De novo asymmetric synthesis of milbemycin β3 via an iterative asymmetric hydration approach

Article abstract of DOI:10.1021/ol061439k

The enantioselective synthesis of the spiroketal/macrolide natural product milbemycin β₃ has been achieved in 22 steps and 2.8% overall yield from an achiral dienoate. The spiroketal ring system was installed by three sequential asymmetric hydr

Full text of DOI:10.1021/ol061439k

ORGANIC
LETTERS
2006
Vol. 8, No. 18
3987-3990
De Novo Asymmetric Synthesis of
Milbemycin â3 via an Iterative
Asymmetric Hydration Approach
Miaosheng Li and George A. O’Doherty*
Department of Chemistry, West Virginia UniVersity, Morgantown, West Virginia 26506
george.odoherty@mail.wVu.edu
Received June 12, 2006
ABSTRACT
The enantioselective synthesis of the spiroketal/macrolide natural product milbemycin â3 has been achieved in 22 steps and 2.8% overall yield
from an achiral dienoate. The spiroketal ring system was installed by three sequential asymmetric hydrations followed by sprioketalization.
Both the absolute and relative stereochemistry of milbemycin â3 was introduced by two Sharpless asymmetric dihydroxylations, two
π
-allylpalladium-catalyzed reductions, and an iridium-catalyzed hydrogen migration/Claisen rearrangement to install the C-12 stereocenter.
Since their initial isolation and structural determination, the
milbemycins1,2 have attracted significant interest for their
potential use as pesticides and pharmaceuticals. In addition
to antibiotic activity, various members of this class of
spiroketal/macrolide natural products have shown significant
activity against various agricultural pests (e.g., mites, beetles,
and tent caterpillars)2,3 and parasites (e.g., nematodes, mites,
ticks, and larvae of biting flies),2,4 while displaying minimal
cytotoxicity to plants and animals.5 Pharmacological interest
in the milbemycins reemerged after it was discovered that
they are also potent efflux pump inhibitors.6
In addition to this array of fascinating biological activities,
the milbemycin structural complexity has also attracted the
attention of the synthetic community.7,8 To date, several total
syntheses of milbemycin â3 have been completed,7 along with
various efforts to the spiroketal ring system.8 While all of
the previous syntheses of the milbemycin (1) derived their
(6) Chamberland, S.; Lee, M.; Lomovskaya, O. Milbemycin class efflux
pump inhibitors for treating microbial infections and cancer. WO 98-
US20916 19981001, 1999 AN1999:244572.
(7) (a) Smith, A. B.; III; Schow, S. R.; Bloom J. D.; Thompson, A. S.;
Winzenberg, K. N. J. Am. Chem. Soc. 1982, 104, 4015-4018. (b) Schow,
S. R.; Bloom J. D.; Thompson, A. S.; Winzenberg, K. N.; Smith, A. B.,
III. J. Am. Chem. Soc. 1986, 108, 2662-2674. (c) Williams, D. R.; Barner,
B. A.; Nishitani, K.; Phillips, J. G. J. Am. Chem. Soc. 1982, 104, 4708-
4710. (d) Street, S. D. A.; Yeater, C.; Kocienski, P.; Campbell, S. F. J.
Chem. Soc., Chem. Commun. 1985, 1386-1388. (e) Attwood, S. V.; Barrett,
A. G. M.; Carr, R. A. E.; Robinson, G. J. Chem. Soc., Chem. Commun.
1986, 479-481. (f) Baker, R.; O′Mahony, M. J.; Swain, C. J. Chem. Soc.,
Chem. Commun. 1985, 1326-1328. (g) Barrett, A. G. M.; Carr, R. A. E.;
Attwood, S. V.; Richardson, G.; Walshe, N. D. A. J. Org. Chem. 1986, 51,
4840-4856.
(8) (a) Yeater, C.; Street, S. D. A.; Kocienski, P.; Campbell, S. F. J.
Chem. Soc., Chem. Commun. 1985, 1388-1389. (b) Baker, R.; O’Mahony,
M. J.; Swain, C. Tetrahedron Lett. 1986, 27, 3059-3062. (c) Crimmins,
M. T.; Bankaitis-Davis, D. M.; Hollis, W. G., Jr. J. Org. Chem. 1988, 53,
652-657. (d) Holoboski, M. A.; Koft, E. J. Org. Chem. 1992, 57, 965-
969.
(1) (a) Mishima, H.; Kurabayashi, M.; Tamura, C.; Sato, S.; Kuwano,
H.; Saito, A. Tetrahedron Lett. 1975, 16, 711-714. (b) Takiguchi, Y.; Ono,
M.; Muramatsu, S.; Ide, J.; Mishima, H.; Terao, M. J. Antibiot. 1983, 36,
502-508. (c) Takiguchi, Y.; Mishima, H.; Okuda, M.; Terao, M.; Aoki,
A.; Fukuda, R. J. Antibiot. 1980, 33, 1120-1127.
(2) For a review see: (a) Shoop, W. L.; Mrozik, H.; Fisher, M. H. Vet.
Parasitol. 1995, 59, 139-156. (b) Aoki, A. J. Pestic. Sci. 1992, 19, 245-
247.
(3) Thamsborg S, Roepstorff A, Larsen M. Vet. Parasitol. 1999, 84, 169-
186.
(4) Bowman, D. D.; Parsons, J. C.; Grieve, R. B.; Hepler, D. I. Am. J.
Vet. Res. 1988, 49, 1986-1989.
(5) (a) Material Safety Data Sheet: Moxidectin: Fort Dodge Animal
Health. (b) Fisher, M. H. Pure Appl. Chem. 1990, 62, 1231-1240.
10.1021/ol061439k CCC: $33.50
© 2006 American Chemical Society
Published on Web 08/08/2006

asymmetry from the chiral pool, we were interested in a de
novo asymmetric approach that would use asymmetric
catalysis to install the six stereocenters in milbemycin â3
from achiral starting materials (5 and 10, Scheme 1).9 Herein
step asymmetric hydration protocol (dihydroxylation, carbon-
ate formation, and palladium-catalyzed reduction),12 dienoate
10 was regio- and enantioselectively transformed into δ-
hydroxy enoate 14, which in turn was diastereoselectively
hydrated to form the protected 3,5-dihydroxy ester 15 using
Evans’ procedure.14 The ester 15 was then converted into
the â-ketophosphonate 8 via Weinreb amide 16 (75% for
the two steps) (Scheme 3).
Scheme 1. Milbemycin â3 (1) and Its Retrosynthesis
Scheme 3. Synthesis and Bis-hydration of Dienoate 10
we describe our successful efforts to implement this strategy
for the de novo synthesis of milbemycin â3.
Retrosynthetically, we envisioned milbemycin â3 (1) being
prepared by an olefination/macrolactonization strategy. This
transform divided the molecule into two halves, an achiral
phosphine oxide 3, which was first prepared by Smith,10 and
a silyl-protected hydroxyaldehyde 2, which possessed both
the spiroketal and the five chiral centers of milbemycin.
Following the Barrett precedent, we planned to install the
sixth chiral center during a Mitsunobu macrocyclization.11
Using a transition-metal variant of the Smith’s vinyl anion
addition Claisen rearrangement, we planned to prepare 2 from
spiroketal 4, which in turn could be prepared from the
partially protected tetraol 6. Finally, we hoped to establish
the four chiral centers and triol functionality of 6 by the
iterative use of our asymmetric hydration protocol (i.e., 10
to 8 and 7 to 6) (Scheme 2).12
The last five carbons of the spiroketal portion of milbe-
mycin â3 came from angelaldehyde (9) and were installed
via a Horner-Wadsworth-Emmons reaction with 8. Expo-
sure of ketophosphonate 8 to aldehyde 9 with Cs2CO3 as
base produced the E,Z-dienone 7 in 82% yield (Scheme 4).
Using a similar three-step sequence, as on dienoate 10,
dienone 7 was diastereoselectively hydrated (7 to 17a).12
Regioselective dihydroxylation of 7 gave an inseparable diols
17a/b in 58% yield and diastereocontrol.15 Because of the
distance between the relevant stereocenters, it was difficult
Scheme 2. Retrosynthesis of Spiroketal 2
1
to distinguish diastereomers 17a and 17b by H NMR and
(9) While it did not exclusively use asymmetric catalysis, Emil Koft had
previous demonstrated that the spiroketal portion of Milbemycin â3 could
be prepared from an achiral starting material, see: ref 8d.
(10) We prepared phosphine oxide 3 by a slightly modified variant of
the Smith route, see Supporting Information and ref 7b.
(11) Barrett had previously shown that the C-19 carboxylate could be
installed by a Mitsunobu reaction, see: ref 7g and Mitsunobu, O. Synthesis
1981, 1-28.
(12) For examples of the asymmetric hydration of unsubstituted dienoates
(e.g. 10 to 8), see: Hunter, T. J.; O’Doherty, G. A. Org. Lett. 2001, 3(7),
1049-1052. Our study of the asymmetric hydration of substituted dienoates,
will be published in due course.
(13) (a) Rychnovsky, S. D.; Kim, J. J. Org. Chem. 1994, 59, 2659-
2660. (b) Trost, B.; Kazmaier, U. J. Am. Chem. Soc. 1992, 114, 7933-35.
(14) Evans, D. A.; Gauchet-Prunet, J. A. J. Org. Chem. 1993, 58, 2446-
2453.
In practice, dienoate 10 was prepared by a three-step
protocol from commercially available 11 via protection,
carboxylation, and ynoate isomerization.13 Using our three-
3988
Org. Lett., Vol. 8, No. 18, 2006

spiroketalization procedure (92%). The minor diastereomeric
impurity in 19 was easily removed by recrystallization from
a mixture of EtOAc/hexanes (9:1). The C-19 alcohol was
protected as the TBDPS-ether, the PMP-group was oxida-
tively removed with CAN and the C-15 alcohol was oxidized
to the aldehyde with the Dess-Martin reagent in an 83%
overall yield.
Scheme 4. Synthesis of Spiroketal 4 via Dienone Hydration
We next looked to extend the C-15 aldehyde to the C-11
aldehyde. Based on the Smith synthesis, we planned to estab-
lish the C-14/C-15 E-double bond and the C-12 stereocenter
by a Claisen rearrangement. In contrast to Smith’s use of an
Ireland enolate rearrangement, we chose to use the isomer-
ization-Claisen rearrangement (ICR) developed by Nelson.17
This procedure has the added advantage of providing the
aldehyde 2 directly.
When aldehyde 4 was exposed to a vinyl cuprate reagent,
an exceedingly diastereoselective addition (dr > 20:1)
occurred to give allylic alcohol 20 in good yield (78%). The
allylic alcohol 20 was allylated with KH/AllylBr to give
allylic ether 21 in nearly quantitative yield (99%). Following
the Nelson protocol, allylic ether 21 was exposed to catalytic
iridium and tricyclohexylphosphine. Under these conditions,
allylic ether 21 cleanly rearranged to the E-enol ether 22, at
which point 6 mol % of PPh3 was added and the dichloro-
ethylene solution was refluxed. After heating for 24 h, an
83% yield of aldehyde 2 was isolated. While aldehyde 2 can
be purified by SiO2 chromatography, this leads to lower
diastereoselectivity (dr ) 4:1). The preferred procedure was
to use the crude aldehyde in the subsequent transformation.
Analysis of the crude 1H NMR indicated that the diastereo-
meric ratio of crude 2 was on the order of 10:1 (Scheme 5).
Finally, with aldehyde 2 in hand, we set out to stitch the
two fragments together via an olefination and lactonization.
The E,E-diene of 23 was stereoselectively installed upon
Scheme 5. Synthesis of Aldehyde 2 via an ICR Reaction
TLC. Thus, Mosher ester analysis (17a/b to 18a/b, see the
Supporting Information) was used to determine the diaster-
eomeric ratio of 17a to 17b (dr ) 11:1). The mixture of
diastereomers 17a/b was converted into cyclic carbonates
and stereoselectively reduced with HCO2H/Et3N and catalytic
palladium(0) in CH2Cl2/hexane to give alcohols 6a/b (93%).16
As with 17a/b, the diastereomers 6a/b were not easily
differentiated by 1H NMR or separated by chromatography.
The diastereomeric alcohols 6a/b were cleanly converted into
spiroketal 19 via a one-pot hydrogenation/hydrogenolysis/
(15) Because most of the enantiomeric impurity in 7 is converted into
the minor diastereomer ent-17b during the asymmetric dihydroxylation (7
to 17a/b), the major diastereomer diol 17a was isolated in essentially
enantiomeric pure form. Consequently, the minor diastereomer 17b must
be formed with lower enantiopurity. For other examples of this enantio-
enriching phenomena, see: Ahmed, Md. M.; Berry, B. P.; Hunter, T. J.;
Tomcik, D. J.; O’Doherty, G. A. Org. Lett. 2005, 7, 745-748.
(16) The CH2Cl2/hexane solvent mixture was critically important to
ensure high yields for the palladium catalyzed reduction. For instance, use
of THF as solvent gave a 1: 1 mixture of double bond regioisomers.
Org. Lett., Vol. 8, No. 18, 2006
3989

lactonized with DIAD/PPh3 in good yield (79%). Using
NaSEt the methyl protecting group was removed, providing
30 mg of synthetic material that was physically (mp, optical
rotation)18 and spectroscopically (1H NMR, 13C NMR, IR,
and MS) identical to natural milbemycin â3 (1) (Scheme 6).
In conclusion, a short de novo asymmetric synthesis of
milbemycin â3 (1) has been developed. This highly enantio-
and diastereocontrolled route illustrates the utility of our
dienoate/dienone asymmetric hydration strategy for natural
product synthesis. In addition, it features the use of Nelson’s
isomerization-Claisen rearrangement (ICR) in a structurally
complex setting. This approach provided milbemycin â3 (1)
in 2.3% overall yields from 5-hexyn-1-ol (11), which should
be amenable to the preparation of its enantiomer. Further
application of this approach to the synthesis of structurally
more complex members of this class of compounds and
biological testing is ongoing.
Scheme 6. Synthesis of Milbemycin â3 (1) via a
Macrolactonization
Acknowledgment. We are grateful to the NIH (GM63150)
and NSF (CHE-0415469) for the support of our research
program and NSF-EPSCoR (0314742) for a 600 MHz NMR
and an LTQ-FT mass spectrometer at WVU.
Supporting Information Available: Complete experi-
mental procedures and spectral data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
exposure of a crude solution of aldehyde 2 with the sodium
salt of phosphine oxide 3. Before the lactonization could
commence, the silyl ether was removed (TBAF, 95%) and
the methyl ester was hydrolyzed (LiOH, 78%). Then fol-
lowing the Barrett procedure the 19-epi-seco acid 24 was
OL061439K
(18) The synthetic material had a melting point range of 182-184 °C
and an optical rotation of +99 (c ) 0.25, MeOH) which is in good
agreement with the literature value (Mp ) 181-183 °C.; [R]D ) +102 (c
) 0.17, MeOH)). The actual value for the optical rotation of milbemycin
â3 has been a source of disagreement. For an interesting discussion of this
issue along with its resolution, see: ref 7 g.
(17) (a) Nelson, S. G.; Bungard, C. J.; Wang, K. J. Am. Chem. Soc.
2003, 125, 13000-13001. (b) Nelson, S. G.; Wang, K. J. Am. Chem. Soc.
2006, 128, 4232-4233.
3990
Org. Lett., Vol. 8, No. 18, 2006