Total synthesis of phorboxazole B

Article abstract of DOI:10.1002/anie.200603656

(Chemical Equation Presented) Challenge met: In the synthesis of phorboxazole B, a highly efficient hetero-Diels-Alder reaction was used to construct the key C33-C39 linchpin, allowing for the completion of the C18-C46 fragment (see picture). Coupling wit

Full text of DOI:10.1002/anie.200603656

Angewandte
Chemie
DOI: 10.1002/anie.200603656
Natural Product Synthesis
Total Synthesis of Phorboxazole B**
Brian S. Lucas, Vijayagopal Gopalsamuthiram, and Steven D. Burke*
Since their isolation by Molinski and Searle in 1995,[1] the
structurally complex phorboxazoles A and B (Scheme 1)
have sparked a flurry of interest from the synthetic commun-
ity. Comprising 15 asymmetric centers, four diversely sub-
stituted hydropyran rings, a 21-membered macrolactone, two
oxazole rings, and a variety of unsaturations, the phorbox-
azoles present a considerable challenge. A total of six
research groups have reported syntheses of the phorboxa-
zoles, beginning with Forsythꢀs synthesis of phorboxazole A
(1)in 1998, [2] followed by Evansꢀs synthesis of phorboxazole B
[3]
(2)in 2000. Subsequent syntheses of phorboxazole A have
been completed by the research groups led by Smith (2001,[4]
2005[5]), Pattenden (2003),[6] and Williams (2003),[7] and of
phorboxazole B by Zhou et al. (2006).[8] In addition, numer-
ous other synthetic efforts have been described.[9]
The unique biological activity of the phorboxazoles is
another driving force for their study. The phorboxazoles are
among the most potent cytostatic agents yet discovered,
exhibiting a mean GI50 < 1.58 ” 10À9 m against the NCI panel
of 60 tumor cell lines.[10] While it is known that the
phorboxazoles induce S-phase cell cycle arrest without
interference of the microtubules, the exact mode of activity
is not fully understood. Recently, Forsyth, La Clair et al. have
shown that fluorescently labeled phorboxazole derivatives
induce association of cell-cycle-dependent kinase 4 (cdk4)
with extranuclear cytokeratin intermediate filaments
(KRT10),[11] and perturbation of cdk4 is known to inhibit
cell cycle progression at the G1/S phase.[12] In addition, new,
potent analogues are beginning to shed some light on the
phorboxazole pharmacophore.[13]
Our interest in the phorboxazoles was kindled by the
hidden symmetry we identified in the C5–C15 bis(oxane)
segment of the phorboxazoles, for which we pursued a two-
Scheme 1. Phorboxazole retrosynthesis. Protecting groups are defined
directional synthesis/desymmetrization strategy to the C3–
C15 cores of phorboxazole A and B.[14] Further investigations
led to a functionalized C1–C17 fragment of phorboxazo-
le B,[15] and a catalytic enantioselective hetero-Diels–Alder
in Scheme 2.
approach to the stereochemically complex C20–C32 seg-
ment.[16] We report herein the culmination of our efforts
toward the total synthesis of phorboxazole B.
Our retrosynthesis for phorboxazole B (2)is shown in
Scheme 1. As in the Forsyth example, the C16–C18 oxazole
ring would be constructed by a coupling of a C18 carboxylic
acid to a C16 amine, followed by cyclodehydration according
to the Wipf procedure.[17] In an effort to minimize steps, the
cyclodehydration of 3 would be performed in the presence of
[*] Dr. B. S. Lucas, Dr. V. Gopalsamuthiram, Prof. S. D. Burke
Department of Chemistry
The University of Wisconsin—Madison
1101 University Avenue, Madison, WI 53706 (USA)
Fax: (+1)608-265-4534
E-mail: burke@chem.wisc.edu
[**] The NIH (grants CA108488 andCA74394) graciously fundedthis
research. We thank M. Dodge, L. Wysocki, J. Schuster, L. Boyle, L.
Luther, andL. Konkel for providing starting materials crucial to this
work. We are grateful to Dr. I. Guzei for the X-ray crystal structure
determination of 19. Professors C. J. Forsyth (University of Minne-
sota) andD. R. Williams (Indiana University) are acknowledgedfor
helpful discussions regarding the hydrolysis of 39.
À
a C3 aldehyde. The C2 C3 Z double bond would be
constructed by a well-precedented intramolecular Horner–
Wadsworth–Emmons reaction (HWE),[2,4–8] necessitating a
C3–C17 fragment 4. Application of Evansꢀs conditions for
thermodynamic oxazole metalation[3] would serve to join the
C20–C32 fragment 5[16] with a C33–C39 lactone 6. Julia–
Kocienski olefination[18] using sulfone 7,[19] as in the Wil-
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 769 –772
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
769

Communications
liams,[7] Pattenden,[6] and Zhou[8] examples, would install the
C42–C46 side chain.
As a result of some late-stage difficulties with our
incorporation of the C1–C3 Z enoate early on,[20] we revised
our synthesis and required a new C3–C17 phorboxazole B
fragment 4. As shown in Scheme 2, the previously constructed
Scheme 3. Synthesis of the C33–C39 lactone 6. a)[Eu(fod)3], CH2Cl2,
08C, 71%; b) Pd/C, H2, EtOAc, 85%; c) MMTr-Cl, pyridine, DMAP,
100%; d) TBDPS-Cl, AgNO3, pyridine, 508C, 71%; fod=tris(6,6,7,7,-
8,8,8-heptafluoro-2,2-dimethyl-3,5-octanedione), MMTr=monomethox-
ytriphenylmethyl, DMAP=N,N-dimethylaminopyridine.
17 to afford 18 proceeded smoothly under Eu3+ catalysis.[30]
Catalytic hydrogenation, serving to both saturate the ring and
liberate the C38 and C39 hydroxy groups, provided crystalline
diol 19, and the stereochemistry at C35 and C37 was
unambiguously determined by X-ray crystallography. Selec-
tive protection of the primary hydroxy group as the mono-
methoxytrityl ether (MMTr)[31] was accomplished in quanti-
tative yield to afford 20, followed by installation of the
TBDPS ether at C38 to give the complete C33–C39 lactone 6
in four steps from Brassardꢀs diene 17.
Phorboxazole B (2)was assembled from subunits 4–7 as
shown in Scheme 4. Coupling of 5 and 6 was achieved by the
strategy of Evans for thermodynamic oxazole metalation[3]
under conditions of high concentration to afford hemiketal 21
in 75% yield (91% based on recovered starting material 5).
Treatment of 21 with PPTS in MeOH served to cleave the
extremely acid-labile MMTr ether at C39[31] and form the
mixed methyl ketal 22 in a single operation. Oxidation to the
C39 aldehyde 23 using Dess–Martin periodinane[21] was
followed by installation of the trisubstituted double bond
using the stabilized Wittig reagent 24 to afford E (a,b)-
unsaturated ethyl ester 25. Dibal reduction to give the allylic
alcohol 26 was followed by careful oxidation yielding the
Julia–Kocienski coupling partner 27. Application of Wil-
liamsꢀs conditions[19] for olefination using sulfone 7 delivered
28 (86%)as a 92:8 mixture of E/Z isomers at C41–C42.[32]
Methanolic p-TsOH proved suitable for the selective
removal of the primary TBDPS ether at C20 in the presence
of the secondary allylic TBDPS ether at C38 to provide the
primary alcohol 29. Oxidation to aldehyde 30 was followed by
homologation using stabilized Wittig reagent 31 to give ethyl
ester 32. Finally, saponification using LiOH afforded the
complete C18–C46 fragment 33.
Scheme 2. Synthesis of the C3–C17 fragment 4. a) LiBH4, THF, 08C,
97%; b) TES-Cl, 2,6-lutidine, CH2Cl2, À788C, 93%; c) Dess–Martin
periodinane, CH2Cl2, 94%; d) [Cp2TiMe2], 1:1 THF/toluene, 808C,
77%; e) p-TsOH, MeOH, 408C, 93%; f) TES-Cl, imidazole, CH2Cl2,
À788C, 76%; g) PPh3, DEAD, DPPA, 88%; h) PPh3, H2O, THF, reflux,
76%. TBDPS=tert-butyldiphenylsilyl, TES=triethylsilyl, p-TsOH=p-tol-
uenesulfonic acid, DEAD=diethylazodicarboxylate, DPPA=diphenyl-
phosphorylazide.
bicyclic lactone 8[15] was reduced to diol 9 with LiBH4,
followed by selective protection of the primary hydroxy
function. Dess–Martin[21] oxidation of the secondary alcohol
in 10 yielded ketone 11, which was olefinated using the Petasis
reagent[22] to install the methylene unit in 12. Acidic removal
of the C16–C17 acetonide and the TES ether yielded triol 13,
which was selectively silylated at the primary sites at C3 and
C17 to provide bis(TES ether) 14. Mitsunobu[23] reaction of
the free secondary alcohol at C16 with diphenylphosphor-
ylazide (DPPA)[24] yielded azide 15, which underwent Stau-
dinger reduction[25] to provide the desired C3–C17 amine 4.
We anticipated that an efficient entry to the C33–C39
lactone 6 could arise from a chelation-controlled hetero-
Diels–Alder reaction of the mannitol-derived aldehyde 16[26]
with Brassard diene 17[27] (Scheme 3). The stereochemical
outcome in 18 was predicted based on the work of Midland
et al.,[28] although the presence of a b-alkoxy substituent could
potentially interfere with the chelation control involving the
stereogenic center at the a position.[29] Combination of 16 and
EDCI-mediated coupling of amine 4 and carboxylic acid
33 produced amide 34 in 85% yield. Removal of the TES
ethers at C3 and C17 proceeded in near-quantitative yield to
give diol 35. The key Dess–Martin oxidation[21]/cyclodehydra-
tion[17] sequence[2] ultimately afforded oxazole 36 in 71%
overall yield with the C3 aldehyde remaining intact. In
preparation for the HWE closure of the macrocycle, oxidative
770
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 769 –772

Angewandte
Chemie
Scheme 4. Completion of the total synthesis of phorboxazole B (2). a) LiNEt2, 6, THF À788C 75%, 91% basedon recoveredstarting material;
b) PPTS, MeOH, 80%; c) Dess–Martin periodinane, CH2Cl2, 92%; d )24, toluene, 808C, 95%; e) Dibal-H, CH2Cl2,À788C, 95%; f) Dess–Martin
periodinane, 2,6-lutidine, CH2Cl2, 92%; g) 7, NaHMDS, THF, À788C to RT, 86% 92:8 E/Z; h) p-TsOH, MeOH, 90% after one recycle; i) Dess–
Martin periodinane, CH2Cl2, 90%; j) 31, toluene, 808C, 90%; k) LiOH, THF/MeOH/H2O (4:1:1), 80%; l) 4, EDCI·MeI, HOBt, CH2Cl2, 85%;
m) 0.1n HCl, MeOH, 08C, 5 min, 97%; n) Dess–Martin periodinane, CH2Cl2; o) 2,6-DBMP, PPh3, (CCl2Br)2, CH2Cl2, 08C then DBU, MeCN, RT,
71% (2 steps); p) DDQ, CH2Cl2, pH 7 buffer, 76%; q) DIC, dimethylphosphonoacetic acid, CH2Cl2, 93%; r) K2CO3, [18]crown-6, toluene, À408C to
RT, 93% 3:1 Z/E; s) TBAF, THF, 74%; t) 0.72n HCl, THF, RT, 60 h, 83%. PPTS=pyridinium p-toluenesulfonic acid, Dibal-H=diisobutylaluminum
hydride, NaHMDS=sodium bis(trimethylsilyl)amide, EDCI·MeI=1-ethyl-3-(3’-dimethylaminopropyl)carbodiimide methiodide, HOBt=1-hydroxy-
benzotriazole, DBMP=di-tert-butyl-4-methylpyridine, DBU=1,8-diazabicyclo[5.4.0]undec-7-ene, DDQ=2,3-dichloro-5,6-dicyano-1,4-benzoquinone;
DIC=diisopropylcarbodiimide, TBAF=tetrabutylammonium fluoride.
deprotection of the PMB ether at C24 to give secondary
alcohol 37 was followed by installation of the dimethylphos-
phonoacetate moiety in 38 using DIC. Ring closure ensued
upon treatment with K2CO3/[18]crown-6 to afford 39 as a 3:1
mixture of Z/E isomers at C2–C3, which were separable by
silica gel column chromatography. Lastly, removal of the silyl
protecting groups using TBAF gave the C33-methoxy-pro-
tected derivative 40 of phorboxazole B which was subse-
quently hydrolyzed to provide phorboxazole B (2)in 83%
yield by stirring in 0.72n HCl[33] for 60 h.[2,4,5,7,8]
Angew. Chem. Int. Ed. 2007, 46, 769 –772
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
771

Communications
[12] a)J. M. Paramio, M. L. Casanova, C. Segrelles, S. Mittnacht,
E. B. Lane, J. L. Jorcano, Mol. Cell. Biol. 1999, 19, 3086; b)G.-Z.
Tao, L. S. Rott, A. W. Lowe, M. B. Omary, J. Biol. Chem. 2002,
277, 19295.
[13] a)F. M. Uckun, C. J. Forsyth, Bioorg. Med. Chem. Lett. 2001, 11,
1181; b)A. B. Smith III, T. M. Razler, G. R. Pettit, J.-C. Chapuis,
Org. Lett. 2005, 7, 4403.
[14] B. S. Lucas, S. D. Burke, Org. Lett. 2003, 5, 3915.
[15] B. S. Lucas, L. M. Luther, S. D. Burke, Org. Lett. 2004, 6, 2965.
[16] B. S. Lucas, L. M. Luther, S. D. Burke, J. Org. Chem. 2005, 70,
3757.
[17] a)P. Wipf, S. Lim, J. Am. Chem. Soc. 1995, 117, 558; b)P. Wipf,
T. H. Graham, J. Org. Chem. 2001, 66, 3242.
In conclusion, modification of our previously reported
synthesis of the C1–C17 fragment afforded the required C3–
C17 fragment 4. A rapid synthesis of the C33–C39 lactone 6
by diastereoselective hetero-Diels–Alder reaction facilitated
construction of the C18–C46 acid segment 33 from previously
reported fragments. Coupling of the subunits 4 and 33 by
amidation was followed by a highly efficient oxidation/
cyclodehydration sequence to install the C16–C18 oxazole
ring. The synthesis of phorboxazole B was thus completed in
28 steps along the longest linear sequence (55 total steps)
from readily available starting materials. The phorboxazole B
produced by this route provides 1H NMR, HRMS, IR,
Rf(TLC), and optical rotation data[1,34] that match those of
naturally occurring phorboxazole B.
[18] For an excellent review on the modified Julia olefination, see
P. R. Blakemore, J. Chem. Soc. Perkin Trans. 1 2002, 2563.
[19] D. R. Williams, A. A. Kiryanov, U. Emde, M. P. Clark, M. A.
Berliner, J. T. Reeves, Proc. Natl. Acad. Sci. USA 2004, 101,
12058.
Received: September 6, 2006
[20] Our difficulties with the sensitive C1–C3 Z enoate will be
described in a forthcoming full account of this research.
[21] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277.
[22] N. A. Petasis, E. I. Bzowej, J. Am. Chem. Soc. 1990, 112, 6392.
[23] O. Mitsunobu, Synthesis 1981, 1.
[24] B. Lal, B. N. Pramanik, M. S. Manhas, A. K. Bose, Tetrahedron
Lett. 1977, 18,1977.
[25] M. Vaultier, N. Knouzi, R. CarriØ, Tetrahedron Lett. 1983, 24,
763.
Published online: December 13, 2006
Keywords: cycloaddition · cyclodehydration · natural products ·
.
symmetry · total synthesis
[1] P. A. Searle, T. F. Molinski, J. Am. Chem. Soc. 1995, 117, 8126.
[2] C. J. Forsyth, F. Ahmed, R. D. Cink, C. S. Lee, J. Am. Chem. Soc.
1998, 120, 5597.
[26] R. Badorrey, C. Cativiela, M. D. Díaz-de-Villegas, J. A. Gµlvez,
Tetrahedron 2002, 58, 341.
[3] D. A. Evans, D. M. Fitch, T. E. Smith, V. J. Cee, J. Am. Chem.
Soc. 2000, 122, 10033.
[4] A. B. Smith III, K. P. Minbiole, P. R. Verhoest, M. Schelhass, J.
Am. Chem. Soc. 2001, 123, 10942.
[5] A. B. Smith III, T. M. Razler, J. P. Ciavarri, T. Hirose, T.
Ishikawa, Org. Lett. 2005, 7, 4399.
[6] M. A. Gonzµlez, G. Pattenden, Angew. Chem. 2003, 115, 1293;
Angew. Chem. Int. Ed. 2003, 42, 1255.
[27] a)J. Savard, P. Brassard, Tetrahedron Lett. 1979, 20, 4911; b)Q.
Fan, L. Lin, Y. Huang, X. Feng, G. Zhang, Org. Lett. 2004, 6,
2185; c)J. D. Winkler, K. Oh, Org. Lett. 2005, 7, 2421.
[28] a)M. M. Midland, R. W. Koops, J. Org. Chem. 1990, 55, 5058;
b)M. M. Midland, R. S. Graham, J. Am. Chem. Soc. 1984, 106,
4294.
[29] K. Mead, T. L. Macdonald, J. Org. Chem. 1985, 50, 422.
[30] M. Bednarski, S. Danishefsky, J. Am. Chem. Soc. 1983, 105, 3716.
[31] M. Smith, D. H. Rammler, I. H. Goldberg, H. G. Khorana, J.
Am. Chem. Soc. 1962, 84, 430.
[7] D. R. Williams, A. A. Kiryanov, U. Emde, M. P. Clark, M. A.
Berliner, J. T. Reeves, Angew. Chem. 2003, 115, 1296; Angew.
Chem. Int. Ed. 2003, 42, 1258.
[8] D.-R. Li, D.-H. Zhang, C.-Y. Sun, J. W. Zhang, L. Yang, J. Chen,
B. Liu, C. Su, W.-S. Zhou, G.-Q. Lin, Chem. Eur. J. 2006, 12, 1185.
[9] a)For a brief review, see L. O. Haustedt, I. V. Hartung, H. M. R.
Hoffmann, Angew. Chem. 2003, 115, 2815; Angew. Chem. Int.
Ed. 2003, 42, 2711; b)For a comprehensive list of synthetic
studies directed at phorboxazole subunits, see the citations of
reference [16].
[10] T. F. Molinski, Tetrahedron Lett. 1996, 37, 7879.
[11] C. J. Forsyth, L. Ying, J. Chen, J. J. La Clair, J. Am. Chem. Soc.
2006, 128, 3858.
[32] The undesired Z isomer was most easily removed by preparative
HPLC at this stage. See the Supporting Information.
[33] References [2,4,5,7,8] report this as “6% aqueous HCl”, which
is ambiguous. The concentration of aqueous HCl used in the
Forsyth and Williams work can also be described more precisely
as 0.72n. We thank Profs. Forsyth and Williams for clarification.
[34] [a]2D2 = + 31 degcm3 gÀ1 dmÀ1 (c = 0.1 gcmÀ3, MeOH). The opti-
cal rotation of natural phorboxazole B is reported as [a]D =
+ 44.4 degcm3 gÀ1 dmÀ1 (c = 1.0 gcmÀ3, MeOH)(see Ref. [1]).
772
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 769 –772