Total Synthesis of Ionomycin Using Ring-Opening Strategies

Article abstract of DOI:10.1021/ol025872f

(Matrix Presented) The total synthesis of the polyether antibiotic ionomycin, a calcium ionophore, is described. The synthesis demonstrates the utility of ring-opening methodologies as applied to the synthesis of polypropionate and deoxypolypropionate subunits, which are found in two of the four fragments in the synthesis.

Full text of DOI:10.1021/ol025872f

ORGANIC
LETTERS
2002
Vol. 4, No. 11
1879-1882
Total Synthesis of Ionomycin Using
Ring-Opening Strategies
Mark Lautens,* John T. Colucci, Sheldon Hiebert, Nicholas D. Smith, and
Giliane Bouchain
DaVenport Research Laboratories, Department of Chemistry, UniVersity of Toronto,
80 St. George St., Toronto, Ontario M5S 3H6, Canada
mlautens@chem.utoronto.ca
Received March 13, 2002
ABSTRACT
The total synthesis of the polyether antibiotic ionomycin, a calcium ionophore, is described. The synthesis demonstrates the utility of ring-
opening methodologies as applied to the synthesis of polypropionate and deoxypolypropionate subunits, which are found in two of the four
fragments in the synthesis.
Ionomycin, isolated in 1978 from Streptomyces conglobatus,1
belongs to a class of natural products known as polyether
antibiotics. The ability to chelate inorganic cations and
transport them across lipid membranes has made these
“ionophores” of significant interest. Ionomycin is unique
among this family because of its high affinity for binding
calcium and as such has been used extensively in neuro-
chemistry as a tool for studying the effects of increasing
intracellular Ca2+ concentrations.1c
The structure of calcium-bound ionomycin was elucidated
by Gougoutas and co-workers in 1979.2 It is highly oxygen-
ated, as is typical of the polyether antibiotics. The presence
of the two acidic functionalities, the terminal carboxylic acid
and the â-diketone, allows the formation of the 1:1 complex
with Ca2+. It also contains three hydroxyl groups and two
tetrahydrofuran rings in addition to 14 stereogenic centers,
making ionomycin a challenging synthetic target.
Two previous total syntheses have been published by the
Evans and Hanessian labs.3 Both groups used synthetic
strategies they developed for the synthesis of polypropionate
and deoxypolypropionate subunits. Evans used the well-
developed chiral enolate chemistry, and Hanessian applied
the chiron approach using L-glutamic acid as the intial source
of chiral information. Our goal was to show the utility of
the ring-opening of oxabicyclic alkenes as an alternative and/
or complementary approach to the synthesis of these
structural motifs.
Our retrosynthetic analysis of ionomycin is similar to that
of the Evans and Hanessian syntheses. The three points of
disconnection are at the â-diketone, the trans olefin, and the
first tetrahydrofuran ring (Scheme 1). We realized that the
presence of the 1,3 syn-dimethyl moiety, which is present
in three of the four fragments, made this molecule ideal for
(2) Toeplitz, B. K.; Cohen, A. I.; Funke, P. T.; Parker, W. L.; Gougoutas,
J. Z. J. Am. Chem. Soc. 1979, 101, 3344.
(1) (a) Liu, C.-M.; Hermann, T. E. J. Biol. Chem. 1978, 253, 5892. (b)
Liu, W.-C.; Slusarchyk, D. S.; Astle, G.; Trejo, W. H.; Brown, W.; Meyers,
E. J. Antibiot. 1978, 31, 815. (c) Parpura, V.; Basrsky, T. A.; Liu, F.;
Jeftinija, K.; Jeftinija, S.; Haydon, P. G. Nature 1994, 369, 744.
(3) (a) Dow, R. L. Ph.D. Dissertation, Harvard University, 1985. (b)
Evans, D. A.; Dow, R. L.; Shih, T. L.; Takacs, J. M.; Zahler, R. J. Am.
Chem. Soc. 1990, 112, 5290. (c) Hanessian, S.; Cooke, N. G.; DeHoff, B.;
Sakito, Y. J. Am. Chem. Soc. 1990, 112, 5276.
10.1021/ol025872f CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/08/2002

the bridging oxygen bond, cycloheptenes with four or five
contiguous stereocenters can be obtained.
Scheme 1
The utility of the asymmetric reductive ring opening is
powerfully shown in the synthesis of the C17-C23 fragment
(Scheme 2), where three of the four stereocenters in the
Scheme 2a
a (A) 5 mol % Ni(COD)2, 10 mol % (S)-BINAP, toluene, 65 °C,
1.1 equiv of DIBAL-H (added over 20 h); (B) (a) DMSO, (COCl)2,
NEt3, CH2Cl2, -78 °C; (b) PhMe, DIBAL-H, -78 °C; (c) THF,
KHMDS, PMBCl; (C) O3, MeOH/CH2Cl2, -78 °C then NaBH4,
rt; (D) CH2Cl2, DDQ, mol sieves, rt; (E) (a) TrCl, NEt3, DMAP,
CH2Cl2; (b) CH2Cl2, DIBAL-H, -78 to O °C; (F) DMSO, (COCl)2,
NEt3, CH2Cl2, -78 °C.
the use of the ring-opening methodologies developed in our
labs. In particular, fragments 3 and 5 can both be obtained
from the [3.2.1] oxabicyclic alkene 8.
Fragment 3 can be synthesized from 6, which is readily
prepared from an asymmetric addition of a hydride to 8. The
asymmetric reductive ring opening was developed in our labs
using a nickel-catalyzed addition of DIBAL-H to the
oxabicycle.4 Likewise, fragment 5 can be obtained from 7,
which comes from an asymmetric addition of a methyl group
to 8. A palladium-catalyzed asymmetric addition of a
dialkylzinc to an oxabicyclic alkene was recently developed
in our group5 and could be used to carry out this transforma-
tion.
The [3.2.1] oxabicyclic alkenes are easily prepared by a
[4 + 3] cycloaddition between furan and an oxyallylcation.6
These substrates have been widely used to gain access to
both cyclic and acyclic molecules.7 Their rigid bicyclic
structure can be used to introduce functional groups in a
highly stereoselective manner. If this functional group
introduction occurs in an SN2′ fashion with ring opening of
fragment are set using the ring-opening reaction. A slow
addition of DIBAL-H to 9 and the Ni(COD)2/(S)-BINAP
catalyst in toluene at 65 °C gave 10 in 95% yield and 93-
95% ee. The configuration of the hydroxyl at C21 needed to
be inverted in order to set the final stereocenter. Swern
oxidation and then reduction with DIBAL-H at -78 °C,
which proceeded with >30:1 selectivity, readily achieved
this goal. The hydroxyl was then protected as a PMB ether
to give 11 in 82% for the three steps.
The cycloheptene ring was then cleaved by ozonolysis to
the dialdehyde, and after reductive workup diol 12 was
obtained in 85% yield. The critical task of differentiating
the two ends of the acyclic subunit was then addressed. The
problem was solved by employing a strategy previously used
by Oikawa,8 who showed that oxidation of the benzylic
position of a p-methoxybenzyl ether under anhydrous condi-
tions in the presence of an appropriately positioned hydroxyl
group would result in cyclization onto the cation to form an
acetal. When diol 12 was treated with DDQ in anhydrous
dichloromethane, selective cyclization occurred to afford the
PMP acetal 13 in 93% yield.
(4) (a) Lautens, M.; Chiu, P.; Ma, S.; Rovis, T. J. Am. Chem. Soc. 1995,
117, 532. (b) Lautens, M.; Rovis, T. J. Am. Chem. Soc. 1997, 119, 11090.
(5) (a) Lautens, M.; Renaud, J.-L.; Hiebert, S. J. Am. Chem. Soc. 2000,
122, 1804. (b) Lautens, M.; Hiebert, S.; Renaud, J.-L. Org. Lett. 2000, 2,
1971. (c) Lautens, M.; Hiebert, S.; Renaud, J.-L. J. Am. Chem. Soc. 2001,
123, 6834.
(6) Lubineau, A.; Bouchain, G. Tetrahedron Lett. 1997, 38, 8031. For
reviews on [4 + 3] cycloadditions, see: (a) Hosomi, A.; Tominaga, Y.
ComprehensiVe Organic Synthesis; Trost, B., Fleming, I., Paquette, L. A.,
Eds.; Pergamon: Oxford, U.K., 1991; Vol. 5, Chapter 5.1, p 593. (b)
Hoffmann, H. Angew. Chem., Int. Ed. Engl. 1984, 23, 1. (c) Noyori, R.;
Hayakawa, Y. Org. React. 1983, 29, 163.
With the ends of the acyclic fragment differentiated the
other hydroxyl could now be protected as a trityl ether.
(7) For reviews on SN2′ and SN2′-like ring opening of oxa-n-cyclo
systems, see: (a) Vogel, P.; Cossy, J.; Plumet, P.; Arjona, O. Tetrahedron
1999, 55, 13521. (b) Woo, S.; Keay, B. Synthesis 1996, 669. (c) Lautens,
M. Synlett 1993, 177.
(8) Oikawa, Y.; Yoshioka, T.; Yonemitsu, O. Tetrahedron Lett. 1982,
23, 885.
1880
Org. Lett., Vol. 4, No. 11, 2002

Regioselective reduction of the acetal under Takano’s
protocol9 using DIBAL-H gave the alcohol 14 in 94% yield
for the two steps. Finally, Swern oxidation of the primary
alcohol gave aldehyde 3 in 90% yield and completed the
synthesis of the C17-C23 fragment in 12 steps from furan in
19.6% overall yield.
Scheme 3a
Whereas the stereochemistry in the C17-C23 fragment was
set by an asymmetric addition of hydride to the oxabicycle,
the Cl-C10 fragment requires the introduction of a methyl
group at the C4 position. This is accomplished using the
asymmetric nucleophillic ring opening catalyzed by a chiral
palladium catalyst.
The chiral catalyst, formed from a 1:1 mixture of Pd(CH3-
CN)2Cl2 and the chiral ligand (R)-iPr-(R)DIPOF,10 was used
together with 10 mol % Zn(OTf)2 for the addition of Me2Zn
to oxabicycle 15 in refluxing dichloroethane. The product
cycloheptendiol 16 containing the 1,3,5-trimethyl subunit
with complete diastereocontrol was obtained in 80% yield
and 94% ee (Scheme 3). The two hydroxyl groups could
now be selectively protected as the tert-butyldimethylsilyl
ether and p-methoxybenzyl ether in 92% and 95% yields,
respectively, to give 17.
The cycloheptene olefin was now cleaved by ozonolysis
to give the acyclic fragment 18 in 90% yield. The dif-
ferentiation of the two hydroxyl groups was carried out as
in the C17-C23 fragment. Treatment of 18 with DDQ in
anhydrous dichloromethane gave selective formation of a
single PMP acetal and after TBAF deprotection of the silyl
ether gave diol 19 in 90% for the two steps. This diol was
treated with thiocarbonyl diimidazole (TCDI) to give a 95%
yield of the cyclic thiocarbonate 20 in order to set up the
deoxygenation. Barton has shown that these cyclic thiocar-
bonates undergo radical-induced reduction to yield prefer-
entially the primary alcohol as a result of the greater stability
of the secondary radical upon collapse of the intermediate
during the deoxygenation.11 Employing Barton’s protocol the
primary alcohol 21 was obtained in 84% yield.
A two-carbon extension was then carried out using a
stabilized Wittig reagent on the aldehyde obtained by TPAP
oxidation of alcohol 21. The overall yield of 22 was 77%.
Hydrogenation catalyzed by Pd/C led to reduction of the
olefin and removal of the PMP acetal to give diol 23 in 98%
yield.
Removal of the remaining secondary hydroxyl group was
achieved in an identical fashion as before using Barton’s
protocol. Formation of the thiocarbonate took place in 90%
yield, while reduction gave the primary alcohol 24 in 80%
yield. Finally, conversion of the primary alcohol to the
methyl ketone 5 was achieved in three steps. Oxidation to
the aldehyde by TPAP/NMO was followed by addition of
MeMgBr in THF at -78 °C to give a mixture of isomers. A
a (A) 5 mol % Pd[(R)-iPr-(R)-DIPOF]Cl2, 10 mol % Zn(OTf)2,
2.5 equiv of Me2Zn, dichloroethane, reflux; (B) (a) TBDMSCl,
imidizole, DMF; (b) NaH, PMBBr, Bu4NI (cat.), DMF; (C) O3,
MeOH/CH2Cl2, -78 °C then NaBH4, rt; (D) (a) CH2Cl2, DDQ,
mol sieves, rt; (b) TBAF, THF, rt; (E) TCDI, THF, 50 °C; (F)
Bu3SnH, AIBN (cat.), toluene, reflux; (G) (a) TPAP (cat.), NMO,
CH2Cl2, rt; (b) MeO2CCHdPPh3; (H) H2, Pd(OH)2, MeOH; (I)
(a) TCDI, THF, 50 °C; (b) Bu3SnH, AIBN (cat.), toluene, reflux;
(J) (a) TPAP (cat.), NMO; (b) MeMgBr, THF; (c) Dess-Martin
oxidation.
Dess-Martin oxidation gave ketone 5 and completed the
synthesis of the Cl-C10 fragment in 18 steps from furan in
6.9% yield.
With all four fragments in hand efficient methods of
coupling were explored.12 The formation of the C23-C24 bond
was carried out using a sulfone anion addition to aldehyde
3 (Scheme 4). The product 25 was obtained as a mixture of
diastereomers, and the phenyl sulfone could now be removed
by a two-step oxidation/R-reduction sequence. Oxidation to
ketone 26 was accomplished by a TPAP/NMO oxidation,
and then SmI2 was used to reductively remove the R-phenyl
sulfone13 to give 27 in 61% for the three steps.
Formation of the trans tetrahydrofuranyl ring was pro-
jected to occur via a diastereoselective substrate-controlled
reduction of the ketone followed by activation and intramo-
lecular SN2 etherification. For the reduction we chose to use
a diastereoselective samarium-mediated Tishchenko reaction
developed by Evans.14 This reaction involves the reduction
of â-hydroxy ketones using SmI2 in the presence of benzal-
dehyde (as the hydride source) to yield benzyloxy alcohols
with high anti selectivities (typically 90-99%).
(9) Takano, S.; Akiyama, M.; Sato, S.; Ogasawara, K. Chem. Lett. 1983,
1593.
(10) For the synthesis of 2(R)-(diphenylphosphino)-1-[4(R)-isopropyl-
2-oxazolinyl]ferrocene (abbreviated as i-Pr-DIPOF), see: (a) Richards, C.;
Locke, A. Tetrahedron: Asymmetry 1998, 9, 2377. (b) Nishibayashi, Y.;
Uemura, S. Synlett 1995, 79.
(11) (a) Barton, D. H. R.; Subramanian, R. J. Chem. Soc., Perkin Trans.
1 1977, 1718. (b) Barton, D. H. R.; McCombie, S. J. Chem. Soc., Perkin
Trans. 1 1975, 1574.
(12) The synthesis of the other fragments along with other approaches
to the coupling strategy will be discussed in a forthcoming paper.
(13) Molander, G. A.; Hahn, G. J. Org. Chem. 1986, 51, 1135.
(14) Evans, D. A.; Hoveyda, A. H. J. Am. Chem. Soc. 1990, 112, 6447.
Org. Lett., Vol. 4, No. 11, 2002
1881

Scheme 4a
a (A) 2 + BuLi then 3, THF, -78 °C; (B) TPAP (cat.), NMO, mol sieves, CH2Cl2; (C) SmI2, MeOH, THF, -78 °C; (D) DDQ, H2O,
CH2Cl2; (E) 26 mol % SmI2, 5 equiv of PhCHO, THF, -10 °C; (F) (a) TsCl, DMAP (cat.), CH2Cl2; (b) 1 M HCl, THF, 0 °C; (c) NaH,
THF, 50 °C; (G) (a) Pd(OH)2, C6H12, EtOH, reflux; (b) TPAP (cat.), NMO, mol sieves, CH2Cl2; (H) KHMDS, 4, then add to 31, THF, -78
°C to rt; (I) DDQ, H2O, CH2Cl2, O°C; (J) (a) DMSO, (COCl)2, NEt3, CH2Cl2, -78 °C; (b) Bu2BOTf, (i-Pr)2(Et)N, 5, then add to aldehyde
from 33, CH2Cl2, -78 °C; (c) CrO3‚Pyr, Celite, CH2Cl2; (K) (a) HF, H2O, MeCN, rt; (b) LiOH, H2O, DME, 80 °C, 7 h.; (c) CaCl2, H2O,
pH 9.
The â-hydroxy ketone 28 was obtained in 86% yield by
an oxidative removal of the PMB ether with DDQ. Applica-
tion of Evans' protocol on 28 using 26 mol % SmI2 provided
hydroxy benzoate 29 in 90% yield with no other isomer
detectable by 1H NMR. The alcohol could now be activated
by formation of the tosylate under standard conditions. The
TMS ether was then removed with dilute HCl, and the crude
hydroxy tosylate was treated with NaH in THF at 50 °C to
yield the bis tetrahydrofuranyl product 30 in 76% yield for
the three steps. Finally, the trityl ether was removed, and
oxidation of the alcohol to the aldehyde 31 was carried out
in order to set up coupling with the C11-C16 fragment.
Incorportation of the C11-C16 fragment required selective
formation of the trans C16-C17 double bond. A Julia-
Lythgoe olefination15 is known to provide predominantly
trans olefins, particularily when there is R branching on the
aldehyde and/or sulfone. Both the Hanessian and Evans’
syntheses utilized this protocol, but we found that using the
same reaction with 31 and the phenyl sulfone analogue of 4
gave poor yields and selectivities. We therefore looked at
newer variants of the Julia reaction and found that the use
of the 1-phenyl-1H-tetrazol-5-yl sulfone developed by Ko-
ceinski and co-workers16 gave the best results. When sulfone
4 was treated with KHMDS and then added to aldehyde 31,
formation of olefin 32 took place in 85% yield (exclusively
trans olefin by H NMR).
1
The final coupling involved an aldol reaction between
fragment 5 and the aldehyde obtained from alcohol 33. The
boron aldol reaction developed by Evans and Dow in their
synthesis provided us with the aldol product accompanied
by loss of the TBS protecting group. When the same reaction
was used with a modified workup procedure, however, the
silyl ether remained intact. After a Collins oxidation the
â-diketone 34 was obtained in 63% yield for the three steps.
Deprotection of the silyl ethers and hyrolysis of the two
esters were also carried out using procedures slightly
modified from those used in the Evans synthesis. Final
treatment with CaCl2 gave ionomycin as the calcium salt,
which was identical to a commercially obtained sample.
Acknowledgment. We thank the Ontario Research and
Development Fund, NSERC (Canada), Merck Frosst, and
the University of Toronto for financial support of this work.
S.H. thanks NSERC for financial support in the form of a
postgraduate fellowship.
Supporting Information Available: Experimental pro-
cedures and full characterization for compounds 1-34, as
well as syntheses of fragments 2 and 4. This material is
available free of charge via the Internet at http://pubs.acs.org.
(15) (a) Julia, M.; Paris, J.-M. Tetrahedron Lett. 1973, 14, 4833. (b)
Kocienski, P. J.; Lythgoe, B.; Ruston, S. J. Chem. Soc., Perkin Trans. 1
1978, 829. (c) Kocienski, P. J.; Lythgoe, B.; Waterhouse, I. J. Chem. Soc.,
Perkin Trans. 1 1980, 1045.
(16) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morely, A. Synlett
1998, 26.
OL025872F
1882
Org. Lett., Vol. 4, No. 11, 2002