Asymmetric C-N bond constructions via crotylsilane addition reactions: A stereocontrolled route to dipeptide isosteres

Article abstract of DOI:10.1021/ja964364w

A new approach to the asymmetric synthesis of (E)-olefin dipeptide isosteres is described based on asymmetric C-N bond constructions resulting from nitronium tetrafluoroborate (NO₂BF₄) promoted electrophilic nitrations of chiral (E)-crotylsilanes and from a copper(I)-catalyzed enantioselective aziridination of chiral (E)-crotylsilanes. The silane reagents undergo efficient anti-S(E)' additions to the nitrogen-based electrophiles to give the (E)-olefin isosteves in > 96% de. The topological bias is principally controlled by the facial bias of the silane reagent. The scope of the methodology was explored via several related crotylsilane derivatives. The (E)-olefin isosteres are nonhydrolyzable, rigid analogs of the peptide linkage in biologically active peptides. The new methodology will facilitate the preparation and study of peptidomimetics since the crotylsilane reagents allow for incorporation of a wide range of functionality on the resulting isosteres.

Full text of DOI:10.1021/ja964364w

6040
J. Am. Chem. Soc. 1997, 119, 6040-6047
Asymmetric C-N Bond Constructions via Crotylsilane
Addition Reactions: A Stereocontrolled Route to Dipeptide
Isosteres
Craig E. Masse,† Bradford S. Knight, Pericles Stavropoulos, and James S. Panek*
Contribution from the Department of Chemistry, Metcalf Center for Science and Engineering,
Boston UniVersity, Boston, Massachusetts 02215
ReceiVed December 19, 1996X
Abstract: A new approach to the asymmetric synthesis of (E)-olefin dipeptide isosteres is described based on
asymmetric C-N bond constructions resulting from nitronium tetrafluoroborate (NO2BF4) promoted electrophilic
nitrations of chiral (E)-crotylsilanes and from a copper(I)-catalyzed enantioselective aziridination of chiral (E)-
crotylsilanes. The silane reagents undergo efficient anti-SE′ additions to the nitrogen-based electrophiles to give the
(E)-olefin isosteres in >96% de. The topological bias is principally controlled by the facial bias of the silane reagent.
The scope of the methodology was explored via several related crotylsilane derivatives. The (E)-olefin isosteres are
nonhydrolyzable, rigid analogs of the peptide linkage in biologically active peptides. The new methodology will
facilitate the preparation and study of peptidomimetics since the crotylsilane reagents allow for incorporation of a
wide range of functionality on the resulting isosteres.
Introduction
Replacement of the amide bond linkage with an (E)-olefin is
a proven useful configurationally biased structural mimic for
the construction of peptide linkages in a number of different
enzyme inhibitors.1 These analogs of biologically active
peptides offer distinct advantages over the naturally occurring
compounds including lower enzymatic degradation, increased
oral bioavailability, and a prolonged duration of action.2 The
amide linkage is the primary target for enzymatic degradation
of peptide and peptide-like substances; therefore structural
modifications at this site may lead to enhanced metabolic
stability. A combination of factors make such suitably func-
tionalized molecules less susceptible to unwanted biodegrada-
tion. These factors include the absence of hydrogen bonding
potential and unlike an amide bond inability to undergo
proteolytic cleavage at the peptide linkage. In addition, it is
known that key factors in the degree of bioactivity of the
peptides are the chemical nature and stereochemistry of the
associated R-amino acid side chains.1a Although hydrogen-
bonding involving the amide linkage to the enzyme active site
is important in bioactivity, the olefinic isostere may still achieve
excellent binding recognition with a target enzyme. A com-
parison between the contributing resonance structures of the
amide linkage and the incorporated (E)-olefin isostere produced
is illustrated in Figure 1. Recently, interest has emerged in
peptidomimetics of inverso D-peptides which are resistant to
Figure 1. Comparison of contributing resonance structures of the amide
linkage with the olefinic isostere.
Figure 2. Comparison of the three-dimensional structure of the amide
linkage with the olefin isostere.4i
the action of cellular proteases thereby making these isosteres
attractive for incorporation into potential drug candidates.3
(E)-Olefin isosteres have been shown to be useful replace-
ments for the amide linkage in drug candidates as the (E)-
CRdCH group closely approximates the bond lengths, angles,
and rigidities of the natural parent amide (Figure 2). These
peptide mimetics require the generation of a stereocenter at the
R-position of a 5-amino-3-hexenoic acid derivative; therefore
any successful chemical synthesis of dipeptide isosteres must
† Recipient of a Graduate Fellowship from the Organic Chemistry
Division of American Chemical Society 1996-1997, sponsored by Organic
Syntheses Inc.
(4) (a) Spaltenstein, A.; Carpino, P. A.; Miyake, F.; Hopkins, P. B. J.
Org. Chem. 1987, 52, 3759-3766. (b) Lehman de Gaeta, L. S.; Czarniecki,
M.; Spaltenstein, A. J. Org. Chem. 1989, 54, 4004-4005. (c) Ibuka, T.;
Habashita, H.; Funakoshi, S.; Fujii, N.; Oguchi, Y.; Uyehara, T.; Yamamoto,
Y. Angew. Chem., Int. Ed. Engl. 1990, 29, 801-803. (d) Allimendinger,
T.; Felder, E.; Hungerbu¨hler, E. Tetrahedron Lett. 1990, 31, 7301-7304.
(e) Kempf, D.; Wang, X. C.; Spanton, S. G. Int. J. Pept. Protein Res. 1991,
38, 237-241. (f) Fujii, N.; Nakai, K.; Habashita, H.; Yoshizawa, H.; Ibuka,
T.; Garrido, F.; Mann, A.; Chounan, Y.; Yamamoto, Y. Tetrahedron Lett.
1993, 34, 4227-4230. (g) Bohnstet, A. C.; Vara Prasad, J. V. N.; Rich, D.
H. Tetrahedron Lett. 1993, 34, 5217-5220. (h) Yong, Y. F.; Lipton, M.
A. Bioorg. Med. Chem. Lett. 1993, 3, 2879-2884. (i) Wipf, P.; Fritch, P.
C. J. Org. Chem. 1994, 59, 4875-4886. (j) Bartlett, P. A.; Otake, A. J.
Org. Chem. 1995, 60, 3107-3111. (k) Norman, B. H.; Kroin, J. S.
Tetrahedron Lett. 1995, 36, 4151-4154.
X Abstract published in AdVance ACS Abstracts, June 1, 1997.
(1) (a) Gante, J. Angew. Chem., Int. Ed. Engl. 1994, 33, 1699-1720
(review). (c) Spatola, A. In Chemistry and Biochemistry of Amino Acids,
Peptides and Protein; Weinstein, B., Ed.; Marcel Dekker: New York, 1983;
Vol. 7, pp 267-357. (d) Also, see: Kaltenbronn, J. S.; Hudspeth, J. P.;
Lunney, E. A.; Michniewicz, B. M.; Nicolaides, E. D.; Repine, J. T.; Roark,
W. H.; Stier, M. A.; Tinney, F. J.; Woo, P. K.; Essenburg, A. D. J. Med.
Chem. 1990, 33, 838-845 and references cited therein.
(2) Hirschmann, R. Angew. Chem., Int. Ed. Engl. 1991, 30, 1278-1301.
(3) (a) Brady, L.; Dodson, G. Nature 1994, 368, 692-693. (b) Guichard,
G.; Benkirane, N.; Graff, R.; Muller, S.; Briand, J.-P. Pept. Res. 1994, 7,
308-321. (c) Muller, S.; Guichard, G.; Benkirane, N.; Brown, F.; Van
Regenmortel, M. H. V.; Briand, J.-P. Pept. Res. 1995, 8, 138-144.
S0002-7863(96)04364-8 CCC: $14.00 © 1997 American Chemical Society

C-N Bond Constructions Via Crotylsilane Addition Reactions
Figure 3. Access of isostere core via crotylsilylation reaction.
J. Am. Chem. Soc., Vol. 119, No. 26, 1997 6041
Scheme 1a
address an (E)-selective olefination process and also allow for
the stereoselective introduction of the R-side chains of the amino
acids associated with P1-P1′ positions of the dipeptide. The
lack of general efficient syntheses of (E)-olefins bearing bis-
allylic stereocenters has impeded their development as pepti-
domimetics and reports describing highly stereoselective ap-
proaches to these isosteric units are rare.4 These methods
generally rely on the use of modified amino acid derivatives as
the source of absolute chirality which typically leads to
difficulties in the introduction of the second R-side chain of
the associated amino acid. Ibuka and co-workers5 have reported
that boron trifluoride‚organocopper reagents can be used for
the introduction of the R-side chains of the (E)-olefin isostere.
The reaction is thought to proceed by an anti-SN2′ displacement
pathway on the δ-amino-γ-mesyloxy-R,â-unsaturated esters and
represents a highly stereoselective process for the introduction
of one of the isostere side chains. However, a multistep process
was required to access the starting unsaturated ester. Recently,
Wipf and co-workers4i have accessed this class of peptide
mimetics by an anti-SN2′ alkylation of chiral N-acylaziridines
with organocuprate reagents. In those cases, the cuprate
additions were found to be prone to several side reactions and
varying levels of diastereoselection.
a The absolute stereochemistry of the silanes used in these examples
is indicated in parentheses, although either antipode is accessible. For
a detailed preparation of the silanes, see ref 8.
to organic synthetic methodology. This advance would also
constitute an important contribution to the general design
considerations for peptide mimetics.
Recent efforts in our laboratories have demonstrated that
chiral (E)-crotylsilanes act as useful carbon nucleophile equiva-
lents in highly diastereo- and enantiospecific condensation
reactions with aldehydes, acetals, and certain electrophilic
alkenes. These experiments have culminated in efficient
methods for the asymmetric synthesis of functionalized ho-
moallylic ethers, tetrahydrofurans, γ-alkoxy-R-amino acid syn-
thons, and tetrasubstituted cyclopentanes.6 The dipeptide
isostere core, characterized by an (E)-olefin bordered on either
side by alkyl substituted stereocenters on the peptide backbone,
closely resembles the general structural type of products
accessed by allylsilane/electrophile additions (Figure 3). The
reaction results in the simultaneous introduction of a new (E)-
double bond and both side chains (R1 and R2) with high levels
of diastereo- and enantioselection. Due to the facile access to
the isostere core that the crotylsilylation methodology affords,
we became interested in the asymmetric synthesis of dipeptide
isosteres. Although we had previously reported the construction
of pseudo C2-symmetric isosteres by asymmetric Lewis acid
promoted additions to s-trioxane,7 a more general and direct
approach to the isosteres would be realized by asymmetric C-N
bond construction.
Our exploration has led to new asymmetric syntheses of (E)-
olefin dipeptide isosteres which utilize a chiral allylsilane bond
construction methodology involving additions of chiral (E)-
crotylsilanes to nitrogen-based electrophiles. This methodology
is capable of solving the problems of stereocontrol in the
installation of the (E)-double bond and introduction of a variety
of alkyl, alkoxy, and allyl groups (amino acid side chains) in
either a syn or anti stereochemical relationship. Thus, the
development of a highly stereoselective general route to these
(E)-olefin isosteres would be considered a valuable contribution
Synthesis of the Chiral Crotylsilanes. The synthesis of the
silane reagents (1a-f), as summarized in Scheme 1, centers
around an ortho-ester Claisen rearrangement for the elaboration
of the illustrated (E)-vinylsilanes to the crotyl derivatives. The
entire process begins with enantiomerically enriched (R,E)- and
(S,E)-vinylsilanes derived from enzymatic resolutions (Pseudomo-
nas lipase) of the corresponding racemic secondary alcohols.8
The sigmatropic rearrangements are highly enantio- and dias-
tereoselective and stereoselective with respect to the configu-
ration of the disubstituted (E)-double bond, which is produced
exclusively. The corresponding anti- and syn-â-substituted
crotylsilanes are produced by either a diastereoselective anti-
alkylation of the unsubstituted crotylsilane (eq 1) or an Ireland-
Claisen rearrangement of the appropriate (E)-vinyl silane (eq
2).
Results and Discussion
Asymmetric C-N Bond Construction. In order to expand
this methodology to encompass the preparation of a general class
of dipeptide isosteres, we needed to address the issue of
asymmetric C-N bond construction in the chiral (E)-crotylsilane
system to prepare the required allylic nitrogen compounds. In
addition to the synthesis of dipeptide isosteres, such a trans-
formation would constitute a facile synthetic route to chiral
allylic amine synthons. We have previously disclosed a method
to these synthons based on nitronium tetrafluoroborate (NO2-
BF4) electrophilic nitrations of chiral (E)-crotylsilanes.9 Forma-
tion of the allylic nitro group takes place with complete
stereocontrol. The chirality of the emerging nitro-bearing center
solely originates from and is controlled by the nature of the
silyl stereocenter and is consistent with an anti-SE′ addition
(5) (a) Ibuka, T.; Habashita, H.; Otaka, A.; Fujii, N.; Oguchi, Y.; Uyehara,
T.; Yamamoto, Y. J. Org. Chem. 1991, 56, 4370-4382.
(6) Masse, C. E.; Panek, J. S. Chem. ReV. 1995, 95, 1293-1316.
(7) Beresis, R.; Panek, J. S. Bioorg. Med. Chem. Lett. 1993, 3, 1609-
1614.
(8) Synthesis of the chiral silane reagents: Beresis, R. T.; Solomon, J.
S.; Yang, M. J.; Jain, N. F.; Panek, J. S. Org. Synth. In press.
(9) Beresis, R. T.; Masse, C. E.; Panek, J. S. J. Org. Chem. 1995, 60,
7714-7715.

6042 J. Am. Chem. Soc., Vol. 119, No. 26, 1997
Masse et al.
Table 1. Asymmetric Synthesis of (E)-Olefin Dipeptide Isosteres
Scheme 2
process. The absolute stereochemical assignment of the nitro
compounds was established by conversion of the appropriate
isostere to L-Cbz-alanine.10 Subsequent reduction of the allylic
nitro compounds using Zn/HCl furnishes the allylic amines.11
The crude amines were N-acylated with carbobenzoxy chloride
to yield the allylic carbamates as the desired isosteres 3a-f in
good overall yields with high levels of stereoselectivity12 and
excellent levels of 1,4-remote asymmetric induction (Scheme
2). This concise stereocontrolled route to these isosteres may
increase their availability for drug discovery programs. The
results of this dipeptide isostere synthesis are summarized in
Table 1. The overall yield for the three-step sequence is
comparable to existing methods for the asymmetric synthesis
of this class of isostere,4 and the synthetic sequence is short
and operationally straightforward.
a Overall yield for the 3 step-sequence based on pure materials
isolated by chromatography (SiO2). b Ratios of products were deter-
1
mined by H NMR (400 MHz) operating at S/N of >200:1. c Silane
1d was prepared by an analogous Claisen strategy starting from
4-methyl-1-pentyn-3-ol, see experimental for details.
Scheme 3
Prior to our work in this area, only one example of the
reaction of a nitronium cation with an allyl metal reagent has
been reported. Olah and Rochin have previously demonstrated
the feasibility of this type of electrophilic substitution through
nitration of unfunctionalized and achiral allylsilanes.13 In our
examples, solid NO2BF414 (1.1 equiv) in anhydrous CH2Cl2 was
determined to be the most effective nitrating agent-solvent
combination for efficient electrophilic substitution and formation
of the allylic nitro compounds in yields ranging between 50
and 66%. The modest yields in the nitration step were primarily
due to partial desilylation of the starting crotylsilane induced
by the presence of free fluoride ion in solution. The amount
of competing desilylation was partially reduced by allowing the
mixture to slowly warm to room temperature over 10 h and
diluting with an aqueous saturated NaHCO3 solution. The
allylic nitro compounds proved to be especially labile and were
subject to partial decomposition during column chromatography
on silica gel and under basic conditions. Due to competing
nitration of the aromatic ring,15 attempts to carry out the
asymmetric nitration reaction on substrates containing benzyl
substituents failed. Reduction was ultimately accomplished
using Zn/HCl in methanol at 0 °C which cleanly afforded the
allylic amines in crude yields on the order of 95%. The crude
amines were immediately N-acylated with carbobenzoxy chlo-
ride (Cbz-Cl) followed by purification by chromatography on
silica gel to give the allylic carbamate isosteres.
(10) The absolute stereochemistry of the allylic nitro compounds has
been established by conversion of the (5S)-isostere to synthetic Cbz-L-alanine
via the following sequence. Comparison to the natural amino acid (Fluka)
derivative unambiguously established the absolute stereochemistry of these
compounds (see Supporting Information for experimental details).
Application to the Synthesis of a Thermolysin Peptido-
mimetic. The potential utility of this technology was illustrated
with a short synthesis of a structural mimetic of the tripeptide
sequence Cbz-Ala[Ψ(E)-CHdCH]Gly-Leu-OMe. Our interest
in this structural class arises from a recent report demonstrating
potent inhibition of the zinc endopeptidase thermolysin by
similar peptidomimetics.4j The synthesis of the tripeptide analog
5 was accomplished by methyl ester hydrolysis of isostere 3a
using lithium hydroxide followed by coupling of the derived
acid 4a with the methyl ester of L-leucine as illustrated in
Scheme 3.
(11) Numerous conditions were surveyed to cleanly effect nitro group
reduction including transfer hydrogenation, CoCl2‚6H20/NaBH4, iron/acetic
acid, and NaBH4/S, all of which resulted in complex mixtures.
(12) All new compounds were isolated as chromatographically pure
materials and exhibited acceptable 1H NMR, 13C NMR, IR, MS, and HRMS
spectral data.
Asymmetric Aziridination. This asymmetric nitration of
chiral (E)-crotylsilanes with NO2BF4 provides access to both
natural and unnatural analogs of peptide sequences by choice
(15) The crude 1H NMR of these substrates indicated substantial
desilation and displayed a para-substituted splitting pattern of the aromatic
protons.
(13) Olah, G.; Rochin, C. J. Org. Chem. 1987, 52, 701-702.
(14) NO2BF4 is commercially available from Aldrich Chemical Co.

C-N Bond Constructions Via Crotylsilane Addition Reactions
J. Am. Chem. Soc., Vol. 119, No. 26, 1997 6043
Scheme 4
Scheme 5
We now wish to disclose the scope of a copper (I)-catalyzed
olefin aziridination of chiral (E)-crotylslianes using PhIdNTs23
as the nitrene precursor. Our initial investigation into the
aziridination was centered around the optimization of reaction
solvent and catalyst as well as reaction temperature. In accord
with the observations of Evans and co-workers,18 the use of
acetonitrile as solvent at room temperature proved to be
optimum; however, the choice of a copper catalyst was
unexpectedly critical. Copper(I) triflate (Cu(I)OTf) was found
to be the optimal catalyst in terms of reaction turnover as
compared to other copper(I) and copper(II) sources such as Cu-
(MeCN)4ClO4,24 Cu(MeCN)4PF6,25 Cu(acac)2,25 and Cu(OTf)2.25
Achiral copper(I) bis-imine catalysts were also found to be less
effective than Cu(I)OTf despite the fact that such bis-imine
ligands have been shown to accelerate the aziridination of certain
olefins.19 The use of 10 mol % Cu(I)OTf in acetonitrile at room
temperature was found to be the optimal catalyst load-solvent
combination for the aziridination of the chiral (E)-crotylsilanes.
Initial attempts at the aziridination using the chiral (E)-crotylsilyl
esters (1) resulted in poor levels of diastereoselection in the
resulting N-tosyl allylic amines. The lack of diastereoselectivity
was attributed to poor olefin facial bias in the binding of the
ester-bearing silanes to the copper species.26 In order to increase
the binding affinity of the crotylsilanes to the copper species
and therefore increase the diastereoselectivity of the aziridina-
tion, the ester functionality of the silanes was reduced to the
corresponding chiral silyl alcohols (6a-f) as illustrated in
Scheme 5 for a substrate possessing anti stereochemistry.
Subjection of the silyl alcohols to the aziridination conditions
described above afforded the N-tosylallylic amines with high
levels of diastereoselection and in moderate yields. Presumably,
the stronger binding affinity of the silyl alcohols to the copper
species allows for superior facial bias in the delivery of the
nitrene source to the olefin of the silane. In this case, the silane
may function as an internal ligand/substrate that binds to the
copper-nitrenoid intermediate thereby directing the tosylnitrene
to the olefin of the silyl alcohol. The present study illustrates
the utility of these chiral silane reagents in the development of
an effective one-step method for the asymmetric synthesis of
(E)-olefin dipeptide isosteres. Five (E)-crotylsilanes (6a-e)
were examined to establish the viability of this approach for
the synthesis of a series of dipeptide isosteres. The synthesis
of the individual dipeptide isosteres is shown in Scheme 6 and
is illustrated with a substrate possessing anti stereochemistry.
The chiral silane reagents undergo a rapid copper(I)-catalyzed
aziridination with PhIdNTs at room temperature to directly
afford the N-tosylallylic amines (7a-e) which are isolated as
chromatographically pure materials. The absolute stereochem-
istry of the allylic N-tosylamines was established by correlation
to the known allylic nitro compounds.27 Presumably, the first
addition products are the aziridines which undergo desilylation
to afford the allylamines; however, analysis of the crude reaction
mixtures showed only the presence of the desired allylic amines
along with starting silane and small amounts of desilylated
of the chirality of the starting allylsilane reagent and provides
a highly diastereo- and enantiospecific method for the asym-
metric synthesis of (E)-olefin dipeptide isosteres. Despite the
concise nature of the asymmetric nitration protocol, we had
envisioned a more efficient approach to this class of isosteres
via a direct asymmetric aziridination of the chiral (E)-crotyl-
silanes. Our expectation for high levels of acyclic stereocontrol
in the aziridination process stems from the ability of the
stereocenter bearing the silicon group to direct addition to one
of the π-faces of the adjacent olefin. Additionally, the σ-donat-
ing silicon group will activate the aziridine ring by σ f σ*
orbital overlap, which will stabilize the emerging â-carbocation
in the ring opening process to afford the allylic amine derivative.
The development of catalytic enantiospecific nitrogen-group
transfer processes to alkenes has recently emerged as a promis-
ing method for the construction of functionalized aziridines,
whose utility has been well documented in organic synthesis.16
In the seminal publication, Kwart and Kahn17 reported the
copper-bronze catalyzed aziridination reactions of cyclohexene
with benzenesulfonylazide. Evans and co-workers then made
the critical discovery that low-valent copper complexes catalyze
the aziridination of various types of olefins using the nitrene
source
of
(N-(p-toluenesulfonyl)imino)phenyliodinane
(PhIdNTs).18 Subsequently, the Jacobsen group19 as well as
the Evans group20 disclosed methods for the asymmetric
aziridination of unfunctionalized olefins utilizing chiral copper-
(I)-based catalyst systems. These methodologies demonstrated
the feasibility of a catalytic asymmetric variant of the reaction
analogous to the well-known methods for enantioselective
cyclopropanation.21 The potential utility of allylsilane substrates
in an aziridination reaction had been reported by Fioravanti and
co-workers who reported the successful addition of the nitrene
precursor N-{[4-nitrophenyl)sulfonyl]oxy}carbamate to achiral
allyltrimethylsilanes to produce allylic amine derivatives (Scheme
4).22 However, an asymmetric variant of this allylsilane
aziridination has not been reported to our knowledge despite
the utility that such a methodology would enjoy. The chiral
allylic amines produced after subsequent desilylation of the
chiral aziridine intermediates are of significant biological interest
as peptidomimetics for pharmaceutical applications. The pos-
sibility of a catalytic, enantiospecific method for the construction
of dipeptide isosteres led us to explore this idea as a viable
method for the efficient aziridination of (E)-crotylsilanes.
(16) (a) Padwa, A; Woolhouse, A. D. Aziridines, Azines and fused ring
derivatives. In ComprehensiVe Hetereocyclic Chemistry; Lwowski, W., Ed.;
Pergamon Press: Oxford, 1984; Vol. 7. (b) Deyrup, J. A. In The Chemistry
of Heterocyclic Compounds; Hassner, A., Ed.; John Wiley and Sons: New
York, 1983; Vol. 42, Part 1. Ring opening of aziridines: (c) Martens, J.;
Scheunemann, M. Tetrahedron Lett. 1991, 32, 1417-1418. (d) Tanner, D.;
Birgersson, C.; Dhaliwal, H. K. Tetrahedron Lett. 1990, 31, 1903-1906.
(17) Kwart, H.; Kahn, A. A. J. Am. Chem. Soc. 1967, 89, 1951-1953.
(18) Evans, D. A.; Faul, M. M.; Bilodeau, M. T. J. Org. Chem. 1991,
56, 6744-6746.
(19) Li, Z.; Conser, K. R.; Jacobsen, E. N. J. Am. Chem. Soc. 1993,
115, 5326-5327.
(20) Evans, D. A.; Faul, M. M.; Bilodeau, M. T.; Anderson, B. A.;
Barnes, D. M. J. Am. Chem. Soc. 1993, 115, 5328-5329.
(21) (a) Lowenthal, R. E.; Abiko, A.; Masamune, S. Tetrahedron Lett.
1990, 31, 6005-6008. (b) Evans, D. A.; Woerpel, K. A.; Hinman, M. M.;
Faul, M. M. J. Am. Chem. Soc. 1991, 113, 726-728. (c) Lowenthal, R. E.;
Masamune, S. Tetrahedron Lett. 1991, 32, 7373-7376.
(22) Fioravanti, S.; Loreto, M. A.; Pellacani, L.; Raimondi, S.; Tardella,
P. A. Tetrahedron Lett. 1993, 34, 4101-4104.
(23) The PhIdNTs was prepared according to Yamada et al. (Yamada,
Y.; Yamamoto, T.; Okawara, M. Chem Lett. 1975, 361-362) with
modifications as described in the Experimental Section.
(24) Kubas, G. J. Inorg. Synth. 1979, 19, 90-92.
(25) Commercially available from Aldrich Chemical Co.
(26) Li, Z.; Quan, R. W.; Jacobsen, E. N. J. Am. Chem. Soc. 1995, 117,
5889-5890.

6044 J. Am. Chem. Soc., Vol. 119, No. 26, 1997
Masse et al.
Table 2. Asymmetric Aziridination of (E)-Crotylsilanes
Scheme 6
a Overall yield for the sequence based on recovered silane. b Ratios
of products were determined by 1H NMR (400 MHz) operating at S/N
of >200:1. c Silane 6e was prepared by an analogous Claisen strategy
starting from 4-methyl-1-pentyn-3-ol, see experimental section for
details.
Scheme 7a
substrate. The overall sequence yields the desired isosteres in
moderate yields with high levels of stereoselectivity and
excellent levels of 1,4-remote asymmetric induction. The results
of the asymmetric aziridination reactions are summarized in
Table 2.
The modest yields in the aziridination reaction were primarily
due to the catalyst turnover rate which resulted in the recovery
of silyl alcohol as well as small amounts of desilylation product
induced by the presence of triflate ion in solution. In these
examples, solid PhIdNTs (1.0 equiv) in anhydrous acetonitrile
(0.4 M) with 10 mol % Cu(I)OTf was determined to be the
most effective aziridinating agent concentration-catalyst load
for the efficient preparation of the N-tosylallylic amines. The
starting silyl alcohols were readily recovered by flash chroma-
tography and could be recycled into the aziridination reaction
after being recharged with catalyst. The aziridination methodol-
ogy allows for the facile introduction of a variety of alkyl or
alkoxy side chains into the dipeptidomimetic. It should also
be pointed out that this approach can be applied to the
preparation of all four possible diastereomers of the Ψ[CHdCH]
dipeptides in a stereocontrolled fashion by the proper selection
of the silane reagent.
1
a Ratio determined by H NMR of crude reaction mixture. b Yield
based on recovered silane.
dination process have been reported using an N-aminoquinazolo-
ne nitrene source.28 In the examples shown in Table 2, the
hydroxyl group seems to reinforce the sense of diastereoselection
by working synergistically with the topology of the (E)-olefin
substrate. We have previously noted similiar hydroxyl directing
effects in the epoxidations of these silyl alcohols with m-CPBA
or VO(acac)2-TBHP.29 In an effort to determine the effect of
the coordination ability of the silane substrate on the diaste-
reoselectivity of the aziridination, several derivatives of silyl
alcohol 6c were prepared and subjected to the aziridination
conditions (Scheme 7). The tert-butyldiphenylsilyl (TBDPS)
ether and the methyl ether derivatives (6f-g) of the silyl alcohol-
6c were prepared to probe the mechanism of the aziridination.30
Aziridination of the TBDPS-ether substrate (6f) resulted in a
2:1 mixture of diastereomeric N-tosylallylic amine isosteres. This
result was consistent with the level of diastereoselection obtained
for the ester-bearing silane reagents (1). Aziridination of the
methyl ether substrate (6g) which is capable of more effective
coordination than the TBDPS ether resulted in a 4:1 mixture of
diastereomeric N-tosylallylic amine isosteres. Clearly, the
alcohol functionality has a pronounced effect on the diastereo-
selectivity of the aziridination which may implicate a need for
a secondary coordination of the silane substrates to the copper-
A Substrate Directed Process. The observation of signifi-
cantly enhanced diastereoselectivity in the aziridination process
upon introduction of a bis-homoallylic alcohol functionality
implicated a preassociation of the silane with the aziridinating
reagent. Several examples of such a hydroxyl-directed aziri-
(27) Proof of the absolute stereochemistry of the allylic N-tosylamines
was obtained by correlation of the known allylic nitro compound 2a to the
allylic N-tosylamine 7a via the following sequence: (see Supporting
Information for experimental details).
(28) (a) Atkinson, R. S.; Kelley, B. J. J. Chem. Soc., Chem. Commun.
1988, 624-625. (b) Atkinson, R. S.; Kelley, B. J.; McNocolas, C. J. Chem
Soc., Chem. Commun. 1988, 562-564. (c) Atkinson, R. S.; Kelley, B. J. J.
Chem. Soc., Perkin Trans. I 1989, 1515-1519.
(29) Panek, J. S.; Garbaccio, R. M.; Jain, N. F. Tetrahedron Lett. 1994,
35, 6453-6456.
(30) See the Experimental Section for the preparation of silanes 6f and
6g.

C-N Bond Constructions Via Crotylsilane Addition Reactions
J. Am. Chem. Soc., Vol. 119, No. 26, 1997 6045
NMR (400 MHz, CDCl3) δ 5.98-5.79 (m, 2H), 5.03 (m, 1H), 3.67 (s,
3H), 3.12 (d, 2H, J ) 6.4 Hz), 1.62 (d, 3H, J ) 6.4 Hz).
Representative Procedure for the Reduction and N-Acylation of
the Allylic Nitro Compounds. (5R)-(E)-Methyl-5-(N-benzyloxy)-
amino-hex-3-enoate (3a). A solution of 2a (0.5 g, 2.9 mmol) in
absolute methanol (10 mL) at 0 °C was treated with zinc dust (0.38 g,
5.8 mmol, 2.0 equiv). Under vigorous stirring, concentrated aqueous
HCl (1.5 mL) was added dropwise to the cooled solution (0 °C). The
reaction mixture was subsequently filtered through Celite, washed
thoroughly with CH2Cl2 (15 mL), and neutralized with 3 N NaOH.
The neutralized solution was extracted further with CH2Cl2 (2 × 15
mL), dried (MgSO4), and concentrated in Vacuo to afford a light yellow
oil, 0.4 g. To a solution of the crude amine (0.4 g, 2.9 mmol) in 15
mL of CH2Cl2 was added pyridine (5.8 mmol, 0.5 mL, 2.0 equiv), and
the solution was cooled to 0 °C. A solution of benzylchloroformate
(5.8 mmol, 0.83 mL, 2.0 equiv) in 6 mL of CH2Cl2 was added via a
dropping funnel to the cooled solution over a 1 h period. The reaction
was quenched with a saturated aqueous NH4Cl solution (10 mL),
extracted with CH2Cl2 (3 × 10 mL), dried (MgSO4), and concentrated
in Vacuo. Purification on SiO2 (15% EtOAc/PE) afforded 3a as a
yellow oil, 0.58 g (78%, two steps). 1H NMR (400 MHz, CDCl3) δ
7.36-7.27 (m, 5H), 5.80-5.63 (m, 2H), 5.21 (s, br, 2H), 4.69 (br,
1H), 3.64 (s, 3H), 2.97 (d, 2H, J ) 6.4 Hz), 1.29 (d, 3H, J ) 6.8 Hz);
13C NMR (67.5 MHz, CDCl3) δ 171.6, 155.7 (d, J ) 83.0 Hz), 154.5,
135.3, 128.6, 128.5, 128.4, 128.3, 128.2, 71.1, 68.4, 65.3, 51.8, 37.3;
IR (neat) νmax 3056, 2980, 2953, 1795, 1734, 1499, 1217; CIMS (NH3
gas) 279, 278, 234, 162, 142, 127, 108, 91, 35; CIHRMS M + H+
Figure 4. Representation of the hydroxyl-assisted aziridination of the
silane reagents.
nitrenoid species. Our experiments seem to be consistent with
a substrate-directable aziridination process.31
In an attempt to characterize the silane bound to the copper
catalyst, crystals were grown from an acetonitrile solution of
1:1 Cu(I)OTf and silane 6d by diffusion of ether into the
acetonitrile solution at -20 °C. Subsequent X-ray crystal-
lographic analysis revealed the crystals consisted of [Cu-
(MeCN)4]OTf which were surface coated with the silane ligand.
This result allows us to conclude that the silyl alcohols (6a-e)
do not participate in binding to the copper(I) species in the solid
state. Although no mechanistic interpretation can be made at
this point, it is possible that the silane ligand binds to the Cu-
(III)-nitrenoid intermediate prior to aziridination of the olefin.
A plausible representation of the hydroxyl-assisted aziridination
transition state is shown in Figure 4 and involves simultaneous
coordination of the (E)-olefin and alcohol functionalities of the
silane reagent in a chair-like conformation.
(calculated for C15H20NO4): 278.1393, found: 278.1392; [R]23
+1.72° (c ) 1.5, CHCl3).
)
D
Conclusions
(2S,5R)-(E)-Methyl-5-(N-benzyloxy)amino-2-methyl-hex-3-
enoate (3b): 1H NMR (400 MHz, CDCl3) δ 7.35-7.27 (m, 5H), 5.73-
5.56 (m, 2H), 5.22 (s, br, 2H), 4.79 (br, 1H), 3.63 (s, 3H), 3.04 (m,
1H), 1.28 (d, 3H, J ) 7.6 Hz), 1.16 (d, 3H, J ) 6.8 Hz); 13C NMR
(67.5 MHz, CDCl3) δ 171.6, 155.7 (d, J ) 83.0 Hz), 154.4, 135.3,
128.8, 128.6, 128.5, 128.3, 128.2, 71.0, 68.3, 58.0, 51.8, 42.3, 17.0;
IR (neat) νmax 3035, 2981, 2953, 1795, 1736, 1499, 1217; CIMS (NH3
gas) 293, 292, 264, 248, 156 141, 108, 91; CIHRMS M + H+
In conclusion, asymmetric C-N bond constructions resulting
from nitronium tetrafluoroborate (NO2BF4) promoted electro-
philic nitrations of chiral (E)-crotylsilanes and by a PhIdNTs/
copper (I)-catalyzed enantiospecific aziridination of chiral (E)-
crotylsilanes provide a highly diastereo- and enantiospecific
method for the asymmetric synthesis of (E)-olefin dipeptide
isosteres and continue to expand the scope and utility of this
developing chiral allylsilane methodology. Further experiments
aimed at the characterization of the silane bound to the copper
species will be reported in due course.
(calculated for C16H22NO4): 292.1549, found: 292.1538; [R]23
-11.3° (c ) 1.4, CHCl3).
)
D
(2S,5R)-(E)-Methyl-5-(N-benzyloxy)amino-2-allyl-hex-3-enoate
(3c): 1H NMR (400 MHz, CDCl3) δ 7.36-7.28 (m, 5H), 5.68-5.60
(m, br, 2H), 5.22-5.16 (m, 4H), 5.02-4.96 (m, 1H), 4.78 (br, 1H)
3.63 (s, 3H), 3.02 (m, br, 1H), 2.40 (m, br, 1H), 2.19 (m, br, 1H), 1.28
(d, 3H, J ) 6.4 Hz); 13C NMR (67.5 MHz, CDCl3) δ 173.5, 155.7 (d,
J ) 83.0 Hz), 154.4, 135.3, 130.5, 128.8, 128.6, 128.3, 117.2, 71.1,
68.4, 51.8, 48.5, 36.5, 17.0; IR (neat) νmax 3046, 2981, 2953, 1795,
1734, 1642, 1588, 1499, 1224; CIMS (NH3 gas) 334, 318, 290, 248,
272, 269, 268, 258, 252, 227, 226, 225, 224, 182 167, 108; CIHRMS
M + H+ (calculated for C18H24NO4): 318.1706, found: 318.1713;
Experimental Section
General Procedures. 1H and 13C NMR spectra were recorded in
CDCl3 at 400 MHz and 67.5 MHz, respectively, unless specified
otherwise. Methylene chloride (CH2Cl2) and pyridine were distilled
from calcium hydride prior to use. The zinc dust utilized had a mesh
size of 325. Anhydrous acetonitrile (CH3CN) was purchased from
Aldrich and used as received. Anhydrous methanol (MeOH) was
purchased from J. T. Baker and used as received. Copper(I) trifluo-
romethanesulfonate benzene complex (Cu(I)OTf‚C6H6) was purchased
from Aldrich and used as received. All reactions were carried out in
oven-dried glassware under a dry argon atmosphere. Analytical thin
layer chromatography was performed on Whatman Reagent 0.25 mm
silica gel 60-Å plates.
Representative Experimental Procedure for the NO2BF4 Addi-
tions Illustrated for the Reaction of (3R)-(E)-Methyl-3-(dimeth-
ylphenyl)silyl-hex-4-enoate with Nitronium Tetrafluoroborate (2a).
A solution of 1a8 (6.3 g, 24.0 mmol) in 120 mL of CH2Cl2 (0.2 M)
was slowly added to a stirred suspension of nitronium tetrafluoroborate
(NO2BF4) (3.65 g, 26.4 mmol, 1.1 equiv) in 260 mL of CH2Cl2 (0.1
M) at -78 °C. The reaction mixture was allowed to warm to room
temperature over 10 h under vigorous stirring and subsequently
quenched with saturated aqueous NaHCO3 (20 mL). The mixture was
extracted with CH2Cl2 (3 × 25 mL), dried (MgSO4), and concentrated
in Vacuo to afford crude 2a as an oil. Purification on SiO2 (10% EtOAc/
PE) afforded 2a as a yellow oil, 2.0 g (50%, 4.1 g theoretical): 1H
[R]23 ) +22.1° (c ) 1.0, CHCl3).
D
(2S,5R)-(E)-Methyl-5-(N-benzyloxy)amino-2-methyl-hept-3-
enoate (3d): 1H NMR (400 MHz, CDCl3) 4d exists as a 2:1 mixture
of rotamers at room temperature. (The major rotamer is reported) δ
7.34-7.25 (m, 5H), 5.66 (dd, 1H, J ) 15.6 Hz, 8.0 Hz), 5.60 (s, 1H),
5.40 (dd, 1H, J ) 14.8 Hz, 8.0 Hz), 5.25-5.12 (m, 2H), 4.07 (s, 1H),
3.61 (s, 3H), 3.08-3.01 (m, 1H), 1.85-1.76 (m, 1H), 0.90 (d, 3H, J )
6.4 Hz), 0.86 (d, 3H, J ) 6.8 Hz), 0.81 (d, 3H, J ) 6.8 Hz); 13C NMR
(67.5 MHz, CDCl3) δ 174.6, 155.1 (d, J ) 145.6 Hz), 134.1, 128.8,
128.6, 128.5, 128.4, 128.3, 127.9, 70.2, 69.1, 51.8, 42.6, 29.4, 19.4,
17.0; IR (neat) νmax 3000, 1805, 1750, 1470, 1400, 1240; CIMS (NH3
gas) 109.1, 184.2, 214.2, 248.2, 274.2, 338.2; CIHRMS M + H+
(calculated for C18H26NO4) 320.1862, found 320.1889; [R]23D ) +1.0°
(c ) 0.1, CHCl3).
(2S,5S)-(E)-Methyl-5-(N-benzyloxy)amino-2-methyl-hex-3-
enoate (3e): 1H NMR (400 MHz, CDCl3) δ 7.33-7.30 (m, 5H), 5.72-
5.59 (m, 2H), 5.20 (s, br, 2H), 4.80 (br, 1H), 3.63 (s, 3H), 3.06 (m,
1H), 1.28 (d, 3H, J ) 6.8 Hz), 1.15 (m, 2H); 13C NMR (67.5 MHz,
CDCl3) δ 174.6, 155.9 (d, J ) 83.0 Hz), 154.6, 135.3, 128.8, 128.6,
128.5, 128.4, 128.3, 71.1, 68.3, 58.0, 51.9, 42.3, 17.0; IR (neat) νmax
3035, 2981, 2953, 1794, 1737, 1499, 1218; CIMS (NH3 gas) 308, 292,
288, 262, 246, 165, 156, 141, 108, 105, 91, 35; CIHRMS M + H+
(31) For an excellent and comprehensive review of substrate-directable
reactions, see: Hoveyda, A. H.; Evans, D. A.; Fu, G. C. Chem. ReV. 1993,
93, 1307-1370.
(32) Tedeschi, R. J.; Casey, A. W.; Clark, G. S.; Huckel, R. W.; Kindley,
L. M.; Russel, J. P. J. Org. Chem. 1963, 28, 1740-1743.

6046 J. Am. Chem. Soc., Vol. 119, No. 26, 1997
Masse et al.
(calculated for C16H21NO4): 292.1548, found: 292.1515; [R]23
+2.6° (c ) 0.38, CHCl3).
)
209.1, 189.1, 152.1, 135.1, 96.1; CIHRMS M + H+ (calculated for
D
C15H25SiO): 249.1675, found: 249.1693; [R]23 ) -5.5° (c ) 1.21,
D
CHCl3).
(2S,5S)-(E)-Methyl-5-(N-benzyloxy)amino-2-methoxy-hex-3-
enoate (3f): 1H NMR (400 MHz, C6D6) δ 7.08-6.93 (m, 5H), 6.05
(s, 1H), 5.72 (dd, 1H, J ) 15.6 Hz, 6.0 Hz), 4.94-4.74 (m, 4H), 4.21
(s, 1H), 3.19 (s, 3H), 3.00 (s, 3H), 1.02 (d, 3H, J ) 6.4 Hz); 13C NMR
(67.5 MHz, CDCl3) δ 170.5, 154.9, 135.2, 134.1, 128.7, 128.5, 128.4,
128.2, 127.8, 80.1, 71.0, 68.3, 57.1, 56.7, 52.2; IR (neat) νmax 3000,
1810, 1760, 1470, 1400, 1230; CIMS (NH3 gas) 91, 157, 262, 304,
338; CIHRMS M + H+ (calculated for C16H22NO5) 308.1498, found
(2R,3R)-(E)-3-(Dimethylphenylsilyl)-2-methoxy-hex-4-enol (6c):
1H NMR (400 MHz, CDCl3) δ 7.50-7.47 (m, 2H), 7.34-7.32 (m,
3H), 5.28-5.21 (m, 2H), 3.64-3.60 (dd, 1H, J1 ) 2.8 Hz, J2 ) 2.8
Hz), 3.51-3.46 (dd, 1H, J1 ) 5.2 Hz, J2 ) 5.2 Hz), 3.20 (s, 3H), 3.19-
3.17 (m, 1H), 2.24 (t, 1H), 1.85 (br, 1H), 1.63 (d, 3H, J ) 5.2 Hz),
0.32 (s, 3H), 0.28 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 138.1, 134.0,
128.8, 127.6, 127.5, 126.0, 82.3, 61.7, 56.3, 35.3, 18.1, -3.0, -3.2; IR
(neat) νmax 3432, 3014, 2960, 2094, 1653, 1246; CIMS (NH3 gas) 264.2,
236.1, 181.1, 152.1, 135.1, 98.1; CIHRMS M + H+ (calculated for
C15H25SiO2): 265.1546, found: 265.1594; [R]23D ) +25.4° (c ) 1.14,
CHCl3).
308.1505; [R]23 ) +8.3° (c ) 0.40, CHCl3).
D
Experimental Procedure for Peptide Coupling: N-((E)-(5R)-5-
(N-benzyloxycarbonyl)amino-3-hexenoyl)-L-leucine Methyl Ester
(5). Acid 4a (0.014 g, 0.053 mmol) was dissolved in 1 mL of CH2Cl2
(0.05 M) and cooled to 0 °C. L-Leucine methyl ester (0.011 g, 0.06
mmol, 1.1 equiv) was added to the reaction mixture followed by HOBT
(0.72 mg, 0.0053 mmol, 0.1 equiv) and DCC (0.013 g, 0.06 mmol, 1.1
equiv). The resulting white suspension was allowed to warm to room
temperature under stirring over 24 h. The suspension was then diluted
with CH2Cl2, recooled to 0 °C, filtered through Celite, and concentrated
in Vacuo. Purification on SiO2 (70% EtOAc/PE) afforded 5 as a white
solid, 0.015 g (70%): 1H NMR (400 MHz, CDCl3) δ 7.34-7.29 (m,
5H); 6.61 (br, 1H); 6.09 (d, 1H, J ) 8.4 Hz); 5.73 (m, 2H); 5.17 (s,
2H); 4.72 (m, 1H); 4.58 (m , 1H); 3.70 (s, 3H); 3.43 (m, 1H); 2.96 (d,
2H, J ) 4.8 Hz); 1.70-1.41 (m, 2H); 1.31 (d, 3H, J ) 7.6 Hz); 1.20-
1.02 (m, 1H); 0.90 (d, 6H, J ) 4.0 Hz); 13C NMR (67.5 MHz, CDCl3)
δ 173.8, 170.6, 157.2, 135.9, 134.1, 128.5, 128.3, 128.1, 124.9, 67.9,
52.4, 41.4, 39.6, 33.8, 24.9, 24.8, 22.8, 21.9, 16.4; IR (KBr) νmax 3306,
2955, 2929, 1744, 1717, 1651; CIMS (NH3 gas) 391, 386, 364, 324,
308, 240, 181, 156, 141, 108, 92, 91; CIHRMS M+ (calculated for
C21H30N2O4): 374.2206, found: 374.2217; [R]23D ) +22.5° (c ) 0.36,
CHCl3).
(2R,3R)-(E)-3-(Dimethylphenylsilyl)-2-methylhex-4-enol (6d): 1H
NMR (400 MHz, CDCl3) δ 7.49-7.47 (m, 2H), 7.38-7.31 (m, 3H),
5.29-5.26 (m, 2H), 3.52-3.48 (dd, 1H, J1 ) 4.0 Hz, J2 ) 4.8 Hz),
3.38-3.34 (dd, 1H, J1 ) 6.0 Hz, J2 ) 6.0 Hz), 1.77-1.65 (m, 2H),
1.64 (d, 3H, J ) 4.4 Hz), 0.86 (d, 3H, J ) 6.0 Hz), 0.29 (s, 3H), 0.26
(s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 139.0, 133.9, 130.1, 128.8,
127.6, 124.9, 67.5, 37.3, 36.5, 18.0, 0.0, -2.4, -3.5; IR (neat) νmax
3371, 2958, 2930, 1726, 1249; CIMS (NH3 gas) 247.1, 209.1, 171.1,
152.1, 135.1, 96.1; CIHRMS M + H+ (calculated for C15H25SiO):
249.1675, found: 249.1700; [R]23 ) -2.75° (c ) 1.20, CHCl3).
D
(3S)-(E)-3-(Dimethylphenylsilyl)-6-methylhept-4-enol (6e): 1H
NMR (400 MHz, CDCl3) δ 7.47-7.45 (m, 2H), 7.34-7.31 (m, 3H),
5.27-5.12 (m, 2H), 3.63-3.51 (m, 2H), 2.24-2.19 (m, 1H), 1.77-
1.61 (m, 2H), 1.56-1.50 (m, 1H), 1.36 (br, 1H), 0.92 (d, 6H, J ) 6.4
Hz), 0.26 (s, 3H), 0.24 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 137.5,
137.0, 134.0, 128.9, 127.6, 127.1, 62.9, 31.7, 31.3, 29.2, 22.9, 22.8,
-4.6, -5.4; IR (neat) νmax 3389, 2958, 2929, 2082, 1653, 1248; CIMS
(NH3 gas) 259.2, 217.1, 189.1, 185.1, 152.1, 135.0; CIHRMS M +
H+ (calculated for C16H27SiO): 263.1831, found: 263.1807; [R]23
-9.80° (c ) 1.53, CHCl3).
)
Modified Experimental Procedure for the Synthesis of (N-(p-
Toluenesulfonyl)imino)phenyliodinane. A solution of p-toluene-
sulfonamide (1.71 g, 10.0 mmol) and potassium hydroxide (1.40 g,
25.0 mmol, 2.5 equiv) in 40 mL of anhydrous MeOH (0.25 M) was
cooled to -10 °C. To this solution was added (diacetoxy)iodobenzene
(3.87 g, 12.0 mmol, 1.2 equiv), and the reaction mixture was allowed
to warm to room temperature over 3 h under vigorous stirring. After
3 h, the reaction mixture was diluted with 150 mL of distilled water to
give a precipitate which was filtered and washed with Et2O (20 mL)
to yield yellow crystals. The yellow crystals (2.80 g, 75%) were dried
under reduced pressure. No further purification steps were performed
D
Note: Silane 6e is prepared from an analogous hydrosilation/Lipase
resolution/ortho-ester Claisen methodology as reported in ref 8 starting
from racemic 4-methyl-1-pentyn-3-ol.32
Experimental Procedure for the Synthesis of (2R, 3R)-(E)-1-tert-
Butyl-diphenylsilyloxy-3-(dimethylphenylsilyl)-2-methoxyhex-4-
ene (6f). A solution of 6c (0.2 g, 0.76 mmol) in 1.5 mL of DMF (0.5
M) was cooled to 0 °C. To this solution was added imidazole (0.16 g,
2.28 mmol, 3.0 equiv) followed by tert-butyldiphenylsilyl chloride (0.21
g, 0.76 mmol, 1.0 equiv). The reaction mixture was allowed to warm
to room temperature over 12 h of vigorous stirring and subsequently
quenched with H2O. The mixture was extracted with Et2O (3 × 25
mL), washed with saturated aqueous NaCl (3 × 25 mL), dried (MgSO4),
1
as the iodinane exhibited an acceptable H NMR and FTIR spectral
properties.23
Representative Experimental Procedure for the Reduction of the
Chiral (E)-Crotylsilyl Esters. (3R)-(E)-3-(Dimethylphenylsilyl)-4-
hexenol (6a). A solution of 1a8 (2.0 g, 8.06 mmol) in 30 mL of
anhydrous Et2O (0.25 M) was cooled to 0 °C. To this solution was
added lithium aluminum hydride (LAH) (0.31 g, 8.06 mmol, 1.0 equiv)
and the reaction mixture was stirred vigorously at 0 °C for 0.5 h. The
reaction mixture was quenched dropwise with saturated aqueous NH4-
Cl (30 mL), and the mixture was extracted with Et2O (2 × 20 mL),
dried (MgSO4), and concentrated in Vacuo. Purification on SiO2 (5%
EtOAc/PE f 20% EtOAc/PE gradient elution) afforded 6a as a pale
yellow oil, 1.85 g (98%, 1.89 g theoretical); 1H NMR (400 MHz,
CDCl3) δ 7.47-7.45 (m, 2H), 7.34-7.31 (m, 3H), 5.31-5.18 (m, 2H),
3.64-3.59 (m, 1H), 3.55-3.49 (m, 1H), 1.80-1.74 (m, 1H), 1.68-
1.60 (m, 1H), 1.63 (d, 3H, J ) 5.6 Hz), 1.55-1.45 (m, 1H), 1.34 (br,
1H), 0.25 (s, 3H), 0.24 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 137.6,
134.0, 131.2, 128.9, 127.6, 123.8, 62.7, 31.7, 29.2, 18.0, -4.5, -5.4;
IR (neat) νmax 3446, 3069, 2959, 1952, 1653, 1249; CIMS (NH3 gas)
250.1, 233.1, 205.0, 189.1, 152.0, 135.0, 91.1; CIHRMS M+ (calculated
and concentrated in Vacuo. Purification on SiO
2 (PE) afforded 6f as a
colorless oil, 0.372 g (97%, 0.382 g theoretical): 1H NMR (400 MHz,
CDCl
1H, J1
3) δ 7.72-7.30 (m, 15H), 5.12-5.10 (m, 2H), 3.70-3.67 (dd,
) 2.4 Hz, J2 ) 2.4 Hz), 3.59-3.55 (dd, 1H, J1 ) 5.6 Hz, J2 )
5.2 Hz), 3.28 (s, 3H), 3.22-3.21 (m, 1H), 2.09 (t, 1H), 1.53 (d, 3H, J
) 4.0 Hz), 1.01 (s, 9H), 0.28 (s, 3H), 0.22 (s, 3H); 13C NMR (67.5
MHz, CDCl3) δ 138.9, 135.7, 135.6, 134.8, 134.2, 133.7, 129.7, 129.5,
128.9, 128.5, 127.7, 127.6, 127.3, 124.7, 83.8, 65.7, 57.9, 36.5, 31.6,
26.8, 22.7, 19.2, 18.1, 14.1, -2.7; IR (neat) νmax 3441, 3071, 2931, 2109,
1653, 1245; CIMS (NH3 gas) 555.2, 520.2, 391.2, 377.1, 274.1, 196.2;
CIHRMS M + H+
(calculated for C31H43Si2O2): 503.2801, found:
) -8.3° (c ) 0.68, CHCl3).
503.2829; [R]23
D
Experimental Procedure for the Synthesis of (2R,3R)-(E)-1-
Methoxy-3-(dimethylphenylsilyl)-2-methoxyhex-4-ene (6g). A solu-
tion of 6c (0.2 g, 0.76 mmol) in 7.6 mL of CH2Cl2 (0.1 M) was cooled
to 0 °C. To this solution was added proton sponge (0.49 g, 2.28 mmol,
3.0 equiv) and trimethyloxonium tetrafluoroborate (0.34 g, 2.28 mmol,
3.0 equiv). The reaction mixture was allowed to warm to room
temperature over 4 h of vigorous stirring and subsequently diluted with
1 N NaHSO4 (20 mL). The mixture was filtered through Celite and
washed thoroughly with Et2O. The filtrate was extracted with Et2O (3
× 15 mL), dried (MgSO4), and concentrated in Vacuo. Purification
on SiO2 (5% EtOAc/PE) afforded 6g as a yellow oil, 0.148 g (70%,
0.212 g theoretical): 1H NMR (400 MHz, CDCl3) δ 7.49-7.47 (m,
2H), 7.32-7.30 (m, 3H), 5.22-5.20 (m, 2H), 3.43-3.40 (m, 1H), 3.30-
3.26 (m, 2H), 3.25 (s, 3H), 3.24 (s, 3H), 2.12 (t, 1H), 1.62 (d, 3H, J )
for C14H22SiO): 234.1440, found: 234.1423; [R]23 ) +12.57° (c )
D
1.40, CHCl3).
(2R,3S)-(E)-3-(Dimethylphenylsilyl)-2-methylhex-4-enol (6b): 1H
NMR (400 MHz, CDCl3) δ 7.50-7.47 (m, 2H), 7.35-7.32 (m, 3H),
5.38-5.27 (m, 2H), 3.37-3.32 (m, 2H), 1.98-1.95 (m, 1H), 1.80-
1.78 (m, 1H), 1.67 (d, 3H, J ) 4.8 Hz), 0.82 (d, 3H, J ) 6.8 Hz), 0.31
(s, 3H), 0.27 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ 138.5, 133.9,
128.8, 127.6, 127.4, 125.6, 68.1, 35.5, 34.7, 18.1, 14.0, -3.3, -3.9; IR
(neat) νmax 3382, 3069, 2960, 1952, 1653, 1248; CIMS (NH3 gas) 249.1,

C-N Bond Constructions Via Crotylsilane Addition Reactions
J. Am. Chem. Soc., Vol. 119, No. 26, 1997 6047
2.0 Hz), 0.31 (s, 3H), 0.26 (s, 3H); 13C NMR (67.5 MHz, CDCl3) δ
138.2, 134.2, 128.7, 128.3, 127.4, 125.3, 81.4, 74.1, 58.9, 57.0, 36.3,
18.1, -2.8, -2.9; IR (neat) νmax 3427, 3070, 2958, 2110, 1712, 1251;
CIMS (NH3 gas) 301.1, 296.1, 233.1, 193.1, 175.1, 161.1; CIHRMS
M+NH4+ (calculated for C16H30SiNO2): 296.2046, found: 296.2081;
97.1; CIHRMS M+ (calculated for C14H22NO4S): 300.1269, found:
300.1265; [R]23 ) -3.4° (c ) 0.35, CHCl3).
D
(2S,5S)-(E)-5-(N-(p-toluenesulfonyl)amino)-2-methylhex-3-enol (7d):
1H NMR (400 MHz, CDCl3) δ 7.72 (d, 2H, J ) 8.4 Hz), 7.28 (d, 2H,
J ) 8.0 Hz), 5.33-5.26 (m, 2H), 4.52 (d, 1H, J ) 6.0 Hz), 3.78-3.77
(m, 1H), 3.42-3.37 (m, 1H), 3.30-3.24 (m, 1H), 2.40 (s, 3H), 2.21-
2.20 (m, 1H), 1.63 (br, 1H), 1.15 (d, 3H, J ) 6.8 Hz), 0.85 (d, 3H, J
) 6.8 Hz); 13C NMR (67.5 MHz, CDCl3) δ 143.5, 138.5, 134.0, 133.5,
129.6, 127.2, 67.1, 51.6, 39.3, 22.1, 21.5, 16.0; IR (neat) νmax 3430,
2094, 1643, 1600, 1322, 1158; CIMS (NH3 gas) 283.1, 219.1, 181.1,
153.1, 151.1, 135.1, 82.9; CIHRMS M+ (calculated for C14H21NO3S):
[R]23 ) +14.6 °C (c ) 0.51, CHCl3).
D
Representative Experimental Procedure for the Aziridination of
the Chiral (E)-Crotylsilyl Alcohols. (5R)-(E)-(N-(p-Toluenesulfo-
nyl)amino)hex-3-enol (7a). To a solution of 6a (0.136 g, 0.581 mmol)
in 1.5 mL of CH3CN (0.4 M) was added (N-(p-toluenesulfonyl)imino)-
phenyliodinane (PhIdNTs) (0.217 g, 0.581 mmol, 1.0 equiv), and the
heterogeneous mixture was stirred under argon. To this heterogeneous
mixture was added Cu(I)OTf‚C6H6 (0.029 g, 0.0581 mmol, 10 mol%).
The reaction mixture was stirred at room temperature for 1 h to yield
a homogeneous yellow solution. Direct purification of reaction mixture
on SiO2 (PE f 50% EtOAc/PE gradient elution) afforded 7a as a yellow
oil, 0.048 g (31%, 0.156 g theoretical, 65% based on recovered
silane): 1H NMR (400 MHz, CDCl3) δ 7.71 (d, 2H, J ) 8.8 Hz), 7.27
(d, 2H, J ) 8.0 Hz), 5.41-5.35 (m, 2H), 4.48 (br, 1H), 3.82-3.80 (m,
1H), 3.55-3.51 (m, 2H), 2.40 (s, 3H), 2.16-2.11 (m, 2H), 1.14 (d,
3H, J ) 6.4 Hz); 13C NMR (67.5 MHz, CDCl3) δ 143.3, 137.8, 134.0,
129.6, 127.7, 127.1, 61.5, 51.3, 35.3, 21.9, 21.5; IR (neat) νmax 3448,
2109, 1684, 1653, 1635, 1319, 1159; CIMS (NH3 gas) 287.1, 270.1,
130.1, 81.1; CIHRMS M+ (calculated for C13H19NO3S): 270.1086,
283.1242, found: 283.1608; [R]23 ) -32.4o (c ) 0.25, CHCl3).
D
(5S)-(E)-5-(N-(p-Toluenesulfonyl)amino)-6-methylhept-3-enol (7e):
1H NMR (400 MHz, CDCl3) δ 7.72 (d, 2H, J ) 8.0 Hz), 7.25 (d, 2H,
J ) 8.0 Hz), 5.37-5.35 (m, 1H), 5.25-5.20 (m, 1H), 4.68 (d, 1H, J )
7.6 Hz), 3.82-3.80 (m, 1H), 3.59-3.43 (m, 2H), 2.39 (s, 3H), 2.20-
2.11 (m, 1H), 1.73-1.64 (m, 2H), 0.81 (d, 3H, J ) 1.6 Hz), 0.79 (d,
3H, J ) 2.0 Hz); 13C NMR (67.5 MHz, CDCl3) δ 143.2, 130.3, 129.4,
128.9, 127.2, 61.8, 56.2, 33.0, 31.2, 21.5, 18.5, 18.3, 17.9; IR (neat)
νmax 3431, 2961, 2927, 2089, 1646, 1321, 1305, 1159; CIMS (NH3
gas) 315.1, 298.1, 255.0, 254.0, 224.0, 189.0, 127.1; CIHRMS M +
H+ (calculated for C15H24NO3S): 298.1478, found: 298.1445; [R]23
) -8.5° (c ) 0.71, CHCl3).
D
(5R)-(E)-Methyl-5-(N-(p-toluenesulfonyl)amino)hex-3-enoate
(8): 1H NMR (400 MHz, CDCl3) δ 7.78 (d, 2H, J ) 8.0 Hz), 7.25 (d,
2H, J ) 8.0 Hz), 6.60 (br, 1H), 5.70-5.30 (m, 1H), 5.45-5.30 (m,
1H), 4.38 (m, 1H), 3.65 (s, 3H), 2.80 (d, 2H, J ) 6.4 Hz), 2.40 (s,
3H), 1.14 (d, 3H, J ) 6.4 Hz); 13C NMR (67.5 MHz, CDCl3) δ 144.4,
131.2, 129.5, 129.3, 124.4, 58.3, 51.8, 37.3, 21.6, 18.0; IR (neat) νmax
3369, 1738, 1597, 1348, 1168, 1090; CIMS (NH3 gas) 298.2, 282.2,
142.1, 127.1, 110.1, 91.1; CIHRMS M + H+ (calculated for C14H20-
found: 270.1185; [R]23 ) +9.2° (c ) 0.19, CHCl3).
D
(2S,5R)-(E)-5-(N-(p-toluenesulfonyl)amino)-2-methylhex-3-enol (7b):
1H NMR (400 MHz, CDCl3) δ 7.72 (d, 2H, J ) 8.0 Hz), 7.27 (d, 2H,
J ) 8.0 Hz), 5.33-5.31 (m, 2H), 4.56 (d, 1H, J ) 7.2 Hz), 3.86-3.82
(m, 1H), 3.41-3.37 (m, 1H), 3.30-3.25 (m, 1H), 2.40 (s, 3H), 2.21-
2.18 (m, 1H), 1.63 (br, 1H), 1.15 (d, 3H, J ) 6.4 Hz), 0.86 (d, 3H, J
) 6.8 Hz); 13C NMR (67.5 MHz, CDCl3) δ 143.3, 138.0, 133.5, 132.1,
129.6, 127.2, 67.0, 51.2, 38.9, 22.1, 21.5, 16.0; IR (neat) νmax 3456,
2090, 1653, 1600, 1319; CIMS (NH3 gas) 284.1, 213.0, 198.0, 189.0,
155.0, 95.1; CIHRMS M + H+ (calculated for C14H22NO3S): 284.1320,
NO4S): 298.1113, found: 298.1093; [R]23 ) -29.5° (c ) 0.21,
D
CHCl3).
found: 284.1340; [R]23 ) +9.0° (c ) 0.48, CHCl3).
D
Acknowledgment. This work has been financially supported
by the National Institutes of Health (RO1CA56304).
(2S,5S)-(E)-5-(N-(p-Toluenesulfonyl)amino)-2-methoxyhex-3-
1
enol (7c): H NMR (400 MHz, CDCl3) δ 7.71 (d, 2H, J ) 8.0 Hz),
7.27 (d, 2H, J ) 8.4 Hz), 5.58-5.53 (dd, 1H, J1 ) 6.0 Hz, J2 ) 6.4
Hz), 5.32-5.26 (dd, 1H, J1 ) 7.6 Hz, J2 ) 7.2 Hz), 4.63 (d, 1H, J )
7.2 Hz), 3.91-3.87 (m, 1H), 3.58-3.53 (m, 1H), 3.39-3.36 (m, 2H),
Supporting Information Available: Stereochemical as-
signments as well as spectral data for all intermediates and final
products (52 pages). See any current masthead page for ordering
and Internet access instructions.
3.19 (s, 3H), 2.40 (s, 3H), 2.12 (br, 1H), 1.17 (d, 3H, J ) 6.4 Hz); 13
C
NMR (67.5 MHz, CDCl3) δ 143.5, 137.8, 135.8, 129.7, 127.6, 127.1,
81.9, 64.9, 56.5, 50.8, 21.9, 21.5; IR (neat) νmax 3471, 2088, 1718,
1636, 1326, 1160; CIMS (NH3 gas) 300.1, 268.1, 250.1, 189.1, 129.1,
JA964364W