Asymmetric induction of conduritols via AAA reactions: Synthesis of the aminocyclohexitol of hygromycin A

Article abstract of DOI:10.1002/1521-3765(20010417)7:8<1619::AID-CHEM16190>3.0.CO;2-4

Two synthetic routes towards the construction of the aminocyclohexitol moiety of hygromycin A have been developed based on palladium-catalyzed asymmetric alkylation of conduritol derivatives. A protocol has been established whereby this biologically relev

Full text of DOI:10.1002/1521-3765(20010417)7:8<1619::AID-CHEM16190>3.0.CO;2-4

FULL PAPER
Asymmetric Induction of Conduritols via AAA Reactions:
Synthesis of the Aminocyclohexitol of Hygromycin A
Barry M. Trost,* Joseph Dudash, Jr., and Erik J. Hembre[a]
Abstract: Two synthetic routes towards
the construction of the aminocyclohex-
itol moiety of hygromycin A have been
developed based on palladium-cata-
lyzed asymmetric alkylation of condur-
itol derivatives. A protocol has been
established whereby this biologically
relevant molecule is formed from ben-
zoquinone. A conduritol A derivative is
synthesized in eight steps from benzo-
quinone and is then subjected to the
palladium reaction. From this flexible
intermediate, four epimers of the ami-
nocyclitol, including the natural one, can
be obtained with complete stereoselec-
tivity. Racemic conduritol B derivatives
are available in four steps from benzo-
quinone, and these are then made enan-
tiomerically pure by a palladium-cata-
lyzed dynamic kinetic resolution. From
the chiral conduritol B, the aminocycli-
tol is available in six steps. Excellent
levels of enantio- and diastereoselectiv-
ity highlight these strategies.
Keywords: amino alcohols ´ asym-
metric catalysis ´ asymmetric syn-
thesis ´ palladium ´ total synthesis
enable a broader investigation of the utility of such com-
pounds. Previous work in these laboratories developed a
general strategy for the asymmetric synthesis of the amino-
cyclopentitols 3 and 4[4] which was further improved by our
ability to utilize the asymmetric allylic alkylation (AAA)
reaction[5] to desymmetrize meso-diols with >95% ee.[6] In
this paper, we develop two general strategies to amino-
cyclohexitols based upon the utilization of the AAA reaction
with achiral or racemic conduritols.
Introduction
Aminocyclitols represent a continuing and growing class of
important compounds for biological function. The amino-
cyclohexitols represented by streptomycin [containing strept-
amine moiety (1)] and kanamycin [containing 2-deoxy-strept-
amine moiety (2)] are significant antibiotics.[1] Recently, the
aminocyclopentitols represented by allosamidin [containing
allosamizoline (3)][2] and mannostatin (4)[3] illustrate the five-
Conduritols, the various diastereomers of cyclohexene
tetraols, have been of interest both because of their biological
activity and of their utility as building blocks.[7] For the latter
to be most useful, these compounds must be accessible
asymmetrically. Most approaches rely on the use of chiral pool
starting materials (typically sugars) or on resolution of
racemic material.[8] In the case of conduritols A and D which
are meso-compounds, enzymatic desymmetrization has been
successful.[9] A fascinating strategy employs the asymmetric
microbial dihydroxylation of substituted aromatic rings.[10] An
advantage of the AAA reaction is the ability to easily access
either enantiomeric series simply by change of ligand. Thus,
the range of applications should be significantly enhanced by
such ease of defining the absolute configuration.
Hygromycin A (5) is a fermentation derived natural
product first isolated from Streptomyces hygroscopicus[11] in
1953. The mode of action of hygromycin A is peptidyltrans-
ferase inhibition and the compound shares the same binding
site on the ribosome as chloramphenicol.[12] It has been shown
that inhibition occurs specifically by interfering with peptide
bond formation, resembling chloramphenicol, and is closely
related as 5 inhibits the effects of the latter. It has been
NHR
NHR
HO
O (Sugar)
OH
O (Sugar)
OH
RNH
RNH
O (Sugar)
O (Sugar)
1
2
HO
RO
SCH3
NH2
HO
O
N
CH3
N
HO
HO
OH
CH3
3
4
membered ring analogues also have promise for biological
applications. Developing general synthetic strategies to these
families of compounds constitutes an important objective to
[a] Prof. B. M. Trost, Dr. J. Dudash, Jr., Dr. E. J. Hembre
Department of Chemistry
Stanford University
Stanford, CA 94305-5080 (USA)
Fax : (‡1)650-725-0002
E-mail: bmtrost@leland.stanford.edu
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1619 $ 17.50+.50/0
1619

B. M. Trost et al.
FULL PAPER
OH
HO
OH
O
O
O
H2N
HO
N3
OH
OH
O
O
O
O
O
N
O
O
H
HO
HO
OH
OH
6
7
Hygromycin A
5
OCO2CH3
O
OR
O
O
reported that 5 has a relatively broad spectrum of activity
against Gram-positive and Gram-negative bacteria.[11a,b] Re-
cently, 5 has attracted renewed interest due to the discovery of
its hemaglutination inactivation activity[13] as well as its high
antitreponemal activity,[14] especially as an effective agent for
the control of swine dysentery, a mucohemorrhagic disease of
economic importance to swine producers. Furthermore the
antibiotic also demonstrates efficacy in the treatment of an
induced dysentery infection model of swine at a level of 5 ±
20 g per ton feed.[11b]
The structural studies of 5, using a degradation method by
Mann[15] and careful spectral analyses by Kakinuma,[16]
revealed that it has quite a unique structure among the
aminocyclohexitol antibiotic family. The cyclohexitol struc-
ture of 5, assigned as 1l-4,5-O-methylene-2-amino-2-deoxy-
neo-inositol (6), is much different from that of common and
general cyclitols found in other aminocyclohexitol antibiot-
ics.[1]
O
OCO2CH3
N3
9
8
Scheme 1. Retrosynthetic analysis.
The ability to discriminate between two prochiral centers of
a meso-compound in an allylic alkylation reaction has been
made possible by the use of C2-symmetric ligands chelated to
a palladium(0) source developed in these laboratories.[6] The
mechanism for the allylic alkylation of a cyclic substrate is
shown in Scheme 2. The enantiodiscriminating step occurs
L*
L*
OR
RO
Nu
RO
Pd
In spite of the unique structure and interesting biological
activity of 5, only a few reports have appeared on the synthesis
of the structural components of 5, and only one report of its
total synthesis has been described. The total synthesis utilized
d-glucose as the starting material for both the furanoside and
cyclohexitol moieties of 5.[17] Overall, twenty steps were
necessary for the preparation of the cyclitol moiety, including
a separation of a 3:2 mixture of diastereomers at a late stage in
the synthesis due to an unselective dihydroxylation. Arjona
and co-workers reported an improvement of a similiar
dihydroxylation step, achieving a diastereoselectivity of 6:1
in favor of the natural epimer.[8e] We therefore chose the
aminocyclohexitol moiety of hygromycin 6 as a target to
develop asymmetric syntheses to such a family of compounds.
Decomplexation
L*
Complexation
L*
L*
L*
Pd
Pd
OR
RO
Nu-
Nu
RO
L*
L*
Selective
Ionization
Pd
Addition
RO
Scheme 2. Conceptualization of the palladium-catalyzed desymmetriza-
tion.
when the catalyst selectively ionizes one of the two possible
enantiotopic allylic esters, depending on the chirality of the
ligand. The energy difference between the two diastereomeric
transition states arising from the steric interactions between
the leaving group and the phenyl rings on phosphine accounts
for the enantioselectivity.
The conduritol A derivative 10, previously available to us
from our work directed toward the total synthesis of
pancratistatin,[18] was a logical choice of precursor to acetal
9 by simple exchange of the acetonide for the methylenedioxy
unit. Surprisingly, all attempts to convert the diol 11 to acetal 9
failed in contrast to the facility of acetonide formation. The
only identifiable formaldehyde adduct is the diadduct 12
which was obtained only in 18% yield in the presence of
Montmorillonite clay (Scheme 3). A more efficient synthesis
would install the methylenedioxy unit early, in lieu of the
acetonide (see Scheme 4). Thus, the diol 13a was dihydroxy-
lated to give tetraol 14a as previously described.[20]
Results and Discussion
Plan A: Desymmetrization strategy
In our first approach, we envisioned the aminocyclohexitol 6
to be derived from the suitably protected hydroxy azide 7 by
trans-dihydroxylation of the cyclohexene (Scheme 1). This
compound would be obtained from the palladium-catalyzed
desymmetrization of the dicarbonate derivative of condur-
itol A 9 using an azide nucleophile to give 8 followed by [3,3]-
sigmatropic rearrangement of the allylic azide.[18, 19] Advan-
tages of this approach include the ability to set the relative
stereochemistry of four chiral centers before asymmetry is
induced, to obtain either enantiomer from the palladium
reaction, to access different regioisomeric aminocyclohexitols,
and to access a variety of such compounds by differential
functionalization of the double bond.
1620
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1620 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 8

Aminocyclohexitol Synthesis
1619±1629
reaction could be performed, the need to remove the
carbonate might be avoided. Taking advantage of the known
charge accelerated retro-Diels ± Alder reaction which may
occur even at room temperature,[21] the diol 13c was
converted in completely analogous fashion through 13d ±
14c (51% for two steps) and 15c into the desired precursor
15d (68% for two steps). Oxy-anion accelerated retro-Diels ±
Alder reaction was thwarted by migration of the carbonate to
the bridgehead alcohol under base conditions.
CH3O2CO
O
CH3O2CO
OH
OH
CH3OH, TsOH
CH2Cl2
62%
O
CH3O2CO
CH3O2CO
10
11
CH3O2CO
O
O
O
(CH2O)4
With the requisite substrate 9 in hand, the palladium-
catalyzed substitution with azide was pursued as shown in
Scheme 5. The initial product 8 was directly hydrolyzed in
CH3O2CO
12
Scheme 3. Formaldehyde adduct side reaction.
H
H
O
O
(17)
OR
OR
H
H
NH HN
OH
OCO2CH3
O
OH
b
PPh2 Ph2P
c
OR
OR
9
TMSN3
(dba)3Pd2·CHCl3
O
X
X
N3
RT
13a: R = H, X = H
14a: R = H, X = H
8: R = CO2CH3
a
a
OR
13b: R = CO2CH3, X = H
13c: R = H, X = OBOM
14b: R = CO2CH3, X = H
K2CO3
N3
O
O
CH3OH
50o
14c: R = CO2CH3, X = OBOM
13d: R = CO2CH3, X = OBOM
7a: R = H
7b: R = TBDMS
7c: R = Ac
OR
O
H
OR
H
O
O
Scheme 5. Palladium-catalyzed substitution.
f
O
OR
OR
16: R = H
X
a
basic methanol at 508C to give hydroxy azide 7a in 70% yield.
Using the cyclohexyl ligand 17, chiral HPLC analysis indicated
the ee of 7a to be 93%. Silylation of the alcohol proceeded
readily under standard conditions to give silyl ether 7b in
73% yield.
9: R = CO2CH3
15a: R = CO2CH3, X = H
15b: R = H, X = H
d
15c: R = CO2CH3, X = OBOM
e
The next stage requires a diastereoselective trans dihydrox-
ylation of the double bond. The first strategy envisioned a
diastereoselective epoxidation from the b face and a regiose-
lective ring opening being dictated by the electronics of the
neighboring substituents. Epoxidation of 7a with trifluoroper-
acetic acid[22] proceeded slowly but diastereoselectively to
form a single epoxide assigned as b, that is 18a (Scheme 6).
15d: R = CO2CH3, X = OH
Scheme 4. Synthesis of conduritol A substrate 9: a) n-C4H9Li, ClCO2,
CH3, THF, 08C. b) OsO4, NMO, THF, t-C4H9OH, H2O or CH2Cl2, H2O,
RT. c) CH3OCH2OCH3, TMSOSO2CF3, 2,6-lutidine, 08C, RT. d) K2CO3,
CH3OH, RT. e) TsOH, CH3OH, RT. f) FVT, 5008C.
While chemoselective formation of the acetonide of 14a
occurred nearly quantitatively, formation of the methylene-
dioxy derivative 15b simply failed. On the other hand, the
differentiated tetraol derivative 14b, obtained from cis-
dihydroxylation of 13b in 69% yield which, in turn derived
from diol 13a in 90% yield, smoothly gave the desired
methylenedioxy compound 15a in 91% yield. Unfortunately,
attempted direct thermolysis of 15a to form 9 failed.
However, the diol 15b readily obtained by simple base
hydrolysis in 82% yield smoothly underwent retro-Diels ±
Alder reaction to produce diol 16 in 73% yield which is
converted to the required dicarbonate in 81% yield under
standard conditions.
OR
OR
Ha
OR
O
N3
N3
N3
O
O
O
O
H
O
7a
O2NO
RO
H
61%
43%
O2NO
Hb
Hc
O
OR
18a: R = H
18b: R = Ac
19a: R = H
19b: R = Ac
OH
N3
H
H
a HO
N3
HO
O
O
Hb
HO
O
H
H
O
H
O2NO
ONO2
The decomposition of the dicarbonate 15a under FVT
conditions was envisioned to derive from the high temper-
atures required. If a low temperature retro-Diels ± Alder
20
Scheme 6. Diastereoselective epoxidation of 7a.
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1621 $ 17.50+.50/0
1621

B. M. Trost et al.
FULL PAPER
On the other hand, the acetate 7c failed to react under the
same conditions. This observation suggests hydroxy partic-
ipation in the epoxidation which would direct the stereo-
chemistry to give the depicted product. When the correspond-
ing acetate 18b was subjected to conditions to open the
epoxide, acetate removal occurred first with no further
reaction observed. The hydroxy epoxide 19a proved resistant
to ring opening with carboxylate nucleophiles. On the other
hand, tetra-n-butylammonium nitrate did lead to ring opening
in the presence of boron trifluoride/etherate. The proton, Ha,
adjacent to the nitrate appeared as a double doublet with J ˆ
9.1 and 8.5 Hz showing coupling to the proton, Hb, adjacent to
azide at d ˆ 3.20. The large vicinal couplings to both adjacent
hydrogens also indicate a trans,trans relation as in 19 but not
depicted. Ultimate conversion of 23 to the aminocyclohexitol
of hygromycin confirms this assignment.
The dramatic difference in facial selectivity by choice of
nitrogen substituent cannot be understood on the basis of
conformational analysis. Steric effects would appear to be
dominated by the methylenedioxy group which would favor b
attack as observed with 21. In this sense, the unusual
diastereoselectivity derives from the reaction of the azide 7.
In this case, an orbital distortion argument would seem
appropriate,[24] that is the azide group would distort the p-
system to favor electrophilic attack anti to itself as observed.
The dramatic difference makes further investigation of this
phenomenon highly worthwhile.
The synthesis of the diastereomer of the aminocyclohexitol
of hygromycin requires inversion at C-6 of diol 23. The
completion of the synthesis is illustrated in Scheme 8.
1
20. The H NMR spectrum of the corresponding acetate 19b
shows Ha (d ˆ 5.37) as a double doublet with J ˆ 9.6 and
9.3 Hz coupled to Hb (d ˆ 3.47, dd, J ˆ 9.2, 3.2 Hz) and Hc (d ˆ
5.28, dd, J ˆ 9.6, 7.3 Hz) consistent with 19b but not 20. Thus,
the neighboring oxygen of the methylenedioxy ring rather
than the azide group controls the regioselectivity of the ring
opening.
An alternative strategy envisions regio- and diastereose-
lective delivery of oxygen by electrophilic initiated cyclization
of the carbamate 21 (available by a Staudinger reduction) but
to no avail.[23] Dihydroxylation of azide 7b using catalytic
osmium tetraoxide in moist methylene chloride gave a single
diol 22 (Scheme 7). Catalytic hydrogenation in the presence
OTBDMS
O
OTBDMS
O
H
N
BocNH
23
O
94%
55%
O
O
O
O
O
OH
X
26
25a: X = SO
25b: X = SO2
88%
Scheme 8. Cyclization yielding oxazolidinone 26.
OTBDMS
O
OTBDMS
Exposure to thionyl chloride ((C2H5)3N, THF, 08C) forms the
cyclic sulfite 25a which proved too unreactive to cyclize to the
oxazolidinone 26. On the other hand, standard oxidation
(NaIO4, RuCl3 ´ 3H2O, H2O, CH3CN, CCl4)[25] gave the cyclic
sulfate which did cyclize in THF at 508C to give the desired
oxazolidinone 26.
N3
BocNH
HO
O
2
HA
HB
7b
5
72%
71%
O
O
HO
OH
OH
22
23
OTBDMS
O
OTBDMS
Plan B: Dynamic kinetic asymmetric transformation of
conduritol B
BocNH
BocNH
HO
O
2
a, b
HA
HB
7b
5
84%
O
O
A second strategy emanates from a conceptually different
process of asymmetric induction as illustrated in Scheme 9.
In this process, a racemic substrate is converted to an
OH
21
24
Scheme 7. Reductive amine protection. a) (CH3)3P, H2O (96%). b) Boc2O
(84%).
L*
L*
OR
Complexation
Nu
Decomplexation
of Boc anhydride gave the carbamate 23 as a single isomer.
On the other hand, dihydroxylation of the carbamate 21 also
gave a single isomer 24 which was different from that obtained
above (see Scheme 7). Previous work suggested a simple
method to determine stereochemistry at C-2 and C-5 based
upon the chemical shifts of the methylenedioxy protons HA
and HB. When the groups at C-2 and C-5 are cis, the
absorptions for HA and HB are typically around d ˆ 5.0 and
d ˆ 5.3 with Dd < 0.3. When the groups are trans, the chemical
shifts are around d ˆ 4.8 and d ˆ 5.3 with Dd > 0.4. In the case
of diol 23, the observed shifts are d ˆ 5.27 and 4.82 (Dd ˆ 0.45)
and for diol 24 the observed shifts are d ˆ 5.19 and 5.02 (Dd ˆ
0.17). These shifts are in accord with the assigned structures as
Pd
L*
L*
*L
L*
Pd
Pd
Pd
H
RO
OR
Nu
Selective
addition
L*
L*
Nu
Ionization
OR
Pd
Scheme 9. Conceptualization of a dynamic kinetic asymmetric transfor-
mation.
1622
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1622 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 8

Aminocyclohexitol Synthesis
1619±1629
intermediate which is formally
a meso complex thereby losing
the stereochemical identity of
the precursor.[5, 26] Asymmetric
induction occurs in the nucleo-
philic addition step. For this
process, conduritol B tetraest-
ers 31/ent-31[27, 28] become the
starting materials as depicted in
the retrosynthetic analysis of
Scheme 10.
This analysis raises the inter-
esting question of the effect of
the symmetry of 31/ent-31 on a
dynamic kinetic asymmetric
transformation (DYKAT) reac-
tion. As depicted in Scheme 11
with a given ligand such as S,S-
isomer 32, ionization of 31 con-
stitutes an energetically favor-
able process (i.e., a ªmatchedº
event); whereas that of ent-31 is
a disfavored one (i.e., a ªmis-
matchedº event). Indeed, the
O
O
NH HN
32
PPh2 Ph2P
"matched"
ionization
31
O
O
O
OCOR
O
OCOR
fast
"matched"
addition
ROCO
O
ROCO
O
L*
Pd
L*
Nu
ROCO
ROCO
slow
Nu
ent-31
33
"mismatched"
ionization
O
O
O
Nu
Nu
Nu
matched
addition
ROCO
O
ROCO
ROCO
"matched"
ionization
L*
Nu
Pd
ROCO
ROCO
O
ROCO
O
L*
OCOR
Nu
33
34
O
Scheme 11. Deracemization reaction of conduritol B tetracarbonate.
energy differences between the matched and mismatched
events is sufficient that a nearly perfect kinetic resolution may
be achieved. Interestingly, the product of the kinetic reso-
lution, 33, still possesses an allylic ester with the same
orientation with respect to the double bond as that in 31. Thus,
its ionization also constitutes a ªmatchedº event. The result
may be a double substitution to form 34. If a carboxylate
nucleophile is employed, the formation of enantiomerically
pure tetraester product 34 is equivalent to a deracemization of
conduritol B tetracarbonate accompanied by simultaneously
differentiation of the allylic esters from the non-allylic esters.
Such
a
process then allows formation of 30 (R ˆ H,
Scheme 10) which sets the stage through 29 for formation of
the requisite cis vicinal aminoalcohol 28 and thus completion
of an efficient approach to the aminocyclohexitol.
Racemic conduritol B tetraacetate (Æ)-36 was obtained
using a modification (Scheme 12) of the method reported by
Guo et al.[26b] Guo et al. had reported that the diacetoxy-
dibromide 35b, available from p-benzoquinone in three steps,
O
O
OH
OCPh
OCPh
H
N
H
N
H2N
HO
O
O
O
O
O
O
OR
O
O
OR
O
O
RO
RO
Br
Br
a,b
c
OH
OCPh
OCPh
O
O
6
OR
OR
27
O
28
O
(±)36
OCPh
OCPh
35a: R = H
35b: R = Ac
36a: R = Ac
R'SO2O
RNH
RO
RO
d,e
d,f
36b: R = CO2Me
O
O
36c: R = CO2CH2CCl3
OCPh
OCPh
Scheme 12. Synthesis of conduritol B derivative 36 from p-benzoquinone:
a) Br2, CH2Cl2, 76%. b) NaBH4, H2O/Et2O, 82%. c) Ac2O, K2CO3, 08C !
RT; then HOAc, reflux, 71%. d) (C2H5)3N, MeOH/H2O, 100%.
e) ClCO2CH3; pyridine, DMAP, CH2Cl2, 93%. f) ClCO2CH2CCl3, pyri-
dine, DMAP, CH2Cl2, 90%.
O
O
29
30
O
O
O
O
OCOR
OCOR
ROCO
ROCO
OCOR
could be converted to (Æ)-36a by heating with potassium
acetate and acetic anhydride in refluxing acetic acid. We
found that 35b could be generated in situ by adding potassium
carbonate to a suspension of the diol-dibromide 35a[30, 31] in
acetic anhydride. Addition of acetic acid to this mixture
followed by warming to reflux then led to formation of (Æ)-36.
Using this route, (Æ)-36 could be prepared in 44% yield over
ROCO
O
OCOR
O
ROCO
O
O
ent-31
31
Scheme 10. Retrosynthetic analysis through DYKAT reaction of condur-
itol B.
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1623 $ 17.50+.50/0
1623

B. M. Trost et al.
FULL PAPER
three steps from p-benzoquinone in relatively large amounts
(>20 g in a single run). A further benefit of this route was that
no chromatographic purifications were required since the
intermediates and (Æ)-36 were readily crystallized solids.
When the racemic tetraacetate (Æ)-36a was submitted to
the palladium-catalyzed reaction with carboxylate nucleo-
philes in the presence of the R,R isomer of the chiral ligand 32
an excellent kinetic resolution was observed, however, greater
than 50% conversion of the starting material was difficult to
obtain due to the relative inactivity of the ªmismatchedº
enantiomer of the racemic tetraacetate starting material. By
simply converting the tetraacetate to the more reactive
tetra(methylcarbonate) derivative 36b a dynamic kinetic
asymmetric transformation was realized using benzoate as
the nucleophile, yielding the dibenzoate 37a in 80% yield and
>99% ee (see Scheme 13). Selective cleavage of the methyl
carbonate groups in the presence of the benzoates to provide
the diol 38 proved difficult. Replacing the methyl carbonate
groups in 36b with 2,2,2-trichloroethyl carbonates gave 36c
(see Scheme 12) which also reacted smoothly in the palla-
dium-catalyzed reaction to give the dibenzoate 37 in excellent
yield and >99% ee (see Scheme 13). The trichloroethyl
carbonate groups could now be easily cleaved in the presence
of the benzoates by simply treating with zinc metal and acetic
acid to give the desired conduritol B 2,3-diol 38 in 77% yield.
OCOPh
O
OCOPh
OCOPh
OCOPh
RO
(+)-38
40
R'O
OCOPh
R
N
39a: R = Ts, R' = H
O
O
39b: R = R' = Ts
OCOPh
39c: R = Ts, R' = CONHCOPh
41a: R = COPh
41b: R = H
Scheme 14. Hydroxy group activation.
to form 40. Initial success for forming the oxazolidinone was
observed with sodium hydride in THF (08C to reflux). While
the expected N-benzoyloxazolidin-2-one (41a) was not ob-
served, the more satisfactory debenzoylated derivative 41b
was obtained in 34% yield. Use of sodium hexamethyldisi-
lamide proved more satisfactory but the maximum yield >
was still only 52%.
A more satisfactory solution was devised when we discov-
ered that treatment of the diol (‡)-38 with benzylisocyanate
provided the mono-carbamate 42 in consistently good yield
(Scheme 15) with only minor amounts of the dicarbamate and
OCOPh
OCO2R
PhCO2H, NaOH
THAB (20 mol%)
ROCO2
ROCO2
ROCO2
ROCO2
path a
(+)-38
(S,S)-ligand 32 (7.5 mol%)
[η3-C3H5)PdCl]2 (2.5 mol%)
OCOPh
OCOPh
OCO2R
Ph
(C2H5)3N
N C O
CH2Cl2/H2O
RO
O
(+)-37
path b
path c
(±)-36b or c
40
(+)-38
R = CH3
80%
90%
>99% ee
84%
Ph
N
O
OCOPh
OCOPh
Ph
O
R = CH2Cl3
H
37a: R = CH3
OCOPh
N
O
37b: R = CH2CCl3
42a: R = H
42b: R = SO2CH3
42c: R = SO2CF3
OCOPh
R = CH2CCl3
Zn
HO
HO
43
HOAc/THF
77%
Scheme 15. Successful oxazolidinone cyclization (path c).
OCOPh
(+)-38
recovered diol. An attempt to cyclize this hydroxy carbamate
directly under Mitsunobu conditions (DIAD, PPh3) only led
to cleavage of the carbamate and return of the diol 38 (path a).
An attempt at oxazolidinone formation involving cyclizing the
mesylate 42b with sodium hydride led to carbamate elimi-
nation and formation of the epoxide 40 (path b).
The key to forming the oxazolidinone was the combination
of the use of a more reactive leaving group and a hydroxide
free base. Thus, reaction of the mono-carbamate 42a with
triflic anyhdride, followed by treatment of the crude triflate
42c with potassium bis(trimethylsily)amide at low temper-
ature ( 408C) provided the desired benzyl oxazolidinone 43
in a gratifying 70% yield over the two steps (path c).
Scheme 13. Selective deprotection.
Installation of the amino functionality of 6 envisioned an
N-alkylative cyclization of a suitable carbamate. Initial
attempts to selectively activate one of the hydroxy groups of
38 by conversion to the mesylate or tosylate met with mixed
results. Generally, mixtures of mono- and di-sulfonylation
were observed in ratios that varied from experiment to
experiment. Under the best conditions (TsCl, DMAP,
CH2Cl2, 08C), yields of a monotosylate 39a were as high as
67% with the bis-tosylate 39b, a significant by-product (16 ±
20%), see Scheme 14. The crude mixture was treated directly
with benzoylisocyanate to give the imide 39c in a 67% overall
yield from diol 38. Curiously, treatment of the tosylate 39c
with DBU in THF (room temperature to reflux) gave only the
epoxide 40 in 68% yield. This product arises by loss of
benzoylisocyanate to form an alkoxide which simply cyclizes
The remainder of the aminocylitol synthesis was relatively
straightforward as summarized in Scheme 16. Dihydroxyla-
tion of 43 with catalytic osmium tetraoxide occurred predom-
inantly from the convex side of the bicyclic alkene (greater
than 10:1 diastereoselectivity) to provide the desired diol 44
1624
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1624 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 8

Aminocyclohexitol Synthesis
1619±1629
tistatin and the other to the aminocyclohexitols herein.
Further, from this latter allylic azide, four different epimers
of the aminocyclitol moiety can be accessed with complete
stereoselectivity. Such flexibility lends itself to the possibility
of synthesizing other analogues for biological evaluation.
When beginning with racemic conduritol B, an advantage is
the ease of obtaining the substrate since gram quantities are
readily available in three steps from p-benzoquinone. Upon
formation of enantiomerically pure conduritol B in a complex
palladium-catalyzed dynamic kinetic resolution, the system
can be manipulated in a straightforward, seven-step sequence
to give the desired natural product. Both routes circumvent
the need for starting from the chiral pool as is typical in most
reported aminocyclitol syntheses and, thus, allow for the
opportunity to easily obtain either enantiomer of the natural
product by simply changing the ligand. These findings should
lead to expanding the use of palladium-catalyzed asymmetric
alkylation reactions in the synthesis of cyclohexitols and other
carbohydrate derivatives.
OCOPh
O
OCOPh
Ph
Ph
O
N
O
N
OH
a
b
O
43
80%
quant.
O
O
OH
OCOPh
OCOPh
44
45
OH
H2N
O
O
c
77%
HO
OCOPh
6
Scheme 16. Final stage of the hygromycin aminocyclitol synthesis:
a) OsO4, NMO, THF, H2O, RT. b) CH3OCH2OCH3, CF3SO3TMS, 2,6-
lutidine, 08C ! RT. c) Li, NH3, THF, 78 to 338C then H2O, 788C to
reflux.
in 80% yield. Conversion of the diol to the methylene acetal
45 proceeded quantitatively using dimethoxymethane and
trimethylsilyl triflate. Final deprotection of the benzyl amine,
the benzoate esters, and the oxazolidinone was accomplished
in a single step using lithium/ammonia followed by an
aqueous quench and warming to reflux. Purification by ion-
Experimental Section
‡
exchange chromatography (Amberlite IRC-76, H form)
General methods: NMR spectra were recorded at room temperature using
either a Varian Gemini-300, Gemini-200, or EM-400 instrument. IR spectra
were recorded on a Perkin ± Elmer FT-IR as neat films on NaCl plates or as
KBr pellets for solid products. Mass spectra were recorded by the Mass
Spectrometer Facility of the School of Pharmacy, University of California,
San Francisco. Optical rotations were measured with a Jasco DIP-360
digital polarimeter. Microanalyses were performed by M-H-W Laborato-
ries, Phoenix AZ. Melting points were not corrected.
provided the desired amino-triol 6 in 77% yield. Comparison
of the data for 6 to that of a known sample indicated their
identity.
The conduritol B approach to the hygromycin aminocycli-
tol was performed in 23% overall yield requiring 10 steps
from the readily available racemic conduritol B tetraacetate
36a. The highlight of the synthesis was the dynamic kinetic
asymmetric transformation of the racemic tetracarbonate 36c
into the dibenzoate (‡)-37b in 90% yield and 99% ee. A
further benefit of this synthesis was the fact that all of the
compounds except 43 and 6 were solids that were purified by
crystallization.
Reactions were generally conducted under a positive pressure of dry argon
or nitrogen in flame-dried glassware. THF was distilled from sodium/
benzophenone. Methylene chloride and acetonitrile were distilled from
calcium hydride. Methanol and ethanol were distilled from magnesium
methoxide and magnesium ethoxide, respectively. Acetone was distilled
from calcium sulfate. Common reagents and materials were purchased
from commercial sources and purified by distillation or recrystallization.
Anhydrous solvents and reaction mixtures were transferred by oven-dried
syringe or cannula. Flash chromatography employed ICN silica gel
(Kieselgel 60, 230 ± 400 mesh), analytical TLC was performed with
0.2 mm silica-coated glass plates (E. Merck, DC-Platten Kieselgel 60 F254).
Conclusion
The AAA reaction constitutes a potentially powerful tool in
asymmetric synthesis because of the diverse mechanisms for
asymmetric induction. The conduritols illustrate the potential.
Conduritol A, a meso compound, can be desymmetrized;
whereas, conduritol B, a chiral compound that generates an
achiral p-allyl complex, can be ªderacemizedº by a dynamic
kinetic asymmetric transformation. Both conduritols, the
latter as a racemate, are readily available from p-benzoqui-
none. Thus, great flexibility derives from the resultant relative
stereochemistry and the differential functionality for further
elaboration. These points are nicely illustrated by the syn-
thesis of the hygromycin aminocyclohexitol.
1,4-Bis(methoxycarboxy)-1b,4b,4ab,9a,9ab,10a-hexahydro-9,10[1',2']-ben-
zenoanthracene (13b): n-Butyllithium (1.4m in hexanes, 64 mmol,
2.2 equiv) was added dropwise to a solution of diol 13a (8.4 g, 29 mmol,
1 equiv) in THF (530 mL) at 08C. The slurry was stirred at 08C for 30 min,
then at room temperature for 30 min. The mixture was recooled to 08C, at
which point methyl chloroformate (10.9 g, 116 mmol, 4 equiv) was added
dropwise. The reaction was stirred at room temperature for 1 h, diluted
with ethyl acetate, washed with saturated ammonium chloride solution and
brine, dried (MgSO4) and evaporated in vacuo. The dicarbonate 13b
(10.77 g, 90%) was obtained as a white solid. M.p. 167 ± 1698C (decomp);
1
IR (neat): nÄ ˆ 1749, 1442, 1261 cm
;
1H NMR (300 MHz, CDCl3): d ˆ
7.29 ± 7.25 (m, 2H), 7.22 ± 7.19 (m, 2H), 7.11 ± 7.04 (m, 4H), 5.14 (d, J ˆ
5.5 Hz, 2H), 4.84 (s, 2H), 4.43 (s, 2H), 3.89 (s, 6H), 2.86 (d, J ˆ 6.0 Hz, 2H);
13C NMR (75.5 MHz, CDCl3): d ˆ 155.4, 143.8, 141.6, 126.4, 126.1, 124.8,
‡
124.3, 123.7, 73.3, 55.0, 44.2, 40.9; HRMS: calcd for C24H22O6 [M] :
When starting from the meso-conduritol A derivative, the
methylene acetal, which is critical for biological activity, is
already in place, while the introduction of a nitrogen synthon
and asymmetric induction occur in a single operation. Using
an azide as a nucleophile allows ready access to two different
regioisomeric products taking advantage of a [3.3]-sigmatrop-
ic rearrangement of an allylic azide. The utility of this aspect is
clearly demonstrated in that one regioisomer led to pancra-
406.1416; found: 406.1397.
1,4-Bis(methoxycarboxy)-2,3-dihydroxy-(1b,2a,3a,4b,4ab,9a,9ab,10a)-oc-
tahydro-9,10[1',2']-benzenoanthracene (14b): NMO (5.4 g, 39.6 mmol,
1.5 equiv) followed by osmium tetroxide (4% solution in water, 0.4 mL,
2.5 mol%) was added to a solution of olefin 13b (10.77 g, 26.4 mmol,
1 equiv) in THF (100 mL), tert-butanol (49 mL), and water (26 mL). The
reaction was stirred for 16 h at room temperature. A 15% aqueous sodium
bisulfite solution was added and the mixture stirred for 1 h. The reaction
was diluted with ethyl acetate and the aqueous layer was washed with ethyl
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1625 $ 17.50+.50/0
1625

B. M. Trost et al.
FULL PAPER
acetate. The combined organic layers were washed with brine, dried
(MgSO4), and evaporated in vacuo. After column chromatography on silica
gel (50% ethyl acetate/pentane) followed by precipitation of the resulting
ligand 17 (39 mg, 0.059 mmol, 4 mol%) in THF (1.4 mL). After 10 min,
freshly distilled trimethylsilyl azide (232 mg, 2.1 mmol, 1.5 equiv) was
added and the reaction stirred for 6 h. The unreacted starting material
(52 mg, 15%) was separated from the desired product by chromatography
on silica gel (15% ethyl acetate/pentane). The resulting product oil was
dissolved in methanol (5 mL) and K2CO3 (58 mg, 0.42 mmol, 40 mol%)
was added. The mixture was heated at 508C for 48 h, the solid was collected
by filtration, and the filtrate was concentrated in vacuo. After column
chromatography on silica gel (30% ethyl acetate/pentane) of the resulting
oil, the azido alcohol 7a (127 mg, 56% two steps; 70% yield based on
oil with diethyl ether, diol 14b (7.2 g, 69%) was produced as a white solid.
1
M.p. 160 ± 1628C; IR (neat): nÄ ˆ 3468, 1750, 1443, 1266 cm
;
1H NMR
(300 MHz, CDCl3): d ˆ 7.37 (dd, J ˆ 5.5, 3.2 Hz, 2H), 7.26 (dd, J ˆ 5.1,
3.5 Hz, 2H), 7.14 (dd, J ˆ 5.5, 3.2 Hz, 2H), 7.08 (dd, J ˆ 5.4, 3.2 Hz, 2H),
4.95 ± 4.90 (m, 2H), 4.52 (s, 2H), 3.88 (s, 6H), 2.95 (appt, J ˆ 5.3 Hz, 2H),
2.87 (d, J ˆ 4.5 Hz, 2H), 2.79 (appt, J ˆ 2.3 Hz, 2H); 13C NMR (75.5 MHz,
CDCl3): d ˆ 156.7, 143.2, 142.0, 127.4, 126.5, 125.8, 123.7, 79.1, 66.2, 55.4,
‡
43.7, 39.9; HRMS: calcd for C24H24O8 [M] : 440.1453; found: 440.1471.
recovered starting material) was obtained as a clear oil. [a]D20
ˆ
318.448
1
(c ˆ 2.06, CH2Cl2); IR (neat): nÄ ˆ 3447, 2106, 1250 cm
;
1H NMR
1,4-Bis(methoxycarboxy)-2,3-methylenedioxy-(1b,2a,3a,4b,4ab,9a,9a-
(300 MHz, CDCl3): d ˆ 6.02 (ddd, J ˆ 10.0, 3.4, 0.9 Hz, 1H), 5.93 (dd, J ˆ
10.0, 4.6 Hz, 1H), 5.04 (s, 1H), 4.92 (s, 1H), 4.55 (dd, J ˆ 6.3, 3.3 Hz, 1H),
4.24 (appt, J ˆ 6.6 Hz, 1H), 4.13 (appt, J ˆ 4.0 Hz, 1H), 4.04 (dd, J ˆ 7.1,
3.8 Hz, 1H), 2.92 (brs, 1H); 13C NMR (75.5 MHz, CDCl3): d ˆ 127.9, 126.9,
b,10a)-octahydro-9,10[1',2']-benzenoanthracene
(15a):
2,6-Lutidine
(3.68 g, 35.5 mmol, 2.3 equiv) followed by trimethylsilyl trifluoromethane-
sulfonate (13.8 g, 63 mmol, 4.1 equiv) was added dropwise at 08C to a
solution of diol 14b (6.7g, 15.41 mmol, 1 equiv) in dimethoxymethane
(114 mL). The reaction was stirred at room temperature for 1.5 h, then
quenched with saturated NaHCO3. The reaction was diluted with ethyl
acetate, the organic layer was washed with 1n sodium hydrogen sulfate and
brine, dried (MgSO4), and evaporated in vacuo. Methylene acetal 15a
(6.3 g, 91%) was obtained as a white solid. M.p. 220 ± 2218C; IR (neat): nÄ ˆ
2958, 1751, 1442, 1264, 1166, 1082, 996, 973, 915 cm 1; 1H NMR (300 MHz,
CDCl3): d ˆ 7.40 (dd, J ˆ 5.4, 3.26 Hz, 2H), 7.26 (dd, J ˆ 5.3, 3.15 Hz, 2H),
7.17 (dd, J ˆ 5.4, 3.2 Hz, 2H), 7.08 (dd, J ˆ 5.4, 3.2 Hz, 2H), 4.99 ± 4.94 (m,
2H), 4.82 (s, 1H), 4.54 (s, 1H), 4.52 (s, 2H), 3.89 (s, 6H), 3.22 (dd, J ˆ 7.2,
1.9 Hz, 2H), 2.85 (appt, J ˆ 2.4 Hz, 2H); 13C NMR (75.5 MHz, CDCl3): d ˆ
155.2, 143.3, 141.9, 127.1, 126.2, 125.7, 123.6, 92.6, 75.9, 74.5, 55.1, 43.5, 40.5;
‡
94.4, 75.0, 72.2, 69.9, 57.9; HRMS: calcd for C7H9N3O3 [M] : 182.0565;
found: 182.0564. Enantiomeric excess 93% determined by HPLC [Chir-
alcel AD column, eluting with 96:4 heptane/iPrOH, 1 mLmin 1: major
enantiomer tR ˆ 25.5 min, minor enantiomer tR ˆ 28.6 min].
(1R,3R,4R,5R)-5-Azido-4-tert-butyldimethylsiloxy-3a,4,5,7a-tetrahydro-
benzo-[1,3]-dioxole (7b): 2,6-Lutidine (0.21 g, 1.81 mmol, 3.4 equiv) fol-
lowed by tert-butyldimethylsilyl trifluoromethanesulfonate (0.28 g,
1.11 mmol, 2.1 equiv) was added to alcohol 7a (100 mg, 0.54 mmol,
1 equiv) in methylene chloride (0.54 mL) at 08C. The reaction was stirred
for 1.5 h at room temperature, then diluted with methylene chloride. The
organic layer was extracted with 1n sodium hydrogen sulfate, saturated
NaHCO3, and brine, dried (MgSO4), and evaporated in vacuo. After
column chromatography on silica gel (15% ethyl acetate/pentane), silyl
‡
HRMS: calcd for C25H24O8 [M] : 452.1471; found: 452.1474.
1,4-Dihydroxy-2,3-methylidene-(1b,2a,3a,4b,4ab,9a,9ab,10a)-octahydro-
9,10[1',2']-benzenoanthracene (15b): K2CO3 (5.22 g, 37.2 mmol, 2.5 equiv)
was added at room temperature to a solution of dicarbonate 15a (6.75 g,
15 mmol, 1 equiv) in methanol (50 mL) and methylene chloride (50 mL).
The reaction mixture was stirred for 4 h, the solid was removed by
filtration, and the filtrate was evaporated in vacuo. After column
chromatography on silica (60% ethyl acetate/pentane), the diol 15b
(4.1 g, 82%) was obtained as a white solid. M.p. 211 ± 2128C; IR (KBr): nÄ ˆ
3418, 3020, 2918, 2850, 1466, 1164, 1091, 1066, 978, 902, 751 cm 1; 1H NMR
(300 MHz, [D6]DMSO): d ˆ 7.32 ± 7.26 (m, 4H), 7.11 ± 7.09 (m, 2H), 7.06 ±
7.02 (m, 2H), 5.31 (d, J ˆ 4.1 Hz, 2H), 4.67 (brs, 3H), 4.36 (s, 1H), 3.73 (brs,
2H), 2.74 (d, J ˆ 7.0 Hz, 2H), 2.36 (brs, 2H); 13C NMR (75.5 MHz,
[D6]DMSO): d ˆ 145.1, 143.3, 125.9, 125.4, 125.3, 123.2, 91.1, 77.2, 67.8, 43.1,
43.0; elemental analysis calcd (%) for C21H20O4‡0.5H2O: C 73.02, H 6.13;
found: C 73.21, H 6.20.
ether 31 was obtained as a clear oil (116 mg, 73%). [a]D25
ˆ
135.188 (c ˆ
1.19, CH2Cl2); IR (neat): nÄ ˆ 2954, 2858, 2102, 1472, 1256, 1128, 906,
1
839 cm
;
1H NMR (300 MHz, CDCl3): d ˆ 5.95 ± 5.87 (m, 2H), 4.97 (s,
1H), 4.92 (s, 1H), 4.55 (dd, J ˆ 5.7, 1.8 Hz, 1H), 4.18 (appt, J ˆ 6.4 Hz, 1H),
4.10 (dd, J ˆ 6.4, 2.0 Hz, 1H), 3.89 (apps, 1H), 0.89 (s, 9H), 0.13 (s, 6H);
13C NMR (75.5 MHz, CDCl3): d ˆ 127.6, 127.2, 94.2, 75.5, 72.3, 71.6, 58.1,
25.6, 17.9, 4.91, 5.20; elemental analysis calcd (%) for C13H23N3Si: C
52.50, H 7.79, N 14.13; found: C 52.39, H 7.60, N 14.06.
(1R,3R,4R,5R,6R,7R)-4-tert-Butyldimethylsiloxy-5-tert-butyloxycarbonyl-
amino-6,7-dihydroxy-1,3,4,5,6,7-hexahydrobenzo[1,3]dioxole (23): NMO
(63 mg, 0.53 mmol, 1.5 equiv) followed by osmium tetroxide (4% solution
in water, 0.2 mL, 0.018 mmol, 5 mol%) was added to a solution of olefin 7b
(106 mg, 0.36 mmol, 1 equiv) in methylene chloride (3.6 mL) at room
temperature. The reaction was stirred for 16 h, then concentrated in vacuo
and chromatographed on silica gel (50% ethyl acetate/pentane) to give the
1,4-Bis(methoxycarboxy)-2,3-methylenedioxy-(1a,2b,3b,4a)-cyclohex-5-
ene (9): The diol 15b (605 mg, 0.82 mmol, 1 equiv) was flash vacuum
thermolyzed (5008C, 0.025 mmHg) to yield a mixture of anthracene and
the desired diol 16. After column chromatography on silica gel (50% ethyl
diol 23 (84 mg, 71%) as a clear oil. IR (neat): nÄ ˆ 3441, 2106, 1472,
1
1257 cm
;
1H NMR (300 MHz, CDCl3): d ˆ 5.20 (s, 1H), 4.84 (s, 1H),
4.22 ± 4.18 (m, 2H), 4.13 ± 4.06 (m, 2H), 4.03 (appt, J ˆ 6.8 Hz, 1H), 3.81
(dd, J ˆ 6.4, 2.2 Hz, 1H), 3.21 (brs, 1H), 3.08 (brs, 1H), 0.88 (s, 9H), 0.12 (s,
6H); 13C NMR (75.5 MHz, CDCl3): d ˆ 95.2, 77.5, 76.3, 70.3, 69.7, 66.8, 64.0,
25.6, 17.9, 4.98, 5.2.
acetate/pentane), the diol 16 (208 mg, 73%) was obtained as a white solid.
1
M.p. 111 ± 1128C; IR (neat): nÄ ˆ 3386, 1642, 1377 cm
;
1H NMR
(300 MHz, CD3OD): d ˆ 5.67 (s, 2H), 5.07 (s, 1H), 4.86 (s, 2H), 4.83 (s,
1H), 4.10 (d, J ˆ 3.8 Hz, 2H), 3.91 (d, J ˆ 3.0 Hz, 2H); 13C NMR
(75.5 MHz, CD3OD): d ˆ 132.0, 95.1, 81.2, 70.5.
A suspension of palladium on carbon (10%, 7.2 mg) in ethyl acetate
(0.7 mL) was stirred under an atmosphere of hydrogen at room temper-
ature for 20 min. A solution of di-tert-butyl dicarbonate (65 mg, 0.3 mmol,
1.2 equiv) and azide 22 (83 mg, 0.25 mmol, 1 equiv) in ethyl acetate
(0.7 mL) was then added and the reaction stirred for 22 h. The mixture was
diluted with ethyl acetate, filtered through celite, and evaporated in vacuo.
After column chromatography on silica gel (50% ethyl acetate/pentane),
n-Butyllithium (6.85 mL, 2.5m in hexanes, 2.2 equiv) was added dropwise
to a solution of diol 16 (1.22 g, 7.72 mmol, 1 equiv) in THF (33 mL). The
slurry was stirred at 08C for 10 min, then at room temperature for 20 min.
The mixture was recooled to 08C and methyl chloroformate (2.93 g,
30.8 mmol, 4 equiv) was added. The reaction was stirred at room temper-
ature for 2 h, then diluted with ethyl acetate. The organic layer was washed
twice with water then brine, dried (MgSO4), and evaporated in vacuo. After
column chromatography on silica gel (30% ethyl acetate/petroleum ether),
carbamate 23 was obtained as a clear oil (86 mg, 72%). [a]2D0 ˆ16.658 (c ˆ
1
1.40, CH2Cl2); IR (neat): nÄ ˆ 3453, 1712, 1510, 1472, 1391, 1366, 1254, cm
;
1H NMR (300 MHz, CDCl3): d ˆ 5.27 (s, 1H), 4.89 (brs, 1H), 4.82 (s, 1H),
4.23 ± 4.21 (m, 2H), 4.14 ± 4.13 (m, 1H), 3.97 ± 3.95 (m, 2H), 3.76 (dd, J ˆ
6.4, 2.0 Hz, 1H), 3.23 (brs, 2H), 1.43 (s, 9H), 0.86 (s, 9H), 0.08 (s, 6H);
13C NMR (75.5 MHz, CDCl3): d ˆ 156.5, 95.7, 79.9, 77.3, 75.2, 69.7, 68.2, 52.9,
30.6, 28.3, 25.6, 17.8, 4.89, 5.13; elemental analysis calcd (%) for
C18H35NO7Si: C 53.31, H 8.70, N 3.45; found: C 53.53, H 8.72, N 3.48.
biscarbonate 9 (1.71 g, 81%) was obtained as a white solid. M.p. 89 ± 908C;
1
IR (neat): nÄ ˆ 2960, 2862, 1755, 1447, 1326, 1254, 1086, 1009, 960 cm
;
1H NMR (300 MHz, CDCl3): d ˆ 5.81 (s, 2H), 5.19 (d, J ˆ 4.1 Hz, 2H), 5.14
(s, 1H), 4.91 (s, 1H), 4.20 (d, J ˆ 4.6 Hz, 2H), 3.82 (s, 6H); 13C NMR
(75.5 MHz, CDCl3): d ˆ 155.2, 127.8, 94.5, 75.5, 73.9, 55.1; HRMS: calcd for
Oxazolidinone 26: Triethylamine (71 mg, 0.7 mmol, 4 equiv) followed by
thionyl chloride (23 mg, 0.19 mmol, 1.1 equiv) was added at 08C to a
solution of diol 23 (71 mg, 0.18 mmol, 1 equiv) in THF (1.7 mL). The
reaction was stirred at 08C for 1 h, then diluted with ethyl acetate. The
organic layer was washed with water and brine, dried (MgSO4), and
evaporated in vacuo. The cyclic sulfite 25a (71 mg, 90%) was obtained as a
‡
C9H11O6 [M C2H3O2] : 215.0556; found: 215.0548.
(1R,3R,4R,5R)-5-Azido-4-hydroxy-3a,4,5,7a-tetrahydrobenzo-[1,3]-diox-
ole (7a): Dicarbonate 9 (390 mg, 1.4 mmol, 1 equiv) in THF (10mL) was
added to a degassed (Ar) solution of tris(dibenzylideneacetone)dipalla-
dium-(chloroform) (14 mg, 0.014 mmol, 1 mol%) and the S,S-cyclohexyl
1626
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1626 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 8

Aminocyclohexitol Synthesis
1619±1629
clear oil and used without purification. Data was reported for a 1:1 mixture
(0.95 g, 77%) in three batches. Rf ˆ 0.40 (1:1 hexanes/EtOAc); m.p. 1538C;
[a]2D3 ˆ ‡ 206.88 (c ˆ 2.11, CHCl3); IR (KBr): nÄ ˆ 3384 (brm), 1716 (s), 1265
(s) cm 1; 1H NMR (CDCl3, 300 MHz): d ˆ 8.08 (d, J ˆ 8.0 Hz, 4H), 7.61 (t,
J ˆ 7.6 Hz, 2H), 7.45 (t, J ˆ 7.6 Hz, 4H), 5.84 (s, 2H), 5.70 (dd, J ˆ 5.4,
2.4 Hz, 2H), 4.05 (dd, J ˆ 5.6, 2.2 Hz, 2H), 3.39 (s, 2H); 13C NMR (CDCl3,
75 MHz): d ˆ 166.9, 133.4, 129.9, 129.5, 128.5, 127.8, 74.9, 74.0; elemental
analysis calcd (%) for C20H18O6: C 67.79, H 5.12; found: C 67.70, H 5.30.
(by 1H NMR) of diastereomers at sulfur. IR (neat): nÄ ˆ 3375, 2930, 2858,
1
1710, 1505, 1471, 1391, 1366, 1257, 1169, 1015 cm
;
1H NMR (300 MHz,
CDCl3): d ˆ 5.37 (brs, 1H), 5.29 (s, 1H), 5.26 ± 5.22 (m, 2H), 4.95 (dd, J ˆ
7.6, 3.2 Hz, 1H), 4.79 (s, 2H), 4.71 (apps, 2H), 4.51 ± 4.48 (m, 1H), 4.43 (dd,
J ˆ 6.9, 3.2 Hz, 2H), 4.32 (dd, J ˆ 6.9, 2.9 Hz, 2H), 4.13 ± 4.04 (m, 3H), 3.91
(apps, 1H), 1.44 (s, 9H), 0.83 (s, 9H), 0.05 (s, 6H).
Ruthenium trichloride (0.5 mg, 0.0024 mmol, 2 mol%) and sodium period-
ate (32 mg, 0.15 mmol, 1.2 equiv) was added at room temperature in one
portion to a solution of cyclic sulfite 25a (55 mg, 0.12 mmol, 1 equiv) in
acetonitrile (1.4 mL) and CCl4 (0.55 mL). Water (0.71 mL) was then added
and the reaction was stirred for 1.5 h. The mixture was diluted with ethyl
acetate. The organic layer was washed with water and brine, dried
(MgSO4), filtered through a silica plug, and evaporated in vacuo. The cyclic
sulfate 25b (49 mg, 88%) was obtained as a clear oil and used without
further purification. IR (neat): nÄ ˆ 3381, 1711, 1519, 1471, 1393, 1367, 1212,
Oxazolidinone of (‡)-(1S,2R,3R,4S)-1,4-dibenzoyloxy-2-(N-benzyl)car-
bamyl-5-cyclohexene-3-ol (43): Triethylamine (0.560 mL, 407 mg,
4.02 mmol) and benzylisocyanate (0.248 mL, 268 mg, 2.01 mmol) were
added to a solution of the diol 38 (475 mg, 1.34 mmol) in methylene
chloride (7 mL) and the resulting mixture was stirred at room temperature.
After 2.5 h, water (1 mL) was added. The solution was diluted with ethyl
acetate (30 mL) and washed with 1m HCl (2 Â 15 mL), sat. NaHCO3
(15 mL) and brine (10 mL). The organic phase was then dried (Na2SO4)
and concentrated. Precipitation from ethyl acetate with petroleum ether
provided the mono-carbamate 42a as an amorphous white solid (509 mg,
78%). The mother liquor was concentrated and the residue was purified by
flash chromatography (3:1 to 1:1 petroleum ether/EtOAc gradient) to
provide the dicarbamate (50 mg, 6%) [Rf ˆ 0.40 (2:1 hexanes/EtOAc)],
followed by 42a [43 mg, combined yield: 552 mg, 84%] followed by diol 38
(50 mg, 10%). Data for 42a: Rf ˆ 0.23 (2:1 hexanes/EtOAc); m.p. 150 ±
1528C; [a]2D3 ˆ ‡ 202.88 (c ˆ 0.91, CHCl3); IR (neat): nÄ ˆ 3335 (brm), 1724
(s), 1690 (s), 1551 (m), 1264 (s) cm 1; 1H NMR (CDCl3, 300 MHz): d ˆ 8.07
(m, 4H), 7.58 (m, 2H), 7.44 (m, 4H), 7.13 (m, 5H), 5.68 (m, 4H), 5.40 (dd,
J ˆ 11.0, 8.5 Hz, 1H), 5.20 (t, J ˆ 5.6 Hz, 1H), 4.36 (dd, J ˆ 15.2, 6.4 Hz,
1H), 4.27 (dd, J ˆ 5.9, 15.3 Hz, 1H), 4.13 (m, 1H), 3.23 (d, J ˆ 5.4 Hz, 1H);
13C NMR (CDCl3, 75 MHz): d ˆ 166.7, 166.1, 156.5, 137.9, 133.4, 129.9,
129.4, 128.6, 128.5, 128.4, 127.8, 127.4, 127.2, 75.5, 75.0, 72.9, 72.2, 45.0.
1
1165, 1100 cm 1; H NMR (300 MHz, CDCl3): d ˆ 5.39 (brs, 1H), 5.29 (s,
1H), 5.20 (dd, J ˆ 7.3, 3.4 Hz, 1H), 4.82 ± 4.76 (m, 2H), 4.46 (dd, J ˆ 6.8,
2.9 Hz, 2H), 4.14 (dd, J ˆ 7.3, 2.0 Hz, 1H), 3.92 (apps, 1H), 1.42 (s, 9H),
0.82 (s, 9H), 0.08 (s, 6H).
A solution of cyclic sulfate 25b (22 mg, 0.04 mmol, 1 equiv) was heated in
THF (0.4 mL) at 508C for 6 h. The solution was cooled to 08C and treated
with 20% aq. sulfuric acid solution. After stirring for 4.5 h, the mixture was
diluted with ethyl acetate. The combined organic layers were washed with
water and brine, dried (MgSO4), and evaporated in vacuo. After column
chromatography on silica (50% ethyl acetate/pentane), the oxazolidinone
26 (7 mg, 55%) was obtained as a white solid. M.p. 180 ± 1828C; [a]D20
ˆ
15.388 (c ˆ 0.59, CH2Cl2); IR (CH2Cl2): nÄ ˆ 3573, 3458, 1769, 1471, 1232,
1086 cm 1; 1H NMR (300 MHz, CDCl3): d ˆ 5.18 (s, 1H), 5.09 (s, 1H), 4.85
(dd, J ˆ 9.3, 5.4 Hz, 1H), 4.69 (s, 1H), 4.43 (appt, J ˆ 4.9, 2.9 Hz, 1H), 4.30
(dd, J ˆ 7.8, 2.9 Hz, 1H), 4.14 ± 4.09 (m, 2H), 4.05 (dd, J ˆ 4.9, 3.4 Hz, 1H),
2.53 (brs, 1H), 0.88 (s, 9H), 0.12 (s, 3H), 0.10 (s, 3H); 13C NMR (75 MHz,
CDCl3): d ˆ 157.8, 94.59, 74.46, 69.51, 67.79, 53.12, 25.63, 17.8, 4.72, 4.98;
Pyridine (33 mL, 32 mg, 0.410 mmol) was added to a solution of the alcohol
42a (100 mg, 0.205 mmol) in methylene chloride (1 mL) and the solution
was cooled to 508C. Triflic anhydride (52 mL, 87 mg, 0.308 mmol) was
added in dropwise fashion and the resulting mixture was stirred at 508C
for 10 min, then warmed to 08C over 20 min and quenched by adding one
drop of water. The mixture was diluted with diethyl ether (3 mL), MgSO4
(ꢀ100 mg) was added, and the mixture was filtered through a plug of SiO2,
rinsing with diethyl ether (5 mL). The filtrate was concentrated to provide
‡
HRMS: calcd for C14H26NO6Si [M ‡ H] : 332.1529; found: 332.1525.
(‡)-(1S,2R,3R,4S)-1,4-Dibenzoyloxy-2,3-di[(2,2,2-trichloroethoxy)carbo-
nyl]oxy-5-cyclohexene (37): Racemic conduritol B tetracarbonate 36c
(424 mg, 0.50 mmol), tetrahexylammonium bromide (43 mg, 0.10 mmol),
benzoic acid (214 mg, 1.75 mmol), [h3-(C3H5)PdCl]2 (4.6 mg, 0.0125 mmol,
2.5 mol%), and the ligand (S,S)-32 (25.8 mg, 0.0375 mmol, 7.5 mol%) were
thoroughly degassed. The flask was evacuated and purged with Ar (3 Â ).
Freshly distilled methylene chloride (1.5 mL) was added, followed by 1n
sodium hydroxide solution (1.5 mL) that had been previously degassed by
sparging with argon. The biphasic mixture was stirred vigorously at room
temperature under an atmosphere of Ar (balloon). After 18 h, the layers
were separated and the aqueous layer was extracted with methylene
chloride (2 Â 2 mL). The combined organic layers were dried (MgSO4),
diluted with an equal volume of diethyl ether and filtered through a plug of
silica gel (ꢀ10 g). The filtrate was concentrated to give a colorless oil that
was treated with methanol (5 mL); the flask was placed in a refrigerator.
After 1 h, the solid was filtered and rinsed with cold methanol. The filtrate
was concentrated and this procedure repeated two more times to provide
dibenzoate 37b as a crystalline white solid (317 mg, 90%). Rf ˆ 0.47 (4:1
hexanes/EtOAc); m.p. 141 ± 1438C; [a]2D3 ˆ ‡177.18 (c ˆ 1.22, CHCl3); IR
the crude triflate as
a yellow foam that was used without further
purification (134 mg, 100%). Rf ˆ 0.61 (2:1 hexanes/EtOAc); 1H NMR
(CDCl3, 300 MHz): d ˆ 8.08 (t, J ˆ 7.1 Hz, 4H), 7.62 (t, J ˆ 7.3 Hz, 2H), 7.47
(t, J ˆ 7.6 Hz, 4H), 7.12 (m, 5H), 6.09 (d, J ˆ 6.6 Hz, 1H), 5.91 (m, 3H), 5.73
(dd, J ˆ 11.0, 8.3 Hz, 1H), 5.37 (dd, J ˆ 11.0, 8.3 Hz, 1H), 5.23 (t, J ˆ 5.4 Hz,
1H), 4.40 (dd, J ˆ 6.3, 15.1 Hz, 4.25 (dd, J ˆ 4.9, 14.9 Hz, 1H).
The crude triflate was dissolved in THF (4 mL), the resulting solution was
cooled to 408C, and potassium bis(trimethylsilyl)amide (0.51 mL, 0.5m in
toluene, 0.256 mmol) was added in dropwise fashion. After 15 min, 15%
aqueous NaHCO3 solution (5 mL) was added. The mixture was extracted
with diethyl ether (2 Â 10 mL), the organic layers were washed with sat.
NaHCO3 (5 mL), and brine (5 mL), then dried (MgSO4), and concentrated
to give a brown oil. Chromatography (2:1 to 1:1 petroleum ether/ether)
provided the oxazolidinone 43 as a white foam (68 mg, 70%). Rf ˆ 0.37 (2:1
hexanes/EtOAc); [a]2D3 ˆ ‡ 235.98 (c ˆ 0.79, CHCl3); IR (neat): nÄ ˆ 1760
(s), 1723 (s), 1269 (s), 1107 (s) cm 1; 1H NMR (CDCl3, 300 MHz): d ˆ 8.04
(d, J ˆ 7.6 Hz, 2H), 8.00 (d, J ˆ 7.3 Hz, 2H), 7.60 (m, 2H), 7.47 (m, 4H), 6.32
(ddd, J ˆ 9.5, 5.9, 2.2 Hz, 1H), 6.13 (dd, J ˆ 9.8, 2.2 Hz, 1H), 5.96 (m, 1H),
5.56 (dd, J ˆ 5.6, 3.7 Hz, 1H), 5.00 (dd, J ˆ 9.5, 4.6 Hz, 1H), 4.87 (d, J ˆ
15.1 Hz, 1H), 4.07 (d, J ˆ 15.1 Hz, 1H), 3.97 (dd, J ˆ 9.3, 3.4 Hz, 1H);
13C NMR (CDCl3, 75 MHz): d ˆ 165.5, 165.4, 157.5, 135.2, 133.8, 133.7,
133.6, 130.1, 129.8, 129.7, 129.2, 128.8, 128.5, 128.5, 128.2, 128.1, 127.3, 74.6,
1
(KBr): nÄ ˆ 1771 (s), 1724 (s), 1248 (s) cm 1; H NMR (CDCl3, 300 MHz):
d ˆ 8.03 (d, J ˆ 7.1 Hz, 4H), 7.59 (t, J ˆ 7.6 Hz, 2H), 7.45 (t, J ˆ 7.6 Hz, 4H),
5.96 (m, 4H), 5.59 (dd, J ˆ 5.3, 2.6 Hz, 2H), 4.81 (d, J ˆ 12.0 Hz, 2H), 4.66
(d, J ˆ 11.7 Hz, 2H); 13C NMR (CDCl3, 75 MHz): d ˆ 165.5, 153.5, 133.7,
129.9, 128.8, 128.5, 127.3, 94.1, 76.8, 75.6, 72.1; elemental analysis calcd (%)
for C26H20Cl6O10: C 44.29, H 2.86; found: C 43.39, H 3.04. Enantiomeric
‡
excess 99% determined by HPLC [Chiralcel AD column, eluting with
71.0, 63.4, 54.2, 46.8; HRMS (EI): calcd for C27H23NO4 [M CO2] :
1
90:10 heptane/iPrOH, 1 mLmin
enantiomer tR ˆ 29.1 min].
:
(‡)-enantiomer tR ˆ 13.5 min,
(
)-
425.1627; found: 425.1625.
Oxazolidinone of (‡)-(1S,2R,3R,4S,5R,6R)-2-amino-2-deoxy-neo-inositol
(44): NMO (70 mg, 0.60 mmol) and osmium tetraoxide (38 mL, 4% in H2O,
0.006 mmol) were added to a solution of the alkene 43 (145 mg, 0.30 mmol)
in THF/water (10:1, 1.5 mL) and the resulting solution was stirred at room
temperature. After 4 h, 15% aq. NaHCO3 was added (2 mL) and the
mixture was stirred for 30 min. The mixture was then extracted with
methylene chloride (5 Â 5 mL), the combined organic layers were washed
with brine (5 mL), then dried (Na2SO4), and concentrated to give a white
solid. Recrystallization from methanol provided the purified diol 44
(121 mg, 80%) in three crops. Rf ˆ 0.31 (1:1 hexanes/EtOAc); m.p. 2248C;
(‡)-(1S,2R,3R,4S)-1,4-Dibenzoyloxy-5-cyclohexene-2,3-diol (38): Freshly
acid washed zinc dust (1.36 g, 20.8 mmol) was suspended in THF/acetic acid
(14 mL) and the mixture was cooled to 08C. The dicarbonate 37b (2.45 g,
3.47 mmol) was added in four batches. After the addition was complete the
solution was warmed to room temperature. After 2 h, the mixture was
diluted with ethyl acetate (100 mL) and washed with water (2 Â 50 mL) and
then sat. K2CO3 until the organic layer was no longer acidic. The organic
layer was then washed with brine, dried (Na2SO4) and concentrated.
Crystallization from ethyl acetate/petroleum ether ꢀ1:5 provided diol 38
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1627 $ 17.50+.50/0
1627

B. M. Trost et al.
FULL PAPER
[a]2D3 ˆ ‡ 85.08 (c ˆ 0.95, THF); IR (neat): nÄ ˆ 3422 (brm), 1761 (s), 1725
(s), 1268 (s) cm 1; 1H NMR (CDCl3, 300 MHz): d ˆ 8.08 (d, J ˆ 7.3 Hz, 2H),
7.99 (d, J ˆ 7.6 Hz, 2H), 7.60 (m, 2H), 7.47 (m, 4H), 7.25 (m, 5H), 5.56 (dd,
J ˆ 6.8, 1.5 Hz, 1H), 5.23 (brs, 1H), 5.07 (t, J ˆ 7.6 Hz, 1H), 4.70 (d, J ˆ
15.1 Hz, 1H), 4.43 (dd, J ˆ 5.1, 1.5 Hz, 1H), 4.30 (dd, J ˆ 8.3, 3.4 Hz, 1H),
4.17 (dd, J ˆ 5.1, 1.5 Hz, 1H), 4.06 (d, J ˆ 15.4 Hz, 1H), 2.68 (brs, 2H);
13C NMR ([D6]acetone, 75 MHz): d ˆ 166.1, 165.6, 158.6, 137.4, 134.4, 134.3,
130.5, 129.53, 129.47, 129.44, 128.5, 128.4, 74.89, 74.85, 72.3, 69.6, 68.6, 54.8,
47.2; elemental analysis calcd (%) for C28H25NO8: C 66.78, H 5.01, N 2.78;
found: C 66.74, H 5.03, N 2.86.
Y. Yamada, Agric. Biol. Chem. 1988, 52, 1615; Y. Nishimoto, S.
Sakuda, S. Takayama, Y. Yamada, J. Antibiot. 1991, 44, 716.
[3] T. Aoyagi, T. Yamamoto, K. Kojiri, H. Morishima, M. Nagai, M.
Hamada, T. Takeuchi, H. Umezawa, J. Antibiot. 1989, 42, 883.
[4] B. M. Trost, D. L. Van Vranken, J. Am. Chem. Soc. 1993, 115,
444.
[5] B. M. Trost, Acc. Chem. Res. 1996, 29, 355; B. M. Trost, D. L.
Van Vranken, Chem. Rev. 1996, 96, 395.
[6] B. M. Trost, D. E. Patterson, J. Org. Chem. 1998, 63, 1339; B. M. Trost,
D. L. Van Vranken, C. Bingel, J. Am. Chem. Soc. 1992, 114, 9327.
[7] J. J. Kiddle, Chem. Rev. 1995, 95, 2189; T. Suami, S. Ogawa, Adv.
Carbohydr. Chem. Biochem. 1990, 48, 21; M. Balci, H. Sütbeyaz, H.
Seçen, Tetrahedron 1990, 46, 3715; M. Balci, Pure Appl. Chem. 1997,
69, 97.
Oxazolidinone of (1R,2R,3R,4S,5R,6R)-2-amino-2-deoxy-4,5-methylene-
neo-inositol (45): Trimethylsilyl triflate (94 mL, 115 mg, 0.51 mmol) was
added to a cooled (08C) suspension of the diol 44 (64 mg, 0.127 mmol) and
2,6-lutidine (37 mL, 34 mg, 0.32 mmol) in dimethoxymethane (0.65 mL).
The resulting solution was allowed to warm to room temperature. After
12 h, the mixture was diluted with diethyl ether (10 mL), washed with sat.
NaHCO3 (5 mL), 15% sodium bisulfate (2 Â 5 mL), sat. NaHCO3 again
(5 mL), and brine (5 mL), then dried (MgSO4) and concentrated to give
acetal 45 (70 mg, 100%) as a white foam that was used without further
purification. Rf ˆ 0.32 (2:1 hexanes/EtOAc); [a]2D3 ˆ ‡ 56.28 (c ˆ 0.55,
CHCl3); IR (neat): nÄ ˆ 1762 (s), 1726 (s), 1262 (s) cm 1; 1H NMR (CDCl3,
300 MHz): d ˆ 8.10 (d, J ˆ 7.7 Hz, 2H), 8.00 (d, J ˆ 7.7 Hz, 2H), 7.61 (m,
2H), 7.48 (m, 4H), 7.29 (m, 5H), 5.96 (dd, J ˆ 7.6, 2.2 Hz, 1H), 5.40 (t, J ˆ
2.5 Hz, 1H), 5.13 (s, 1H), 5.08 (dd, J ˆ 9.0, 7.8 Hz, 1H), 4.71 (d, J ˆ 15.1 Hz,
1H), 4.64 (s, 1H), 4.56 (dd, J ˆ 7.3, 1.7 Hz, 1H), 4.50 (dd, J ˆ 7.3, 2.2 Hz,
1H), 4.21 (dd, J ˆ 9.0, 2.9 Hz, 1H), 4.07 (d, J ˆ 15.1 Hz, 1H); 13C NMR
(CDCl3, 75 MHz): d ˆ 165.7, 164.8, 157.1, 135.3, 134.0, 133.6, 129.9, 129.0,
128.9, 128.5, 128.3, 128.2, 128.1, 95.4, 74.0, 73.0, 71.6, 65.8, 52.8, 46.4; HRMS
[8] For some examples towards conduritol B, see: a) H. Paulsen, W.
von Deyn, Liebigs Ann. Chem. 1987, 133; b) T. Akiyama, H. Shima,
M. Ohnari, T. Okazaki, S. Ozaki, Bull. Chem. Soc. Jpn. 1993, 66, 3760;
c) T. K. Park, S. J. Danishefsky, Tetrahedron Lett. 1994, 35, 2667; d) N.
Chida, M. Ohtsuka, K. Nakazawa, S. Ogawa, J. Org. Chem. 1991, 56,
2976; e) O. Arjona, A. de Dios, J. Plumet, B. Saez, Tetrahedron Lett.
1995, 36, 1319; f) Y. Watanabe, M. Mitani, S. Ozaki, Chem. Lett. 1987,
123; g) J. E. Innes, P. J. Edwards, S. V. Ley, J. Chem. Soc. Perkin Trans.
1 1997, 795; h) C. LeDrian, J. P. Vionnet, P. Vogel, Helv. Chim. Acta
1990, 73, 161; i) C. Sanfilippo, A. Patti, G. Nicolosi, Tetrahedron:
Asymmetry 1999, 10, 3273; j) M. Honzumi, K. Hiroya, T. Taniguchi, K.
Ogasawara, Chem. Commun. 1999, 1985. Also see: C. Sanfilippo, A.
Patti, G. Nicolosi, Tetrahedron: Asymmetry 2000, 11, 1043; O.
Plettenburg, S. Adelt, G. Vogel, H. J. Altenbach, Tetrahedron:
Asymmetry 2000, 11, 1057.
‡
calcd for C29H25NO8 [M] : 515.1580; found: 515.1563.
[9] C. R. Johnson, Acc. Chem. Res. 1998, 31, 3330. Also see: H. B.
Mereyala, B. R. Gaddam, J. Chem. Soc. Perkin Trans. 1 1994, 2187; L.
Dumortier, P. Luu, S. Bobbelaere, J. Van der Eycken, M. Vanderwalle,
Synlett 1992, 243.
[10] T. Hudlicky, H. Luna, H. F. Olivo, C. Andersen, T. Nugent, J. D. Price,
J. Chem. Soc. Perkin Trans. 1 1991, 2907.
[11] a) R. C. Pittenger, R. N. Wolfe, M. M. Hohen, P. N. Marks, W. A.
Daily, M. McGuire, Antibiot. Chemother. 1953, 3, 1268; b) R. L. Mann,
R. M. Gale, F. R. Van Abeele, Antibiot. Chemother. 1953, 3, 1279;
c) Y. Sumiki, G. Nakamura, M. Kawasaki, S. Yamashita, K. Anzai, K.
Isono, Y. Serizawa, Y. Tomiyama, S. Suzuki, J.Antibiot. 1955, 8, 170;
d) K. Isono, S. Yamashita, Y. Tomiyama, S. Suzuki, J. Antibiot. 1957,
10, 21; e) Y. Wakisaka, K. Koizumi, Y. Nishimoto, M. Kobayashi, N.
Tsuji, J. Antibiot. 1980, 33, 695.
[12] M. C. Guerrer, J. Modolell, Eur. J. Biochem. 1980, 107, 409.
[13] M. Yoshida, E. Takahashi, T. Uozumi, T. Beppu, Agric. Biol. Chem.
1986, 50, 143.
[14] a) S. Omura, A. Nakagawa, T. Fujimoto, K. Saito, K. Otoguro, J.
Antibiot. 1987, 40, 1619; b) A. Nakagawa, T. Fujimoto, S. Omura, J. C.
Walsh, R. L. Stotish, B. George, J. Antibiot. 1987, 40, 1627.
[15] R. L. Mann, D. O. Woolf, J. Am. Chem. Soc. 1957, 79, 120.
[16] K. Kakinuma, Y. Sakagami, Agric. Biol. Chem. 1978, 42, 279.
[17] N. Chida, M. Ohtsuka, K. Nakazawa, S. Ogawa, J. Org. Chem. 1991,
56, 2976.
(1S,2R,3R,4S,5R,6R)-2-Amino-2-deoxy-4,5-methylene-neo-inositol
(6):
Freshly cut Li metal (11.5 mg, 1.63 mmol, three pieces) was added to a
cooled ( 788C) solution of the oxazolidinone 45 (42 mg, 0.082 mmol) in
THF/NH3 (1:2, 1.5 mL). The flask was removed from the cold bath and
allowed to warm to 338C. Blue streaks started to form from the metal and
after ꢀ5 min. the solution turned completely blue. After another 5 min, the
solution was recooled to
788C and the reaction was quenched by
carefully adding water (0.5 mL). TLC analysis showed disappearance of the
starting material and formation of a new lower spot corresponding to the
debenzylated diol-oxazolidinone. Rf ˆ 0.30 (10:1 EtOAc/MeOH); IR
1
(neat): nÄ ˆ 3347 (brs), 1739 (s) cm
;
1H NMR (CD3OD, 300 MHz): d ˆ
5.12 (s, 1H), 4.71 (dd, J ˆ 6.9, 9.0 Hz, 1H), 4.66 (s, 1H), 4.47 (d, J ˆ 6.9 Hz,
1H), 4.23 (brs, 2H), 4.06 (dd, J ˆ 2.9, 9.0 Hz, 1H), 3.78 (brs, 1H).
The mixture was warmed to room temperature and the ammonia was
allowed to evaporate. The mixture was then heated in a 908C oil bath. After
1 h, no oxazolidinone was observed by TLC. The solution was cooled to
room temperature and applied directly to an ion-exchange column
‡
(Amberlite IRC-76, H form, ꢀ5 mL wet) that was eluted with water
and then 5% ammonium hydroxide to give the amino cyclitol 6 as an
amorphous solid (12 mg, 77%). Rf ˆ 0.37 (10:10:1 CH2Cl2/MeOH/
NH4OH); [a]2D2
ˆ
28.98 (c ˆ 0.85, H2O) [lit. [a]D22
ˆ
338 (c ˆ 1.97,
H2O),[11] [a]2D1
ˆ
368 (c ˆ 0.49, H2O)[16]]; IR (neat): nÄ ˆ 3347 (brs), 2884
(w), 1098 (s) cm 1; 1H NMR (D2O, with acetone as an internal standard at
d ˆ 2.08, 300 MHz): d ˆ 5.07 (s, 1H), 4.83 (s, 1H), 4.17 (dd, J ˆ 7.8, 5.1 Hz,
1H), 4.03 (m, 2H), 3.74 (dd, J ˆ 9.8, 3.7 Hz, 1H), 3.61 (dd, J ˆ 7.8, 3.4 Hz,
1H), 3.26 (t, J ˆ 3.4 Hz, 1H); 13C NMR (CD3OD, 75 MHz): d ˆ 95.7, 79.4,
78.7, 71.9, 71.7, 70.3, 56.1. These data were consistent with literature values.
[18] B. M. Trost, S. R. Pulley, J. Am. Chem. Soc. 1995, 117, 10143.
[19] a) B. M. Trost, E. J. Hembre, Tetrahedron Lett. 1999, 40, 219; b) B. M.
Trost, D. E. Patterson, E. J. Hembre, J. Am. Chem. Soc. 1999, 121,
10834.
[20] R. C. Cambie, N. D. Renner, P. S. Rutledge, P. D. Woodgate, Synth.
Commun. 1989, 19, 537.
[21] S. Knapp, R. M. Ornaf, K. E. Rodriques, J. Am. Chem. Soc. 1983, 105,
5494.
[22] B. A. McKittrick, B. Ganem, Tetrahedron Lett. 1985, 26, 4895.
[23] For an example directed towards aminocyclopentitols, see: W.
Shrader, B. Imperiali, Tetrahedron Lett. 1996, 37, 599.
[24] E. M. Burgess, C. L. Liotta, J. Org. Chem. 1981, 46, 1703. Also see:
A. S. Cieplak, Chem. Rev. 1999, 99, 1265.
Acknowledgement
We thank the National Institutes of Health, General Medical Sciences, for
their generous support of our programs. E.J.H. thanks the NIH for a
postdoctoral fellowship. Mass spectra were provided by the Mass Spec-
trometry Facility at the University of California/San Francisco supported
by the NIH Division of Research Resources.
[25] Y. Gao, K. B. Sharpless, J. Am. Chem. Soc. 1988, 110, 7538.
[26] B. M. Trost, R. C. Bunt, J. Am. Chem. Soc. 1994, 116, 4089.
[27] a) K. J. Lee, S. A. Boyd, N. S. Radin, Carbohydr. Res. 1985, 144, 148;
b) D. H. R. Barton, P. Dalko, S. D. Gero, Tetrahedron Lett. 1991, 32,
2471; c) R. Downham, P. J. Edwards, D. A. Entwistle, A. B. Hughes,
K. S. Kim, S. V. Ley, Tetrahedron: Asymmetry 1995, 6, 2403.
[1] D. A. Cox, K. Richardson, B. C. Ross in Topics in Antibiotic
Chemistry, Vol. 1 (Ed.: P. G. Sammes), Ellis Horwood, Sussex, UK,
1977, p. 5.
[2] S. Sakuda, A. Isogai, S.Matsumoto, A.Suzuki, Tetrahedron Lett. 1986,
27, 2475; S. Sakuda, A. Isogai, S. Matsumoto, K. Koseki, H. Kodama,
1628
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1628 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 8

Aminocyclohexitol Synthesis
1619±1629
Â
[28] a) A. E. Gal, J. P. Voorstad, J. Labelled Compd. Radiopharm. 1987, 24,
397; b) Z.-X. Guo, A. H. Haines, S. M. Pyke, S. G. Pyke, R. J. K.
Taylor, Carbohydr. Res. 1994, 264, 147.
[30] T. Esser, F. Farkas, S. Mangholz, U. Sequin, Tetrahedron 1994, 50,
3709.
[31] H. J. Altenbach, H. Setgelmeier, E. Vogel, Tetrahedron Lett. 1978,
[29] a) B. M. Trost, E. J. Hembre, Tetrahedron Lett. 1999, 40, 219; b) B. M.
Trost, D. E. Patterson, E. J. Hembre, J. Am. Chem. Soc. 1999, 121,
10834.
3333.
Received: October 23, 2000 [F2818]
Chem. Eur. J. 2001, 7, No. 8
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0708-1629 $ 17.50+.50/0
1629