Optically pure N-hydroxy-O-triisopropylsilyl-α-L-amino acid methyl esters from AlCl3-assisted ring opening of chiral oxaziridines by nitrogen containing nucleophiles

Article abstract of DOI:10.1021/jo051890+

This article reports a straightforward and unprecedented process of AlCl₃-assisted oxaziridine ring opening by nitrogen containing nucleophiles, in a totally anhydrous milieu. Under these conditions, nucleophiles exclusively attack the carbon atom of the three-membered heterocycles, obtained from methyl esters of natural α-amino acids, generating N-hydroxy-α-L-amino acid methyl esters. No nitrones, amides, or other side products, either from unwanted rearrangements or due to the attack of the nucleophile on the N atom of the oxaziridine systems, are formed. The hydroxylamine compounds are recovered in excellent yields, after their site-specific conversion into the corresponding O-triisopropylsilyl derivatives, by exposure to triisopropylsilyl triflate in the presence of 1H-imidazole. Derivatization, performed immediately after the recovery of the N-hydroxylated precursors, allows the chiral integrity of the asymmetric α-carbon atoms in the amino acid methyl esters to be retained. It also protects the obtained compounds from frame degradation by disproportionation. N-Hydroxy-O- triisopropylsilyl-α-L-amino acid methyl esters are important intermediates in the study of natural α-L-amino acid metabolic pathways and are ideal candidates as starting materials in the synthesis of biologically, pharmacologically, and nutritionally important N-hydroxy peptides.

Full text of DOI:10.1021/jo051890+

Optically Pure N-Hydroxy-O-triisopropylsilyl-r-L-amino Acid
Methyl Esters from AlCl3-Assisted Ring Opening of Chiral
Oxaziridines by Nitrogen Containing Nucleophiles
Maria Luisa Di Gioia, Antonella Leggio, Adolfo Le Pera, Angelo Liguori,* and Carlo Siciliano
Dipartimento di Scienze Farmaceutiche, Universita` della Calabria, Via P. Bucci,
I-87036 Arcavacata di Rende (CS), Italy
A.Liguori@unical.it
Received September 8, 2005
This article reports a straightforward and unprecedented process of AlCl3-assisted oxaziridine ring
opening by nitrogen containing nucleophiles, in a totally anhydrous milieu. Under these conditions,
nucleophiles exclusively attack the carbon atom of the three-membered heterocycles, obtained from
methyl esters of natural R-amino acids, generating N-hydroxy-R-L-amino acid methyl esters. No
nitrones, amides, or other side products, either from unwanted rearrangements or due to the attack
of the nucleophile on the N atom of the oxaziridine systems, are formed. The hydroxylamine
compounds are recovered in excellent yields, after their site-specific conversion into the corre-
sponding O-triisopropylsilyl derivatives, by exposure to triisopropylsilyl triflate in the presence of
1H-imidazole. Derivatization, performed immediately after the recovery of the N-hydroxylated
precursors, allows the chiral integrity of the asymmetric R-carbon atoms in the amino acid methyl
esters to be retained. It also protects the obtained compounds from frame degradation by
disproportionation. N-Hydroxy-O-triisopropylsilyl-R-L-amino acid methyl esters are important
intermediates in the study of natural R-L-amino acid metabolic pathways and are ideal candidates
as starting materials in the synthesis of biologically, pharmacologically, and nutritionally important
N-hydroxy peptides.
Introduction
chiral ligands or chelators of iron(III). Peptide segments
containing N-hydroxy-R-L-amino acids are also useful
molecular devices, for example, in regulating the protein
ubiquitin-proteasome pathway in Parkinson’s desease.3
In addition, peptides related to N-hydroxy-R-L-amino acid
derivatives are of biological importance, because they
show good activities as low molecular weight sidero-
phores.4 They can also mimic desferrioxamines,5 acting
as carrier systems which are vital to the microorganism’s
acquisition of iron from food and the environment.
Finally, N-hydroxy-R-L-amino acid derivatives are im-
portant molecular signals that indicate pathologies re-
lated to some human and animal tumors.6
Oxaziridines belong to one of the most appropriate
families of molecules potentially offering easy synthetic
routes to a broad range of chiral nitrogenated com-
pounds.1 Enantiomerically pure oxaziridines displaying
natural R-L-amino acid frames can play a very central
role as candidates in preparing N-hydroxy-R-L-amino acid
derivatives: these compounds are relatively uncommon
in nature, but possess attractive biological properties and
occur as residues in several simple and more complex
natural compounds.2 They are often present as constitu-
ents in tri- and tetrapeptide segments which are useful
scaffolds for making peptide sequences able to act as
(3) Zhang, X.; Xie, W.; Qu, S.; Pan, T.; Wang, X.; Le, W. Biochem.
Biophys. Res. Commun. 2005, 333, 544.
(1) (a) Schmitz, E. Adv. Heterocycl. Chem. 1963, 2, 83. (b) Dupin, J.
F. Bull. Soc. Chim. Fr. 1967, 3085. (c) Spence, G. G.; Taylor, E. C.;
Buchardt, O. Chem. Rev. 1970, 70, 231. (d) Bellamy, F.; Streith, J.
Heterocycles 1976, 4, 1391. (e) Schmitz, E. Adv. Heterocycl. Chem.
1979, 24, 63. (f) Davis, F. A.; Sheppard, A. C. Tetrahedron 1989, 45,
5703.
(4) Drechsel, H.; Jung, G. J. Pept. Sci. 1998, 4, 147.
(5) (a) Neilands, J. B. Annu. Rev. Microbiol. 1981, 36, 285. (b) Crosa,
J. H. Annu. Rev. Microbiol. 1984, 38, 69. (c) Hara, Y.; Akiyama, M.
Inorg. Chem. 1996, 35, 5173. (d) Roosenberg, J. M.; Lin, Y. M.; Lu, Y.;
Miller, M. J. Curr. Med. Chem. 2000, 7, 159. (e) Surman, M. D.; Miller,
M. J. Org. Lett. 2001, 3, 519.
(2) Ottenheim, H. C. J.; Herscheid, J. D. M. Chem. Rev. 1986, 86,
697.
(6) Neunhoeffer, O. Z. Naturforsch., B: Chem. Sci. 1970, 25, 299.
10.1021/jo051890+ CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/11/2005
10494
J. Org. Chem. 2005, 70, 10494-10501

AlCl3-Assisted Ring Opening of Chiral Oxaziridines
A large number of syntheses of N-hydroxy-R-L-amino
acid derivatives have been reported, but most have shown
that the desired compounds can be realized only through
nonstereoselective methods, and the final compounds are
isolated in overall yields that are invariably low.7 Dis-
proportionation and the formation of nitrones and amides
as side products are two crucial factors hampering the
satisfactory synthesis of N-hydroxylated R-L-amino acid
analogues.8
synthesis of N-hydroxy-R-L-amino acid derivatives, but
all suffer from drawbacks. In many cases, in fact, the loss
of the chiral integrity of the R-carbon atom in the
N-hydroxy-R-L-amino acid derivatives cannot be con-
trolled.
This work focuses on a novel, clean, and very efficient
ring opening of chiral oxaziridines obtained from R-L-
amino acid methyl esters. The method uses AlCl3, in a
totally anhydrous environment, to assist the ring opening
of the heterocycle by nitrogen containing nucleophiles,
provoking the desired C-O bond cleavage and directing
the nucleophilic attack specifically to the carbon atom of
the three-membered structure. The reaction proceeds
with no production of unwanted nitrone, aldehyde,
ketone, amide, or hydrazine side products. Site-specific
derivatization, carried out after ring opening, is proposed
to prepare the title compounds in excellent yields. This
method opens a new pathway in the synthesis of optically
pure N-hydroxy-R-L-amino acid derivatives, which are
useful precursors in the preparation of N-hydroxy pep-
tides of biological interest.
Ring opening of oxaziridines obtained from methyl
esters of natural R-L-amino acids can offer a straightfor-
ward solution to the problems above. A series of biologi-
cally active N-hydroxy-R-L-amino acid derivatives has
been prepared by transforming R-L-amino acids into
oxaziridines and subsequent acid-catalyzed hydrolysis.
In aqueous and anhydrous organic acids, oxaziridines
were shown to undergo ring opening and cleavage to form
carbonyl compounds and N-substituted hydroxylamines.9
Hydrolitic methods previously employed to obtain pure
N-hydroxy analogues of common R-L-amino acids, be-
cause of the presence of water, also produce nitrone side
products accompanied by disproportionated products,
thus seriously limiting overall yields.10 Upon exposure
to aqueous acidic media, oxaziridines displaying a natural
R-L-amino acid methyl ester framework undergo struc-
tural rearrangement to the corresponding nitrones. Low
temperatures reduce but do not eliminate this rearrange-
ment.11 Any increase in reaction temperature results in
the formation of amides, and at higher T values and with
prolonged reaction times, a notable lowering of the
N-hydroxylated derivative yields has been observed, due
to the activation of thermal rearrangement pathways.
Formation of side products must be avoided if the aim of
the oxaziridine ring opening is to obtain pure N-hydroxy-
L-amino acid derivatives in high yields. Nonhydrolytic
methods are also available in the literature12 for the
Results and Discussion
We initially prepared the optically pure chiral imines
1a and 1b (Scheme 1) in nearly quantitative yields from
a TiCl4-mediated condensation between 4-methoxyben-
zaldehyde and methyl esters of R-L-natural amino acids.13
Imines were then subjected to treatment with anhydrous
m-chloroperoxybenzoic acid (m-CPBA) in dry ethanol-free
methylene chloride at -40 °C, -15 °C, and room tem-
perature (rt). Conversion of the starting imines was
monitored by GC-MS and TLC. Reactions were complete
in 12-23 h, depending on the temperature chosen for the
experiment. The oxidation was studied using imines 1a
and 1b (Scheme 1) as models. The reaction showed good
stereoselectivity: 1a and 1b were each converted into a
mixture of four stable nonracemic oxaziridines, 2a-5a
and 2b-5b, respectively.
The diastereomers were easily separated by flash
column chromatography (FCC) techniques and fully
characterized by 1H NMR. Assignment was unequivocally
confirmed by comparison with data already reported in
the literature14 for a series of diastereomeric oxaziridines
bearing a chiral amine residue on the N-atom of the ring.
Absolute configuration assignments were also supported
by the already reported14c,e,g finding that the absolute
configuration of the nitrogen atom in the three-membered
heterocycle is the opposite of the absolute configuration
(7) (a) Snow, G. A. J. Chem. Soc. 1954, 2588. (b) Emery, T. F.
Biochemistry 1963, 2, 1041. (c) Zinner, G.; Moderhack, D.; Kliegel, W.
Chem. Ber. 1969, 101, 2536. (d) Kneifel, H.; Bayer, E. Angew. Chem.
1973, 85, 542. (e) Buechi, G.; Luk, K. C.; Kobbe, B.; Townsend, J. M.
J. Org. Chem. 1977, 42, 244. (f) Møller, B. L. J. Labelled Compd.
Radiopharm. 1978, 14, 663. (g) Herscheid, J. D. M.; Ottenheijm, H.
C. J. Tetrahedron Lett. 1978, 5143. (h) Cooper, A. J. L.; Griffith, O. W.
J. Biol. Chem. 1979, 254, 2748. (i) Lee, V. J.; Woodward, R. B. J. Org.
Chem. 1979, 44, 2487. (j) Tijhuis, M. W.; Herscheid, J. D. M.;
Ottenheijm, H. C. J. Synthesis 1980, 890. (k) Herscheid, J. D. M.;
Colstee, J. H.; Ottenheijm, H. C. J. J. Org. Chem. 1981, 46, 3346. (l)
Ottenheijm, H. C. J.; Plate, R.; Noordik, J. H.; Herscheid, J. D. M. J.
Org. Chem. 1982, 47, 2147.
(8) (a) Emmons, W. D. J. Am. Chem. Soc. 1957, 79, 5739. (b) Horner,
L.; Ju¨rgens, E. Chem. Ber. 1957, 90, 2184.
(9) (a) Widmer, J.; Keller-Schierlein, W. Helv. Chim. Acta 1974, 57,
657. (b) Duhamel, P.; Be´nard, D.; Plaquevent, J.-C. Tetrahedron Lett.
1985, 26, 6065. (c) Duhamel, P.; Goument, B.; Plaquevent, J.-C.
Tetrahedron Lett. 1987, 28, 2595. (d) Grundke, G.; Keese, W.; Rimpler,
M. Synthesis 1987, 1115. (e) Plaquevent, J. C.; Be´nard, D.; Goument,
B. New J. Chem. 1991, 15, 579. (f) Giard, T.; Be´nard, D.; Plaquevent,
J.-C. Synthesis 1998, 297.
(12) (a) Shin, C.; Masaki, M.; Ohta, M. Bull. Chem. Soc. Jpn. 1970,
43, 3219. (b) Novikov, V. T.; Avrutskaya, I. A.; Fioshin, M. Y.; Belikov,
V. M.; Babievskii, K. K. Elektrokhimiya 1976, 12, 1061. (c) Wittman,
M. D.; Halcomb, R. L.; Danishefsky, S. J. J. Org. Chem. 1990, 55, 1981.
(d) Detomaso, A.; Curci, R. Tetrahedron Lett. 2001, 42, 755.
(13) Leggio, A.; Le Pera, A.; Liguori, A.; Napoli, A.; Romeo, C.;
Siciliano, C.; Sindona, G. Synth. Commun. 2003, 33, 4331.
(14) (a) Bełz˘ ecki, C.; Mostowicz, D. J. Org. Chem. 1975, 40, 3878.
(b) Bogucka-Ledochowska, M.; Konitz, A.; Hempel, A.; Dauter, Z.;
Bełz˘ ecki, C.; Mostowicz, D. Tetrahedron Lett. 1976, 1025. (c) Buccia-
relli, M.; Moretti, I.; Torre, G. J. Chem. Soc., Chem. Commun. 1976,
60. (d) Mostowicz, D.; Bełz˘ ecki, C. J. Org. Chem. 1977, 42, 3917. (e)
Bucciarelli, M.; Forni, A.; Moretti, I.; Torre, G. J. Chem. Soc., Perkin
Trans. 2 1977, 1339. (f) Bełz˘ ecki, C.; Mostowicz, D. J. Chem. Soc.,
Chem. Commun. 1977, 51. (g) Forni, A.; Garati, G.; Moretti, I.; Torre,
G.; Andreetti, G. D.; Bocelli, G.; Sgorabotto, P. J. Chem. Soc., Perkin
Trans. 2 1978, 401. (h) Bogucka-Ledochowska, M.; Konetz, A.; Hempel,
A.; Dantor, Z.; Borowski, E. Z. Kristallogr. 1979, 149, 49.
(10) (a) O’Connor, C. J.; Fendler, E. J.; Fendler, J. H. J. Chem. Soc.,
Perkin Trans. 2 1973, 1744. (b) Połon´ski, T.; Chimiak, A. Tetrahedron
Lett. 1974, 28, 2453. (c) Połon´ski, T.; Chimiak, A. Bull. Acad. Pol. Sci.,
Ser. Sci. Chim. 1979, 27, 459. (d) Butler, A. R.; Challis, B. C.; Lobo, A.
M. J. Chem. Soc., Perkin Trans. 2 1979, 1035. (e) Butler, A. R.; White,
J. G.; Challis, B. C.; Lobo, A. M. J. Chem. Soc., Perkin Trans. 2 1979,
1039.
(11) (a) Sternbach, L. H.; Keechlin, B. A.; Reeder, E. J. Org. Chem.
1962, 27, 4671. (b) Bigot, B.; Roux, D.; Sevin, A.; Devaquet, A. J. Am.
Chem. Soc. 1979, 101, 2560. (c) Sauer, J. Tetrahedron 1979, 35, 2109.
(d) Boyd, D. R.; Coulter, P. B.; Hamilton, W. J.; Jennings, W. B.; Wilson,
V. E. Tetrahedron Lett. 1981, 22, 2287. (e) Davis, F. A.; Jenkins, R.
H., Jr. In Asymmetric Synthesis; Morrison, J. D., Ed.; Academic
Press: New York, 1984; Vol. 4.
J. Org. Chem, Vol. 70, No. 25, 2005 10495

Di Gioia et al.
SCHEME 1. Diastereoselective Oxidation of Imines 1a,b: Preparation of Oxaziridines 2a-5a and 2b-5b
TABLE 1. Yields of the Oxaziridines 2a-5a and 2b-5b
total yielda (%)
(2+3+4+5)
product distributionb
trans:cis ratiob
(2+3):(4+5)
recovered product ratioc
(2:3:4:5)
T (°C)
t (h)
(2:3:4:5)
a: R ) CH2Ph
-40
-15
rt
-40
-15
rt
23
19
12
23
19
12
78
75
71
70
68
63
69:19:7:5
67:19:8:6
68:17:9:6
71:19:7:3
61:22:10:7
60:20:12:8
88:12
86:14
85:15
90:10
83:17
80:20
68:18:7:3
68:18:6:4
67:14:8:5
70:18:6:3
59:22:10:5
58:19:12:7
b: R ) CH2CH(CH3)2
a Expressed as oxaziridine products recovered by FCC. b Determined by 1H NMR. c Obtained for each diastereomer after recovery by
FCC.
of the chiral substituent connected to the nitrogen atom
were formed at any of the reaction temperatures em-
ployed, unlike previously reported methods.16 Selection
of the appropriate thermal conditions can improve the
total yields of oxaziridine products. Yields were lower at
room temperature than at -40 °C, varying from 71 to
78%, for oxidation of imine 1a, and from 63 to 70% in
the case of imine 1b. Similar results were obtained for
the oxidation of the imine containing the leucine motif
in its structural framework. In addition, by using anhy-
drous m-CPBA, reactions are clean and there is no
formation of other undesired species such as acids or bis-
peroxides.
When oxaziridines are obtained from natural R-L-
amino acids, of particular interest is the possibility to
conduct the hydrolytic ring opening cleanly to produce
the corresponding N-hydroxylated derivatives. The previ-
ously reported hydrolytic methods exploit the oxaziridine
ring opening by using concentrated acidic media in an
aqueous environment.9,10 We found these systems to be
impractical due to the large amount of nitrone and amide
derivatives produced. Moreover, these procedures did not
facilitate the recovery of the final compounds, whose total
yields were generally low and limited by disproportion-
ation of the N-hydroxylated R-L-amino acid derivatives.
To define a straightforward and efficient synthesis of
the desired N-hydroxy-R-L-amino acid derivatives, we
of the CdN bond in the parent imines. 1H NMR analysis
of the respective crude materials also proved useful in
determining the effective yields of each stereomeric
oxaziridine formed by oxidation of 1a and 1b. The
calculation was made on the basis of the intensities of
the well distinguished resonances related to the methyl
group of the amino acid ester moiety and agreed closely
to the results obtained via FCC (Table 1).
The presence of a second chiral atom, the C3 term of
the heterocyclic ring, proved to be a valuable feature that
facilitated the separation and quantification of all dia-
stereomers from their respective mixture. The trans-
oxaziridines were always the predominant oxidation
products. Treatment of imines 1a and 1b with m-CPBA
at -15 °C proceeded diastereoselectively, giving trans/
cis ratios of 86:14 and 83:17 for oxaziridines 2a-5a and
2b-5b, respectively. At temperatures from -40 °C to rt,
the diastereomeric distributions were quite constant for
oxaziridines containing the L-phenylalanine residue,
while a slight change was observed for oxaxiridines
derived from L-leucine when the process was carried out
at very low temperature. Values, summarized in Table
1, are between 88:12 and 85:15 for oxaziridines 2a-5a
at all temperatures employed. For oxaziridines 2b-5b,
values between 80:20 and 90:10 were found, the higher
ratio being observed when oxidation takes place at -40
°C. The stereoselectivity was in accordance with the data
reported in the literature for other hindered chiral
systems.15
(15) Boyd, D. R.; Neill, D. C.; Watson, C. G.; Jennings, W. B. J.
Chem. Soc., Perkin Trans. 2 1975, 1813.
(16) (a) Krimm, H. Chem. Ber. 1958, 91, 1057. (b) Splitter, J. S.;
Calvin, M. J. Org. Chem. 1965, 30, 3427. (c) Splitter, J. S.; Su, T. M.;
Ono, H.; Calvin, M. J. Am. Chem. Soc. 1971, 93, 4075. (d) Toda, F.;
Tanaka, K. Chem. Lett. 1987, 2283. (e) Aube´, J.; Burgett, P. M.; Wang,
Y. Tetrahedron Lett. 1988, 2283.
The instrumental analysis of the unseparated product
mixtures recovered showed that no nitrones or amides
10496 J. Org. Chem., Vol. 70, No. 25, 2005

AlCl3-Assisted Ring Opening of Chiral Oxaziridines
SCHEME 2. AlCl3-Assisted Ring Opening of Oxaziridines 2a-5a with Hydroxylamine
subjected all the obtained oxaziridines to a ring opening
performed by nitrogenated nucleophiles, in the presence
of AlCl3 and under totally anhydrous conditions. To our
knowledge, only a few applications of Lewis acids in
oxaziridine reactivity studies have been reported.17 Ex-
amples include the use of BF3 to produce the only known
stable boronate derivatives of nitrones in very high yields,
but no formation of N-hydroxylamine compounds by BF3-
assisted C-O bond activation in oxaziridines has been
reported.17a,b Moreover, no data are available in the
literature concerning the role played by AlCl3 in assisting
oxaziridine ring opening in reactions conducted by ni-
trogenated nucleophiles.
The ring opening was initially attempted by exposing
a mixture of 2a-5a to 2 equiv of AlCl3 and 1 equiv of
hydroxylamine (Scheme 2). The 2 equiv of AlCl3 are
required because of the two potential O atom coordination
sites present in the oxaziridine molecule, namely the one
in the carbonyl moiety and of course the target O atom
in the three-membered ring. The nucleophile was pre-
pared by treating the commercially available hydrochlo-
ride with a stoichiometric amount of sodium methoxide
in dry methanol at 0 °C for 10 min and then evaporating
the solvent. Reaction was performed in dry ethanol-free
methylene chloride for 1 h at room temperature. After
this time, reaction was complete and gave a surprisingly
complex mixture of products. To characterize the desired
N-hydroxylated compound 6, formed together with an
equivalent amount of oximes 7 and 8, the crude mixture
recovered from the reaction was treated overnight with
an excess of acetic anhydride at room temperature
(Scheme 2). GC-MS analysis of samples taken from the
reaction mixture revealed acetylation to be unsatisfac-
tory, because it led to the formation of complex mixtures
of the N- and O-monoacetylated derivatives 9 and 10. The
N,O-bisacetyl-R-L-amino acid methyl ester 11 and the
O-acetylated oximes 12 and 13 were also produced during
the derivatization process. This made separation of the
mixture difficult. Despite these partially negative results,
GC-MS and 1H NMR controls of the crude material
recovered from the acetylation demonstrated that the
aromatic moiety on the L-phenylalanine side chain re-
mained totally unaffected by the reaction with AlCl3. This
fact illustrates the total compatibility between aromatic
substrates and the experimental conditions selected for
the Lewis acid-assisted oxaziridine ring opening. The
results also prompted the changing of the nucleophile
used in the process, specifically those not containing free
hydroxyl functions.
We thus selected O-benzylhydroxylamine as the nu-
cleophile. The free base was obtained by reaction of the
commercial hydrochloride salt with 1 equiv of sodium
methoxide dissolved in dry methanol, exactly as previ-
ously discussed for hydroxylamine. One equivalent of the
freshly prepared nucleophile, a sticky white solid, was
added to the dry ethanol-free methylene chloride solution
containing 1 equiv of the substrate and 2 equiv of AlCl3
(Scheme 3).
After 1 h at room temperature, the crude reaction
mixture was chromatographed to isolate the N-hydroxy-
R-L-amino acid methyl ester 6 in a 60% yield, calculated
on the basis of the amount of oxaziridine precursor, and
of the two anti- and syn-O-benzylaldoximates 14 and 15,
obtained in a 95% yield with respect to the starting
(17) (a) Emmons, W. D. Chem. Heterocycl. Compd. 1964, 19, 624.
(b) Milliet, P.; Lusinchi, X. Tetrahedron Lett. 1971, 3763. (c) Schou-
macker, S.; Hamelin, O.; Te´ti, S.; Pe´caut, J.; Fontecave, M. J. Org.
Chem. 2005, 70, 301.
J. Org. Chem, Vol. 70, No. 25, 2005 10497

Di Gioia et al.
SCHEME 3. AlCl3-Assisted Ring Opening of
Oxaziridines 2a-5a with O-Benzylhydroxylamine
the same as those obtained when using a SiO2 stationary
phase.
It is worth noting that the N-hydroxy-R-L-amino acid
methyl ester 6 showed identical optical activity to the
same product obtained by hydrolytic methods.9,10 Since
the stereochemistry of the R-L-amino acid residue is
maintained during the oxidation of the imine starting
materials, we suppose, on the basis of the results
mentioned above for compound 6, that the AlCl3-assisted
ring opening of oxaziridines does not alter the chirality
of the R-carbon atom.
The results obtained from the treatment of oxaziridines
bearing the L-phenylalanine methyl ester frame surpris-
ingly showed that the attack of the nitrogen nucleophile
takes place exclusively at the carbon atom of the hetero-
cyclic ring. No hydrazine-like compounds from possible
interaction between the nitrogen of the nucleophile and
the N term of the three-membered ring are generated
during the process. This evidence is a novel acquisition
in the field of oxaziridine chemistry, since the literature19
reports that nitrogenated nucleophiles preferentially
attack the N atom rather than the C3 term of the
heterocycle, even in the cases of substrates containing
at least one aromatic substituent on the same carbon
atom. Moreover, N-hydroxy-R-L-amino acid methyl esters
can be synthesized without altering the chirality of the
parent R-L-amino acid methyl esters employed in the
preparation of the starting oxaziridines.
SCHEME 4. Mechanism of Disproportionation of
N-Hydroxy-r-L-amino Acid Methyl Ester 6
To study the temperature dependence of the ring-
opening process, a mixture of the model oxaziridines
2a-5a was refluxed with 2 equiv of AlCl3 and 1 equiv of
O-benzylhydroxylamine in dry ethanol-free methylene
chloride. The reaction rate increased and the substrate
was totally converted in 25 min, but no traces of products
coming from thermal transposition or decomposition of
the starting material and/or the formed products were
detected, as demonstrated by GC-MS analysis. The
yields of the final products were not affected by raising
the reaction temperature from the rt value to that of the
refluxing solvent.
material, and as an equimolar mixture of both the
geometric isomers. Chromatography also furnished the
O-methyl aldoximate 16 and L-phenylalanine methyl
ester 17 in 18 and 15% yields, respectively, estimated in
this case also on the basis of the starting substrate
amount. Recovery of both the O-benzylaldoximates 14
and 15 allowed us to calculate the conversion of the initial
substrate. From the chromatography data it was seen
that oxaziridines subjected to the ring opening, under
these conditions, were totally consumed. Nevertheless,
in all the tests performed, yield of the recovered product
6 was rather modest and never exceeded 60%. Compari-
son of GC-MS and TLC analyses performed on the crude
reaction mixture during the treatment and after the
reaction workup step did not show the presence of 16 and
17 as side products. Their formation may be supposed to
happen, probably via the initial dehydration of 6, on the
chromatography column during product purification.
Compounds 16 and 17 could form from 6 via a nitroso
intermediate (Scheme 4), similarly to the mechanism
already reported18 for the disproportionation of N-hy-
droxy-R-L-amino acids.
The study of the process was concluded by demonstrat-
ing that the nucleophile plays a fundamental role in the
ring opening of oxaziridines to produce N-hydroxy de-
rivatives of natural R-L-amino acids. This was shown by
refluxing 2a-5a dissolved in dry ethanol-free methylene
chloride with AlCl3. After 12 h, the mixture of oxaziri-
dines subjected to conversion was recovered totally
1
unmodified as confirmed by GC-MS runs and H NMR
analysis. On the other hand, the necessity for the Lewis
acid with regard to the ring-opening development was
definitively illustrated by carrying out an experiment in
which a mixture of 2a-5a was treated with O-benzyl-
hydroxylamine in the required molar ratio, in refluxing
dry ethanol-free methylene chloride for 12 h. GC-MS and
1H NMR analysis of samples taken during the reaction,
as well as at the end, showed no change in the substrate
(18) (a) Ahmad, A. Bull. Chem. Soc. Jpn. 1974, 47, 2583. (b) Møller,
B. L.; McFarlane, I. J.; Conn, E. E. Acta Chem. Scand. B 1977, 31,
343.
(19) (a) Schirmann, J.-P.; Weiss, F. Tetrahedron Lett. 1972, 635. (b)
Schmitz, E.; Ohme, R.; Schramm, S.; Striegler, H.; Heyne, H. U.;
Rusche, J. J. Prakt. Chem. 1977, 319, 195. (c) Hata, Y.; Watanabe, M.
J. Am. Chem. Soc. 1979, 101, 6671. (d) Hata, Y.; Watanabe, M. J. Org.
Chem. 1981, 46, 610. (e) Rastetter, W. H.; Wagner, W. R.; Findeis, M.
A. J. Org. Chem. 1982, 47, 419.
The disproportionation occurs even using neutral
Al2O3. The amounts of 16 and 17 recovered are practically
10498 J. Org. Chem., Vol. 70, No. 25, 2005

AlCl3-Assisted Ring Opening of Chiral Oxaziridines
SCHEME 5. Mechanism of the AlCl3-Assisted
Oxaziridine Ring Opening with
O-Benzylhydroxylamine
After 25 min, the reaction mixture was cooled to room
temperature, workup was carried out, and the crude
material recovered was immediately subjected to deriva-
tization with a stoichiometric amount of the reagent
system TIPSOTf/1H-imidazole in dry ethanol-free meth-
ylene chloride at room temperature (Scheme 6). Treat-
ment was complete in 3 h, as demonstrated by GC-MS
and TLC analysis of the crude reaction mixture. Chro-
matography on silica gel of the crude product recovered
from the final workup allowed the separation of the anti-
and syn-O-benzylaldoximates 14 and 15, in an amount
equivalent to that obtained for the desired N-hydroxy-
O-triisopropylsilyl-R-L-amino acid methyl ester 21, which
was isolated in a 94% overall yield, calculated on the
basis of the oxaziridine substrate. Compounds 14, 15, and
21 were the only products. The use of TIPSOTf resulted
in the site-specific derivatization at the O atom of the
N-hydroxylated compound. The silylated product 21
obtained was optically pure and isolated with ease. Since
the process of O-masking is performed immediately after
workup of the reaction mixture obtained from the Lewis
acid-assisted oxaziridine ring opening, the overall yield
registered for the silyl derivative at the end of the
treatment is directly related to that actually obtainable
for the nonsilylated product, as could be determined by
1H NMR analysis of the respective crude material
observed before functionalization. Furthermore, O-tri-
isopropylsilylation suppresses the tendency toward spon-
taneous disproportionation of the N-hydroxy-R-L-amino
acid methyl ester, since the OH group of the hydroxy-
lamine moiety is not available to begin dehydration, the
process responsible for the unwanted structural trans-
formation of the R-L-amino acid derivative (Scheme 4).
and no formation of either the desired compound or other
side products. These data confirm the role that the
nucleophile plays in the process and also provide evidence
for the assistance of the Lewis acid to the reaction. The
use of completely anhydrous conditions is strategic in this
ring-opening step, since formation of unwanted side
products arising from the action of water as a nucleophile
can easily be avoided.
The stability of the O-triisopropylsilylated derivative
1
21 was checked by repeating the TLC, GC-MS, and H
NMR analyses on the same sample, both as a solid and
in deuteriochloroform solution, stored at room tempera-
ture and in the light. Even after long periods, up to three
weeks, the sample remained pristine. To test its thermal
stability, one analytical sample of 21 was refluxed in dry
toluene for up to 12 h. In this case also, neither molecular
degradation nor rearrangement into amide products was
observed. The optical purity of the chiral N-hydroxy-O-
triisopropylsilyl-R-L-amino acid methyl ester obtained
was established by 1H NMR in titration experiments
performed with the shift reagent Eu(hfc)3: no residual
resonances attributable to the D optical antipode of 21
were individuated, and only the spin system signals of
the desired molecule were visible in the 1D spectrum.
The kinetics of the ring opening is noticeably quickened
by the intervention of the Lewis acid, which coordinates
the apical O atom of the three-membered heterocycle
activating the C-O linkage and regiospecifically address-
ing the nucleophilic attack onto the C3 atom, the most
electrophilic term of structure I (Scheme 5). The attack
of O-benzylhydroxylamine produces the tetrahedral zwit-
terionic intermediate II, which, in turn, can evolve to
compounds III and IV through C-N bond breakage. The
final compounds V and VI are then generated after the
reaction workup. Our hypothesis of the mechanism is
supported by the evidence that oxaziridines do not
isomerize when they are treated under the conditions
adopted for the experiment, but solely when AlCl3 is used.
Because of the drawbacks discussed above, we coupled
the AlCl3-assisted ring opening to a derivatization of the
N-hydroxylated R-L-amino acid analogues. This last step
must be performed immediately after the total consump-
tion of the starting substrates and a minimal workup to
remove the residual Lewis acid, without purification of
the crude reaction mixture. Triisopropylsilyl triflate
(TIPSOTf) was chosen as the derivatizing reagent.
The capabilities of this methodology, in which the
AlCl3-assisted oxaziridine ring opening is the key step,
were exploited by treating a mixture of 2a-5a with 2
equiv of the Lewis acid and 1 equiv of O-benzylhydroxy-
lamine, in refluxing dry ethanol-free methylene chloride.
To fully exploit the capabilities of the studied meth-
odology, the application of the same synthetic approach
was extended to the preparation of a series of N-hydroxy-
O-triisopropylsilyl-R-L-amino acid methyl esters (Scheme
6). Attention was devoted to the preparation of lipophilic
N-hydroxy-R-L-amino acid methyl esters, as these are
known to occur as constituents of the iron(III)-chelating
crowns in N-hydroxy peptides of biological importance.20
Oxidation with anhydrous m-CPBA of the respective
imine precursors13 gave mixtures of the corresponding
oxaziridines 2b-5b and 18-20 in 70-77% yields. The
heterocycles were refluxed in dry ethanol-free methylene
chloride in the presence of 2 equiv of AlCl3 and 1 equiv
of O-benzylhydroxylamine.
J. Org. Chem, Vol. 70, No. 25, 2005 10499

Di Gioia et al.
SCHEME 6. Synthesis of N-Hydroxy-O-triisopropylsilyl-r-L-amino Acid Methyl Esters
TABLE 2. Yields of the
No side products arising from disproportionation of the
N-hydroxylated R-L-amino acid derivatives or molecular
rearrangement of the oxaziridine precursor are gener-
ated, and no loss of the chiral integrity of the parent
amino acid frames is observed.
N-Hydroxy-O-triisopropylsilyl-r-L-amino Acid Methyl
Esters
compound
R
overall yields (%)a
21
22
23
24
25
CH2Ph
94
94
92
90
88
CH2CH(CH3)2
CH(CH3)2
CH3
Experimental Section
H
Optical rotations of all new chiral compounds (oxaziridines
and N-hydroxy-O-triisopropylsilyl-R-L-amino acid methyl
esters) were measured in a 1-dm (2-mL volume) tube. All
measurements were carried out on chloroform solutions of each
compound (c 2), and [R]D values were determined at 20 °C by
using a digital polarimeter.
a Obtained after FCC and calculated on the basis of the parent
oxaziridine amounts.
All reactions were complete in 25 min, as checked by
TLC and GC-MS analysis of the reaction mixtures. After
treatment with TIPSOTf and 1H-imidazole in dry etha-
nol-free methylene chloride at room temperature for 3
h, the crude products were chromatographed to separate
the O-benzylaldoximates 14 and 15 from the O-triiso-
propylsilyl derivatives 22-25, which were then isolated
in excellent overall yields (88-94%), calculated with
respect to the amount of oxaziridine precursors (Table
2). 1H NMR analysis of the pure compounds, accom-
plished using the shift reagent Eu(hfc)3, showed the
obtained chiral R-L-amino acid derivatives 22-25 to be
enantiomerically pure.
Synthesis of Oxaziridines 2a-5a and 2b-5b. General
Procedure. The appropriate imine (5.75 mmol) was dissolved
in dry ethanol-free methylene chloride (15 mL) and maintained
at 0 °C under magnetic stirring. A solution of anhydrous
m-chloroperoxybenzoic acid (1.1 g, 6.33 mmol) in dry ethanol-
free methylene chloride (15 mL) was added dropwise in 20 min.
After being stirred for 23 h at -40 °C, TLC showed the reaction
to be complete (Et2O/n-hexane 60:40, v/v), and the m-chlo-
robenzoic acid precipitated was rapidly removed by filtration
under vacuum. The mother liquor was treated with a 1 N
NaOH aqueous solution (20 mL) and brine (10 mL) and then
extracted with ethanol-free chloroform (5 × 10 mL). Organic
layers were separated, washed once with brine (10 mL), and
dried on MgSO4. Solvent was then removed under reduced
pressure conditions, and the mixture of diastereomeric oxa-
ziridines recovered was subjected to flash column chromatog-
raphy.
Conclusions
Methyl esters of natural R-L-amino acids can be N-
hydroxylated using a strategy in which the respective
imine derivatives are oxidized stereoselectively to the
corresponding oxaziridines. Ring opening of these het-
erocycles, assisted by the Lewis acid AlCl3 and carried
out by free O-benzylhydroxylamine, is the key step of the
procedure. In the absence of water, and different from
all previously reported methods, reactions are rapid and
clean, giving rise only to N-hydroxy-R-L-amino acid
methyl esters which are generated through the regiospe-
cific attack of the nitrogenated nucleophile at the C3
atom of the parent oxaziridines. The site-specific O-tri-
isopropylsilylation of the N-hydroxy R-L-amino acid ana-
logues obtained is the conclusive step of the proposed
synthetic route. After derivatization, the desired optically
pure compounds are isolated in excellent overall yields.
Oxaziridine 2a: Pale yellow oil (1.224 g. 3.91 mmol; 68%).
[R]20D -48.9 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.68. FT-IR (neat) 3031, 2953, 2862, 1745, 1613, 1517, 1252,
1170, 1032 cm-1. 1H NMR (CDCl3) δ 7.67 (d, 2 H, J ) 8.0 Hz),
6.70-7.39 (m, 7 H), 4.65 (s, 1 H), 3.80 (s, 3 H), 3.77 (s, 3 H),
3.05-3.42 (m, 3 H). 13C NMR (CDCl3) δ 172.6, 158.7, 140.2,
129.3, 128.8, 128.0, 127.4, 126.2, 114.2, 80.1, 68.5, 56.1, 52.0,
32.0. MS (EI) m/z (relative intensity %) 313 (M+•, 2.1), 298 (1.1),
282 (2.1), 254 (1.0), 163 (4.5), 151 (71.2), 135 (100), 120 (12.3),
107 (10.9), 92 (9.8), 77 (12.8), 64 (7.9), 51 (4.2). HRMALDI:
calcd for C18H20NO4 (M + H)+, 314.13923; found 314.13919.
Oxaziridine 3a: Pale yellow oil (0.33 g, 1.04 mmol; 18%).
[R]20D -49.2 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.67. FT-IR (neat) 3031, 2865, 1748, 1612, 1517, 1252, 1168,
1033 cm-1. 1H NMR (CDCl3) δ 7.67 (d, 2 H, J ) 8.0 Hz), 6.69-
7.39 (m, 7 H), 4.64 (s, 1 H), 3.79 (s, 3 H), 3.75 (s, 3 H), 3.04-
3.43 (m, 3 H). 13C NMR (CDCl3) δ 171.9, 158.2, 141.0, 129.1,
128.8, 128.0, 127.1, 126.0, 114.1, 79.2, 68.1, 57.2, 51.8, 31.8.
MS (EI) m/z (relative intensity %) 313 (M+•, 1.0), 298 (1.0),
282 (1.9), 254 (1.0), 163 (4.7), 151 (68.5), 135 (100), 120 (13.3),
107 (11.9), 92 (9.8), 77 (10.8), 64 (7.7), 51 (4.0). HRMALDI:
calcd for C18H20NO4 (M + H)+, 314.13923; found 314.13921.
Oxaziridine 4a: Pale yellow oil (0.13 g, 0.40 mmol; 7%).
[R]20D -22.4 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.65. FT-IR (neat) 3030, 2954, 2838, 1754, 1613, 1516, 1251,
1170 cm-1. 1H NMR (CDCl3) δ 7.68 (d, 2 H, J ) 8.1 Hz), 6.71-
(20) (a) Wagner, I.; Musso, H. Angew. Chem., Int. Ed. Engl. 1983,
22, 816. (b) Shimizu, K.; Nakayama, K.; Akiyama, M. Bull. Chem. Soc.
Jpn. 1984, 57, 2456. (c) Akiyama, M.; Katoh, A.; Iijima, M.; Takagi,
T.; Natori, K.; Kojima, T. Bull. Chem. Soc. Jpn. 1992, 65, 1356. (d)
Yakirevitch, P.; Rochel, N.; Albrecht-Gary, A. M.; Libman, J.; Shanzer,
A. Inorg. Chem. 1993, 32, 1779. (e) Bianco, A.; Zabel, C.; Walden, P.;
Jung, G. J. Pept. Sci. 1998, 4, 471. (f) Braslau, R.; Axon, J. R.; Lee, B.
Org. Lett. 2000, 2, 1399. (g) Ye, Y.; Liu, M.; Kao, J. L.-K.; Marshall, G.
R. Biopolymers 2003, 71, 489.
10500 J. Org. Chem., Vol. 70, No. 25, 2005

AlCl3-Assisted Ring Opening of Chiral Oxaziridines
7.38 (m, 7 H), 4.74 (s, 1 H), 3.81 (s, 3 H), 3.72 (s, 3 H), 3.18-
3.47 (m, 3 H). 13C NMR (CDCl3) δ 172.3, 158.2, 142.0, 129.0,
128.7, 128.0, 127.1, 126.0, 114.4, 78.5, 68.3, 57.8, 52.3, 33.0.
MS (EI) m/z (relative intensity %) 313 (M+•, 0.9), 298 (1.2),
282 (2.2), 254 (1.4), 163 (3.9), 151 (70.1), 135 (100), 120 (13.1),
107 (14.9), 92 (9.3), 77 (11.8), 64 (7.9), 51 (4.2). HRMALDI:
calcd for C18H20NO4 (M + H)+, 314.13923; found 314.13926.
Oxaziridine 5a: Pale yellow oil (53.2 mg, 0.17 mmol; 3%).
[R]20D -23.5 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.63. FT-IR (neat) 3029, 2957, 2833, 1751, 1618, 1517, 1250,
1168 cm-1. 1H NMR (CDCl3) δ 7.68 (d, 2 H, J ) 8.1 Hz), 6.70-
7.39 (m, 7 H), 4.72 (s, 1 H), 3.79 (s, 3 H), 3.70 (s, 3 H), 3.19-
3.47 (m, 3 H). 13C NMR (CDCl3) δ 172.3, 159.0, 142.0, 128.8,
128.5, 128.0, 127.3, 126.3, 114.8, 80.1, 69.0, 57.9, 52.7, 32.1.
MS (EI) m/z (relative intensity %) 313 (M+•, 1.2), 298 (1.2),
282 (1.9), 254 (1.1), 163 (4.9), 151 (72.0), 135 (100), 120 (11.3),
107 (10.7), 92 (10.1), 77 (12.2), 64 (7.2), 51 (3.9). HRMALDI:
calcd for C18H20NO4 (M + H)+, 314.13923; found 314.13920.
Oxaziridine 2b: Pale yellow oil (1.124 g; 4.03 mmol; 70%).
[R]20D -49.2 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.79. FT-IR (neat) 3170, 2958, 2770, 1752, 1621, 1512, 1235,
diastereomers; 2.2 mmol) dissolved in dry ethanol-free meth-
ylene chloride (10 mL), AlCl3 (0.59 g, 4.4 mmol) was added.
After 5 min at rt, a solution of the free O-benzylhydroxylamine
(0.6 g; 2.2 mmol) in dry ethanol-free methylene chloride (10
mL) was added, and the resulting mixture was refluxed under
magnetic stirring for 25 min. After this time, TLC (Et2O/n-
hexane 60:40, v/v) showed the reaction to be complete. The
solution was cooled at rt, then treated with a 1 N HCl aqueous
solution (pH ) 3). The aqueous phase was separated, and the
organic layer, after drying on MgSO4, was filtered and
evaporated to dryness. The residue was treated with a dry
ethanol-free methylene chloride (10 mL) solution containing
1H-imidazole (0.15 g, 2.2 mmol) and triisopropylsilyl triflate
(TIPSOTf; 0.52 mL, 2.2 mmol) under magnetic stirring at rt.
Reaction was complete in 3 h, as checked by TLC (Et2O/n-
hexane 60:40, v/v). The organic solvent was removed under
reduced pressure conditions, and the solid residue was dis-
solved in a 5% NaHCO3 aqueous solution (5 mL). The aqueous
phase was extracted with AcOEt (3 × 5 mL), and the collected
organic layers were washed once with a 5% KHSO4 aqueous
solution (5 mL) and once with brine (5 mL). After drying on
MgSO4, the solvent was removed under reduced pressure
conditions to give a pale yellow crude material, which was
chromatographed to afford the title compounds in 92-94%
overall yields.
1
1170 cm-1. H NMR (CDCl3) δ 7.38 (d, 2 H, J ) 8.1 Hz), 7.04
(d, 2 H, J ) 8.1 Hz), 4.51 (s, 1 H), 3.78 (s, 3 H), 3.76 (s, 3 H),
3.23 (dd, 1 H, J ) 9.2 Hz, J ) 6.3 Hz), 1.61-1.82 (m, 3 H),
0.87 (d, 6 H, J ) 7.4 Hz). 13C NMR (CDCl3) δ 172.6, 159.3,
130.1, 127.6, 115.0, 79.3, 63.5, 56.0, 51.8, 35.0, 23.1, 22.0. MS
(EI) m/z (relative intensity %) 279 (M+•, 0.1), 223 (14.4), 191
(9.3), 151 (4.5), 135 (100), 107 (8.7), 92 (5.4), 76 (7.2), 64 (13.1),
59 (34.0), 43 (72.3). HRMALDI: calcd for C15H22NO4 (M + H)+,
280.15488; found 280.15484.
N-Hydroxy-O-triisopropylsilyl-r-L-amino Acid Methyl
Ester 21: White powder (0.73 g, 2.07 mmol; 94%), mp 122-
124 °C. [R]20 +22.1 (c 2.0, CHCl3). TLC (Et2O/n-hexane 60:
D
40, v/v) Rf 0.72. FT-IR (KBr) 3350, 3310, 3065, 2870, 2810,
1752, 1554, 1352, 1102 cm-1 1H NMR (CDCl3) δ 7.31-7.36
.
Oxaziridine 3b: Pale yellow oil (0.29 g, 1.04 mmol; 18%).
[R]20D -45.3 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.77. FT-IR (neat) 3168, 2949, 2771, 1750, 1622, 1518, 1235,
(m, 5 H), 6.11 (d, 1 H, J ) 10.1 Hz), 3.87 (ddd, 1 H, J ) 10.1
Hz, J ) 9.1 Hz, J ) 6.3 Hz), 3.71 (s, 3 H), 3.02 (dd, 1 H, J )
13.5 Hz, J ) 6.3 Hz), 2.89 (dd, 1 H, J ) 13.5 Hz, J ) 9.1 Hz),
1.81 (septet, 3 H, J ) 7.1 Hz), 0.93 (d, 18 H, J ) 7.1 Hz). 13C
NMR (CDCl3) δ 171.9, 140.1, 128.7, 127.1, 125.9, 67.7, 52.3,
33.4, 17.8, 10.2. MS (EI) m/z (relative intensity %) 351 (M+.,
21.9), 336 (21.6), 320 (3.4), 308 (9.8), 292 (10.8), 274 (24.3),
260 (9.7), 194 (8.8), 188 (21.7), 178 (11.2), 173 (7.8), 163 (9.9),
157 (15.7), 91 (100), 77 (87.5), 65 (56.9), 59 (34.9), 51 (45.3),
43 (87.9), 31 (34.0). HRMALDI: calcd for C19H34NO3Si (M +
H)+, 352.23080; found 352.23075.
1
1171 cm-1. H NMR (CDCl3) δ 7.38 (d, 2 H, J ) 8.0), 7.05 (d,
2 H, J ) 8.0), 4.54 (s, 1 H), 3.81 (s, 3 H), 3.75 (s, 3 H), 3.23
(dd, 1 H, J ) 9.2 Hz, J ) 6.2 Hz), 1.61-1.83 (m, 3 H), 0.88 (d,
6 H, J ) 7.4 Hz). 13C NMR (CDCl3) δ 171.9, 159.1, 130.0, 127.5,
114.8, 78.1, 64.3, 57.1, 52.8, 34.8, 23.0, 21.8. MS (EI) m/z
(relative intensity %) 279 (M+•, 0.1), 223 (14.4), 191 (9.8), 151
(7.8), 135 (100), 107 (10.7), 92 (7.4), 76 (7.1), 64 (12.8), 59 (34.3),
43 (70.0). HRMALDI: calcd for C15H22NO4 (M + H)+, 280.15488;
found 280.1548.
N-Hydroxy-O-triisopropylsilyl-r-L-amino Acid Methyl
Ester 22: White powder (0.64 g, 2.024 mmol; 92%), mp 90-
92 °C. [R]20D +18.5 (c 2.0, CHCl3). TLC (Et2O/n-hexane 60:40,
v/v) Rf 0.74. FT-IR (KBr) 3348, 3299, 2875, 2820, 1751, 1550,
Oxaziridine 4b: Pale yellow oil (97.7 mg, 0.35 mmol; 6%).
[R]20D -23.0 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.76. FT-IR (neat) 3170, 2959, 2770, 1749, 1620, 1510, 1238,
1
1171 cm-1. H NMR (CDCl3) δ 7.36 (d, 2 H, J ) 8.1 Hz), 6.98
1349, 1100 cm-1 1H NMR (CDCl3) δ 6.10 (d, 1 H, J ) 10.4
.
(d, 2 H, J ) 8.1 Hz), 4.62 (s, 1 H), 3.80 (s, 3 H), 3.73 (s, 3 H),
3.19 (dd, 1 H, J ) 9.1 Hz, J ) 6.4 Hz), 1.62-1.81 (m, 3 H),
0.87 (d, 6 H, J ) 7.3 Hz). 13C NMR (CDCl3) δ 172.3, 159.6,
129.8, 127.6, 115.1, 80.1, 65.5, 56.7, 52.3, 35.3, 23.2, 22.1. MS
(EI) m/z (relative intensity %) 279 (M+•, 0.4), 223 (13.4), 191
(10.1), 151 (3.5), 135 (100), 107 (8.9), 92 (8.4), 76 (7.2), 64 (10.1),
59 (34.0), 43 (72.3). HRMALDI: calcd for C15H22NO4 (M + H)+,
280.15488; found 280.15490.
Hz), 3.68 (s, 3 H), 3.40-3.52 (m, 1 H), 1.79 (septet, 3 H, J )
7.2 Hz), 1.53-1.71 (m, 1 H), 1.27-1.36 (m, 2 H), 0.85-0.92
(m, 24 H). 13C NMR (CDCl3) δ 172.2, 65.7, 52.2, 37.8, 23.1,
22.8, 17.8, 10.1. MS (EI) m/z (relative intensity %) 317 (M+•
,
12.2), 302 (10.2), 286 (8.9), 274 (34.2), 260 (12.6), 258 (44.1),
188 (24.0), 173 (28.7), 160 (22.3), 157 (15.8), 144 (11.0), 129
(12.4), 59 (22.1), 57 (100), 43 (87.0), 31 (23.1). HRMALDI: calcd
for C16H36NO3Si (M + H)+, 318.24645; found 318.24640.
Oxaziridine 5b: Pale yellow oil (47.4 mg, 0.17 mmol; 3%).
[R]20D -24.4 (c 1.0, CHCl3). TLC (Et2O/n-hexane 60:40, v/v) Rf
0.75. FT-IR (neat) 3167, 2954, 2772, 1751, 1620, 1518, 1234,
Acknowledgment. This work was supported by
financial grants from Ministero dell’Istruzione, dell’Uni-
versita` e della Ricerca (MIUR).
1
1165 cm-1. H NMR (CDCl3) δ 7.39 (d, 2 H, J ) 8.0 Hz), 6.97
(d, 2 H, J ) 8.0 Hz), 4.60 (s, 1 H), 3.78 (s, 3 H), 3.70 (s, 3 H),
3.16 (dd, 1 H, J ) 9.2 Hz, J ) 6.1 Hz), 1.60-1.83 (m, 3 H),
0.88 (d, 6 H, J ) 7.4 Hz). 13C NMR (CDCl3) δ 171.8, 159.0,
129.5, 127.3, 114.6, 80.4, 64.4, 57.0, 51.9, 34.6, 23.6, 21.7. MS
(EI) m/z (relative intensity %) 279 (M+•, 0.4), 223 (15.9), 191
(11.2), 151 (5.5), 135 (100), 107 (8.9), 92 (10.4), 76 (8.2), 64
(11.8), 59 (32.4), 43 (71.3). HRMALDI: calcd for C15H22NO4
(M + H)+, 280.15488; found 280.15485.
Supporting Information Available: General experimen-
tal methods and detailed descriptions of procedures used for
the synthesis of oxaziridines 18-20 and N-hydroxy-O-triiso-
propylsilyl-R-L-amino acid methyl esters 23-25, spectral
characterization of O-benzylaldoximates 14 and 15, and
elemental analysis of compounds 2a-5a, 2b-5b, 14, 15, and
21-25. This material is available free of charge via the
Internet at http://pubs.acs.org.
Synthesis of N-Hydroxy-O-triisopropylsilyl-r-L-amino
Acid Methyl Esters 21 and 22. General Procedure. To a
solution of the appropriate oxaziridine (used as mixture of
JO051890+
J. Org. Chem, Vol. 70, No. 25, 2005 10501