Design of chiral auxiliaries for the allene ether Nazarov cyclization

Article abstract of DOI:10.1021/ja069342g

A 1,3- or a 1,4-cis axial tert-butyldimethylsilyloxy substituent on the pyranose derived chiral auxiliary for the allene ether Nazarov cyclization leads to products in high optical purity. α-Pyranose and β-pyranose derived auxiliaries lead to enantiomeric products. Copyright

Full text of DOI:10.1021/ja069342g

Published on Web 04/11/2007
Design of Chiral Auxiliaries for the Allene Ether Nazarov Cyclization
April R. Banaag† and Marcus A. Tius*,†,‡
Department of Chemistry, UniVersity of Hawaii, 2545 The Mall, Honolulu, Hawaii 96822, and Cancer Research
Center of Hawaii, 1236 Lauhala Street, Honolulu, Hawaii 96813
Received December 28, 2006; E-mail: tius@gold.chem.hawaii.edu
The allene ether version1 of the Nazarov cyclization2,3 takes place
Scheme 1
under very mild conditions and is distinguished by being highly
suitable for asymmetric synthesis.4 For example, lithioallene 1 (eq
1), prepared by deprotonating the hydrocarbon with n-butyllithium,
adds to morpholino enamide 2. Mild acidic workup presumably
leads to allenyl vinyl ketone 3 that undergoes spontaneous cycliza-
tion to R-hydroxycyclopentenone 4 with loss of R1 as a stable cation.
High levels of asymmetry (er up to 96.5/3.5) can be induced in
products 4 through choice of chiral auxiliary R1.5 Since the auxiliary
is lost during the cyclization and no added step is needed for its
cleavage, it is traceless.
information from auxiliary to product through electron pair donation
to the developing pentadienyl cation.4a,8 This has the effect of
limiting the conformational mobility of the cation and enforcing
proximity between the auxiliary and the developing five-membered
ring. The pyran oxygen atom plays a critical role since the
We recently described our results with 2-deoxy-D-glucose derived
lithioallene 5 (Scheme 1), which is an excellent reagent for the
allene ether Nazarov reaction.5 Cyclic products from 5 were formed
with er’s of 92.5/7.5 to 96.5/3.5. We were puzzled by the fact that
the tri-OTBS (TBS ) tert-butyldimethylsilyl) reagent 5 led to
products of much higher optical purity than the corresponding
cyclohexyl version of 11 leads to (R)-9 in low yield in ca. 55.5/
44.5 er. For this model to be valid, the 3,4,5-triaxial conformation
of the pyran ring in 7 must be energetically accessible,9 and one
trimethoxy lithioallene, or the lithioallene that is derived from
must assume a late transition state for the cyclization.
permethylated R-D-glucose. We rationalized this by postulating a
change in conformation for the pyranose derived chiral auxiliary
during the stereochemistry-determining step.6 In this communica-
If this model is valid, then one can make the following
predictions. The trans equatorial group at C-3 in 7 will not exert
much of an effect on the optical purity of product 9, whereas the
tion, we provide additional evidence to support our hypothesis and
OTBS group at C-4 will. Furthermore, locking the C-4 substituent
have designed greatly improved chiral auxiliaries for the allene ether
in the equatorial position of the pyran ring will lead to erosion of
Nazarov cyclization on this basis.
the optical purity of the product by preventing its close approach
Woerpel has demonstrated that the groups at C-3 and C-4 exert
to the pentadienyl cation from taking place. All three predictions
a large influence on the conformational preferences of tetrahydro-
pyran oxocarbenium ions relative to their uncharged precursors.6,7
Specifically, C-3 and C-4 alkoxy groups have a pseudoaxial
preference in the oxocarbenium ion. In the case of the Nazarov
cyclization that is outlined in Scheme 1, nucleophilic addition of 5
to 6 probably gives intermediate 7 that leads to transition state 8
during workup with acid. The developing positive charge at C-1
causes inversion of the pyranose ring to occur, resulting in axial
are borne out. To test the first prediction, lithioallene 11, lacking
a C-3 OTBS group, was allowed to react with 6 to produce (R)-9
in 94/6 er, within experimental error the same result that was
obtained with 5. It is therefore clear that the silyloxy group at C-3
does not influence the stereochemical outcome of the cyclization.
The same cannot be said for the group at C-4 since the reaction of
2,4-dideoxy compound 12 with 6 led to marked attenuation of the
optical purity of (R)-9 (86.5/13.5 er). Therefore, the equatorial C-4
placement of the C-4 OTBS group. Steric shielding of the back
group must be important, and preventing its close approach to the
face of the pentadienyl cation by the group at C-4 forces conrotation
pentadienyl cation in the stereochemistry-determining transition state
to take place in the counterclockwise direction as shown in Scheme
(see 8) by conformationally locking the pyran ring also leads to
1, leading to (R)-9 in an er of 93/7 with loss of the chiral auxiliary
erosion of the optical purity of 9: Lithioallene 13 led to (R)-9 in
as pyrylium species 10. According to this model, the pyranyl oxygen
atom plays a pivotal role in the transmission of stereochemical
86.5/13.5 er.
These results instill some confidence in the validity of the model.
Since the model requires a 1,3-cis axial interaction in the transition
state to influence the stereochemical outcome of the cyclization,
† University of Hawaii.
‡ Cancer Research Center of Hawaii.
9
5328
J. AM. CHEM. SOC. 2007, 129, 5328-5329
10.1021/ja069342g CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
one can go a step further and predict that conformationally locked
lithioallene 14, which incorporates an axial C-3 OTBS group, should
be an effective reagent. This also proved to be the case, and the
reaction of 14 with 6 led to (R)-9 in 95/5 er.
thorough screen of 14 and 16 against a collection of enamides to
determine the scope of their cyclization chemistry is planned for
future work.
The analysis presented above is overly simple, as it attributes
the stereochemical outcome of the Nazarov cyclization to a single
steric interaction, to the exclusion of all other contributory factors.
Nevertheless, the results strongly suggest that the dominant
stereochemical factor has been identified and that it can be used in
the design of even more effective chiral auxiliaries for the Nazarov
cyclization.
The absolute stereochemistry of the Nazarov cyclization product
can be inverted by simply changing stereochemistry at the anomeric
carbon atom.4c This allows both enantiomeric series of cyclopen-
tenones to be derived from D sugars. We therefore set out to design
â-pyranose derived chiral auxiliaries based on the model that has
been described above. Lithioallene 15, derived from â-2-deoxy-D-
glucose, and lithioallene 16, derived from â-2-deoxy-D-galactose,
would present cis axial groups at both C-3 and at C-5 if
conformational inversion were to take place in the stereochemistry-
determining step. Both axial groups would be expected to influence
the absolute stereochemistry of product in the same way. In fact,
the reaction of 15 with 6 led to (S)-9 in only 11/89 er, whereas
lithioallene 16 led to (S)-9 in 4/96 er. The results from 15 and 16
suggest that inversion of the pyran ring does not take place in these
â-pyrans, or they would have given product of similar optical purity.
The reason for the apparent difference in conformational behavior
of R- and â-pyrans is not obvious but may be due to the absence
of destabilizing 1,3-diaxial interactions involving the large C-1
substituent in the â series. The absolute stereochemistry of product
from 16 can be predicted according to the transition state 17
(Scheme 2). Electron pair donation by the pyran oxygen atom to
the developing pentadienyl cation restricts its conformational
mobility and brings it close to the axial C-4 OTBS group. The
buttressing effect of this group forces the conrotation in 17 to take
place in the counterclockwise direction, as shown in Scheme 2,
leading to cyclopentenone (S)-9 with loss of pyrylium ion 18. These
results appear to be congruent with Woerpel’s conformational
analysis of pyranose derived pyrylium ions.10
Acknowledgment. We thank the Department of Defense Breast
Cancer Research Program (DAMD17-03-1-0685) and the NIH
(GM57873) for generous support.
Supporting Information Available: Complete experimental pro-
1
cedures for the preparation of 11-16 and 9, analytical data, and H
and 13C NMR spectra for 11-16. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
(1) Tius, M. A. Acc. Chem. Res. 2003, 36, 284-290.
(2) For reviews of the Nazarov reaction, see: (a) Harmata, M. Chemtracts
2004, 17, 416-435. (b) Frontier, A. J.; Collison, C. Tetrahedron 2005,
61, 7577-7606. (c) Pellissier, H. Tetrahedron 2005, 61, 6479-6517. (d)
Tius, M. A. Eur. J. Org. Chem. 2005, 2193-2206.
(3) For recent examples of the Nazarov reaction, see: (a) Liang, G.; Trauner,
D. J. Am. Chem. Soc. 2004, 126, 9544-9545. (b) Janka, M.; He, W.;
Haedicke, I. E.; Fronczek, F. R.; Frontier, A. J.; Eisenberg, R. J. Am.
Chem. Soc. 2006, 128, 5312-5313. (c) Grant, T. N.; West, F. G. J. Am.
Chem. Soc. 2006, 128, 9348-9349.
(4) (a) Harrington, P. E.; Murai, T.; Chu, C.; Tius, M. A. J. Am. Chem. Soc.
2002, 124, 10091-10100. (b) Harrington, P. E.; Tius, M. A. J. Am. Chem.
Soc. 2001, 123, 8509-8514. (c) Harrington, P. E.; Tius, M. A. Org. Lett.
2000, 2, 2447-2450.
(5) delos Santos, D. B.; Banaag, A.; Tius, M. A. Org. Lett. 2006, 8, 2579-
2582.
(6) Ayala, L.; Lucero, C. G.; Romero, J. A. C.; Tabacco, S. A.; Woerpel, K.
A. J. Am. Chem. Soc. 2003, 125, 15521-15528.
(7) Shenoy, S. R.; Woerpel, K. A. Org. Lett. 2005, 7, 1157-1160.
(8) Our observations can also be rationalized by invoking a charge-dipole
interaction between cation and pyranyl oxygen atom and the exo anomeric
effect. We thank Professor William R. Roush for suggesting this.
(9) Evidence suggests that this is the case. See: (a) Tius, M. A.; Busch-
Petersen, J. Tetrahedron Lett. 1994, 35, 5181-5184. (b) Yamada, H.;
Tanigakiuchi, K.; Nagao, K.; Okajima, K.; Mukae, T. Tetrahedron Lett.
2004, 45, 5615-5618.
Lithioallene 5 had been screened against morpholino enamide 6
and eight other 2,3-disubstituted morpholino enamides, leading to
products in er’s ranging from 92.5/7.5 to 96.5/3.5. Enamide 6 was
by no means the best substrate for 5, and it is unlikely that it will
prove to be the optimum substrate for lithioallenes 14 and 16. A
(10) Lucero, C. G.; Woerpel, K. A. J. Org. Chem. 2006, 71, 2641-2647.
JA069342G
9
J. AM. CHEM. SOC. VOL. 129, NO. 17, 2007 5329