Total synthesis of the antibiotic kendomycin by maerocyclization using photo-fries rearrangement and ring-closing metathesis

Full text of DOI:10.1002/anie.200900522

Communications
DOI: 10.1002/anie.200900522
Natural Product Synthesis
Total Synthesis of the Antibiotic Kendomycin by
Macrocyclization using Photo-Fries Rearrangement and
Ring-Closing Metathesis**
Thomas Magauer, Harry J. Martin, and Johann Mulzer*
Dedicated to Professor Hans-Ulrich
Reißig on the occasion of his 60th
birthday
Angewandte
Chemie
6032
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 6032 –6036

Angewandte
Chemie
Kendomycin [1, (À)-TAN2162], an ansamycin isolated from
different Streptomyces species, has been shown in studies over
the last decade to be a potent endothelin receptor antagonist
and antiosteoperotic compound with remarkable antibacte-
rial and cytostatic activity.[1] The challenging structure and
diverse pharmacological profile of kendomycin[1,2] has moti-
vated us[3] and, sometime later a number of other groups,[4–6]
to carry out studies towards its synthesis. To date, three total
syntheses[4] and one formal synthesis[5] have been reported,
along with a number of fragment preparations.[6] The main
problem for all the approaches has been the formation of the
strained macrocyclic ansa-ring. For example, macrocycliza-
tions were performed using C-glycosidation,[4a] Barbier-type
organometallic addition,[4d] Prins reaction,[5] and Horner–
Wadsworth–Emmons olefination.[6e] Most strikingly, all
attempts to achieve 13,14-macrocyclization by ring-closing
metathesis (RCM)[4b,6a,e] were plagued by low yields and
formation of the undesired 13,14-Z-olefin. We tested alter-
native locations for RCM connections; however to our
disappointment, both the 9,10- and the 19,20-positions
proved to be unsuited.[7] Nonetheless, we were still convinced
that RCM should be a highly serviceable tool for ring closure.
As we have demonstrated in previous studies,[3,6e] hin-
dered rotation around the C4a/C5 bond connecting the
tetrahydropyran ring and the aromatic system makes ring
closure difficult. Therefore, we decided to postpone tetrahy-
dropyran formation until after macrocyclization, and conse-
quently, we report herein two novel ring closures: the first by
a photo-Fries ring contraction[8] connecting C4a/C5, and the
second by a RCM to form a 10,11-olefin. Both routes would
lead to the known benzofuran intermediate 2.[4a]
Scheme 1. Retrosynthesis of 1: Photo-Fries approach. TBDPS=tert-
butyldiphenylsilyl, Bn=benzyl.
The photo-Fries route (Scheme 1) centers around macro-
lactone 3 as a key intermediate, which was assembled from
building blocks 4, 5, and 6 by a Claisen–Ireland rearrange-
ment (C15/C16 connection) and Evans aldolization (C8/C9-
connection).
The synthesis of the benzofuran fragment 5 (Scheme 2)
started with known aldehyde 7,[4a] which is easily available
from citronellene (see the Supporting Information). A Colvin
C1 chain elongation[9] furnished alkyne 8, which was con-
verted into vinyl iodide 9. Negishi coupling[10] with aryl
bromide 10[6e] led to styrene 11, which after epoxidation was
subjected to palladium(0)-mediated rearrangement[11] to
ketone 12. Acid-catalyzed formation of the furan ring
concomitantly removed the 3-OMOM group, which was
reinstalled. Desilylation delivered alcohol 13, which was
oxidized to carboxylic acid 5.
Scheme 2. Synthesis of carboxylic acid 5. a) TMSCHN2, nBuLi, THF,
À788C to RT, 83%; b) [Cp2ZrHCl], benzene, 508C; I2, 08C, 76%;
c) [Pd(PPh3)4], tBuLi, ZnCl2, Et2O/THF, 08C, 67%; d) DMDO, acetone,
RT, 99% (d.r. 1.1:1); e) Pd(OAc)2, PBu3, tBuOH, reflux, 81% (2 steps);
f) TfOH, toluene/EtOH (4:1), molecular sieves 4 ꢀ, 808C, 5 min;
g) MOMCl, NaH, DMF, 08C, 90% (2 steps); h) TBAF, THF, RT, 89%;
i) IBX, DMSO, RT, 97%; j) NaClO2/NaH2PO4, 2,3-dimethylbut-2-ene,
tBuOH/H2O, 99%. MOM=methoxymethyl, TMS=trimethylsilyl,
Cp=cyclopentadienyl, DMDO=dimethyldioxirane, Tf=trifluorometh-
anesulfonyl, DMF=dimethylformamide, TBAF=tetra-n-butylammo-
nium fluoride, THF=tetrahydrofuran, IBX=o-iodoxybenzoic acid,
DMSO=dimethylsulfoxide.
[*] Mag. T. Magauer, Dr. H. J. Martin, Prof. J. Mulzer
Institute of Organic Chemistry, University of Vienna
Wꢁhringerstrasse 38, 1090 Vienna (Austria)
Fax: (+43)1-4277-52189
E-mail: johann.mulzer@univie.ac.at
Homepage: http://www.univie.ac.at/rg_mulzer/
[**] We thank Dr. Lothar Brecker and Susanne Felsinger for NMR
spectra, and Prof. A. Zeeck for an authentic sample of 1.
Allylic alcohol 4,[12] which was available from aldehyde 7
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200900522.
by Nozaki–Hiyama–Kishi addition[13] of isopropenyl bromide
Angew. Chem. Int. Ed. 2009, 48, 6032 –6036
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6033

Communications
(Scheme 3), was connected with carboxylic acid 5 to furnish
ester 14 as the substrate of an Claisen–Ireland rearrangement,
which, after reduction, led to primary alcohol 15 as an easily
separable 4:1 diastereomeric mixture at C16.[14] Reduction
and desilylation gave alcohol 16, which was oxidized to
aldehyde 17.
Scheme 4. Synthesis of benzofuran 2. a) 6, Sn(OTf)2, CH2Cl2, Et3N,
À208C, then À788C, then 17, 87% (d.r. 6:1); b) Me4NBH(OAc)3,
CH3CN/AcOH (2:1), À328C to 08C, 72% (d.r. 20:1); c) LiOH, H2O2,
THF/H2O (3:1), 96%; d) 3m HCl, dioxane, 508C; e) (CH3)2C(OMe)2,
CSA, RT, 85% (2 steps); f) LiOH, THF/MeOH/H2O (2:1:1), 12 h, RT,
84%; g) EDCI, DMAP, DMAP·HCl, CHCl3, reflux, 20 h, 55%; h) hn
(254 nm), cyclohexane, 50 min, 75%; i) NaBH4, MeOH, RT, then 0.5m
HCl; j) TsOH, toluene, 608C, 71% (2 steps). CSA=camphorsulfonic
acid, Ts=4-toluenesulfonyl.
Scheme 3. Synthesis of aldehyde 17. a) CrCl2 (4 equiv), NiCl2
(0.04 equiv), DMF, 08C to RT, 86% (d.r. 1.4:1); b) EDCI, DMAP, 5,
CH2Cl2, 85%; c) LHMDS, HMPA, TBSCl, À788C to reflux; d) LiAlH4,
Et2O, 08C, 84% (d.r. 4:1, 2 steps); e) MsCl, Et3N, CH2Cl2, 08C;
f) LiAlH4, Et2O, 08C, 94% (2 steps); g) TBAF, THF, RT, 93%; h) IBX,
DMSO, RT, 93%. EDCI=1-ethyl-3-(3-dimethylaminopropyl)carbodi-
imide, DMAP=4-(dimethylamino)pyridine, HMDS=hexamethyldisila-
zane, HMPA=hexamethylphosphoramide, TBS=tert-butyldimethyl-
silyl, Ms=methanesulfonyl.
Aldol addition with the Evans ketoimide 6[15] furnished
intermediate 18, which has the full carbon skeleton of
kendomycin. Reduction[16] of the ketone and hydrolytic
removal of the auxiliary led to lactone 19, which was
converted into seco-acid 20 by formation of the acetonide-
protected methylester. Macrolactonization under modified
Boden–Keck conditions[17] gave monomer 3 in 55% yield,
which underwent clean photo-Fries rearrangement to give
ketone 21. Reduction to the secondary alcohol, followed by
removal of the acetonide and SN1 cyclization, furnished key
intermediate 2 (Scheme 4).
Scheme 5. Retrosynthesis of 2: RCM route.
Our second route is outlined in Scheme 5. In contrast to
earlier RCM attempts, RCM of triene 22 should not meet
with major ring strain and thus proceed smoothly, as two
monosubstituted olefins are connected. The tetrahydropyran
ring is formed later, such that the above-mentioned atrop-
isomerism cannot impede the macrocyclization. Triene 22 was
to be constructed from building blocks 23, 24, and 5.
Aldehyde 24 was obtained by Evans aldolization from
ketoimide 6 and acrolein (Scheme 6) to give adduct 25 with
good diastereoselectivity, which was converted into lactone 26
by reduction and removal of the auxiliary. For the conversion
of 26 into aldehyde 24, an analogous sequence was used as for
the preparation of 20 from 19.
For the preparation of allylic alcohol 23, a Duthaler–
Hafner crotylation[18] of methacrolein with titanate 27 was the
method of choice (Scheme 7). Esterification of acid 5 with
alcohol 23 paved the way for the Claisen–Ireland rearrange-
ment, by which ester 28 was smoothly converted into the acid,
and after reduction, into alcohol 29. Deoxygenation furnished
diene 30. Subsequent ortho-directed metalation and addition
of aldehyde 24 gave alcohol 22 as a mixture of diastereomers.
RCM with second-generation Grubbs catalyst[19] induced
smooth ring closure to form the 10,11-E-olefin 31 exclusively.
1
The extremely broad H NMR signal of the 8-CH3 group,
which sharpens to a doublet on raising the temperature to
6034 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 6032 –6036

Angewandte
Chemie
350 K (see Supporting Information), indicates that the con-
formational mobility in 31 is significantly restricted. Selective
reduction of the double bond with diimide,[20] followed by
acid-catalyzed formation of the tetrahydropyran ring and
removal of the OMOM group, led to intermediate 2, which
was oxidized to 1.[4a] The analytical data of the synthetic
material were in full agreement with those of an authentic
sample kindly provided by Professor Zeeck.
In conclusion, we have presented two novel approaches to
the antibiotic kendomycin (RCM route: 23 linear steps,
photo-Fries route: 29 linear steps). Apart from Claisen–
Ireland rearrangements of unusual complexity, this work not
only demonstrates the hitherto unrecognized capability of the
photo-Fries ring contraction for the formation of macrocycles,
but also reemphasizes the unparalleled potential of RCM for
connecting monosubstituted olefin residues.
Scheme 6. Synthesis of aldehyde 24. a) Sn(OTf)2, CH2Cl2, Et3N,
À208C, À788C, then acrolein, 91% (d.r. 5:1); b) Me4NBH(OAc)3,
CH3CN/AcOH (2:1), À328C to 08C, 70% (d.r. 6:1); c) LiOH, H2O2,
THF/H2O (2:1), RT, 72%; d) (CH3)2C(OMe)2, CSA, RT, 91%; e) LiAlH4,
Et2O, 08C, 96%; f) pyridine·SO3, Et3N, CH2Cl2/DMSO, À58C, 99%.
Received: January 28, 2009
Published online: April 6, 2009
Keywords: aldol reactions · antibiotics · benzofurans ·
.
Claisen–Ireland rearrangement · total synthesis
[1] a) H. B. Bode, A. Zeeck, J. Chem. Soc. Perkin Trans. 1 2000, 323;
b) H. B. Bode, A. Zeeck, J. Chem. Soc. Perkin Trans. 1 2000,
2665.
[2] a) S. C. Wenzel, H. B. Bode, I. Kochems, R. Mꢀller, ChemBio-
Chem 2008, 9, 2711; b) C. O. Janssen, S. Lim, E. P. Lo, K. F. Wan,
V. C. Yu, M. A. Lee, S. B. Ng, M. J. Everett, A. D. Buss, D. P.
Lane, R. S. Boyce, Bioorg. Med. Chem. Lett. 2008, 18, 5771.
[3] H. J. Martin, M. Drescher, H. Kꢁhlig, S. Schneider, J. Mulzer,
Angew. Chem. 2001, 113, 3287; Angew. Chem. Int. Ed. 2001, 40,
3186.
[4] a) Y. Yu, H. Men, C. Lee, J. Am. Chem. Soc. 2004, 126, 14720;
b) A. B. Smith III, E. F. Mesaros, E. Meyer, J. Am. Chem. Soc.
2006, 128, 5292; c) A. B. Smith III, E. F. Mesaros, E. Meyer, J.
Am. Chem. Soc. 2005, 127, 6948; d) J. T. Lowe, J. S. Panek, Org.
Lett. 2008, 10, 3813.
[5] K. B. Bahnck, S. D. Rychnovsky, J. Am. Chem. Soc. 2008, 130,
13177.
[6] a) T. Sengoku, D. Uemura, H. Arimoto, Chem. Lett. 2007, 36,
726; b) T. Sengoku, H. Arimoto, D. Uemura, Chem. Commun.
2004, 10, 1220; c) J. D. White, H. Smits, Org. Lett. 2005, 7, 235;
d) D. Williams, K. Shamim, Org. Lett. 2005, 7, 4161; e) J. Mulzer,
S. Pichlmair, M. P. Green, M. M. B. Marques, H. J. Martin, Proc.
Natl. Acad. Sci. USA 2004, 101, 11980.
[7] T. Magauer, Towards the Total Synthesis of the ansa-Macrocycle
Kendomycin, Diploma Thesis, University of Vienna, 2007.
[8] P. Magnus, C. Lescop, Tetrahedron Lett. 2001, 41, 7193.
[9] E. W. Colvin, B. J. Hamill, J. Chem. Soc. Chem. Commun. 1973,
151.
Scheme 7. RCM and synthesis of 1. a) Reaction of but-2-enyl-MgBr
with 27, then methacrolein, Et2O, À788C, 52% (d.r. 50:1, 86% ee);
b) DMAP, EDCI, CH2Cl2, RT, then 23, 81%; c) LHMDS (4 equiv),
HMPA, THF, then 28 dissolved in TBSCl, À788C to RT, then DMF,
microwave irradiation (10 min, 1808C); d) LiAlH4, Et2O, 08C, 89% (d.r.
4:1, 2 steps); e) MsCl, CH2Cl2, 08C; f) LiAlH4, Et2O, 08C, 89%
(2 steps); g) nBuLi, TMEDA, THF, then 24, À788C to À308C, 90%
(d.r. 3.2:1); h) Grubbs II catalyst, 20 mol%, CH2Cl2, reflux, 16 h, 62%
(E only); i) N2(COOK)2, AcOH, CH2Cl2, 40 h, reflux, 60%; j) 3m HCl,
MeOH, RT, 96%; k) TESOTf, Et3N, CH2Cl2, 08C, 82%; l) IBX, DMF, RT,
24 h; m) 0.1m HF, MeCN, RT, 30% (2 steps). TMEDA=tetramethyl-
ethylendiamine, TES=triethylsilyl.
[10] a) E. Negishi, C. Kotora, C. Xu, J. Org. Chem. 1997, 62, 8957;
b) Review: E. Negishi, Q. Hu, Z. Huang, M. Qian, G. Wang,
Aldrichimica Acta 2005, 38, 71.
[11] S. Kulasegaram, R. J. Kulawiec, J. Org. Chem. 1997, 62, 6547.
[12] Compound 4 and its C13 epimer were obtained as an easily
separable 1.4:1 mixture. The epimer was recycled by oxidation to
the ketone and reduction with NaBH4 (d.r. 2:1).
[13] a) Y. Okude, S. Hirano, T. Hiyama, H. Nozaki, J. Am. Chem. Soc.
1976, 98, 3179; b) A. Fꢀrstner, N. Shi, J. Am. Chem. Soc. 1996,
118, 12349.
[14] Claisen–Ireland rearrangements of this dimension are rare in the
literature; we are aware of three examples of similar complexity:
Angew. Chem. Int. Ed. 2009, 48, 6032 –6036
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6035

Communications
a) C. E. Stivala, A. Zakarian, J. Am. Chem. Soc. 2008, 130, 3774;
[17] E. P. Boden, G. E. Keck, J. Org. Chem. 1985, 50, 2394.
[18] a) A. Hafner, R. O. Duthaler, R. Marti, G. Rihs, P. Rothe-Streit,
F. Schwarzenbach, J. Am. Chem. Soc. 1992, 114, 2321; b) Review:
A. Hafner, R. O. Duthaler, Chem. Rev. 1992, 92, 807.
[19] a) M. Scholl, S. Ding, C. W. Lee, R. H. Grubbs, Org. Lett. 1999, 1,
953; b) B. Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc.
1996, 118, 100.
b) M. J. Fisher, C. D. Myers, J. Joglar, S. H. Chen, S. J. Dani-
shefsky, J. Org. Chem. 1991, 56, 5826; c) R. E. Ireland, D.
Hꢁbich, D. W. Norbeck, J. Am. Chem. Soc. 1985, 107, 3271.
[15] a) J. S. Clark, R. Metternich, V. J. Novack, G. S. Sheppard, D. A.
Evans, J. Am. Chem. Soc. 1990, 112, 866; b) D. A. Evans, J. S.
Clark, D. L. Rieger, Tetrahedron 1992, 48, 2127.
[16] D. A. Evans, K. T. Chapman, E. M. Carreira, J. Am. Chem. Soc.
1988, 110, 3578.
[20] D. J. Pasto, R. T. Taylor, Org. React. 1991, 40, 91 – 155.
6036 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 6032 –6036