Stereocontrol of 5,5-spiroketals in the synthesis of cephalosporolide H epimers

Article abstract of DOI:10.1021/ol102201z

A blueprint for controlling the stereochemistry of oxygenated 5,5-spiroketals using chelation effects is provided. Chelation specifically of zinc salts (other protic and Lewis acids were less effective) between the spiroketal oxygen and an appropriately positioned alcohol group overrides normal biases in the preparation of 5,5-spiroketals, as illustrated by the stereocontrolled synthesis of epimeric cephalosporolide H isomers. This study provides new and valuable information for prescribing the chirality of the stereogenic core of 5,5-spiroketals.

Full text of DOI:10.1021/ol102201z

ORGANIC
LETTERS
2010
Vol. 12, No. 20
4698-4701
Stereocontrol of 5,5-Spiroketals in the
Synthesis of Cephalosporolide H
Epimers
Sami F. Tlais and Gregory B. Dudley*
Department of Chemistry and Biochemistry, Florida State UniVersity, Tallahassee,
Florida 32306-4390, United States
gdudley@chem.fsu.edu
Received September 14, 2010
ABSTRACT
A blueprint for controlling the stereochemistry of oxygenated 5,5-spiroketals using chelation effects is provided. Chelation specifically of zinc
salts (other protic and Lewis acids were less effective) between the spiroketal oxygen and an appropriately positioned alcohol group overrides
normal biases in the preparation of 5,5-spiroketals, as illustrated by the stereocontrolled synthesis of epimeric cephalosporolide H isomers.
This study provides new and valuable information for prescribing the chirality of the stereogenic core of 5,5-spiroketals.
Spiroketals are found in many biologically active compounds
(cf. Figure 1).1 The rigid spiroketal framework offers
precision orientation of pendant functional groups, which
likely supports precise interactions in complex biochemical
systems. However, one must be able to install the core
spiroketal stereochemistry efficiently in order to take advan-
tage of its rigid three-dimensional shape. For stereocontrol
of spiroketals, anomeric effects can be predictably exploited
in 6,6- and 5,6-systems,2 but they are less reliable in 5,5-
systems.3-5
Consequently, natural 5,5-spiroketals can be extremely
challenging targets for stereoselective synthesis. In some
cases, such as halichondrin B (Figure 1), the natural epimer
(3) Anomeric stabilization can be often achieved by either diastereoi-
somer, compromising the predictive value and preparative utility of anomeric
effects for stereocontrol of 5,5-spiroketals. For thoughtful analyses of 5,5-
spiroketal stereochemistry in the ritterazine and cephalostatin natural
products, see: (a) Taber, D. F.; Joerger, J.-M. J. Org. Chem. 2007, 72, 3454.
(b) Frotner, K. C.; Kato, D.; Tanaka, Y.; Shair, M. D. J. Am. Chem. Soc.
2010, 132, 275. (c) Kim, S.; Sutton, S. C.; Guo, C.; LaCour, T. G.; Fuchs,
P. L. J. Am. Chem. Soc. 1999, 121, 2056.
(4) Other approaches to controlling the stereochemistry of 5,5-spiroket-
als: (a) Bueno, A. B.; Hegedus, L. S. J. Org. Chem. 1998, 63, 684. (b)
Cubero, I. I.; Plaza Lopez-Espinosa, M. T.; Kari, N. Carbohydr. Res. 1994,
261, 231. (c) Sharma, G. V. M.; Chander, A. S.; Goverdhan Reddy, V.;
Krishnudu, K.; Ramana Rao, M. H. V.; Kunwar, A. C. Tetrahedron Lett.
2000, 41, 1997. (d) Yin, B.-L.; Hu, T.-S.; Yue, H.-J.; Gao, Y.; Wu, W.-M.;
Wu, Y.-L. Synlett 2004, 306.
(1) (a) Perron, F.; Albizati, K. F. Chem. ReV. 1989, 89, 1617. (b) Aho,
J. E.; Pihko, P. M.; Rissa, T. K. Chem. ReV. 2005, 105, 4406. (c) Raju,
B. R.; Saikia, A. K. Molecules 2008, 13, 1942. (d) Blunt, J. W.; Copp,
B. R.; Hu, W.-P.; Munro, M. H. G.; Northcote, P. T.; Prinsep, M. R. Nat.
Prod. Rep. 2009, 26, 170.
(2) Reviews on anomeric effects: (a) Deslongchamps, P. Stereoelectronic
Effects in Organic Chemistry; Pergamon Press: Oxford, 1983. (b) Juaristi,
E.; Cuevas, G. The Anomeric Effect; CRC Press: Boca Raton, FL, 1994.
Selected methods for making 6,6-spiroketals under kinetic and/or chelation
control: (c) Reference 1b. (d) Lau, C. K.; Crumpler, S.; Macfarlane, K.;
Lee, F.; Berthelette, C. Synlett 2004, 2281. (e) Evans, D. A.; Coleman,
P. J.; Dias, L. C. Angew. Chem., Int. Ed. 1997, 36, 2738.
(5) There is some inconsistency relating to common spiroketal numbering
(i.e., [4.4]-spiroketal vs 5,5-spiroketal), likely arising from conflicting desires
to conform partially to IUPAC spirocycle nomenclature versus to adhere
to the common practice of counting the ring sizes independently. We are
following the common practice of identifying the ring sizes independently,
separated by a comma: 5,5-spiroketal refers to spiro-fused THF rings, and
6,6-spiroketal refers to spiro-fused THP (oxane) rings.
10.1021/ol102201z  2010 American Chemical Society
Published on Web 09/22/2010

facilitate the synthesis of complex and contra-thermodynamic
5,5-spiroketals. The ability to design substrate preferences
for either 5,5-spiroketal epimer will offer entry into new
regions of chemical space.9
We envisioned a strategy in which chelation of metal salts
across the two rings is used to control spiroketal stereo-
chemistry. Such metal-chelation effects have been docu-
mented in 6,6-spiroketals2 but to our knowledge not in 5,5-
spiroketals.10 Preliminary development of such a strategy for
controlling the stereochemistry of 5,5-spiroketals is described
herein. Finally, application of this strategy to the synthesis
of both spiroketal epimers of cephalosporolide H is pre-
sented.11
The central hypothesis, that chelation between a free
alcohol and the spiroketal oxygen of the adjoining ring will
override steric effects and determine which core diastereomer
predominates, and initial corroborating observations are
outlined in Scheme 1.
Scheme 1.
Epimerization of ꢀ-Hydroxylated 5,5-Spiroketalsa
Figure 1. Reported (truncated) structures of some 5,5-spiroketal
natural products.3-5
a Spiroketals I and II were epimerized using zinc chloride and
equilibrium ratios established by 1H NMR analysis. Other protic and Lewis
acids either promoted decomposition (camphorsulfonic acid, TiCl4,
BF3·OEt2) or failed to induce epimerization (MgCl2).
is preferred, but such substrate biases are not always clear
at the outset. The related cephalostatins and ritterazines, for
example, feature contra-thermodynamic 5,5-spiroketals, as
addressed in thoughtful studies by Shair, Taber, and Fuchs.3
Deslongchamps just reported the synthesis of the natural 5,5-
spiroketal hippuristanol, the thermodynamics of which are
solvent-dependent.6 In many cases, however, both 5,5-
spiroketals epimers are encountered within the same natural
product family (cf. symbiospirols, ascospiroketals, and
cephalosporolides).
Our studies focus on cephalosporolide H, which is isolated
as a single spiroketal epimer from the culture broth of the
marine fungus Penicillium sp.7 Cephalosporolide H has
potential as an anti-inflammatory agent by virtue of its
inhibitory activity against 3R-hydroxysteroid dehydrogenase
(3R-HSD).8
Initial experiments involved treating ꢀ-oxygenated 5,5-
spiroketals12 I and II with various protic and Lewis acids
(Scheme 1). Mild Lewis acids such as MgCl2 did not induce
epimerization, and strong Lewis acids including BF3·OEt2
and TiCl4 promoted double elimination to furan and/or
general decomposition. Protic acids has similar effects. In
contrast, zinc chloride (ZnCl2) effectively catalyzed the
equilibration between isomers a and b. In each case, the
major isomer (Ia and IIb)13 was the one capable of
participating in the hypothesized chelation interaction.
The role of chelation clearly emerges from experiments
on spiroketal I (Table 1), the alcohol stereochemistry of
which corresponds to cephalosporolide H. Either spiroketal
epimer (a or b) can be produced in g15:1 dr by enabling or
This Letter provides (i) a blueprint for controlling the
stereochemistry of oxygenated 5,5-spiroketals using chelation
effects and (ii) practical demonstration in target-oriented
synthesis. Prescribed access to 5,5-spiroketals will greatly
(9) For diversity-oriented synthesis of 6,6-spiroketals in which chelation
effects are cited, see: Moilanen, S. B.; Potuzak, J. S.; Tan, D. S. J. Am.
Chem. Soc. 2006, 128, 1792, and references therein.
(10) We speculate that this discrepancy arises from the tendency of
oxygenated 5,5-spiroketals to undergo acid-catalyzed dehydration to furans,
an issue that we address herein.
(6) Ravindar, K.; Reddy, M. S.; Lindqvist, L.; Pelletier, J.; Deslong-
champs, P. Org. Lett. 2010, published ASAP ahead of print, DOI: 10.102/
ol1019663.
(11) Synthesis of the spiroketal core of the cephalosporolides has been
reported twice, each time in a roughly 2:1 mixture: (a) Ramana, C. V.;
Suryawanshi, S. B.; Gonnade, R. G. J. Org. Chem. 2009, 74, 2842. (b)
Fernandes, R. A.; Ingle, A. B. Synlett 2010, 158.
(7) Li, X.; Yao, Y.; Zheng, Y.; Sattler, I.; Lin, W. Arch. Pharm. Res.
2007, 30, 812.
(12) The synthesis of spiroketals I and II follows closely the synthesis
of cephalosporolide H (vide infra). See Supporting Information.
(8) Penning, T. M. J. Pharm. Sci. 1985, 74, 651.
Org. Lett., Vol. 12, No. 20, 2010
4699

reduction with (S)-CBS reagent17 provided alcohol 3 exclu-
sively; the overall yield for the four-step sequence 2 f 3
was 63%.
Table 1. Role of Chelation in the Stereoselective Epimerization
of Ia
Propargyl alcohol 3 was subjected to the alkyne zipper
reaction18 and protected as a TBS ether to give terminal
alkyne 4. Coupling19 with (R)-1,2-epoxynonane (5)20 gave
internal alkyne 6, a key intermediate in the synthesis.
After considerable optimization (the details of which will
be reported elsewhere), it was found that gold(I) chloride in
MeOH21c induced cycloisomerization21 of alkyne 6 with
concomitant cleavage of the PMP acetal and TBS ether to
give 5,5-spiroketal 7 in 80% yield, but as a ca. 1:1 mixture
with spiroketal epimer 9. Notably, exposure of this mixture
to zinc chloride chelation effects for 8 h deliVered spiroketal
7 as a single diastereomer in 86% yield. Magnesium oxide
was included in this case as a protic acid scavenger.22 The
protic acid scavanger helps suppress acid-catalyzed dehydra-
tion to furan byproduct, which otherwise were detected in
small quantities.
entry
substrate
R1
R2
ratio (a:b)
1
2
3
4
I
H
H
H
TBS
TBS
15:1
3:1
1:10
1:>20
I-2
I-3
I-4
TBDPS
H
TBDPS
a Spiroketal I was selectively silylated at either or both hydroxy groups
and then epimerized using zinc chloride, and the equilibrium ratios were
established by H NMR analysis. See Supporting Information and ref 13.
1
blocking chelation (entries 1 and 4). Note that isomer a is
not observed when silyl groups are positioned to disrupt
chelation effects (entry 4), indicating that the reported
structure of cephalosporolide H may be contra-thermody-
namic.14 Considering the experiments recounted in entries
2 and 3, we conclude that a free hydroxyl group on the
spiroketal ring (R2 ) H) provides the dominant chelation
interaction, with a side-chain alcohol providing a secondary,
reinforcing role. Perhaps most instructive is to compare
entries 2 and 4: silylation changes the selectivity from
predominantly isomer a to exclusively isomer b. What
emerges from this study is a strategy for preparing either
isomer of the reported cephalosporolide H spiroketal core,
as described below.
Oxidation of spiroketal diol 7 (cf. Scheme 1, isomer a)
was expected to provide cephalosporolide H, but spectro-
scopic data for lactone 1 did not match that reported for the
natural product.7
Retreating in the synthetic sequence to internal alkyne 6,
cycloisomerization with bis-acetonitrile palladium(II) chlo-
ride in CH3CN provided 5,5-spiroketal 8 with the TBS ether
intact, this time as a 9:1 mixture favoring the opposite
(thermodynamic) spiroketal stereochemistry (cf. Scheme 1,
isomer b). Desilylation with TBAF provided spiroketal 9,
still in a 9:1 ratio over 7. TEMPO-catalyzed oxidation gave
The synthesis of epimeric cephalosporolide H spiroketals
began with alcohol 215 (Scheme 2): Swern oxidation and
propynyl Grignard addition gave alcohol 3 as the minor
diastereomer in a 1:3 mixture. Note that this Grignard
addition is an unusual example of Felkin selectivity in which
the quaternary carbon is acting as the “medium” substituent,
despite the potential for chelation-control involving the acetal
oxygen (Figure 2).16 Reoxidation and matched asymmetric
1
(13) Spiroketal stereochemistry assigned by H NMR using diagnostic
NMR resonances of the spiroketal methyl group. To quote from ref 13c
(below), “CH3... when cis to the oxygen of the second ring resonates at
lower field than when trans to the oxygen... of the second ring.” For further
discussion and analysis of this spectroscopic trend, see Supporting Informa-
tion, ref 11a and (a) Brimble, M. A.; Bryant, C. J. Chem. Commum. 2006,
4506. (b) Occhiato, E. G.; Guarna, A.; De Sarlo, F.; Scarpi, D. Tetrahedron:
Asymmetry 1995, 6, 2971. (c) Nishiyama, T.; Woodhall, J. F.; Lawson,
E. N.; Kitching, W. J. Org. Chem. 1989, 54, 2183. For the seminal
identification of this NMR correlation, see: (d) Francke, W.; Reith, W.;
Sinnwell, V. Chem. Ber. 1980, 113, 2686. For independent confirmation
of this trend by X-ray analysis, see ref 11a.
(14) This preference for the large tert-alkyl group to orient trans to the
spiroketal oxygen is not unusual in 5,5-spiroketal systems, although this
trend is not sufficiently reliable as to be predictive. See ref 13a-c.
(15) Alcohol 2 is available in two steps from (-)-pantolactone; see:
Shiina, I.; Shibata, J.; Ibuka, R.; Imai, Y.; Mukaiyama, T. Bull. Chem. Soc.
Jpn. 2001, 74, 113.
(16) (a) Tlais, S. F.; Clark, R. J.; Dudley, G. B. Molecules 2009, 14,
5216. (b) Changing the reaction solvent from THF to ethyl ether did not
alter the product ratio significantly.
(17) Parker, K. A.; Ledeboer, M. W. J. Org. Chem. 1996, 61, 3214.
(18) Scheerer, J. R.; Lawrence, J. F.; Wang, G. C.; Evans, D. A. J. Am.
Chem. Soc. 2007, 129, 8968.
(19) Trost, B. M.; Weiss, A. H. Angew. Chem., Int. Ed. 2007, 46, 7664.
(20) (a) Frost, C. G.; Penrose, S. D.; Gleave, R. Org. Biomol. Chem.
2008, 6, 4340. (b) Savle, P. S.; Lamoreaux, M. J.; Berry, F.; Gandour, R. D.
Tetrahedron: Asymmetry 1998, 9, 1843.
(21) (a) Utimoto, K. Pure. Appl. Chem. 1983, 55, 1845. (b) Liu, B.; De
Brabander, J. K. Org. Lett. 2006, 8, 4907. (c) Li, Y.; Zhou, F.; Forsyth, C.
Angew. Chem., Int. Ed. 2007, 46, 279. (d) Aponick, A.; Li, C.-Y.; Palmes,
J. A. Org. Lett. 2008, 11, 121.
(22) For previous applications of magnesium oxide as an acid scavenger,
see: (a) Espino, C. G.; Wehn, P. M.; Chow, J.; DuBois, J. J. Am. Chem.
Soc. 2001, 123, 6935. (b) Poon, K. W. C.; Dudley, G. B. J. Org. Chem.
2006, 71, 3923.
Figure 2. Felkin-Anh models for formation of 3 (desired) and
epi-3.
4700
Org. Lett., Vol. 12, No. 20, 2010

Scheme 2.
Synthesis of the Reported Structure of Cephalosporolide H (1) and Its Spiroketal Epimer (10)a
a See Supporting Information for details.
rise to diastereomerically pure lactone 1023 in 68% yield;
spectroscopic data aligned better with data reported for
cephalosporolide H.24,25 Accordingly, we suggest that the
reported stereochemical assignment for cephalosporolide H
is incorrect and that the spiroketal stereochemistry for
cephalosporolide I7 and the penisporolides26 should also be
reconsidered.
To summarize, we have shown that chelation of zinc salts
between the spiroketal oxygen and an appropriately posi-
tioned alcohol group overrides normal biases in the prepara-
tion of 5,5-spiroketals, as illustrated by the stereocontrolled
synthesis of both the reported structure of cephalosporolide
H (1) and its spiroketal epimer (10). The use of zinc salts is
important: weaker acids do not enable isomerization, and
stronger acids promote decomposition, including especially
elimination to furans. This study provides new and valuable
information for prescribing the chirality of the stereogenic
core of 5,5-spiroketals.
(23) The oxidation 9 f 10 is significantly faster than the oxidation
7 f 1, facilitating production of the more stable lactone (10). For a related
example, see: Dudley, G. B.; Engel, D. A.; Ghiviriga, I.; Lam, H.; Poon,
K. W. C.; Singletary, J. A. Org. Lett. 2007, 9, 2839.
(24) Our data for 10 are in close agreement with the data reported for
the natural product, and our data for synthetic 1 are not. However, attempts
to secure an authentic sample and/or copies of original NMR spectra were
unsuccessful, so positive conclusions on the correct structure of natural
cephalosporolide H cannot be made at this time. See Supporting Information
for a table comparing the various NMR data.
Acknowledgment. This research is supported by a grant
from the National Science Foundation (NSF-CHE 0749918).
We thank Dr. Tom Gedris (FSU) for assistance with NMR
analysis.
(25) Our stereochemical assignment of spiroketals 1 and 10 is based on
NMR correlations with the spiroketal resonances of cephalosporolides E
and F, for which X-ray crystallographic data is available, and by analogy
to the model systems I and II (see ref 13). Direct spectroscopic analysis
using NOE and NOESY experiments conducted on both spiroketal epimers
was inconclusive. The original spiroketal stereochemical assignment of
cephalosporolide H (ref 7) was made on the basis of NOESY crosspeaks
that we observe in both epimers. A full analysis of the factors supporting
our structural conclusions is provided in Supporting Information.
(26) Li, X.; Sattler, I.; Lin, W. J. Antibiot. 2007, 60, 191.
Supporting Information Available: Experimental pro-
cedures, characterization data, and copies of NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL102201Z
Org. Lett., Vol. 12, No. 20, 2010
4701