the crude product: ds 13a/

14a ϭ 99.7:0.3; 13a: t_Rϭ 14.75 min, 14a: t_Rϭ 16.77 min.

(S)-6-tert-Butyl-6-methyl-1,4-oxazine-2,5-dione [(S)-11]: Compound

(S)-10 (100 mg, 0.490 mmol) was suspended in CH₂Cl₂(10 mL)

and treated with 2-chloro-1-methylpyridinium iodide (138 mg,

0.54 mmol, 1.1 equiv.). After addition of ethyldiisopropylamine

(255 µL, 1.47 mmol, 3.0 equiv.), the reaction mixture was kept un-

der reflux for 4 h. Finally the solvent was removed in vacuo and the

residue was purified by CC (petroleum ether/ethyl acetate, 20:80) to

yield (S)-11 (75 mg, 82%) as colorless crystals, m.p. 124 °C. [α]²_D⁰

ϭ

ϩ37.3 (c ϭ 0.57, CHCl₃). TLC: R_fϭ 0.37 (petroleum ether/ethyl Compound 13a: Colorless crystals, m.p. 52 °C. [α]²_D⁰ϭ ϩ29.9 (c ϭ

acetate, 20:80). ¹H NMR (CDCl₃): δ ϭ 1.12 [s, 9 H, (CH₃)₃C], 1.60 0.60, CHCl₃). TLC: R_fϭ 0.22 (n-pentane/ethyl acetate, 96:4). ¹H

(s, 3 H, CH₃), 4.22 (d, J ϭ 2.6 Hz, 1 H, NHCH₂CO), 4.23 (d, J ϭ NMR (CDCl₃): δ ϭ 0.92 (s, 3 H, CH₃), 0.94 [s, 9 H, (CH₃)₃C],

2.6 Hz, 1 H, NHCH₂CO), 6.53 (s, 1 H, NH) ppm. IR: ν˜ ϭ 3321

3.23 (dd, J ϭ 13.2/ 4.7 Hz, 1 H, CH₂C₆H₅), 3.30 (dd, J ϭ 13.2/

5.5 Hz, 1 H, CH₂C₆H₅), 3.71 (s, 3 H, OCH₃), 4.47 (dd, J ϭ 5.5/

cm^Ϫ1, 2967, 1751, 1682, 1374, 1342, 1283. MS (CH₄; CI): m/z

(%) ϭ 186 (100) [M ϩ H^ϩ]. C₉H₁₅NO₃(185.2): calcd. C 58.36, H 4.7 Hz, 1 H, NCHCO), 7.20Ϫ7.27 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ

8.16, N 7.56; found C 58.33, H 8.21, N 7.43.

2964 cm^Ϫ1, 1740, 1695, 1376, 1344, 1308. MS (CH₄; CI): m/z (%) ϭ

290 (100) [M ϩ H^ϩ]. C₁₇H₂₃NO₃(289.4): calcd. C 70.56, H 8.01,

N 4.84; found C 70.55, H 8.08, N 4.83.

(R)-6-tert-Butyl-6-methyl-1,4-oxazine-2,5-dione [(R)-11]: This com-

pound was produced by the procedure described for the synthesis

of (S)-11, from (R)-10 (550 mg, 2.71 mmol). Yield: 416 mg (83%)

as colorless crystals, m.p. 125 °C. C₉H₁₅NO₃(185.2): calcd. C

58.36, H 8.16, N 7.56; found C 58.48, H 8.20, N 7.40.

Since the amount of 14a formed in the synthesis described above

was small, 14a was synthesized separately by epimerization of 13a:

13a (14 mg, 0.07 mmol) in THF (0.7 mL) was treated with sBuLi

(60 µL, 0.078 mmol) and the reaction was quenched after 30 min

at Ϫ78 °C with a solution of citric acid (29 mg, 0.15 mmol, 2.2

equiv., in 0.3 mL of H₂O). The workup was performed as in GP

A. The diastereoselectivity was determined by analytical HPLC (n-

heptane/ethyl acetate, 95:5, 1.5 mL/min) on the crude product: ds

13a/14a ϭ 26.7:73.3; 13a: t_rϭ 19.20 min, 14a: t_rϭ 21.45 min. The

diastereomers were separated by preparative HPLC (n-heptane/

ethyl acetate, 95:5; 16.5 mL/min).

(S)-6-tert-Butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-

one [(S)-2]: A solution of (S)-11 (73 mg, 0.39 mmol) in CH₂Cl₂

(3 mL) was added to a suspension of Me₃O^ϩBF₄(92 mg,

Ϫ

0.64 mmol) in CH₂Cl₂(3 mL). The reaction mixture was stirred at

room temp. for 24 h. After hydrolysis with a phosphate buffer (pH

7, 2 mL) the organic layer was separated and the aqueous layer was

extracted with Et₂O (3 ϫ 5 mL). The combined organic layers were

dried over MgSO₄and filtered, and the solvent was removed in

vacuo, which after CC (petroleum ether/ethyl acetate, 90:10) af-

Compound 14a: Colorless crystals, m.p. 48 °C. Yield: 6 mg (60%).

TLC: R_fϭ 0.35 (petroleum ether/ethyl acetate, 90:10). ¹H NMR

(CDCl₃): δ ϭ 0.83 [s, 9 H, (CH₃)₃C], 1.47 (s, 3 H, CH₃), 3.16 (dd,

forded (S)-2 (69 mg, 89%) as colorless crystals, m.p. 83 °C. [α]²_D⁰

ϩ121.4 (c ϭ 0.58, CHCl₃). TLC: R_fϭ 0.18 (petroleum ether/ethyl

ϭ

acetate, 90:10). ¹H NMR (CDCl₃): δ ϭ 1.04 (s, 9 H, [CH₃)₃C], 1.56 J ϭ 13.3/7.5 Hz, 1 H, CH₂C₆H₅), 3.35 (dd, J ϭ 13.3/4.5 Hz, 1 H,

(s, 3 H, CH₃), 3.85 (s, 3 H, OCH₃), 4.26 (d, J ϭ 21.6 Hz, 1 H, CH₂C₆H₅), 3.68 (s, 3 H, OCH₃), 4.44 (dd, J ϭ 7.5/4.5 Hz, 1 H,

1254

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

NCHCO), 7.20Ϫ7.27 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ 2966 cm^Ϫ1

,

(156 mg, 109 µL, 1.29 mmol), the reaction mixture being allowed

to warm to 0 °C and quenched after 4 h at 0 °C. Workup was by

1749, 1695, 1370, 1312, 1250. MS (CH₄; CI): m/z (%) ϭ 290 (100)

[M ϩ H^ϩ]. C₁₇H₂₃NO₃(289.4): calcd. C 70.56, H 8.01, N 4.84; GP A. Preparative HPLC (n-heptane/ethyl acetate, 95:5; 16.5 mL/

found C 70.85, H 8.24, N 4.64.

min) gave a diastereomeric mixture of 13b and 14b (76%) and 15b

(5 mg, 4%). The diastereoselectivity 13b/14b was determined as de-

scribed above by analytical HPLC on the crude product: ds 13b/

14b ϭ 95.8:4.2.

B: The compounds were produced by GP A, from (S)-2 (46 mg,

0.23 mmol), sBuLi (195 µL, 0.25 mmol), and benzyl bromide

(118 mg, 82 µL, 0.690 mmol); the reaction mixture was allowed to

warm to 0 °C; quench after 4 h at 0 °C; workup by GP A. Preparat-

ive HPLC (n-heptane/ethyl acetate, 95:5; 16.5 mL/min) gave a dia-

stereomeric mixture of 13a and 14a (66%) and 15a (2.8 mg, 3%).

The diastereoselectivity 13a/14a was determined by analytical

HPLC on the crude product as described above: ds 13a/14a ϭ

95.2:4.8.

Compound 15b: Colorless crystals, m.p. 68 °C. [α]²_D⁰ϭ Ϫ1.0 (c ϭ

0.59, CHCl₃). TLC: R_fϭ 0.69 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 1.02 [s, 9 H, (CH₃)₃C], 1.50 (s, 3

H, CH₃), 2.41 (dd, J ϭ 13.2/7.3 Hz, 1 H, CH₂CHϭCH₂),

2.52Ϫ2.66 (m, 3 H CH₂CHϭCH₂), 3.70 (s, 3 H, OCH₃), 5.08Ϫ5.16

(m, 4 H, CH₂CHϭCH₂), 5.64 (ddt, J ϭ 16.1/11.0/8.4 Hz, 1 H,

CH₂CHϭCH₂), 5.83 (ddt, J ϭ 17.2/10.1/7.5 Hz, 1 H, CH₂CHϭ

CH₂) ppm. IR: ν˜ ϭ 3080 cm^Ϫ1, 2957, 1736, 1686, 1440, 1348, 1297.

MS (CH₄; CI): m/z (%) ϭ 280 (100) [M ϩ H^ϩ]. C₁₆H₂₅NO₃(279.4):

calcd. C 68.79, H 9.02, N 5.01; found C 68.90, H 9.13, N 4.80.

Compound 15a: Colorless oil. TLC: R_fϭ 0.65 (petroleum ether/

ethyl acetate, 90:10). ¹H NMR (CDCl₃): δ ϭ 0.46 [s, 9 H, (CH₃)₃C],

1.55 (s, 3 H, CH₃), 2.94 (d, J ϭ 12.6 Hz, 1 H, CH₂C₆H₅), 3.12 (d,

J ϭ 12.8 Hz, 1 H, CH₂C₆H₅), 3.43 (d, J ϭ 12.6 Hz, 1 H,

CH₂C₆H₅), 3.51 (d, J ϭ 12.8 Hz, 1 H, CH₂C₆H₅), 3.77 (s, 3 H,

OCH₃), 7.13Ϫ7.26 (m, 10 H, C₆H₅) ppm. IR: ν˜ ϭ 2957 cm^Ϫ1, 1737,

1688, 1348, 1299, 1225. MS (CH₄; CI): m/z (%) ϭ 380 (64) [M ϩ

H^ϩ], 83 (100). C₂₄H₂₉NO₃(379.4): calcd. C 75.98, H 7.71, N 3.69;

found C 75.69, H 7.54, N 3.98.

(3R,6S)-6-tert-Butyl-5-methoxy-3,6-dimethyl-3,6-dihydro-2H-1,4-

oxazin-2-one (13c) and (3S,6S)-6-tert-Butyl-5-methoxy-3,6-di-

methyl-3,6-dihydro-2H-1,4-oxazin-2-one (14c): The compounds

were obtained by GP A, from (S)-2 (60 mg, 0.301 mmol), sBuLi

(331 µL, 0.255 mmol), and methyl iodide (128 mg, 56 µL,

0.903 mmol). Purification by CC (petroleum ether/ethyl acetate,

90:10) without separation of the diastereomers yielded 13c/14c

(46 mg, 72%). The diastereoselectivity was determined by analytical

HPLC (n-heptane/ethyl acetate, 95:5; 1.5 mL/min) on the crude

product: ds 13c/14c ϭ 96.8:3.2; 13c: t_Rϭ 20.05 min, 14c: t_Rϭ

23.95 min. 13c was isolated by preparative HPLC (n-heptane/ethyl

acetate, 95:5; 15 mL/min).

(3R,6S)-3-Allyl-6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-

1,4-oxazin-2-one (13b), (3S,6S)-3-Allyl-6-tert-butyl-5-methoxy-6-

methyl-3,6-dihydro-2H-1,4-oxazin-2-one (14b), and (S)-3,3-Diallyl-

6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

(15b). A: These compounds were produced by GP A; the reaction

was conducted at Ϫ90 °C, with (S)-2 (60 mg, 0.301 mmol), sBuLi

(255 µL, 0.331 mmol), and allyl bromide (109 mg, 77 µL,

0.903 mmol); quenching at Ϫ90 °C. Purification by CC (petroleum

ether/ethyl acetate, 90:10) yielded 13b/14b (68 mg, 94%) as a dias-

tereomeric mixture. The diastereoselectivity was determined by

analytical HPLC (n-heptane/ethyl acetate, 95:5, 1.5 mL/min) on the

crude product: ds 13b/14b ϭ 99.7:0.3; 13b: t_rϭ 19.84 min, 14b: t_rϭ

23.15 min. The diastereomers were separated by preparative HPLC

(n-heptane/ethyl acetate, 95:5; 16.5 mL/min).

Compound 13c: Colorless crystals, m.p. 50 °C. [α]²_D⁰ϭ ϩ109.6 (c ϭ

0.45, CHCl₃). TLC: R_fϭ 0.32 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 1.01 [s, 9 H, (CH₃)₃C], 1.51 (s, 3

H, CH₃), 1.54 (d, J ϭ 7.5 Hz, 3 H, CH₃CH), 3.71 (s, 3 H, OCH₃),

4.17 (q, J ϭ 7.5 Hz, 1 H, CH₃CH) ppm. IR: ν˜ ϭ 2974 cm^Ϫ1, 1746,

1690, 1475, 1370, 1310. MS (CH₄; CI): m/z (%) ϭ 214 (100) [M ϩ

H^ϩ]. C₁₁H₁₉NO₃(213.3): calcd. C 61.95, H 8.98, N 6.57; found C

62.13, H 9.07, N 6.31.

Compound 13b: Colorless crystals, m.p. 64 °C. [α]²_D⁰ϭ ϩ84.4 (c ϭ

0.61, CHCl₃). TLC: R_fϭ 0.33 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 0.99 [s, 9 H, (CH₃)₃C], 1.49 (s, 3

H, CH₃), 2.64Ϫ2.78 (m, 2 H, CH₂CHϭCH₂), 3.72 (s, 3 H, OCH₃),

4.23 (dd, J ϭ 6.2/4.8 Hz, 1 H, NCHCO), 5.09Ϫ5.18 (m, 2 H,

CH₂CHϭCH₂), 5.79 (ddt, J ϭ 17.2/10.2/7.2 Hz, 1 H, CH₂CHϭ

CH₂) ppm. IR: ν˜ ϭ 3078 cm^Ϫ1, 1739, 1690, 1441, 1348, 1328, 1223.

MS (CH₄; CI): m/z (%) ϭ 240 (100) [M ϩ H^ϩ]. C₁₃H₂₁NO₃(239.3):

calcd. C 65.25, H 8.84, N 5.85; found C 65.42, H 8.94, N 5.59.

Compound 14c: Only characterized by NMR: ¹H NMR (CDCl₃):

δ ϭ 1.03 [s, 9 H, (CH₃)₃C], 1.54 (s, 3 H, CH₃), 1.58 (d, J ϭ 7.5 Hz,

3 H, CH₃CH), 3.83 (s, 3 H, OCH₃), 4.16 (q, J ϭ 7.5 Hz, 1 H,

CH₃CH) ppm.

(3S,6R)-6-tert-Butyl-5-methoxy-[3-¹³C]methyl-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one [(ent)-13d] and (3R,6R)-6-tert-Butyl-5-

methoxy-[3-¹³C]methyl-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

[(ent)-14d]: sBuLi (3.91 mL, 5.08 mmol, 1.1 equiv.) was added at

Ϫ85 °C with stirring to (R)-2 (920 mg, 4.62 mmol) in THF

(10 mL). After 45 min, a precooled solution (Ϫ78 °C) of [¹³C]me-

thyl iodide (0.992 g, 438 µL, 6.94 mmol, 1.5 equiv.) in THF (3 mL)

was added. The reaction mixture was stirred for 24 h at Ϫ85 °C and

subsequently quenched at Ϫ85 °C with phosphate buffer (pH ϭ 7,

5 mL). The organic layer was separated and the aqueous layer was

extracted with Et₂O (4 ϫ 15 mL). The combined organic layers

were washed with brine, dried over MgSO₄, filtered, and concen-

trated. CC (petroleum ether/ethyl acetate, 90:10) afforded (ent)-13d

and (ent)-14d as a colorless oil (815 mg, 82%). The diastereoselec-

tivity was determined by analytical HPLC (n-heptane/ethyl acetate,

95:5, 1.5 mL/min) on the crude product: ds (ent)-13d/(ent)-14d ϭ

99.7:0.3; (ent)-13d: t_Rϭ 16.63 min, (ent)-14d: t_Rϭ 21.27 min.

1

Compound 14b: Only characterized by NMR: H NMR (CDCl₃):

δ ϭ 1.04 [s, 9 H, (CH₃)₃C], 1.52 (s, 3 H, CH₃), 2.66Ϫ2.80 (m, 2 H,

CH₂CHϭCH₂), 3.70 (s, 3 H, OCH₃), 4.30 (dd, J ϭ 6.2/4.8 Hz, 1

H, NCHCO), 5.08Ϫ5.17 (m, 2 H, CH₂CHϭCH₂), 5.90 (ddt, J ϭ

17.2/10.2/7.2 Hz, 1 H, CH₂CHϭCH₂).

B: The compounds were obtained by GP A, from (S)-2 (60 mg,

0.301 mmol), sBuLi (255 µL, 0.331 mmol), and allyl bromide

(109 mg, 77 µL, 0.903 mmol), with quenching at Ϫ78 °C. Purifica-

tion by CC (petroleum ether/ethyl acetate, 90:10) yielded 13b/14b

(55 mg, 76%) as a diastereomeric mixture and 15b (1 mg, 1%). The

diastereoselectivity was determined by analytical HPLC (n-hept-

ane/ethyl acetate, 95:5, 1.75 mL/min) on the crude product: ds 13b/

14b ϭ 99.2:0.8; 13b: t_Rϭ 9.03 min, 14b: t_Rϭ 11.63 min.

C: The compounds were obtained by GP A, from (S)-2 (86 mg,

0.43 mmol), sBuLi (365 µL, 0.475 mmol), and allyl bromide 0.32 (petroleum ether/ethyl acetate, 90:10). H NMR (CDCl₃): δ ϭ

Compound (ent)-13d: [α]²_D⁰ϭ Ϫ103.6 (c ϭ 2.20, CHCl₃). TLC: R_fϭ

1

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1255

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

1.02 [s, 9 H, (CH₃)₃C], 1.51 (s, 3 H, CH₃), 1.55 (dd, J ϭ 129.8/ on the crude product: ds 13f/14f ϭ 92.7:7.3; 13f: t_Rϭ 4.89 min, 14f:

7.3 Hz, 3 H, ¹³CH₃CH), 3.71 (s, 3 H, OCH₃), 4.17 (dq, J ϭ 7.3/ t ϭ 6.05 min. Compound 13f was isolated by preparative HPLC (n-

7.3 Hz, 1 H, CH₃CH) ppm. IR: ν˜ ϭ 2972 cm^Ϫ1, 1747, 1692, 1447, heptane/ethyl acetate, 95:5; 17 mL/min).

1376, 1325, 1122. MS (CH₄; CI): m/z (%) ϭ 215 (100) [M ϩ H^ϩ].

Compound 13f: Colorless crystals, m.p. 52 °C. [α]²_D⁰ϭ ϩ90.2 (c ϭ

C₁₀¹³CH₁₉NO₃(214.3): calcd. C 61.95, H 8.94, N 6.54; found C

61.64, H 9.09, N 6.55.

0.65, CHCl₃). TLC: R_fϭ 0.42 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 0.78 [d, J ϭ 6.9 Hz, 3 H,

Compound (ent)-14d: Only characterized by NMR: ¹H NMR

CH(CH₃)₂], 1.01 [s, 9 H, (CH₃)₃C], 1.12 [d, J ϭ 6.9 Hz, 3 H,

(CDCl₃): δ ϭ 1.03 [s, 9 H, (CH₃)₃C], 1.52 (s, 3 H, CH₃), 1.54 (dd, CH(CH₃)₂], 1.49 (s, 3 H, CH₃), 2.52 [septd, J ϭ 6.9/3.1 Hz, 1 H,

J ϭ 129.8/7.3 Hz, 3 H, ¹³CH₃CH), 3.68 (s, 3 H, OCH₃), 4.17 (dq,

J ϭ 7.3/7.3 Hz, 1 H, CH₃CH) ppm.

CH(CH₃)₂], 3.73 (s, 3 H, OCH₃), 4.00 (d, J ϭ 3.1 Hz, 1 H,

NCHCO) ppm. IR: ν˜ ϭ 2963 cm^Ϫ1, 2874, 1745, 1698, 1462, 1312,

1104. MS (CH₄; CI): m/z (%) ϭ 242 (26) [M ϩ H^ϩ], 149 (100).

C₁₃H₂₃NO₃(241.3): calcd. C 64.70, H 9.61, N 5.80; found C 64.66,

H 9.75, N 5.65.

(3R,6S)-3-n-Butyl-6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-

1,4-oxazin-2-one (13e), (3S,6S)-3-n-Butyl-6-tert-butyl-5-methoxy-6-

methyl-3,6-dihydro-2H-1,4-oxazin-2-one (14e), and (6S)-3,3-Di-n-

Butyl-6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-

2-one (15e): The compounds were obtained by GP A, from (S)-2

(50 mg, 0.25 mmol) in DME (2.5 mL) at ؊50 °C with sBuLi (212

µL, 0.275 mmol) and n-butyl iodide (138 mg, 86 µL, 0.75 mmol).

Compound 14f: Only characterized by NMR: ¹H NMR (CDCl₃):

δ ϭ 0.82 (d, J ϭ 6.9 Hz, 3 H, CH(CH₃)₂], 1.06 [s, 9 H, (CH₃)₃C],

1.14 (d, J ϭ 6.9 Hz, 3 H, CH(CH₃)₂], 1.49 (s, 3 H, CH₃), 2.52

(septd, J ϭ 6.9/3.1 Hz, 1 H, CH(CH₃)₂], 3.72 (s, 3 H, OCH₃), 3.97

Purification by CC (petroleum ether/ethyl acetate, 90:10) without (d, J ϭ 3.1 Hz, 1 H, NCHCO).

separation of the diastereomers yielded 13e/14e (46 mg, 72%), to-

Compound 15f: Colorless crystals, m.p. 74Ϫ77 °C. [α]²_D⁰ϭ ϩ0.5

gether with 15e (4.2 mg, 5%) and (S)-2 (8.4 mg, 17%). The diastereo-

selectivity was determined by analytical HPLC (n-heptane/ethyl

acetate, 95:5, 1.75 mL/min) on the crude product: ds 13e/14e ϭ

97.0:3.0; 13e: t_Rϭ 6.32 min, 14e: t_Rϭ 8.10 min. Compound 13e was

isolated by preparative HPLC (n-heptane/ethyl acetate, 95:5;

17 mL/min).

(c ϭ 0.39, CHCl₃). TLC: R_fϭ 0.49 (n-pentane/ethyl acetate, 97:3).

¹H NMR (CDCl₃): δ ϭ 0.90 [d, J ϭ 7.0 Hz, 3 H, CH(CH₃)₂], 0.93

[d, J ϭ 7.0 Hz, 3 H, CH(CH₃)₂], 0.95 [d, J ϭ 7.0 Hz, 3 H,

CH(CH₃)₂], 0.99 [d, J ϭ 7.0 Hz, 3 H, CH(CH₃)₂], 1.06 [s, 9 H,

(CH₃)₃C], 1.55 (s, 3 H, CH₃), 2.19 [sept, J ϭ 7.0 Hz, 1 H,

CH(CH₃)₂], 2.33 [sept, J ϭ 7.0 Hz, 1 H, CH(CH₃)₂], 3.71 (s, 3 H,

OCH₃) ppm. IR: ν˜ ϭ 2973cm^Ϫ1, 2879, 1728, 1690, 1476, 1296,

Compound 13e: Colorless crystals, m.p. 71 °C. [α]²_D⁰ϭ ϩ90.9 (c ϭ

0.45, CHCl₃). TLC: R_fϭ 0.38 (petroleum ether/ethyl acetate, 1094, 831. MS (CH₄; CI): m/z (%) ϭ 284 (24) [M ϩ H^ϩ], 225 (68),

90:10). ¹H NMR (CDCl₃): δ ϭ 0.88Ϫ0.95 [m, 3 H, CH₃(CH₂)₃], 105 (100). C₁₆H₂₉NO₃(283.4): calcd. C 67.81, H 10.31, N 4.94;

1.01 [s, 9 H, (CH₃)₃C], 1.31Ϫ1.39 [m, 4 H, CH₃(CH₂)₂CH₂], 1.50

(s, 3 H, CH₃), 1.81Ϫ1.89 (m, 1 H, CH₂C₃H₇), 1.93Ϫ2.05 (m, 1 H,

CH₂C₃H₇), 3.72 (s, 3 H, OCH₃), 4.13 (dd, J ϭ 6.6/4.8 Hz, 1 H,

NCHCO) ppm. IR: ν˜ ϭ 2960 cm^Ϫ1, 1743, 1692, 1458, 1344, 1313.

MS (CH₄; CI): m/z (%) ϭ 256 (100) [M ϩ H^ϩ]. C₁₄H₂₅NO₃(255.4):

calcd. C 65.85, H 9.87, N 5.49; found C 65.92, H 9.80, N 5.50.

found C 67.36, H 10.40, N 4.76.

General Procedure B (GP B) for the Alkylation of 13/14a-c, e, f:

Unless otherwise stated, a solution of the monoalkylation product

(0.1 , either pure or as diastereomeric mixture) in THF was

treated at Ϫ78 °C with sBuLi (1.3  solution in hexane, 1.1 equiv.).

After 30 min, alkyl halide (3.0 equiv.) was added and the solution

was stirred at Ϫ78 °C overnight. The mixture was quenched at this

Compound 14e: Only characterized by NMR: ¹H NMR (CDCl₃):

δ ϭ 0.84Ϫ0.91 (m, 3 H, CH₃(CH₂)₃], 1.01 [s, 9 H, (CH₃)₃C], temperature and allowed to warm to room temp. The hydrolysis

1.35Ϫ1.45 (m, 4 H, CH₃(CH₂)₂CH₂), 1.51 (s, 3 H, CH₃), 1.83Ϫ1.88 was performed by addition of phosphate buffer (pH ϭ 7, c ϭ 1.0

(m, 1 H, CH₂C₃H₇), 1.96Ϫ2.01 (m, 1 H, CH₂C₃H₇), 3.70 (s, 3 H, mol/L). The organic layer was separated and the aqueous layer was

OCH₃), 4.13 (dd, J ϭ 6.6/4.8 Hz, 1 H, NCHCO) ppm.

extracted with Et₂O (3 ϫ 5Ϫ10 mL). The combined organic layers

were washed with brine, dried over MgSO₄, filtered, and concen-

trated. The resulting residue was purified by CC.

Compound 15e: Colorless crystals, m.p. 30 °C. [α]²_D⁰ϭ ϩ6.2 (c ϭ

1.035, CHCl₃). TLC: R_fϭ 0.51 (petroleum ether/diethyl ether,

1

90:10). H NMR (CDCl₃): δ ϭ 0.83Ϫ0.92 (m, 6 H, CH₃), 1.04 [s, (3S,6S)-3-Benzyl-6-tert-butyl-5-methoxy-3,6-dimethyl-3,6-dihydro-

9 H, (CH₃)₃C], 1.11Ϫ1.20 (m, 2 H, CH₂), 1.24Ϫ1.35 (m, 6 H, CH₂), 2H-1,4-oxazin-2-one (17a) and (3R,6S)-3-Benzyl-6-tert-butyl-5-me-

1.52 (s, 3 H, CH₃), 1.63Ϫ1.70 (m, 1 H, CH₂), 1.72Ϫ1.88 (m, 3 H,

CH₂), 3.69 (s, 3 H, OCH₃) ppm. IR: ν˜ ϭ 2960 cm^Ϫ1, 2873, 1740, compounds were obtained by GP B, from 13a/14a (30 mg,

1686, 1458, 1303, 1100. MS (CH₄; CI): m/z (%) ϭ 312 (100) [M ϩ 0.104 mmol), sBuLi (88 µL, 0.114 mmol), and methyl iodide

H^ϩ]. C₁₈H₃₃NO₃(311.5): calcd. C 69.41, H 10.68, N 4.50; found (44 mg, 20 µL, 0.312 mmol). Purification by CC (petroleum ether/

thoxy-3,6-dimethyl-3,6-dihydro-2H-1,4-oxazin-2-one (18a): The

C 69.91, H 10.86, N 4.41.

ethyl acetate, 90:10) without separation of the diastereomers

yielded 17a/18a (25 mg, 79%). The diastereoselectivity was deter-

mined by analytical HPLC (n-heptane/ethyl acetate, 98:2; 1.5 mL/

min) on the crude product: ds 17a/18a ϭ 95.3:4.7; 17a: t_rϭ

30.25 min, 18a: t_rϭ 26.29 min. Compound 17a was isolated by pre-

parative HPLC (n-heptane/ethyl acetate, 98:2; 15 mL/min).

(3R,6S)-6-tert-Butyl-3-isopropyl-5-methoxy-6-methyl-3,6-dihydro-

2H-1,4-oxazin-2-one (13f), (3S,6S)-6-tert-Butyl-3-isopropyl-5-me-

thoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one (14f), and (6S)-6-

tert-Butyl-3,3-diisopropyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-

oxazin-2-one (15f): The compounds were obtained by GP A, from

(S)-2 (25 mg, 0.125 mmol), phosphazenic base tBuϪP₄(c ϭ 1  in

Compound 17a: Colorless crystals, m.p. 72 °C. [α]²_D⁰ϭ ϩ49.9 (c ϭ

n-hexane, 138 µL, 0.138 mmol, 1.1 equiv.), and isopropyl iodide 0.225, CHCl₃). TLC: R_fϭ 0.49 (petroleum ether/ethyl acetate,

(64 mg, 38 µL, 0.38 mmol, 3.0 equiv.). Purification by CC (petro-

leum ether/ethyl acetate, 90:10) without separation of the diastereo-

90:10). ¹H NMR (CDCl₃): δ ϭ 0.54 [s, 9 H, (CH₃)₃C], 1.37 (s, 3

H, CH₃), 1.46 (s, 3 H, CH₃), 2.84 (d, J ϭ 12.8 Hz, 1 H, CH₂C₆H₅),

mers yielded 13f/14f (19.2 mg, 63%), together with 15f (1.9 mg, 5%) 3.33 (d, J ϭ 12.8 Hz, 1 H, CH₂C₆H₅), 3.65 (s, 3 H, OCH₃),

and (S)-2 (4.1 mg, 16%). The diastereoselectivity was determined

by analytical HPLC (n-heptane/ethyl acetate, 95:5; 1.75 mL/min)

7.05Ϫ7.15 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ 2972 cm^Ϫ1, 1738, 1694,

1454, 1369, 1308, 1220. MS (CH₄; CI): m/z (%) ϭ 304 (100) [M ϩ

1256

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

H^ϩ]. C₁₈H₂₅NO₃(303.4): calcd. C 71.26, H 8.31, N 4.62; found C

71.41, H 8.45, N 4.50.

(3R,6S)-3-Benzyl-6-tert-butyl-3-isopropyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (17d) and (3S,6S)-3-Benzyl-6-tert-bu-

tyl-3-isopropyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-

one (18d): The compounds were obtained by GP B, from 13a/14a

(29 mg, 0.10 mmol), phosphazenic base tBuϪP₄(c ϭ 1.0  in n-

hexane, 110 µL, 0.11 mmol, 1.1 equiv.), and isopropyl iodide

(53 mg, 30 µL, 0.30 mmol) immediately added, with quenching

after 1 h. Purification by CC (n-pentane/ethyl acetate, 98:2) without

separation of the diastereomers yielded 17d/18d (28.3 mg, 86%).

The diastereoselectivity was determined by analytical HPLC (n-

heptane/ethyl acetate, 98:2, 1.75 mL/min) on the crude product: ds

17d/18d ϭ 95.1:4.9; 17d: t_Rϭ 8.83 min, 18d: t_Rϭ 8.11 min. Com-

pound 17d was isolated by preparative HPLC (n-heptane/ethyl

acetate, 98:2; 16.5 mL/min).

Compound 18a: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see below).

(3S,6S)-3-Allyl-3-benzyl-6-tert-butyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (17b) and (3R,6S)-3-Allyl-3-benzyl-6-

tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

(18b): The compounds were obtained by GP B, from 13a/14a

(30 mg, 0.104 mmol), sBuLi (88 µL, 0.114 mmol), and allyl brom-

ide (38 mg, 27 µL, 0.312 mmol). Purification by CC (petroleum

ether/ethyl acetate, 90:10) without separation of the diastereomers

yielded 17b/18b (20.4 mg, 60%). The diastereoselectivity was deter-

mined by analytical HPLC (n-heptane/ethyl acetate, 98:2, 1.75 mL/

min) on the crude product: ds 17b/18b ϭ 96.6:3.4; 17b: t_Rϭ

15.32 min, 18b: t_Rϭ 14.34 min. Compound 17b was isolated by pre-

parative HPLC (n-heptane/ethyl acetate, 98:2; 16.5 mL/min).

Compound 17d: Colorless crystals, m.p. 74 °C. [α]²_D⁰ϭ ϩ34.7 (c ϭ

2.73, CHCl₃). TLC: R_fϭ 0.45 (n-pentane/ethyl acetate, 96:4). ¹H

NMR (CDCl₃): δ ϭ 0.51 [s, 9 H, (CH₃)₃C], 0.98 [d, J ϭ 6.8 Hz, 3

H, CH(CH₃)₂], 1.08 [d, J ϭ 6.8 Hz, 3 H, CH(CH₃)₂], 1.45 (s, 3 H,

CH₃), 2.21 [sept, J ϭ 6.8 Hz, 1 H, CH(CH₃)₂], 3.06 (d, J ϭ 12.8 Hz,

1 H, CH₂C₆H₅), 3.30 (d, J ϭ 12.8 Hz, 1 H, CH₂C₆H₅), 3.74 (s, 3

Compound 17b: Colorless crystals, m.p. 64 °C. [α]²_D⁰ϭ ϩ27.1 (c ϭ

0.47, CHCl₃). TLC: R_fϭ 0.46 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 0.60 [s, 9 H, (CH₃)₃C], 1.44 (s, 3

H, CH₃), 2.46 (dd, J ϭ 13.2/7.5 Hz, 1 H, CH₂ϪCHϭCH₂), 2.74

(dd, J ϭ 13.2/7.5 Hz, 1 H, CH₂ϪCHϭCH₂), 3.00 (d, J ϭ 13.0 Hz,

1 H, CH₂C₆H₅), 3.35 (d, J ϭ 13.0 Hz, 1 H, CH₂C₆H₅), 3.72 (s, 3

H, OCH₃), 5.11Ϫ5.18 (m, 2 H, CH₂ϪCHϭCH₂), 5.71 (ddt, J ϭ

17.6/9.5/7.5 Hz, 1 H, CH₂ϪCHϭCH₂), 7.14Ϫ7.21 (m, 5 H, C₆H₅)

ppm. IR: ν˜ ϭ 3086 cm^Ϫ1, 2987, 1730, 1685, 1482, 1372, 1299. MS

(CH₄; CI): m/z (%) ϭ 330 (100) [M ϩ H^ϩ]. C₂₀H₂₇NO₃(329.4):

calcd. C 72.92, H 8.26, N 4.25; found C 72.77, H 8.30, N 3.96.

H, OCH₃), 7.12Ϫ7.18 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ 2966 cm^Ϫ1

,

1743, 1689, 1456, 1368, 1311. MS (CH₄; CI): m/z (%) ϭ 332 (100)

[M ϩ H^ϩ]. C₂₀H₂₉NO₃(331.5): calcd. C 72.47, H 8.82, N 4.23;

found C 72.48, H 8.79, N 4.25.

Compound 18d: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see below).

(3R,6S)-3-Benzyl-6-tert-butyl-5-methoxy-3,6-dimethyl-3,6-dihydro-

2H-1,4-oxazin-2-one (18a) and (3S,6S)-3-Benzyl-6-tert-butyl-5-me-

thoxy-3,6-dimethyl-3,6-dihydro-2H-1,4-oxazin-2-one (17a): The

compounds were obtained by GP B, from 13c/14c (48 mg,

0.225 mmol), sBuLi (191 µL, 0.248 mmol), and benzyl bromide

(116 mg, 81 µL, 0.68 mmol). Purification by CC (petroleum ether/

ethyl acetate, 90:10) without separation of the diastereomers

yielded 17a/18a (56 mg, 82%). The diastereoselectivity was deter-

mined by analytical HPLC (n-heptane/ethyl acetate, 98:2, 1.5 mL/

min) on the crude product: ds 18a/17a ϭ 98.5:1.5; 18a: t_Rϭ

25.12 min, 17a: t_Rϭ 33.33 min. Compound 18a was isolated by pre-

parative HPLC (n-heptane/ethyl acetate, 98:2; 15 mL/min).

Compound 18b: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see below).

(3S,6S)-3-Benzyl-3-n-butyl-6-tert-butyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (17c) and (3R,6S)-3-Benzyl-3-n-butyl-

6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

(18c): The compounds were obtained by GP B, from 13a/14a

(29 mg, 0.10 mmol) in DME at Ϫ50 °C, with sBuLi (1.19  in

n-hexane, 92 µL, 0.11 mmol), and n-butyl iodide (55 mg, 35 µL,

0.30 mmol). Purification by CC (petroleum ether/ethyl acetate,

90:10) without separation of the diastereomers yielded 17c/18c

(22.6 mg, 66%), together with 13a/14a (6.1 mg, 21%). The diaster-

eoselectivity was determined by analytical HPLC (n-heptane/ethyl

acetate, 98:2, 1.75 mL/min) on the crude product: ds 17c/18c ϭ

94.1:5.9; 17c: t_Rϭ 8.13 min, 18c: t_Rϭ 9.16 min. Compound 17c was

isolated by preparative HPLC (n-heptane/ethyl acetate, 98:2;

16.5 mL/min).

Compound 18a: Colorless crystals, m.p. 69 °C. [α]²_D⁰ϭ Ϫ35.2 (c ϭ

0.305, CHCl₃). TLC: R_fϭ 0.49 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 0.52 (s, 3 H, CH₃), 0.91 [s, 9 H,

(CH₃)₃C], 1.63 (s, 3 H, CH₃), 2.85 (d, J ϭ 12.6 Hz, 1 H, CH₂C₆H₅),

3.32 (d, J ϭ 12.6 Hz, 1 H, CH₂C₆H₅), 3.72 (s, 3 H, OCH₃),

7.09Ϫ7.23 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ 2970 cm^Ϫ1, 1741, 1692,

1450, 1373, 1311, 1221. MS (CH₄; CI): m/z (%) ϭ 304 (100) [M ϩ

H^ϩ]. C₁₈H₂₅NO₃(303.4): calcd. C 71.26, H 8.31, N 4.62; found C

71.28, H 8.40, N 4.55.

Compound 17c: Colorless oil. [α]²_D⁰ϭ ϩ29.3 (c ϭ 1.15, CHCl₃).

TLC: R_fϭ 0.63 (petroleum ether/ethyl acetate, 90:10). ¹H NMR

(CDCl₃): δ ϭ 0.63 [s, 9 H, (CH₃)₃C], 0.89 [t, J ϭ 7.1 Hz, 3 H,

(CH₂)₃CH₃), 1.19Ϫ1.34 [m, 4 H, CH₂(CH₂)₂CH₃], 1.46 (s, 3 H,

CH₃), 1.67Ϫ1.75 (m, 1 H, CH₂C₃H₇), 1.95Ϫ2.03 (m, 1 H,

CH₂C₃H₇), 2.99 (d, J ϭ 12.9 Hz, 1 H, CH₂C₆H₅), 3.32 (d, J ϭ

12.9 Hz, 1 H, CH₂C₆H₅), 3.72 (s, 3 H, OCH₃), 7.10Ϫ7.22 (m, 5 H,

C₆H₅) ppm. IR: ν˜ ϭ 2959 cm^Ϫ1, 2872, 1736, 1693, 1456, 1302,

1221. MS (CH₄; CI): m/z (%) ϭ 346 (100) [M ϩ H^ϩ]. C₂₁H₃₁NO₃

(345.5): calcd. C 73.01, H 9.04, N 4.05; found C 73.07, H 9.08,

N 3.96.

Compound 17a: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see above).

(3R,6S)-3-Allyl-3-benzyl-6-tert-butyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (18b) and (3S,6S)-3-Allyl-3-benzyl-6-

tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

(17b): The compounds were obtained by GP B, from 13b/14b

(32 mg, 0.135 mmol), sBuLi (114 µL, 0.149 mmol), and benzyl

bromide (69 mg, 48 µL, 0.41 mmol). Purification by CC (petroleum

ether/ethyl acetate, 90:10) without separation of the diastereomers

yielded 17b/18b (37.6 mg, 85%). The diastereoselectivity was deter-

mined by analytical HPLC (n-heptane/ethyl acetate, 98:2; 1.5 mL/

Compound 18c: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see below).

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1257

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

min) on the crude product: ds 18b/17b ϭ 99.0:1.0; 18b: t_Rϭ

13.97 min, 17b: t_Rϭ 15.55 min. Compound 18b was isolated by pre-

parative HPLC (n-heptane/ethyl acetate, 98:2; 16.5 mL/min).

NMR (CDCl₃): δ ϭ 0.18 (s, 3 H, CH₃), 0.91 [d, J ϭ 6.8 Hz, 3 H,

CH(CH₃)₂], 0.92 [s, 9 H, (CH₃)₃C], 1.15 [d, J ϭ 6.8 Hz, 3 H,

CH(CH₃)₂], 2.41 [sept, J ϭ 6.8 Hz, 1 H, CH(CH₃)₂], 3.02 (d, J ϭ

12.7 Hz, 1 H, CH₂C₆H₅), 3.18 (d, J ϭ 12.7 Hz, 1 H, CH₂C₆H₅),

3.76 (s, 3 H, OCH₃), 7.10Ϫ7.25 (m, 5 H, C₆H₅) ppm. IR: ν˜ ϭ 2964

cm^Ϫ1, 2876, 1727, 1692, 1439, 1298, 1097, 1039, 702. MS (CH₄;

CI): m/z (%) ϭ 332 (100) [M ϩ H^ϩ], 184 (12). C₂₀H₂₉NO₃(331.5):

calcd. C 72.47, H 8.82, N 4.23; found C 72.32, H 8.97, N 4.04.

Compound 18b: Colorless crystals, m.p. 70 °C. [α]²_D⁰ϭ Ϫ36.7 (c ϭ

0.51, CHCl₃). TLC: R_fϭ 0.46 (petroleum ether/ethyl acetate,

90:10). ¹H NMR (CDCl₃): δ ϭ 0.38 (s, 3 H, CH₃), 0.89 [s, 9 H,

(CH₃)₃C], 2.64Ϫ2.82 (m, 2 H, CH₂ϪCHϭCH₂), 2.92 (d, J ϭ

12.7 Hz, 1 H, CH₂C₆H₅), 3.26 (d, J ϭ 12.7 Hz, 1 H, CH₂C₆H₅),

3.76 (s, 3 H, OCH₃), 5.12Ϫ5.22 (m, 2 H, CH₂ϪCHϭCH₂), 5.90

(ddt, J ϭ 16.8/10.3/7.7 Hz, 1 H, CH₂ϪCHϭCH₂), 7.14Ϫ7.21 (m,

5 H, C₆H₅) ppm. IR: ν˜ ϭ 3089 cm^Ϫ1, 2985, 1735, 1686, 1477, 1369,

1220. MS (CH₄; CI): m/z (%) ϭ 330 (100) [M ϩ H^ϩ]. C₂₀H₂₇NO₃

(329.4): calcd. C 72.92, H 8.26, N 4.25; found C 73.09, H 8.42,

N 4.50.

Compound 17d: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see above).

(3R,6S)-3-Benzyl-6-tert-butyl-3-hydroperoxy-5-methoxy-6-methyl-

3,6-dihydro-2H-1,4-oxazin-2-one (19). A: Mixture 13a/14a (189 mg,

0.632 mmol) in THF (7 mL) was treated at Ϫ80 °C with sBuLi (c ϭ

1.15  in cyclohexane, 0.60 mL, 0.717 mmol, 1.1 equiv.) and stirred

for 30 min. Dry air (700 mL) was then slowly bubbled at Ϫ78 °C

through the solution over 2 h. Finally, after the reaction mixture

had been quenched with phosphate buffer (pH ϭ 7, 10 mL), it was

extracted with Et₂O (4 ϫ 10 mL) and the combined organic layers

were dried over MgSO₄and concentrated in vacuo. Purification by

CC (petroleum ether/ethyl acetate, 90:10) yielded 19 (170 mg, 81%)

as colorless crystals, m.p. 85 °C. [α]²_D⁰ϭ ϩ25.2 (c ϭ 0.72, CHCl₃).

Compound 17b: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see above).

(3R,6S)-3-Benzyl-3-n-butyl-6-tert-butyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (18c) and (3S,6S)-3-Benzyl-3-n-butyl-

6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one

(17c): The compounds were obtained by GP B, from 13e/14e

1

(38 mg, 0.15 mmol), sBuLi (1.19

 in n-hexane, 139 µL,

TLC: R_fϭ 0.34 (isohexane/ethyl acetate, 3:1). H NMR (CDCl₃):

0.165 mmol), and benzyl bromide (77 mg, 54 µL, 0.45 mmol). Puri-

fication by CC (petroleum ether/ethyl acetate, 90:10) without sep-

aration of the diastereomers yielded 17c/18c (44.1 mg, 85%), to-

gether with 13e/14e (4.1 mg, 11%). The diastereoselectivity was de-

termined by analytical HPLC (n-heptane/ethyl acetate, 98:2;

1.5 mL/min) on the crude product: ds 18c/17c

؍

99.2:0.8; 18c: t_Rϭ

9.29 min, 17c: t_rϭ 8.55 min. Compound 18c was isolated by prepar-

ative HPLC (n-heptane/ethyl acetate, 98:2; 16.5 mL/min).

δ ϭ 0.64 [s, 9 H, (CH₃)₃C], 1.55 (s, 3 H, CH₃), 1.62 (s, broad, 1 H,

OH), 3.08 (d, J ϭ 12.9 Hz, 1 H, CH₂C₆H₅), 3.40 (d, J ϭ 12.9 Hz,

1 H, CH₂C₆H₅), 3.85 (s, 3 H, OCH₃), 7.16Ϫ7.24 (m, 5 H, C₆H₅)

ppm. IR: ν˜ ϭ 3347 cm^Ϫ1, 2965, 1746, 1716, 1685, 1663, 1460, 1348,

1226. MS (CH₄; CI): m/z (%) ϭ 322 (2) [M ϩ H^ϩ], 288 (32), 232

(49), 214 (100). C₁₇H₂₃NO₅(321.4): calcd. C 63.50, H 7.21, N 4.36;

found C 63.50, H 7.15, N 4.36.

B: The compound was obtained by GP B, from 13a/14a (53 mg,

0.18 mmol), sBuLi (154 µL, 0.20 mmol), and isopropyl iodide

(92 mg, 51 µL, 0.54 mmol). Purification by CC (petroleum ether/

ethyl acetate, 90:10) afforded 19 (46 mg, 83%) as colorless crystals,

m.p. 85 °C.

Compound 18c: Colorless oil. [α]²_D⁰ϭ Ϫ33.6 (c ϭ 0.955, CHCl₃).

TLC: R_fϭ 0.60 (petroleum ether/ethyl acetate, 90:10). ¹H NMR

(CDCl₃): δ ϭ 0.32 (s, 3 H, CH₃), 0.85Ϫ0.99 (m, 3 H, (CH₂)₃CH₃),

0.91 [s, 9 H, (CH₃)₃C], 1.30Ϫ1.40 [m, 4 H, CH₂(CH₂)₂CH₃],

1.86Ϫ1.95 (m, 1 H, CH₂C₃H₇), 2.02Ϫ2.10 (m, 1 H, CH₂C₃H₇),

2.90 (d, J ϭ 12.8 Hz, 1 H, CH₂C₆H₅), 3.26 (d, J ϭ 12.8 Hz, 1 H,

CH₂C₆H₅), 3.74 (s, 3 H, OCH₃), 7.10Ϫ7.25 (m, 5 H, C₆H₅) ppm.

IR: ν˜ ϭ 2948 cm^Ϫ1, 2665, 1731, 1689, 1452, 1296, 1121, 1101. MS

(CH₄; CI): m/z (%) ϭ 346 (100) [M ϩ H^ϩ], 290 (18). C₂₁H₃₁NO₃

(345.5): calcd. C 73.01, H 9.04, N 4.05; found C 72.59, H 9.05,

N 3.99.

(3R,6S)-(3-Benzyl-6-tert-butyl-5-methoxy-6-methyl-2-oxo-3,6-

dihydro-2H-1,4-oxazin-3-yl) Peracetate (20): Compound 19 (59 mg,

0.18 mmol) was dissolved in acetic anhydride (18 mL) and stirred

for 2 h at room temp. After addition of H₂O (46 mL) and extrac-

tion with Et₂O (4 ϫ 10 mL), the combined organic layers were

dried over MgSO₄and concentrated in vacuo to yield 20 (54 mg,

81%) as colorless crystals, m.p. 56Ϫ57 °C. [α]²_D⁰ϭ ϩ64.5 (c ϭ 0.75,

CHCl₃). TLC: R_fϭ 0.23 (isohexane/ethyl acetate, 80:20). ¹H NMR

(CDCl₃): δ ϭ 0.60 [s, 9 H, (CH₃)₃C], 1.53 (s, 3 H, CH₃), 2.10 (s, 3

H, CH₃CO), 3.23 (d, J ϭ 12.6 Hz, 1 H, CH₂C₆H₅), 3.52 (d, J ϭ

12.6 Hz, 1 H, CH₂C₆H₅), 3.80 (s, 3 H, OCH₃), 7.19Ϫ7.26 (m, 5 H,

C₆H₅) ppm. IR: ν˜ ϭ 2969 cm^Ϫ1, 1796, 1750, 1676, 1456, 1342,

1187, 1102. MS (CH₄; CI): m/z (%) ϭ 364 (26) [M ϩ H^ϩ], 214

(100). C₁₉H₂₅NO₆(363.4): calcd. C 62.80, H 6.93, N 3.85; found C

62.84, H 6.74, N 3.87.

Compound 17c: A selective synthesis of this compound, not isolated

from the above experiment, was accomplished by an inverse al-

kylation sequence (see above).

(3S,6S)-3-Benzyl-6-tert-butyl-3-isopropyl-5-methoxy-6-methyl-3,6-

dihydro-2H-1,4-oxazin-2-one (18d) and (3R,6S)-3-Benzyl-6-tert-bu-

tyl-3-isopropyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-

one (17d): The compounds were obtained by GP B, from 13f/14f

(31 mg, 0.128 mmol), sBuLi (c ϭ 1.24  in cyclohexane, 114 µL,

0.141 mmol), and benzyl bromide (66 mg, 46 µL, 0.38 mmol). Puri-

fication by CC (n-pentane/ethyl acetate, 98:2) without separation

of the diastereomers yielded 17d/18d (33.6 mg, 79%). The diaster-

eoselectivity was determined by analytical HPLC (n-heptane/ethyl

acetate, 98:2; 1.75 mL/min) on the crude product: ds 18d/17d

؍

99.6:0.4; 18d: t_rϭ 7.87 min, 17d: t_rϭ 8.79 min. Compound 18d was

isolated by preparative HPLC (n-heptane/ethyl acetate, 98:2;

16.5 mL/min).

(1S,3R,6S)-6-tert-Butyl-1-hydroxymethyl-5-methoxy-6-methyl-7-

oxa-4-azaspiro[2.5]oct-4-en-8-one (21): NaN(SiMe₃)₂(c ϭ 1.0  in

THF, 4.42 mL, 4.42 mmol, 2.2 equiv.) was slowly (30 h) added at

Ϫ20 °C to (S)-2 (400 mg, 2.01 mmol) and (S)-(ϩ)-epichlorohydrin

(1.12 g, 946 µL, 12.06 mmol, 6.0 equiv.) in THF (20 mL). After

another 16 h stirring at Ϫ20 °C the reaction mixture was hy-

drolyzed with phosphate buffer (pH ϭ 7, 20 mL) and the organic

solvent was removed in vacuo. The aqueous layer was extracted

with CH₂Cl₂(5 ϫ 50 mL). The combined organic layers were dried

over MgSO₄and filtered, and the solvent was removed in vacuo.

Compound 18d: Colorless crystals, m.p. 70 °C. [α]²_D⁰ϭ Ϫ58.2 (c ϭ

0.52, CHCl₃). TLC: R_fϭ 0.38 (isohexane/ethyl acetate, 98:2). ¹H CC (petroleum ether/ethyl acetate, 70:30) yielded 215 mg (42%) of

1258

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

21 as colorless crystals, m.p. 90 °C. The diastereoselectivity was

determined by analytical HPLC (n-heptane/ethyl acetate, 85:15;

2.0 mL/min) on the crude product: no other diastereomer being

detected, ds Ͼ 99.9:0.1; 21: t_rϭ 25.74 min. [α]²_D⁰ϭ ϩ61.5 (c ϭ

0.183, CHCl₃). TLC: R_fϭ 0.20 (petroleum ether/ethyl acetate,

70:30). ¹H NMR (CDCl₃): δ ϭ 1.04 [s, 9 H, (CH₃)₃C], 1.47 (dd,

J ϭ 7.7/4.6 Hz, 1 H, CH₂CH), 1.56 (s, 3 H, CH₃), 1.78 (dd, J ϭ

9.7/4.6 Hz, 1 H, CH₂CH), 2.20Ϫ2.27 (m, 1 H, CH₂CH), 3.67 (s, 3

H, OCH₃), 3.90 (dd, J ϭ 12.1/5.6 Hz, 1 H, CH₂OH), 4.10 (dd, J ϭ

12.1/2.8 Hz, 1 H, CH₂OH), OH ppm, not located. IR (KBr): ν˜ ϭ

3448 cm^Ϫ1, 1739, 1685, 1370, 1211. MS (CH₄; CI): m/z (%) ϭ 256

(100) [M ϩ H^ϩ]. C₁₃H₂₁NO₄(255.3): calcd. C 61.16, H 8.29, N

5.49; found C 61.23, H 8.26, N 5.44.

1264, 1212. MS (CH₄; CI): m/z (%) ϭ 242 (100) [M ϩ H^ϩ].

C₁₂H₁₉NO₄(241.3): calcd. C 59.73, H 7.94, N 5.81; found C 59.79,

H 8.04, N 5.75.

(1S,3R,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

spiro[2.5]oct-4-ene-1-carbaldehyde (24): Compound 21 (50 mg,

0.196 mmol) in DMSO (2.5 mL) was added to ortho-iodoxybenzoic

acid (67.4 mg, 0.235 mmol, 1.2 equiv.) in DMSO (3 mL). The reac-

tion mixture was stirred for 3 h at room temp. and Et₂O (5 mL)

and H₂O (5 mL) were then added. The organic layer was separated

and the aqueous layer was saturated with NaCl and extracted with

Et₂O (4 ϫ 3 mL). The combined organic layers were dried over

MgSO₄, filtered, and concentrated in vacuo. CC (petroleum ether/

ethyl acetate, 70:30) yielded 24 (31 mg, 63%) as a colorless oil.

[α]²_D⁰ϭ Ϫ19.1 (c ϭ 0.23, CHCl₃). TLC: R_fϭ 0.58 (petroleum ether/

ethyl acetate, 70:30). ¹H NMR (CDCl₃): δ ϭ 1.04 [s, 9 H, (CH₃)₃C],

1.55 (s, 3 H, CH₃), 2.03 (dd, J ϭ 7.1/4.9 Hz, 1 H, CH₂CH), 2.13

(dd, J ϭ 9.2/4.9 Hz, 1 H, CH₂CH), 2.73Ϫ2.79 (m, 1 H, CH₂CH),

3.64 (s, 3 H, OCH₃), 9.40 (d, J ϭ 4.6 Hz, 1 H, CHO) ppm. IR:

ν˜ ϭ 2959 cm^Ϫ1, 1739, 1716, 1684, 1354, 1253. MS (CH₄; CI): m/z

(%) ϭ 254 (100) [M ϩ H^ϩ]. C₁₃H₁₉NO₄(253.3): calcd. C 61.64, H

7.56, N 5.53; found C 61.84, H 7.65, N 5.32.

(1R,3S,6S)-6-tert-Butyl-1-hydroxymethyl-5-methoxy-6-methyl-7-

oxa-4-azaspiro[2.5]oct-4-en-8-one (22). A: The preparation was per-

formed in analogy to the procedure described for 21, from (S)-

2 (669 mg, 3.36 mmol), (R)-(Ϫ)-epichlorohydrin (1.87 g, 1.58 mL,

20.16 mmol, 6.0 equiv.) in THF (30 mL), and NaN(SiMe₃)₂(c ϭ

1.0  in THF, 7.4 mL, 7.4 mmol, 2.2 equiv.). CC (petroleum ether/

ethyl acetate, 70:30) yielded 22 (373 mg, 44%) as colorless crystals,

m.p. 95 °C. The diastereoselectivity was determined by analytical

HPLC (n-heptane/ethyl acetate, 85:15; 2.0 mL/min) on the crude

product, no other diastereomer being detected: ds Ͼ 99.9:0.1; 22:

t_rϭ 29.51 min.Ϫ [α]²_D⁰ϭ Ϫ4.0 (c ϭ 0.56, CHCl₃). TLC: R_fϭ 0.20

(petroleum ether/ethyl acetate, 70:30). ¹H NMR (CDCl₃): δ ϭ 1.04

[s, 9 H, (CH₃)₃C], 1.55 (dd, J ϭ 7.5/4.5 Hz, 1 H, CH₂CH), 1.56 (s,

3 H, CH₃), 1.90 (dd, J ϭ 9.8/4.5 Hz, 1 H, CH₂CH), 2.08Ϫ2.14 (m,

1 H, CH₂CH), 3.67 (s, 3 H, OCH₃), 3.89 (dd, J ϭ 12.2/4.8 Hz, 1

H, CH₂OH), 3.98 (dd, J ϭ 12.2/2.6 Hz, 1 H, CH₂OH), OH ppm,

not located. IR: ν˜ ϭ 3432 cm^Ϫ1, 1736, 1684, 1376, 1219. MS (CH₄;

CI): m/z (%) ϭ 256 (100) [M ϩ H^ϩ]. C₁₃H₂₁NO₄(255.3): calcd. C

61.16, H 8.29, N 5.49; found C 61.13, H 8.32, N 5.49.

(1R,3R,6S)-6-tert-Butyl-5-methoxy-6-methyl-1-phenylaminomethyl-

7-oxa-4-azaspiro[2.5]oct-4-en-8-one (25): Compound 24 (25 mg,

0.10 mmol) in THF (2.5 mL) was treated with aniline (28 mg, 27

µL, 0.30 mmol) and AcOH (6 mg, 4 µL, 0.10 mmol, 1.0 equiv.) and

the mixture was stirred for 30 min. NaBH₃CN (6.3 mg, 0.10 mmol,

1.0 equiv.) was then added at room temp. After 2 h the solvent was

removed in vacuo and Et₂O (5 mL) was added. The solution was

then washed with a NaHCO₃solution (5%, 3 ϫ 3 mL) and the

aqueous layer was re-extracted with Et₂O (3 ϫ 3 mL). The com-

bined organic layers were dried over MgSO₄and filtered, and the

solvent was removed in vacuo. CC (petroleum ether/ethyl acetate,

70:30) afforded 25 (22 mg, 67%) as colorless crystals, m.p. 74 °C.

[α]²_D⁰ϭ ϩ23.9 (c ϭ 0.51, CHCl₃). TLC: R_fϭ 0.69 (petroleum ether/

ethyl acetate, 70:30). ¹H NMR (CDCl₃): δ ϭ 1.03 [s, 9 H, (CH₃)₃C],

1.14 (dd, J ϭ 7.5/4.4 Hz, 1 H, CH₂CH), 1.55 (s, 3 H, CH₃), 1.79

(dd, J ϭ 9.3/4.4 Hz, 1 H, CH₂CH), 2.31 (dddd, J ϭ 9.3/7.6/7.5/

5.1 Hz, 1 H, CH₂CH), 3.41 (dd, J ϭ 13.2/7.6 Hz, 1 H, CH₂NH),

3.52 (dd, J ϭ 13.2/5.1 Hz, 1 H, CH₂NH), 3.65 (s, 3 H, OCH₃), 6.66

(d, J ϭ 7.4 Hz, 2 H, C₆H₅), 6.72 (t, J ϭ 7.4 Hz, 1 H, C₆H₅), 7.18

(t, J ϭ 7.4 Hz, 2 H, C₆H₅), NH ppm, not located. IR: ν˜ ϭ 3379

cm^Ϫ1, 1735, 1689, 1377. MS (CH₄; CI): m/z (%) ϭ 331 (100) [M ϩ

H^ϩ]. C₁₉H₂₆N₂O₃(330.4): calcd. C 69.06, H 7.93, N 8.48; found C

69.15, H 7.89, N 8.59.

B: A Et₂Zn solution (1.0  in n-hexane, 110 µL, 0.11 mmol, 1.5

equiv.) was added at room temp. to 23a/b (as a 96:4 mixture of

Z/E isomers, 18 mg, 0.074 mmol) in Et₂O (0.5 mL), followed after

15 min by diiodomethane (40 mg, 12 µL, 0.149 mmol, 2.0 equiv.).

The mixture was stirred for 18 h at room temp. and then quenched

with phosphate buffer (pH 7, 1 mL). The aqueous layer was ex-

tracted with Et₂O (3 ϫ 3 mL) and the combined organic layers

were dried (MgSO₄), filtered, and concentrated in vacuo. By the

¹H NMR spectrum the crude product (ഠ 20 mg) was composed of

a 4:6 mixture of 21/22 (65%), together with some starting mat-

erial (35%).

(Z)-(S)-6-tert-Butyl-3-(2-hydroxyethylidene)-5-methoxy-6-methyl-

3,6-dihydro-2H-1,4-oxazin-2-one (23a) and (E)-(S)-6-tert-Butyl-3-

(2-hydroxyethylidene)-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-

oxazin-2-one (23b): Powdered KOH (30 mg, 0.544 mmol) was ad-

ded at 0 °C to (S)-2 (102 mg, 0.512 mmol) in EtOH (5 mL), fol-

lowed over 2 h by glycolaldehyde dimer (4 portions each of 61 mg,

1.02 mmol, 2 equiv.). After additional stirring for 30 min the reac-

tion mixture was hydrolyzed with phosphate buffer (pH ϭ 7,

3 mL). The aqueous layer was extracted with Et₂O (3 ϫ 10 mL).

The combined organic layers were dried over MgSO₄, filtered, and

concentrated in vacuo. CC (petroleum ether/ethyl acetate, 70:30)

yielded a 96:4 mixture of 23a/23b (98 mg, 79%) as a colorless oil.

[α]²_D⁰ϭ Ϫ86.7 (c ϭ 1.787, CHCl₃). TLC: R_fϭ 0.24 (petroleum

(1R,3S,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

spiro[2.5]oct-4-ene-1-carbaldehyde (26): The preparation was per-

formed analogously to the procedure described for 24, from ortho-

iodoxybenzoic acid (132 mg, 0.39 mmol, 1.2 equiv.) in DMSO

(6.4 mL) and 22 (100 mg, 0.39 mmol) in DMSO (4.3 mL). CC (pet-

roleum ether/ethyl acetate, 70:30) yielded 26 (87 mg, 88%) as a col-

orless oil. [α]²_D⁰ϭ ϩ99.7 (c ϭ 0.30, CHCl₃). TLC: R_fϭ 0.58 (petro-

1

leum ether/ethyl acetate, 70:30). H NMR (CDCl₃): δ ϭ 1.00 [s, 9

H, (CH₃)₃C], 1.57 (s, 3 H, CH₃), 2.13 (dd, J ϭ 7.1/4.9 Hz, 1 H,

CH₂CH), 2.31 (dd, J ϭ 9.2/4.9 Hz, 1 H, CH₂CH), 2.58Ϫ2.63 (m,

1 H, CH₂CH), 3.67 (s, 3 H, OCH₃), 9.29 (d, J ϭ 6.6 Hz, 1 H,

CHO) ppm. IR: ν˜ ϭ 2958 cm^Ϫ1, 1740, 1715, 1683, 1354, 1252. MS

(CH₄; CI): m/z (%) ϭ 254 (100) [M ϩ H^ϩ]. C₁₃H₁₉NO₄(253.3):

calcd. C 61.64, H 7.56, N 5.53; found C 61.75, H 7.61, N 5.39.

1

ether/ethyl acetate, 70:30). H NMR (CDCl₃): δ ϭ 0.99 [s, 0.04 ϫ

9 H, (CH₃)₃C], 1.02 [s, 0.96 ϫ 9 H, (CH₃)₃C], 1.53 (s, 0.04 ϫ 3 H,

CH₃), 1.56 (s, 0.96 ϫ 3 H, CH₃), 3.82 (s, 0.04 ϫ 3 H, OCH₃), 3.84

(s, 0.96 ϫ 3 H, OCH₃), 4.61Ϫ4.66 (m, 3 H, CHCH₂OH), 6.33 (t,

(1S,3S,6S)-6-tert-Butyl-5-methoxy-6-methyl-1-phenylaminomethyl-

J ϭ 6.0 Hz, 0.04 ϫ 1 H, CHCH₂OH), 6.64 (t, J ϭ 4.0 Hz, 0.96 ϫ 7-oxa-4-azaspiro[2.5]oct-4-en-8-one (27): The preparation was per-

1 H, CHCH₂OH) ppm. IR: ν˜ ϭ 3422 cm^Ϫ1, 2961, 1711, 1656, 1332, formed analogously to the procedure described for 25, from 26

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1259

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

(82 mg, 0.32 mmol) in THF (8 mL), aniline (90 mg, 89 µL, CH₂NCO), 4.52 (dd, J ϭ 11.5/5.5 Hz, 1 H, CH₂NCO), 6.91 (d, J ϭ

0.97 mmol), AcOH (19 mg, 19 µL, 0.32 mmol), and NaBH₃CN 7.5 Hz, 2 H, C₆H₅), 7.02 (t, J ϭ 7.5 Hz, 1 H, C₆H₅), 7.29 (t, J ϭ

(20 mg, 0.32 mmol, 1 equiv.). CC (petroleum ether/ethyl acetate,

70:30) afforded 27 (77 mg, 73%) as colorless crystals, m.p. 76 °C.

[α]²_D⁰ϭ Ϫ11.1 (c ϭ 0.44, CHCl₃). TLC: R_fϭ 0.68 (petroleum ether/

7.5 Hz, 2 H, C₆H₅) ppm. IR: ν˜ ϭ 3357 cm^Ϫ1, 2962, 1734, 1680,

1594, 1372, 1260. MS (CH₄; CI): m/z (%) ϭ 374 (100) [M ϩ H^ϩ].

C₂₀H₂₇N₃O₄(373.5): calcd. C 64.32, H 7.29, N 11.25; found C

ethyl acetate, 70:30). ¹H NMR (CDCl₃): δ ϭ 1.04 [s, 9 H, (CH₃)₃C], 64.40, H 7.40, N 11.11.

1.26 (dd, J ϭ 7.4/4.4 Hz, 1 H, CH₂CH), 1.57 (s, 3 H, CH₃), 1.92

1-[(1S,3S,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

(dd, J ϭ 9.5/4.4 Hz, 1 H, CH₂CH), 2.17Ϫ2.24 (m, 1 H, CH₂CH),

3.39 (d, J ϭ 6.4 Hz, 2 H, CH₂NH), 3.67 (s, 3 H, OCH₃), 6.59 (d,

J ϭ 7.4 Hz, 2 H, C₆H₅), 6.71 (t, J ϭ 7.4 Hz, 1 H, C₆H₅), 7.17 (t,

J ϭ 7.4 Hz, 2 H, C₆H₅), NH signal ppm, not located. IR: ν˜ ϭ 3365

cm^Ϫ1, 1730, 1687, 1370. MS (CH₄; CI): m/z (%) ϭ 331 (100) [M ϩ

H^ϩ]. C₁₉H₂₆N₂O₃(330.4): calcd. C 69.06, H 7.93, N 8.48; found C

68.95, H 7.79, N 8.68.

spiro[2.5]oct-4-en-1-ylmethyl]-3-phenyl-1-(trifluoroacetyl)urea (31):

This preparation was as described for 29, from PPh₃(154 mg,

0.59 mmol, 1.5 equiv.), DIAD (114 µL, 0.59 mmol, 1.5 equiv.) in

THF (3.5 mL), 22 (100 mg, 0.39 mmol), and 1-phenyl-3-(trifluo-

roacetyl)urea (28, 136 mg, 0.59 mmol, 1.5 equiv.) in THF (5.5 mL).

CC (n-heptane/ethyl acetate, 70:30) afforded 31 (116 mg, 63%) as a

colorless oil. [α]²_D⁰ϭ ϩ30.1 (c ϭ 0.59, CHCl₃). TLC: R_fϭ 0.33

(petroleum ether/iPr₂O, 55:45). ¹H NMR (CDCl₃): δ ϭ 0.86 [s, 9

H, (CH₃)₃C], 1.27 (dd, J ϭ 7.5/4.6 Hz, 1 H, CH₂CH), 1.53 (s, 3 H,

CH₃), 1.95 (dd, J ϭ 9.6/4.6 Hz, 1 H, CH₂CH), 2.35Ϫ2.40 (m, 1 H,

CH₂CH), 3.65 (s, 3 H, OCH₃), 4.69 (dd, J ϭ 11.4/8.4 Hz, 1 H,

CH₂NCO), 4.83 (dd, J ϭ 11.4/5.6 Hz, 1 H, CH₂NCO), 7.24Ϫ7.29

(m, 3 H, C₆H₅), 7.35Ϫ7.39 (m, 2 H, C₆H₅), 11.28 (s, 1 H, NH)

1-Phenyl-3-(trifluoroacetyl)urea (28): N-Phenylurea (500 mg,

3.67 mmol) and trifluoroacetic anhydride (7.55 g, 5.00 mL,

36.0 mmol) were stirred at room temp. for 8 h. The reaction mix-

ture was then concentrated in vacuo. CC (petroleum ether/ethyl

acetate, 70:30) yielded 28 (734 mg, 86%) as colorless crystals, m.p.

211 °C (dec.) {ref.^[33]215Ϫ216 °C (dec.)}. TLC: R_fϭ 0.49 (petro-

leum ether/ethyl acetate, 70:30). ¹H NMR (CDCl₃): δ ϭ 7.10Ϫ7.47 ppm. IR: ν˜ ϭ 3296 cm^Ϫ1, 2957, 1738, 1689, 1652, 1614, 1582, 1451.

(m, 5 H, C₆H₅), 9.38 (s, 1 H, NH), 9.77 (s, 1 H, NH) ppm. IR: ν˜ ϭ

MS (CH₄; CI): m/z (%) ϭ 470 (3) [M ϩ H^ϩ], 238 (100).

3306 cm^Ϫ1, 3152, 3000, 1719, 1616, 1568, 1233. MS (CH₄; CI): C₂₂H₂₆F₃N₃O₅(469.5): calcd. C 56.29, H 5.58, N 8.95; found C

m/z (%) ϭ 233 (100) [M ϩ H^ϩ]. C₉H₇N₂O₂F₃(232.2): calcd. C

56.22, H 5.65, N 8.95.

46.56, H 3.04, N 12.07; found C 46.61, H 3.03, N 12.03.

1-[(1S,3S,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

spiro[2.5]oct-4-en-1-ylmethyl]-3-phenylurea (32): This preparation

was as described for 30, from 31 (57 mg, 0.122 mmol) and a K₂CO₃

solution (20%, 9.5 mL). Yield: 34 mg (76%) of 32 as a colorless oil.

[α]²_D⁰ϭ Ϫ3.6 (c ϭ 0.45, CHCl₃). TLC: R_fϭ 0.38 (petroleum ether/

ethyl acetate, 20:80). ¹H NMR (CDCl₃): δ ϭ 1.05 [s, 9 H, (CH₃)₃C],

1-[(1R,3R,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

spiro[2.5]oct-4-en-1-ylmethyl]-3-phenyl-1-(trifluoroacetyl)urea (29):

diisopropyl azodicarboxylate (DIAD) (71 mg, 69 µL, 0.35 mmol,

1.5 equiv.) was added at 0 °C to a solution of PPh₃(93 mg,

0.35 mmol, 1.5 equiv.) in THF (2 mL), followed at 0 °C by a solu-

tion of 21 (60 mg, 0.235 mmol) and 1-phenyl-3-(trifluoroacetyl)- 1.24 (dd, J ϭ 7.7/4.3 Hz, 1 H, CH₂CH), 1.57 (s, 3 H, CH₃), 1.95

urea (28, 82 mg, 0.35 mmol, 1.5 equiv.) in THF (3.3 mL). The mix- (dd, J ϭ 9.6/4.3 Hz, 1 H, CH₂CH), 2.30Ϫ2.39 (m, 1 H, CH₂CH),

ture was allowed to warm to room temp. After 24 h the solvent was 3.70 (s, 3 H, OCH₃), 3.99 (s, 2 H, NH), 4.37 (dd, J ϭ 11.4/9.0 Hz,

removed in vacuo, and the residue was redissolved in CH₂Cl₂and

washed with a K₂CO₃solution (5%). The organic layer was separ-

ated, the solvent was removed, and the residue was purified by

1 H, CH₂NCO), 4.50 (dd, J ϭ 11.4/5.3 Hz, 1 H, CH₂NCO), 6.89

(d, J ϭ 7.5 Hz, 2 H, C₆H₅), 7.02 (t, J ϭ 7.5 Hz, 1 H, C₆H₅), 7.29

(t, J ϭ 7.5 Hz, 2 H, C₆H₅) ppm. IR: ν˜ ϭ 3359 cm^Ϫ1, 2963, 1730,

CC (n-heptane/ethyl acetate, 70:30) to afford 29 (72 mg, 65%) as a 1679, 1590, 1373, 1262. MS (CH₄; CI): m/z (%) ϭ 374 (100) [M ϩ

colorless oil. [α]²_D⁰ϭ Ϫ29.7 (c ϭ 0.20, CHCl₃). TLC: R_fϭ 0.25

(petroleum ether/iPr₂O, 55:45). ¹H NMR (CDCl₃): δ ϭ 1.02 [s, 9

H, (CH₃)₃C], 1.24 (dd, J ϭ 7.2/4.8 Hz, 1 H, CH₂CH), 1.46 (s, 3 H,

CH₃), 1.87 (dd, J ϭ 9.5/4.8 Hz, 1 H, CH₂CH), 2.44Ϫ2.47 (m, 1 H,

CH₂CH), 3.62 (s, 3 H, OCH₃), 4.63 (dd, J ϭ 11.5/9.1 Hz, 1 H,

CH₂NCO), 4.98 (dd, J ϭ 11.5/5.9 Hz, 1 H, CH₂NCO), 7.25Ϫ7.29

(m, 3 H, C₆H₅), 7.33Ϫ7.38 (m, 2 H, C₆H₅), 11.26 (s, 1 H, NH)

ppm. IR: ν˜ ϭ 3296 cm^Ϫ1, 2955, 1735, 1689, 1650, 1612, 1582, 1470.

MS (CH₄; CI): m/z (%) ϭ 470 (8) [M ϩ H^ϩ], 238 (100).

C₂₂H₂₆F₃N₃O₅(469.5): calcd. C 56.29, H 5.58, N 8.95; found C

56.45, H 5.73, N 8.65.

H^ϩ]. C₂₀H₂₇N₃O₄(373.5): calcd. C 64.32, H 7.29, N 11.25; found

C 64.18, H 7.21, N 11.05.

Preparation of the Free Amino Acids

General Procedure (GP C) for the Hydrolysis of the Monoalkylated

Derivatives 13aϪc and for the Hydrolysis of the Cyclopropyl Derivat-

ives of 2 (21, 22, 25, and 27): A solution of the respective monoalky-

lated derivative of 2 in acetonitrile (0.1 ) was added to trifluo-

roacetic acid (0.2  solution in H₂O, 5.0 equiv.) and the mixture

was stirred for 20 h at 60 °C, after which the solvent was removed

in vacuo.

1-[(1R,3R,6S)-6-tert-Butyl-5-methoxy-6-methyl-8-oxo-7-oxa-4-aza-

spiro[2.5]oct-4-en-1-ylmethyl]-3-phenylurea (30): Compound 29

(60 mg, 0.128 mmol) was stirred at room temp. in a 20% K₂CO₃

solution (2.0 g K₂CO₃in a mixture of 4 mL H₂O and 4 mL of

MeOH, 10 mL). After 4 h, MeOH was removed in vacuo and the

residual aqueous layer was extracted with Et₂O (3 ϫ 3 mL). The

combined organic layers were dried over MgSO₄and filtered, and

the solvent was removed in vacuo. CC (n-heptane/ethyl acetate,

20:80) yielded 30 (35 mg, 74%) as a colorless oil. [α]²_D⁰ϭ ϩ43.1

For 13aϪc the residue was dissolved in aqueous NaOH (40%, re-

sulting conc. ഠ 0.1 ) and allowed to stir for 20 h at room temp.

For 21 and 22 the residue was added to NaOH (0.2  solution in

MeOH, 6 equiv.) and heated at 60 °C for 20 h.

For 25 and 27 the residue was dissolved in DME (25) or in 2-

ethoxyethanol (27) (resulting conc. 0.1 ), added to NaOH (10 

aqueous solution, 40 equiv.) and heated at 90 °C for 5 d.

(c ϭ 0.50, CHCl₃). TLC: R_fϭ 0.46 (petroleum ether/ethyl acetate, After the second of the hydrolyses had been finished, the solvent

20:80). ¹H NMR (CDCl₃): δ ϭ 1.04 [s, 9 H, (CH₃)₃C], 1.15 (dd, was removed and the residue was redissolved in a small amount of

J ϭ 7.5/4.5 Hz, 1 H, CH₂CH), 1.56 (s, 3 H, CH₃), 1.84 (dd, J ϭ water. The solution was washed with Et₂O (3 ϫ) and CH₂Cl₂(3

9.7/4.5 Hz, 1 H, CH₂CH), 2.38Ϫ2.45 (m, 1 H, CH₂CH), 3.69 (s, 3 ϫ) and then acidified at 0 °C with HCl (ഠ pH 3). The acidic solu-

H, OCH₃), 4.07 (s, 2 H, NH), 4.45 (dd, J ϭ 11.5/8.6 Hz, 1 H,

tion was washed with Et₂O (3 ϫ), CH₂Cl₂(3 ϫ), and ethyl acetate

1260

Eur. J. Org. Chem. 2003, 1244Ϫ1263

A New Chiral Glycine Equivalent

FULL PAPER

(3 ϫ). The residue obtained by concentration of the acidic aqueous

orless crystals, m.p. Ͼ300 °C (dec.) {ref.^[33]298Ϫ299 °C (dec.)}.

solution was finally (after it had been dissolved in a small amount Yield: 3.6 mg (91%). Analytical data (¹H NMR, IR, MS) were in

of water) subjected to ion-exchange chromatography (Dowex 50 W

X 8 cation exchange resin, elution with H₂O until the eluent was

neutral and free of chloride, then elution with 10% aqueous NH₃)

to afford the free amino acid.

agreement with literature data.^[34][α]²_D⁰ϭ Ϫ16.1 (c ϭ 0.16, H₂O)

{ref.^[31][α]²_D⁰ϭ Ϫ21.5 (c ϭ 1.0, H₂O)}.

(S)-2-Amino-2-benzylpent-4-enoic Acid (34b): This compound was

obtained by GP D, from 17b (17 mg, 0.052 mmol). Colorless crys-

tals, m.p. 218 °C (dec.) {ref.^[4]220 °C (dec.)}. Yield: 9.6 mg (90%).

Analytical data (¹H NMR, IR, MS) were in agreement with literat-

General Procedure (GP D) for the Hydrolysis of the Dialkylated

Derivatives of 2 (17aϪd): A solution of dialkylated 2 in a mixture

consisting of 40% aqueous NaOH and MeOH (4:1, 0.1 ) was

either stirred at 60 °C for 1 to 3 days (17aϪc) or heated at 120 °C

for 2 days (17d). The alkaline solution was washed with Et₂O (3

ϫ) and then acidified at 0 °C with HCl (6 ). The acidic aqueous

phase was concentrated in vacuo, and the residue was dissolved in

a small amount of water and subjected to ion-exchange chromato-

graphy with a Dowex 50 W X 8 resin (see GP C) to give the free

amino acid.

ure values.^[4,35][α]²_D⁰ϭ ϩ22.8 (c ϭ 0.365, H₂O) {ref.^[35][α]²_D⁰

ϩ27.3 (c ϭ 1.0, H₂O)}.

ϭ

(S)-2-Amino-2-benzylhexanoic Acid (34c):^[36]This compound was

obtained by GP D, from 17c (34 mg, 0.099 mmol). Colorless crys-

tals, m.p. 240 °C (dec.). Yield: 17 mg (78%). [α]²_D⁰ϭ Ϫ19.0 (c ϭ

1

0.31, H₂O). H NMR (D₂O): δ ϭ 0.76 (t, J ϭ 7.2 Hz, 3 H, CH₃),

1.05Ϫ1.16 (m, 1 H, CH₂), 1.17Ϫ1.29 (m, 3 H, CH₂), 1.58Ϫ1.66

(m, 1 H, CH₂), 1.82Ϫ1.91 (m, 1 H, CH₂), 2.87 (d, J ϭ 14.4 Hz, 1

H, CH₂C₆H₅), 3.17 (d, J ϭ 14.4 Hz, 1 H, CH₂C₆H₅), 7.14 (d, J ϭ

6.9 Hz, 2 H, C₆H₅), 7.19Ϫ7.30 (m, 3 H, C₆H₅) ppm. IR: ν˜ ϭ 3032

cm^Ϫ1, 2958, 1596, 1395, 702. MS (CH₄; CI): m/z (%) ϭ 222 (77)

[M ϩ H^ϩ], 176 (90), 130 (100). C₁₃H₁₉NO₂·H₂O (239.3): calcd. C

65.25, H 8.84, N 5.85; found C 65.32, H 8.27, N 5.93.

Procedure for the Hydrolysis of 13c under Acidic Conditions: Com-

pound 13c was treated with a HBr solution (3 , 25 equiv.) and

the mixture was stirred at 130 °C in a pressure tube for 24 h. After

removal of solvent in vacuo, the residue was washed with Et₂O (3

ϫ), redissolved in a small amount of water and eluted on a Dowex

50 W X 8 resin (see GP C) to yield 33c (85%). The enantiomeric

excess was determined by analytical HPLC on a Nucleosil^

CHIRALϪ1 column,^[11](H₂O, 0.5 mmol CuSO₄·5 H₂O/L; 1.0 mL/

min), kept at 55Ϫ60 °C. es ϭ 98.1:1.9; 33c: t_Rϭ 5.51 min; (ent)-

33c: t_Rϭ 4.73 min.

(R)-2-Amino-2-benzyl-3-methylbutanoic Acid (34d): This compound

was obtained by GP D, from 17d (30 mg, 0.090 mmol). Colorless

crystals, m.p. 189 °C (dec.). Yield: 9.6 mg (51%). Analytical data

(¹H NMR, IR, MS) were in agreement with literature values.^[34]

[α]²_D⁰ϭ ϩ3.3 (c ϭ 0.48, H₂O). [α]²_D⁰ϭ ϩ2.8 (c ϭ 0.45, MeOH)

{ref.^[34][α]²_D⁰ϭ ϩ3.3 (c ϭ 1.1, MeOH)}.

(R)-Phenylalanine (33a): This compound was obtained by GP C,

from 13a (30 mg, 0.104 mmol). Colorless crystals, m.p. 279 °C

(dec.) {ref.^[28]283Ϫ284 °C (dec.)}. Yield: 4.2 mg (25%). The enanti-

omeric excess was determined by analytical HPLC on a Nucleosil^

CHIRALϪ1 column,^[11](H₂O, 0.5 mmol CuSO₄·5 H₂O/L; 1.0 mL/

min), kept at 55Ϫ60 °C. ee ϭ 87%; 33a: t_Rϭ 16.9 min; (ent)-33a:

t_Rϭ 11.4 min. Analytical data (¹H NMR, IR, MS) were in agree-

ment with literature data. [α]²_D⁰ϭ ϩ32.0 (c ϭ 0.19, H₂O) {ref.^[29]

[α]²_D⁵ϭ ϩ34.8 (c ϭ 2.0, H₂O)}.

(1R,2S)-1-Amino-2-(hydroxymethyl)cyclopropanecarboxylic

Acid

(35): This compound was obtained by GP C, from 21 (60 mg,

0.23 mmol). Colorless crystals, m.p. 198 °C (dec.). Yield: 26 mg

(86%). Analytical data (¹H NMR, IR, MS) were in agreement with

literature values.^[16][α]²_D⁰ϭ ϩ64.4 (c ϭ 0.90, H₂O) {ref.^[16][α]²_D⁵

ϩ73.81 (c ϭ 0.48, H₂O)}.

ϭ

(1S,2R)-1-Amino-2-(hydroxymethyl)cyclopropanecarboxylic

Acid

[(ent)-35]: This compound was obtained by GP C, from 22 (63 mg,

0.25 mmol). Colorless crystals, m.p. 213 °C (dec.). Yield: 32 mg

(97%). Analytical data (¹H NMR, IR, MS) were in agreement with

literature values for the (1R,2S) enantiomer.^[16][α]²_D⁰ϭ Ϫ74.0 (c ϭ

0.50, H₂O) {ref.^[16][α]²_D⁵ϭ Ϫ74.5 (c ϭ 0.184, H₂O) for the

(1R,2S)-enantiomer}.

(R)-2-Aminopent-4-enoic Acid (33b): This compound was obtained

by GP C, from 13b (18.9 mg, 0.079 mmol). Colorless crystals, m.p.

244 °C (dec.) {ref.^[30]240Ϫ241 °C (dec.)}. Yield: 3.5 mg (38%). The

enantiomeric excess was determined by analytical HPLC on a Nu-

cleosil^CHIRALϪ1 column,^[11](H₂O, 0.5 mmol CuSO₄·5 H₂O/L;

1.0 mL/min) that was kept at 55Ϫ60 °C. ee ϭ 90%; 33b: t_R

ϭ

16.8 min; (ent)-33b: t_Rϭ 11.3 min. Analytical data (¹H NMR, IR,

MS) were in agreement with literature values. [α]²_D⁰ϭ ϩ39.7 (c ϭ

0.175, H₂O) {ref.^[31][α]²_D⁰ϭ ϩ37.9 (c ϭ 1.3, H₂O)}.

(1R,2R)-1-Amino-2-(phenylaminomethyl)cyclopropanecarboxylic

Acid (36):^[37]This compound was obtained by GP C, from 25

(20 mg, 0.060 mmol) in DME. Colorless crystals, m.p. 178Ϫ180 °C

(dec.). Yield: 13 mg (52%). [α]²_D⁰ϭ ϩ7.1 (c ϭ 0.50, H₂O). ¹H NMR

(D₂O): δ ϭ 1.21 (dd, J ϭ 7.7/6.5 Hz, 1 H, CH₂), 1.61 (dd, J ϭ 9.7/

6.5 Hz, 1 H, CH₂), 1.99Ϫ2.06 (m, 1 H, CH), 3.36 (dd, J ϭ 13.4/

6.7 Hz, 1 H, NCH₂), 3.39 (dd, J ϭ 13.4/6.7 Hz, 1 H, NCH₂),

6.90Ϫ6.98 (m, 3 H, ArH), 7.32Ϫ7.32 (m, 2 H, ArH) ppm. IR: ν˜ ϭ

3400 cm^Ϫ1, 1602, 1508, 1378, 1253, 750, 691. HRMS (DEIϩ, 70

eV): calcd. for C₁₁H₁₄N₂O₂206.10553; found m/z ϭ 206.1056.

(R)-Alanine (33c): This compound was obtained by GP C, from 13c

(24 mg, 0.11 mmol). Colorless crystals, m.p. 282 °C (dec.) {ref.^[32]

289Ϫ291 °C (dec.)}. Yield: 7 mg (69%). Analytical data (¹H NMR,

IR, MS) were in agreement with literature values. [α]²_D⁰ϭ Ϫ1.9 (c ϭ

0.10, H₂O) {ref.^[29][α]²_D⁵ϭ Ϫ2.6 (c ϭ 6.0, H₂O)}.

(S)-[3؊¹³C]Alanine [(ent)-33d]: This compound was obtained by

the procedure for the hydrolysis of 13c under acidic conditions (see

above), from (ent)-13d (100 mg, 0.467 mmol); es ϭ 98.6:1.4; (ent)-

33d: t_Rϭ 4.51 min; 33d: t_Rϭ 5.28 min. Colorless crystals, m.p. 290

°C (dec.). Yield: 35 mg (83%). [α]²_D⁰ϭ ϩ1.5 (c ϭ 0.80, H₂O). ¹H

NMR (D₂O): δ ϭ 1.49 (dd, J ϭ 129.9/7.2 Hz, 3 H, CH₃), 3.78 (dq,

J ϭ 4.6/7.2 Hz, 1 H, CH) ppm. MS (CH₄; CI): m/z (%) ϭ 91 (100)

[M ϩ H^ϩ], 86 (12). ¹³CC₂H₇NO₂(90.1): calcd. C 40.00, H 7.83, N

15.55; found C 39.01, H 7.72, N 14.88.

(1S,2S)-1-Amino-2-(phenylaminomethyl)cyclopropanecarboxylic

Acid [(ent)-36]:^[37]This compound was obtained by GP C, from 27

(34 mg, 0.10 mmol) in 2-ethoxyethanol. Colorless crystals. Yield:

10 mg (50%). [α]²_D⁰ϭ Ϫ6.8 (c ϭ 0.40, H₂O). The spectroscopic data

(¹H NMR, IR, MS) were identical to those described above for 36.

HRMS (DEIϩ, 70 eV): calcd. for C₁₁H₁₄N₂O₂206.10553; found

m/z ϭ 206.1055.

(S)-2-Amino-2-methyl-3-phenylpropanoic Acid (34a): This com-

pound was obtained by GP D, from 17a (6.7 mg, 0.022 mmol). Col-

(1R,2R)-1-Amino-2-(aminomethyl)cyclopropanecarboxylic

[(ent)-37]: Compound 31 (58 mg, 0.124 mmol) in acetonitrile

Acid

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1261

ˇ

´

C. J. Koch, S. Simonyiova, J. Pabel, A. Kärtner, K. Polborn, K. T. Wanner

FULL PAPER

[8g]

R. Wang, J. Am. Chem. Soc. 1999, 121, 6213Ϫ6219.

D. Z.-

(0.62 mL) was added to trifluoroacetic acid (6 equiv. 0.1  solution

in H₂O) and the mixture was stirred for 20 h at 60 °C. The reaction

mixture was concentrated in vacuo and the residue was dissolved

in a mixture of aqueous NaOH (10 , 0.5 mL, 40 equiv.) and DME

(1.0 mL) and heated to 90 °C. After 7 d the alkaline solution was

washed with CH₂Cl₂(3 ϫ), acidified with HCl (2 , pH 3), and

again washed with CH₂Cl₂(3 ϫ). The acidic aqueous solution was

concentrated in vacuo and the residue was dissolved in a small

amount of water and subjected to ion-exchange chromatography

with a Dowex 50 W X 8 resin to give the free amino acid as color-

less crystals, still contaminated with about 10% of an inseparable

R. Wang, A. Streitwieser, Can. J. Chem. 1999, 77, 654Ϫ658.

[9] [9a]

I. A. Faworskaja, J. Gen. Chem. (U. S. S. R.) 1948, 18,

[9b]

52Ϫ59; Chem. Abstr. 1948, 42, 4905d.

A. Richard, Ann.

Chim. (Paris) 1910, 21, 360Ϫ406. ^[9c]M. B. Green, W. J. Hick-

inbottom, J. Org. Chem. 1957, 22, 3262Ϫ3270. For an already

[9d]

existing asymmetric synthesis of 1 see:

E. L. Eliel, J. E.

Lynch, W. R. Kenan Jr., Tetrahedron Lett. 1987, 28,

[9e]

4813Ϫ4816.

R. J. D. Evans, S. R. Landor, J. Chem. Soc.

1965, 2553Ϫ2559. ^[9f]S. Ahmad, S. H. Spergel, J. C. Barrish, J.

DiMarco, J. Gougoutas, Tetrahedron: Asymmetry 1995, 6,

2893Ϫ2894.

[10]

For the synthesis of phenylalaninol from phenylalanine by re-

duction with NaBH₄/H₂SO₄see: A. Abiko, S. Masamune, Tet-

rahedron Lett. 1992, 33, 5517Ϫ5518.

H. Brückner, Chromatographia 1987, 24, 725Ϫ738.

E. Bald, K. Saigo, T. Mukaiyama, Chem. Lett. 1975,

1163Ϫ1167.

side product; m.p. Ͻ150 °C (dec.). Yield: 3.9 mg (24%). [α]²_D⁰

ϭ

Ϫ6.9 (c ϭ 0.19, H₂O). ¹H NMR (D₂O): δ ϭ 1.19 (dd, J ϭ 7.4/

6.4 Hz, 1 H, CH₂CH), 1.13 (dd, J ϭ 10.0/6.4 Hz, 1 H, CH₂CH),

1.83Ϫ1.92 (m, 1 H, CH₂CH), 3.48 (dd, J ϭ 12.3/6.6 Hz, 1 H,

CH₂NH₂), 3.70 (dd, J ϭ 12.3/4.8 Hz, 1 H, CH₂NH₂) ppm.

[11]

[12]

[13]

U. Schöllkopf, U. Groth, C. Deng, Angew. Chem. 1981, 93,

793Ϫ795; Angew. Chem. Int. Ed. Engl. 1981, 20, 798Ϫ799.

[14] [14a]

J. E. Baldwin, V. Lee, C. J. Schofield, Heterocycles 1992,

Acknowledgments

[14b]

34, 903Ϫ906.

D. Seebach, J. D. Aebi, R. Naef, T. Weber,

Financial support of this work by the Deutsche Forschungsgemein-

schaft, the Fonds der Chemischen Industrie and the BMBF is

gratefully acknowledged.

Helv. Chim. Acta 1985, 68, 144Ϫ154.

[15]

Crystallographic data for structure 21 have been deposited with

the Cambridge Crystallographic Data Centre as supplementary

publication no. CCDC 192162. Copies of the data can be ob-

tained free of charge, on application to CCDC, 12 Union

Road, Cambridge CB2 1EZ, UK. Fax: (internat.) ϩ44 1223

336033 or E-mail: deposit@ccdc.cam.ac.uk. The X-ray diffrac-

tion analysis of 22 was performed by M. Malecka and E. Bud-

zisz at the University of Lodz and will be published by these

authors.

[1]

^[1a]T. Wirth, Angew. Chem. 1997, 109, 235Ϫ237; Angew. Chem.

[1b]

Int. Ed. Engl. 1997, 36, 225Ϫ227.

B. M. Trost, X. Ariza,

Angew. Chem. 1997, 109, 2749Ϫ2751; Angew. Chem. Int. Ed.

Engl. 1997, 36, 2635Ϫ2637. ^[1c]S. Hanessian, G. McNaughton-

Smith, H. G. Lombart, W. D. Lubell, Tetrahedron 1997, 53,

12789Ϫ12854.

[16]

[17]

M. C. Pirrung, S. E. Dunlap, U. P. Trinks, Helv. Chim. Acta

1989, 72, 1301Ϫ1310.

The stereochemistry found for related cyclopropane derivatives

derived from 4^[4]is in disagreement with the results described

in the present paper and will have to be reevaluated, as the

assignment then was based only on the optical rotation of the

final amino acids.

[2] [2a]

M. Goodman, H. Shao, Pure Appl. Chem. 1996, 68,

[2b]

1303Ϫ1308 and references therein.

W. F. Degrado, Adv.

Protein Chem. 1988, 39, 51Ϫ124 and references therein.

[3] [3a]

C. Cativiela, M. D. Diaz-de-Villegas, Tetrahedron: Asym-

[3b]

metry 1998, 9, 3517Ϫ3599.

C. Cativiela, M. D. Diaz-de-

Villegas, Tetrahedron: Asymmetry 2000, 11, 645Ϫ732.

O. Achatz, A. Grandl, K. Th. Wanner, Eur. J. Org. Chem.

1999, 1967Ϫ1978.

Camphanic acid, which is commercially available in both en-

antiomeric forms, serves as a starting material for the synthesis

of the chiral auxiliary. For large-scale synthesis it is more eco-

nomic to prepare it from the corresponding enantiomer of

camphoric acid. However, the (Ϫ)-enantiomer of camphoric

acid is far more expensive than the (ϩ)-enantiomer.

[4]

[5]

[18] [18a]

J. Furukawa, N. Kawabata, J. Nishimura, Tetrahedron

[18b]

1968, 24, 53Ϫ59.

J. Furukawa, N. Kawabata, Adv. Or-

ganomet. Chem. 1974, 12, 83Ϫ134.

[19] [19a]

O. Mitsunobu, M. Yamada, Bull. Chem. Soc. Jpn. 1967,

[19b]

[19c]

40, 2380Ϫ2382.

O. Mitsunobu, M. Wada, T. Sano, J. Am. Chem. Soc. 1972, 94,

679Ϫ680.

O. Mitsunobu, Synthesis 1981, 1Ϫ28.

[19d]

O. Mitsunobu, M. Yamada, T. Mukaiyama,

[19e]

Bull. Chem. Soc. Jpn. 1967, 40, 935Ϫ939.

T. Tsunoda, Y.

[6] [6a]

R. M. Williams, Synthesis of Optically Active α-Amino Ac-

Yamamiya, S. Ito, Tetrahedron Lett. 1993, 34, 1639Ϫ1642.

B. A. Arbuzov, N. N. Zobova, N. R. Rubinova, Ser. Khim.

1976, 1, 202Ϫ204.

ids, Pergamon Press, Oxford, 1989. ^[6b]R. Fitzi, D. Seebach, Tet-

rahedron 1988, 44, 5277Ϫ5292.

[20]

[21]

[7] [7a]

G. Porzi, S. Sandri, P. Verrocchio, Tetrahedron: Asymmetry

For

a

synthesis of the free racemic 1-amino-2-{(3-

[7b]

1998, 9, 119Ϫ132.

metry 1998, 9, 3411Ϫ3420.

G. Porzi, S. Sandri, Tetrahedron: Asym-

phenylureido)methyl}cyclopropanecarboxylic acid see: C. Bal-

samini, A. Bedini, G. Spadoni, G. Tarzia, A. Tontini, Farmaco

1998, 53, 181Ϫ188.

TFA proved to be better than HCl, as the lactam also was

formed to some extent with the latter. This phenomenon is well

known and has been explained in terms of the higher nucleo-

philicity of the chloride ion than of the trifluoroacetate ion,

see: T. Beulshausen, U. Groth, U. Schöllkopf, Liebigs Ann.

Chem. 1991, 1207Ϫ1209.

For examples of hydrolysis of similar substrates under similar

conditions see: G. Kardassis, P. Brungs, C. Nothhelfer, E.

Steckhan, Tetrahedron 1998, 54, 3479Ϫ3488.

For example, in the first step the imidate group was cleaved

with TFA. When the subsequent hydrolysis was then attempted

under basic conditions the urea side chain was destroyed,

whereas under acidic conditions the cyclopropyl ring was

opened. Cleavage of the cyclopropyl ring was also observed

[7c]

R. M. Williams, M.-N. Im, J.

[7d]

Am. Chem. Soc. 1991, 113, 9276Ϫ9286.

K. Hammer, K.

[7e]

[22]

Undheim, Tetrahedron 1997, 53, 5925Ϫ5936.

K. Hammer,

K. Undheim, Tetrahedron 1997, 53, 2309Ϫ2322. ^[7f]T. Abellan,

´

C. Najera, J. M. Sansano, Tetrahedron: Asymmetry 1998, 9,

2211Ϫ2214.

[7g]

For a review on asymmetric synthesis of α-

quaternary aminocyclopropanecarboxyclic acids see: K. Bur-

gess, K.-K. Ho, D. Moye-Sherman, Synlett 1994, 575Ϫ583. ^[7h]

A related chiral glycine equivalent based on atrolactic acid has

recently been reported: S. N. Crane, E. J. Corey, Organic Letters

2001, 3, 1395Ϫ1397.

[23]

[24]

[8]

For further investigations on lithio-aggregates formed in ether-

[8a]

eal solvents see:

D. Z. Wang, Y.-J. Kim, A. Streitwieser, J.

Am. Chem. Soc. 2000, 122, 10754Ϫ10760. ^[8b]J. H. Horner, M.

[8c]

Vera, J. B. Grutzner, J. Org. Chem. 1986, 51, 4212Ϫ4220.

L. M. Jackman, B. C. Lange, Tetrahedron 1977, 33,

2737Ϫ2769. ^[8d]E. M. Arnett, C. A. Palmer, J. Am. Chem. Soc.

with Me₃SiI, which is known to cause this type of reaction;

[8e]

[24a]

1990, 112, 7354Ϫ7360.

A. Streitwieser, S. S.-W. Leung, Y.-

see:

R. D. Miller, D. R. McKean, Tetrahedron Lett. 1979,

[8f]

[24b]

J. Kim, Org. Lett. 1999, 1, 145Ϫ147.

A. Streitwieser, D. Z.-

20, 2305Ϫ2308.

E. W. Logusch, Tetrahedron Lett. 1984,

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1262

A New Chiral Glycine Equivalent

FULL PAPER

A. Studer, D. Seebach, Liebigs Ann. 1995, 217Ϫ222.

T. M. Zydowsky, E. de Lara, S. G. Spanton, J. Org. Chem.

1990, 55, 5437Ϫ5439.

The configuration was the result of a comparison of the data

obtained for the N-acetylation product of 34c with those re-

ported in the literature for this compound. See: U. Schöllkopf,

H.-H. Hausberg, I. Hoppe, M. Segal, U. Reiter, Angew. Chem.

1978, 90, 136Ϫ138; Angew. Chem. Int. Ed. Engl. 1978, 17,

117Ϫ119.

Compound 36 is identical with a derivative already presented

in ref.^[4](compound number 24b). However, in that paper the

stereochemical descriptors for position 2 of 24b and for its iso-

mer 24a as well as for their precursors 23a and 23b (position

1) were assigned incorrectly.

[24c]

[34]

[35]

25, 4195Ϫ4198.

M. Ihara, T. Taniguchi, Y. Tokunaga, K.

Fukumoto, J. Org. Chem. 1994, 59, 8092Ϫ8100.

W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43,

2923Ϫ2925.

C. Glücksmann, Monatsh. Chem. 1889, 10, 770Ϫ782.

M. R. Heilmann, Bull. Soc. Chim. Fr. 1929, 45, 412Ϫ414.

E. Fischer, A. Mouneyrat, Ber. Dtsch. Chem. Ges. 1900, 33,

2384Ϫ2385.

G. Porzi, S. Sandri, Tetrahedron: Asymmetry 1996, 7, 189Ϫ196.

Q. B. Broxterman, B. Kaptein, J. Kamphuis, H. E. Shoemaker,

J. Org. Chem. 1992, 57, 6286Ϫ6294.

J. Mulzer, A. Angermann, B. Schubert, C. Seilz, J. Org. Chem.

1986, 51, 5294Ϫ5299.

R. Amoroso, G. Cardillo, C. Tomasini, P. Tortoreto, J. Org.

Chem. 1992, 57, 1082Ϫ1087.

Y. N. Belokon, V. I. Bakhmutov, N. I. Chernoglazova, K. A.

Kochetkov, S. V. Vitt, N. S. Garbalinskaya, V. M. Belikov, J.

Chem. Soc., Perkin Trans. 1 1988, 305Ϫ312.

[25]

[36]

[26]

[27]

[28]

[29]

[30]

[37]

[31]

[32]

[33]

Received September 2, 2002

[O02498]

Eur. J. Org. Chem. 2003, 1244Ϫ1263

1263

Article Doi

Asymmetric synthesis with 6-tert-butyl-5-methoxy-6-methyl-3,6-dihydro-2H-1,4-oxazin-2-one as a new chiral glycine equivalent: Preparation of enantiomerically pure α-tertiary and α-quaternary α-amino acids

DOI: 10.1002/ejoc.200390179

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejoc.200390179

Full text of DOI:10.1002/ejoc.200390179

Products guided by the article

R&D Labs maybe for 565237-97-0

Relevant to this article

Hot Product