Total synthesis of antascomicin B

Article abstract of DOI:10.1002/anie.200500174

(Chemical Equation Presented) The structurally enticing antascomicin B (1), which exhibits potent immunosuppressant-antagonizing properties, has a complex polyketide structure. Key steps in its enantioselective total synthesis include a transannular catec

Full text of DOI:10.1002/anie.200500174

Communications
Natural Products Synthesis
Total Synthesis of Antascomicin B**
Dominic E. A. Brittain, Charlotte M. Griffiths-Jones,
Michael R. Linder, Martin D. Smith,
Catherine McCusker, Jaqueline S. Barlow,
Ryo Akiyama, Kosuke Yasuda, and Steven V. Ley*
In 1996, a search for novel materials that might elicit similar
biological effects to rapamycin and FK506 was initiated at
Sandoz.[1] The immunophilin FKBP12 (the primary target for
rapamycin and FK506) was used and the degree of binding to
substrates in competition with FK506 was measured to
evaluate metabolites from over 12000 strains of Actino-
mycetes. Alongside the known immunophilin-binding com-
pounds rapamycin, ascomycin,[2] and meridamycin,[3] the
novel antascomicin family was discovered.
It was found that, despite the strong levels of binding to
FKBP12 of all these compounds, only rapamycin and
ascomycin show immunosuppressive properties in vitro and
in vivo, whereas meridamycin and the antascomicins show
none.[1,3] Specifically, the antascomicins bind FKBP12 to the
same degree as do FK506 and rapamycin (1.1 and 0.6 nm,
respectively, in the same binding assay) and antagonize both
immunosuppressants. It has been demonstrated by Schreiber
and co-workers that it is the fate of the ligand–FKBP12
complex, rather than solely ligand binding to FKBP12, that
determines the immunosuppressive effect.[4,5] Subsequently,
the significantly higher levels of FKBP12 found in the brain,
rather than in the immune system,[6] stimulated work on the
effects of rapamycin and FK506 complexes with FKBP12 in
the treatment of neurodegenerative diseases.[7] However, the
immunosuppressive effects of these complexes became a
liability,[8] thus pushing nonimmunosuppressive FKBP ligands
to the fore. Further results demonstrated the ability of
synthetic FKBP ligands to promote not only regrowth of
damaged nerves in the peripheral nervous system, but also the
regeneration of damaged neurons in the central nervous
system.[9]
The wealth of biological and chemical attention attracted
by other FKBP12 ligands prompted us to initiate a program
aimed at the total synthesis of antascomicin B, as it combined
both structural complexity and the most potent FKBP12
binding ability. Antascomicin B is an enticing target for the
synthetic organic chemist as it has a complex polyketide
structure that includes both lactol and lactam functionalities,
12 stereogenic centers, and a masked 1,2,3-tricarbonyl unit. A
recent synthetic study described the stereoselective synthesis
of the C18–C34 fragment of antascomicin A;[10] however, no
total synthesis of any member of the antascomicin family has
been disclosed. Examination of the structure of antascomi-
cin B (1) suggested disconnection to three major fragments
(Scheme 1).
Construction of the C10–C34 carbon framework was
envisaged through the coupling of a C24 anion (via an
intermediate sulfone on fragment III) with the epoxide
functionality at C25 in fragment II, followed by esterification
with pipecolic acid fragment I. The installation of all but two
of the carbon atoms present in the natural product would then
set the stage for the key macrocyclization and tricarbonyl-
generation steps. Enone formation (C16) and deprotection
would then be all that was required to complete the synthesis.
Fragment II could be constructed by a ring-closing meta-
thesis of diene 2, which could be derived by addition of the
allyl stannane 4 to a butanediacetal-protected aldehyde 3
(Scheme 2).
[*] Dr. D. E. A. Brittain, Dr. C. M. Griffiths-Jones, Dr. M. R. Linder,
Dr. M. D. Smith, Dr. C. McCusker, Dr. J. S. Barlow, Dr. R. Akiyama,
Dr. K. Yasuda, Professor Dr. S. V. Ley
Department of Chemistry
University of Cambridge
Lensfield Road, Cambridge CB2 1EW (UK)
Fax: (+44)1223-336-442
E-mail: svl1000@cam.ac.uk
[**] The authors would like to thank Novartis for the Novartis Research
Fellowship (to S.V.L.), Trinity College, Cambridge for a Junior
Research Fellowship (to D.E.A.B.), Pembroke College, Cambridge
and the Royal Society for fellowships (to M.D.S.), the EPSRC (to
C.M.G.-J., M.R.L., C.M., and D.E.A.B.), Tanabe Seiyaku Co., Ltd. for
a postdoctoral fellowship (to K.Y.), and the Japanese Society for the
Promotion of Sciences (to R.A.). The authors would also like to
acknowledge the contribution of Prof. Hidenori Watanabe (Univer-
sity of Tokyo; Cambridge 1998–2000) whose studies on a related
system were invaluable in the development of the macrocyclization
method described herein.
By utilizing work developed in our group on the use of
desymmetrized butanediacetal-protected tartrates as building
blocks for anti-1,2-diols,[11–13] diacetal 6 was easily prepared
from dimethyl d-tartrate 5 in large quantities and converted
into protected aldehyde 3 via protected tetrol 7 (Scheme 3). It
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
2732
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200500174
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737

Angewandte
Chemie
Scheme 1. Disconnection of antascomicin B. Fmoc=9-fluorenylmethyloxycar-
bonyl; MOM=methoxymethyl; TBS=tert-butyldimethylsilyl; PMB=p-methoxy-
benzyl.
Scheme 4. Reagents and conditions: a) 1. (E)-Butene, KOtBu, n-BuLi,
THF, ꢀ788C; 2. (+)-(Ipc)2BOMe, Et2O, BF3·OEt2, aldehyde, ꢀ788C;
b) BnBr, NaH, DMF, 62% over two steps; c) 1. O3, CH2Cl2, ꢀ788C;
then PPh3, 2. NaBH4, MeOH, 90% over two steps; d) I2, PPh3, imida-
zole, CH2Cl2, 90%; e) allylpyridylsulfide, tBuLi, THF, ꢀ788C, 95%;
f) Bu3SnLi, CuBr, THF, ꢀ788C, 98%. Ipc=isopinocampheyl;
DMF=N,N-dimethylformamide.
of the ozonide to reduction by sodium borohydride, it was
necessary to form the aldehyde initially en route to the
alcohol. This was achieved with triphenylphosphane, and the
transformation remained a one-pot procedure. Subsequent
conversion into iodide 11 and displacement by allylpyridyl-
sulfide provided substrate 12. The allyl stannane functionality
was subsequently installed through an anionic cuprate
procedure to give stannane 4.[15]
Scheme 2. Retrosynthetic analysis of fragment II. Bn=benzyl.
A variety of conditions were examined for the aldehyde/
stannane addition, with the optimum selectivity arising from
the use of zinc iodide in CH2Cl2 to give 13 with near total
diastereoselectivity at C30 and in a 7:1 ratio at C29 in favor of
the desired isomer (Scheme 5). It was found that the E/Z ratio
of the stannane had only a minor effect on the diastereo-
selectivity of the subsequent addition.
Modeling studies and precedent for stannane additions
suggest the mechanism and transition state shown in
Figure 1.[16,17] It is believed that chelation between the
carbonyl group and the axial lone pair of electrons on the
a-oxygen atom of the butanediacetal group through the Zn2+
permits attack of the stannane to occur onto only one face of
the carbonyl group. The allyl stannane attacks in an anti-
periplanar orientation, with the vinylic hydrogen substituent
being placed over the chelate ring.
Protection of the newly formed hydroxy group as a MOM
ether and removal of the TBS protecting group yielded
alcohol 14, which upon completion of a Swern–Wittig
protocol, provided access to the ring-closing-metathesis
precursor 2. Portionwise addition of Grubbs second-gener-
ation imidazolidine catalyst[18] to the substrate in refluxing
toluene enabled the substituted cyclohexene to be formed in
95% yield. Treatment with hydrogen in the presence of
palladium on carbon resulted in the reduction of the C33–C34
double bond and the simultaneous removal of both benzyl
Scheme 3. Reagents and conditions: a) MeCOCOMe, CSA, CH(OMe)3,
MeOH, 68%; b) LiAlH4, THF, 99%; c) NaH, TBSCl, 87%; d) (COCl)2,
DMSO, CH2Cl2, Et3N. CSA=(+/ꢀ)-camphorsulfonic acid; DMSO=
dimethyl sulfoxide.
was found that butanediacetal-protected aldehyde 3 was most
effective when used crude in subsequent reactions.
After some encouraging results with model studies on the
addition of allyl stannanes to aldehyde 3, work began on the
synthesis of the fully substituted allyl stannane 4 (Scheme 4).
The two stereogenic centers in allyl stannane 4 were installed
through asymmetric crotylation of commercially available
benzyloxyacetaldehyde (8) to give the secondary alcohol with
97% ee, under conditions originally developed by Brown and
Bhat.[14] Benzylation to provide alkene 9 and reductive
ozonolysis afforded primary alcohol 10. Owing to the stability
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2733

Communications
Scheme 6. Retrosynthetic analysis of fragment III.
Scheme 7. Reagents and conditions: a) AllylMgCl, CuI, THF, 90%;
b) O3, CH2Cl2, ꢀ788C; then PPh3, 88%.
chiral auxiliary.[19] Addition of the boron enolate of oxazoli-
dinone 20 to aldehyde 19 gave aldol adduct 21 in high yield as
a single diastereoisomer after recrystallization. Protection of
the C14 hydroxy group as a silyl ether, and removal of the
auxiliary via thioester 22[20,21] gave the desired aldehyde 16
(Scheme 8).
Scheme 5. Reagents and conditions: a) ZnI2, CH2Cl2 ꢀ788C!RT, 85%
over two steps, syn/anti 6:1 at C29–C30; b) 1. MOMCl, DIPEA, CH2Cl2;
2. TBAF, THF, 89% over two steps; c) (COCl)2, DMSO, Et3N, CH2Cl2;
d) MePPh3Br, BuLi, THF, ꢀ788C!RT, 87% over two steps; e) 1,3-(bis-
(mesityl)-2-imidazolidinylidene) dichloro(phenylmethylene)(tricyclo-
hexylphosphane) ruthenium (3ꢀ3.5 mol%), toluene, reflux, 95%;
f) H2, Pd/C, EtOH, 96%; g) NaH, 2-(2,4,6-triisopropylbenzenesulfonyl)-
imidazole, THF, 88%. DIPEA=diisopropylethylamine; TBAF=tetra-
butylammonium fluoride.
Scheme 8. Reagents and conditions: a) Bu2BOTf, Et3N, CH2Cl2, 19,
ꢀ808C, 60 h; then pH 7.2 phosphate-buffered MeOH, H2O2, 94%
(single isomer); b) TBSCl, imidazole, DMF, 100%; c) EtSH, BuLi, THF,
ꢀ788C, 99%; d) Et3SiH, Pd/C, acetone, 99%. Tf=trifluoromethane-
sulfonyl.
Figure 1. Suggested mechanism and transition state for stannane
additions.
groups to give diol 15, which was transformed in one step to
the epoxide, fragment II (Scheme 5).
The synthesis of fragment III centered on the construction
of aldehyde 16 and alkyne 17 and on their coupling through a
hydrozirconation protocol to create the C10–C24 allylic
alcohol (Scheme 6).
Synthesis of the C10–C16 aldehyde 16 started from iodide
18, which is available in two steps from commercially
available (2S)-3-bromo-2-methylpropan-1-ol by silylation
and halide exchange. Displacement with allylmagnesium
chloride in the presence of copper iodide gave the terminal
olefin-bearing compound that underwent ozonolysis with
reductive workup to afford the aldehyde 19 in an overall yield
of 79% for the two steps (Scheme 7).
The synthesis of alkyne 17 started from primary alcohol
23, which is available in four steps from Roche ester ((S)-3-
hydroxy-2-methylpropionic acid methyl ester). Swern oxida-
tion followed by Wittig reaction was used to establish the
C21–C22 trans olefin 24 with exceptionally high selectivity
(Scheme 9). Reduction with DIBAL-H to alcohol 25 and
conversion into bromide 26 enabled introduction of the
alkyne functionality in its TMS-protected form to give
protected alkyne 27 in 88% yield. A minor allene side
product (8%), which arises from SE2’ attack of the TMS-
propargyl organolithium reagent, was also isolated. Desilyl-
ation under basic conditions completed the synthesis of
alkyne 17 in this high-yielding series of steps.
The C14–C15 stereogenic centers were introduced
through an asymmetric aldol reaction by using the Evans
2734
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737

Angewandte
Chemie
Scheme 9. Reagents and conditions: a) (COCl)2, DMSO, Et3N, CH2Cl2,
95%; b) Ph3PCHCO2Et, CH2Cl2, 08C!RT, 2 days, 98%, E/Z 99:1;
c) DIBAL-H, PhMe, ꢀ788C, 1.5 h, 93%; d) Ph3P, CBr4, CH2Cl2, 95%;
e) BuLi, THF, ꢀ788C, 1 h, 88%; f) aqueous NaOH (10%), 2 days,
99%. DIBAL-H=diisobutylaluminum hydride.
Cis-specific hydrozirconation of alkyne 17 with freshly
prepared Schwartz reagent[22] afforded a vinylzirconium
species[23] that underwent silver perchlorate mediated addi-
tion to aldehyde 16 in 96% yield.[24,25] The resultant secondary
alcohol 28 was formed as a 1:1 mixture of separable
diastereoisomers; the stereoselectivity is unimportant at this
stage as the alcohol functionality at C16 is ultimately
oxidized. After separation (for analytical convenience),
silylation of the secondary alcohol completed the synthesis
of fragment III (Scheme 10).
Scheme 11. Reagents and conditions: a) DDQ, CH2Cl2, H2O;
b) PhSSPh, PBu3, py, 95% over two steps; c) PhSeSePh, H2O2, Et2O,
88%; d) BuLi, HMPA, THF, ꢀ788C; then fragment II at ꢀ208C, 70%;
e) lithium 4,4-di-tert-butylbiphenyl, THF, ꢀ788C, 78%. DDQ=2,3-
dichloro-5,6-dicyano-p-benzoquinone; HMPA=hexamethylphos-
phoramide.
epimeric at C24, which were reductively desulfonylated with
excess lithium in the presence of 1.5 equivalents of 4,4-di-tert-
butylbiphenyl to give alcohol 30.[27–29] This was subsequently
esterified with N-Fmoc-protected pipecolic acid, fragment I,
and the primary TBS group was removed by using camphor-
sulfonic acid in methanol to give late-stage intermediate 31a
(Scheme 12).
As this desilylation was not completely selective, a small
proportion (25%) of bisdeprotected product 31b was also
formed. This was easily recycled by protection with TBSCl to
regenerate the fully protected triol 30. The C10-OH group
was oxidized to carboxylic acid through a two-step TPAP[30]
Pinnick protocol[31] in 76% yield over two steps.
–
Despite employing a disconnection across the C9–C10
carbon–carbon bond, it was envisaged that macrocyclization
could be effected by exploiting the relative ease of carbon–
heteroatom bond formation and the entropic advantage of
using a tether. With this in mind, the Fmoc group on the
pipecolic acid moiety was removed to afford the free amine
32, which was treated with benzo[1,4]dioxin-2-one to form
amide 33 with the first carbon–heteroatom bond. High-
dilution macrolactonization (0.001m) with EDCI then formed
the second carbon–heteroatom bond to generate catechol-
tethered macrocycle 34. Treatment with LHMDS in THF led
to the formation of the required C9–C10 bond, presumably
through a kinetic C9 deprotonation followed by a trans-
annular Dieckmann-type condensation onto the C10 ester to
give macrocycle 35. The effectiveness of this transformation
Scheme 10. Reagents and conditions: a) 1. [Zr(Cp)2(H)(Cl)], CH2Cl2,
ꢀ78!08C; 2. 16, AgClO4, CH2Cl2, 96%, (1:1 mixture, separable);
b) 1. separation; 2. TBSCl, imidazole, DMF, 93%.
To couple with fragment II, the 16R diastereoisomer of
fragment III was converted into the C24 sulfone (Scheme 11).
Oxidative removal of the PMB protecting group and direct
conversion into the sulfide was followed by selective oxida-
tion of the sulfur function by using diphenyldiselenide and
hydrogen peroxide to give sulfone 29.[26] Addition of the a-
lithiated sulfone to fragment II was then carried out in THF
and HMPA to give two diastereoisomeric g-hydroxysulfones,
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2735

Communications
Scheme 12. Reagents and conditions: a) EDCI, DMAP, CH2Cl2, fragment I, ꢀ58C, 95%; b) CSA, MeOH, CH2Cl2, ꢀ158C 62%; c) TBSCl, imidazole,
DMF, 98%; d) 1. TPAP, NMO, CH2Cl2, 2. NaClO2, NaH2PO4, 2-methylbut-2-ene, tBuOH, 76% over two steps; e) piperidine, DMF; f) benzo-
[1,4]dioxin-2-one, DMAP, CH2Cl2, 87% over two steps; g) EDCI, CH2Cl2, 71%; h) LHMDS, THF, 68%; i) DMP, H2O, CH2Cl2, py, 80%; j) 1. HF·py,
py, THF, 4 days; 2. DMP, CH2Cl2, py; 3. TFA, H2O, 10 min, 13% over three steps. EDCI=1-ethyl-3-[3-(dimethylamino)propyl]carbodiimide hydro-
chloride; DMAP=4-dimethylaminopyridine, TPAP=tetra-n-propylammonium perruthenate; NMO=N-methylmorpholine N-oxide;
LHMDS=lithium hexamethyldisilazide; DMP=Dess–Martin periodinane; TFA=trifluoroacetic acid.
lies in the simplified requirements of its six-membered
transition state. Oxidation[32,33] with the Dess–Martin
reagent[34] in the presence of water[35] cleaved the catechol
tether and directly furnished the desired tricarbonyl 36, which
contains the complete carbon framework of antascomicin B
(1). Medium-ring ketones have been synthesized by the
Ohtsuka group in a related approach by using lactam
sulfoxides followed by a reductive tether cleavage;[36–38] this
method has been applied to the preparation of a taxane
skeleton.[39]
a stereodirecting functionality, and the development of novel
transformations including a transannular Dieckmann oxida-
tion protocol to access the vicinal tricarbonyl functionality.
Completion of the synthesis in a total of 52 synthetic steps
from commercially available starting materials (longest linear
sequence: 23 steps) demonstrates the utility of these methods
in the synthesis of complex natural products.
Received: January 17, 2005
Published online: April 1, 2005
Removal of the TBS protecting groups at C16 and C14
with pyridine-buffered HF·pyridine complex was slow but
effective and led to the spontaneous formation of the six-
membered C10 lactol. The C16 allylic alcohol was then
oxidized under Dess–Martin conditions to generate the
corresponding a,b-unsaturated carbonyl compound. Simulta-
neous deprotection of the MOM and butane diacetal groups
under acidic conditions afforded the natural product, which
was demonstrated to be identical to an authentic
sample.[40,41–48]
Keywords: macrocycles · natural products · oxidation ·
total synthesis
.
[1] T. Fehr, J.-J. Sanglier, W. Schuler, L. Gschwind, M. Ponelle, W.
Schilling, C. Wioland, J. Antibiot. 1996, 49, 230.
[2] M. Morisaki, T. Arai, J. Antibiot. 1992, 45, 126.
[3] G. M. Salituro, D. L. Zink, A. Dahl, J. Nielsen, E. Wu, L. Y.
Huang, C. Kastner, F. J. Dumont, Tetrahedron Lett. 1995, 36, 997.
[4] B. E. Bierer, P. K. Somers, T. J. Wandless, S. J. Burakoff, S. L.
Schreiber, Science 1990, 250, 556.
The challenge posed by the total synthesis of antascomi-
cin B has been met through the combination of established
synthetic procedures and our research group methods,
including the use of butanediacetal as both a protecting and
[5] P. K. Somers, T. J. Wandless, S. L. Schreiber, J. Am. Chem. Soc.
1991, 113, 8045.
2736
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737

Angewandte
Chemie
[6] J. P. Steiner, T. M. Dawson, M. Fotuhi, C. E. Glatt, A. M.
Snowman, N. Cohen, S. H. Snyder, Nature 1992, 358, 584.
[7] S. H. Snyder, D. M. Sabatini, M. M. Lai, J. P. Steiner, G. S.
Hamilton, P. D. Suzdak, Trends Pharmacol. Sci. 1998, 19, 21.
[8] J. P. Steiner, M. A. Connolly, H. I. Valentine, G. S. Hamilton,
T. M. Dawson, L. Hester, S. H. Snyder, Nat. Med. 1997, 3, 421.
[9] G. S. Hamilton, W. Huang, M. A. Connolly, D. T. Ross, H. Guo,
H. L. Valentine, P. D. Suzdak, J. P. Steiner, Bioorg. Med. Chem.
Lett. 1997, 7, 1785.
treatment of authentic antascomicin A with catalytic trifluoro-
acetic acid in chloroform generates a second species (a seven-
membered lactol) analogous to that observed in the synthetic
sample of antascomicin B. Over time, this converts back into the
original six-membered lactol system.
[41] K. Baumann, B. Oberhauser, M. A. Grassberger, G. Haidl, G.
Schulz, Tetrahedron Lett. 1995, 36, 2231.
[42] K. Baumann, B. Oberhauser, G. Strnadt, H. Knapp, G. Schulz,
M. A. Grassberger, Synlett 1999, 877.
[43] F. P. Gailliot, T. K. Natishan, J. M. Ballard, R. A. Reamer, D.
Kuczynski, G. J. McManemin, R. S. Egan, B. C. Buckland, J.
Antibiot. 1994, 47, 806.
[10] H. Fuwa, Y. Okamura, H. Natsugari, Tetrahedron 2004, 60, 5341.
[11] J. S. Barlow, D. J. Dixon, A. C. Foster, S. V. Ley, D. J. Reynolds, J.
Chem. Soc. Perkin Trans. 1 1999, 1627.
[44] Y. Namiki, N. Kihara, S. Koda, K. Hane, T. Yasuda, J. Antibiot.
1993, 46, 1149.
[12] D. J. Dixon, A. C. Foster, S. V. Ley, D. J. Reynolds, J. Chem. Soc.
Perkin Trans. 1 1999, 1631.
[45] Y. Namiki, A. Fujiwara, N. Kihara, S. Koda, K. Hane, T. Yasuda,
Drug Dev. Ind. Pharm. 1995, 21, 809.
[46] P. Hughes, J. Musser, M. Conklin, R. Russo, Tetrahedron Lett.
1992, 33, 4739.
[47] P. Karuso, H. Kessler, D. F. Mierke, J. Am. Chem. Soc. 1990, 112,
9434.
[48] H. Kessler, R. Haessner, W. Schꢀler, Helv. Chim. Acta 1993, 76,
117.
[13] S. V. Ley, D. K. Baeschlin, D. J. Dixon, A. C. Foster, S. J. Ince,
H. W. M. Priepke, D. J. Reynolds, Chem. Rev. 2001, 101, 53.
[14] H. C. Brown, K. S. Bhat, J. Am. Chem. Soc. 1986, 108, 5919.
[15] T. Takeda, S. Ogawa, N. Ohta, T. Fujiwara, Chem. Lett. 1987,
1967.
[16] G. E. Keck, K. A. Savin, E. N. K. Cressman, D. E. Abbott, J.
Org. Chem. 1994, 59, 7889.
[17] G. E. Keck, D. E. Abbott, E. P. Boden, E. J. Enholm, Tetrahe-
dron Lett. 1984, 25, 3927.
[18] M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1,
953.
[19] D. A. Evans, J. V. Nelson, E. Vogel, T. R. Taber, J. Am. Chem.
Soc. 1981, 103, 3099.
[20] T. Fukuyama, S.-C. Lin, L. Li, J. Am. Chem. Soc. 1990, 112, 7050.
[21] D. A. Evans, W. C. Black, J. Am. Chem. Soc. 1993, 115, 4497.
[22] S. L. Buchwald, S. J. Lamaire, R. B. Nielsen, B. T. Watson, S. M.
King, Tetrahedron Lett. 1987, 28, 3895.
[23] J. Schwartz, D. W. Hart, J. Am. Chem. Soc. 1974, 96, 8115.
[24] H. Maeta, T. Hashimoto, T. Hasegawa, K. Suzuki, Tetrahedron
Lett. 1992, 33, 5965.
[25] K. Suzuki, T. Hasegawa, T. Imai, H. Maeta, S. Ohba, Tetrahedron
1995, 51, 4483.
[26] H. J. Reich, F. Chow, S. L. Peake, Synthesis 1978, 299.
[27] M. Yus, Chem. Soc. Rev. 1996, 25, 155.
[28] M. Yus, R. P. Herrera, A. Guijarro, Tetrahedron Lett. 2001, 42,
3455.
[29] M. Yus, R. P. Herrera, A. Guijarro, Chem. Eur. J. 2002, 8, 2574.
[30] S. V. Ley, J. Norman, W. P. Griffith, S. P. Marsden, Synthesis
1994, 639.
[31] B. S. Bal, W. E. Childers, Jr., H. W. Pinnick, Tetrahedron 1981,
37, 2091.
[32] M. J. Batchelor, R. J. Gillespie, J. M. C. Golec, C. J. R. Hedge-
cock, Tetrahedron Lett. 1993, 34, 167.
[33] C. M. Hayward, D. Yohannes, S. J. Danishefsky, J. Am. Chem.
Soc. 1993, 115, 9345.
[34] D. B. Dess, J. C. Martin, J. Org. Chem. 1983, 48, 4155.
[35] S. D. Meyer, S. L. Schreiber, J. Org. Chem. 1994, 59, 7549.
[36] Y. Ohtsuka, T. Oishi, Chem. Pharm. Bull. 1983, 31, 443.
[37] Y. Ohtsuka, T. Oishi, Chem. Pharm. Bull. 1983, 31, 454.
[38] Y. Ohtsuka, S. Niitsuma, H. Tadokoro, T. Hayashi, T. Oishi, J.
Org. Chem. 1984, 49, 2326.
[39] K. Takatori, Y. Takeuchi, Y. Shinohara, K. Yamaguchi, M.
Nakamura, T. Hirosawa, T. Shimizu, M. Saito, S. Aizawa, O.
Sugiyama, Y. Ohtsuka, M. Kajiwara, Synlett 1999, 975.
[40] It has been demonstrated that natural products of this nature can
exist as isomeric mixtures, corresponding to the six-membered
lactol and a seven-membered lactol (derived by addition of the
14-OH group to the C9 carbonyl group.[41–48] The synthetic
sample contained a second isomer of antascomicin B, which was
identified as the corresponding seven-membered lactol; the
1H NMR is complicated by the natural product existing in
CDCl3 as a mixture of rotamers that do not interconvert on the
NMR timescale. In a model study, it was demonstrated that
Angew. Chem. Int. Ed. 2005, 44, 2732 –2737
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2737