Synthesis of the C20-C32 tetrahydropyran core of the phorboxazoles and the C22 epimer via a stereodivergent Michael reaction

Article abstract of DOI:10.1021/ol3026523

A stereoselective synthesis of the C20-C32 tetrahydropyran core of the phorboxazoles has been achieved in only seven steps and in a 31% overall yield. The C22 epimer was also synthesized. The key step was a silyl ether deprotection/oxy-Michael cyclization. When this step was conducted under Bronsted acid conditions, the C20-C32 core was formed with the desired 2,6-cis-stereochemistry. However, when the silyl ether deprotection/oxy-Michael cyclization was conducted under fluoride conditions buffered with acetic acid, the C22 epimer of the core was the sole product.

Full text of DOI:10.1021/ol3026523

ORGANIC
LETTERS
2012
Vol. 14, No. 21
5550–5553
Synthesis of the C20ÀC32
Tetrahydropyran Core of the
Phorboxazoles and the C22 Epimer via
a Stereodivergent Michael Reaction
Paul A. Clarke* and Kristaps Ermanis
Department of Chemistry, University of York, Heslington, York YO10 5DD,
United Kingdom
paul.clarke@york.ac.uk
Received September 26, 2012
ABSTRACT
A stereoselective synthesis of the C20ÀC32 tetrahydropyran core of the phorboxazoles has been achieved in only seven steps and in a 31%
overall yield. The C22 epimer was also synthesized. The key step was a silyl ether deprotection/oxy-Michael cyclization. When this step was con-
ducted under Brønsted acid conditions, the C20ÀC32 core was formed with the desired 2,6-cis-stereochemistry. However, when the silyl ether
deprotection/oxy-Michael cyclization was conducted under fluoride conditions buffered with acetic acid, the C22 epimer of the core was the sole product.
Since their isolation and characterization by Searle and
Molinski in 1995, the phorboxazole natural products1
(Scheme 1) have been the subjectofintense synthetic study.
This is in part due to their challenging molecular architec-
ture which contains four differently substituted tetrahy-
dropyran (THP) rings, two oxazole rings, seven alkene
units, and 15 stereogenic centers. It is also due to their
subnanomolar (GI50 < 8 Â 10À10 M) activity against the
entire range of human tumor cell lines held by the NIH.
Due to the scarcity of these molecules from the natural
source, theonlyway to realistically provideenoughmaterial
for a full biological analysis is via an efficient total
synthesis. To date, there have been seven successful total
syntheses of phorboxazole A2 and three of phorboxazole B.3
In addition to these syntheses, several groups have tar-
geted the C20ÀC32 penta-substituted THP core which is
common to both structures.4
(1) Searle, P. A.; Molinski, T. F. J. Am. Chem. Soc. 1995, 117, 8126.
(2) (a) Forsyth, C. J.; Ahmed, F.; Cink, R. D.; Lee, C. S. J. Am. Chem.
Soc. 1998, 120, 5597. (b) Smith, A. B., III; Verhoest, P. R.; Minbiole,
K. P.; Schelhaas, M. J. Am. Chem. Soc. 2001, 123, 4834. (c) Gonzalez,
M. A.; Pattenden, G. Angew. Chem., Int. Ed. 2003, 42, 1255. (d) Williams,
D. R.; Kiryanov, A. A.; Emde, U.; Clark, M. P.; Berliner, M. A.; Reeves,
J. T. Angew. Chem., Int. Ed. 2003, 42, 1258. (e) Pattenden, G.; Gonzalez,
M. A.;Little, P. B.;Milan, D. S.;Plowright, A. T.;Tornos, J. A.;Ye, T.Org.
Biomol. Chem. 2003, 1, 4173. (f) Smith, A. B., III; Razler, T. M.; Ciavarri,
J. P.; Hirose, T.; Ishikawa, T. Org. Lett. 2005, 7, 4399. (g) White, J. D.;
Kuntiyong, P.; Tae, H. L. Org. Lett. 2006, 8, 6039. (h) White, J. D.;
Kuntiyong, P.; Tae, H. L. Org. Lett. 2006, 8, 6043. (i) Smith, A. B., III;
Razler, T. M.; Ciavarri, J. P.; Hirose, T.; Ishikawa, T.; Meis, R. M. J. Org.
Chem. 2008, 73, 1192. (j) Wang, B.; Hansen, T. M.; Wang, T.; Wu, D.;
Weyer, L.; Ying, L.; Engler, M. M.; Sanville, M.; Leitheiser, C.;
Christmann, M.; Lu, Y.; Chen, J.; Zunker, N.; Cink, R. D.; Ahmed,
F.; Lee, C.-S.; Forsyth, C. J. J. Am. Chem. Soc. 2011, 133, 1484. (k)
Wang, B.; Hansen, T. M.; Weyer, L.; Wu, D.; Wang, T.; Christmann,
M.; Lu, Y.; Ying, L.; Engler, M. M.; Cink, R. D.; Lee, C.-S.; Ahmed, F.;
Forsyth, C. J. J. Am. Chem. Soc. 2011, 133, 1506.
Over the past few years, we have spent some effort develop-
ing new and efficient routes, based on the MaitlandÀJapp
reaction, for the synthesis of functionalized THP rings
(Scheme 2).5,6
r
10.1021/ol3026523
Published on Web 10/22/2012
2012 American Chemical Society

fragment 4 of phorboxazole B, our asymmetric diketene
MaitlandÀJapp reaction5e enabled us to complete an
efficient synthesis.7 This was later improved by our devel-
opment of a highly asymmetric Chan’s diene MaitlandÀ
Japp reaction.5f However, when we turned our attention to
the synthesis of the C20ÀC32 core 5, the MaitlandÀJapp
chemistry was not able to provide access to diastereomeri-
cally pure THPs insufficient quantities tocontinue the syn-
thesis.8 Therefore, an alternative strategy for the synthesis
of this fragment was required.
Scheme 1. Retrosynthetic Analysis of the Phorboxazoles:
Phorboxazole A (1), R1 = H, R2 = OH; Phorboxazole B (2),
R1 = OH, R2 = H
The key ring-forming step in the MaitlandÀJapp reac-
tion was a 6-endo-trig oxy-Michael reaction. We decided to
investigate the use of the alternative 6-exo-trig reaction.9
However, there are drawbacks with the use of 6-exo-trig
reactions for the formation of cis-THP rings, in that many
of these cyclizations produce the 2,6-trans-product under
kinetic conditions. The 2,6-cis-product can be favored by
operating under thermodynamic conditions, although of-
tentimes the 2,6-cis-selectivities of these reactions are low
and the conditions forcing.2j,10
Recently, Fuwa speculated that the replacement of an
R,β-unsaturated ester Michael acceptor with an R,β-un-
saturated thioester Michael acceptor may lead toenhanced
2,6-cis-selectivity as it mimics the acyclic polyketide chain
bound via a thioester linkage to the acyl carrier protein in
the pyran synthase-mediated biosynthesis of THPs. Re-
markably, he found that this change led to THP products
being formed with excellent 2,6-cis-selectivities, although
forcing conditions were still required (70 °C for 7.5 h).11
We decided to study the 6-exo-trig oxy-Michael reac-
tion of both an R,β-unsaturated ester Michael acceptor
and an R,β-unsaturated thioester Michael acceptor to
see if there was a significant difference in the diastereo-
selectivities.
Scheme 2. MaitlandÀJapp Reaction
Our synthesis commenced (Scheme 3) with an anti-aldol
reaction using the MasamuneÀAbiko auxiliary 6 and
aldehyde 7,12 promoted by (cy-hex)2BOTf and Et3N. This
produced the desired anti-aldol adduct 8 in 91% yield as a
14:1 mixture of diastereomers, which were separated by
column chromatography. Protection of the allylic alcohol
as a TBS-ether was achieved in 93% using TBSOTf and
2,6-lutidine in CH2Cl2 at 0 °C. Reductive removal of the
We considered phorboxazole a target which would
provide a formidable test of our methodology. Our strat-
egy was to break phorboxazole into three fragments 3, 4,
and 5 of approximately equal complexity (Scheme 1). In
the case of the C1ÀC19 2,6-cis- and 2,6-trans-bispyran
(3) (a) Evans, D. A.; Cee, V. J.; Smith, T. E.; Fitch, D. M.; Cho, P. S.
Angew. Chem., Int. Ed. 2000, 39, 2533. (b) Evans, D. A.; Fitch, D. M.
Angew. Chem., Int. Ed. 2000, 39, 2536. (c) Evans, D. A.; Fitch, D. M.;
Smith, T. E.; Cee, V. J. J. Am. Chem. Soc. 2000, 122, 10033. (d) Li, D.-R.;
Zhang, D.-H.; Sun, C.-Y.; Zhang, J.-W.; Yang, L.; Chen, J.; Liu, B.; Su,
C.; Zhou, W.-S.; Lin, G.-Q. Chem.;Eur. J. 2006, 12, 1185. (e) Lucas,
B. S.; Gopalsamuthiram, V.; Burke, S. D. Angew. Chem., Int. Ed. 2007,
46, 769.
(4) (a) Paterson, I.; Arnott, E. A. Tetrahedron Lett. 1998, 39, 7185.
(b) Rychnovsky, S. D.; Thomas, C. R. Org. Lett. 2000, 2, 1217.
(c) Hoffmann, H. M. R.; Misske, A. M. Tetrahedron 1999, 55, 4315.
(d) Yadav, J. S.; Satyanarayana, M.; Srinivasulu, G.; Kunwar, A. C.
Synlett 2007, 1577. (e) Chakraborty, T. K.; Reddy, V. R.; Reddy, T. J.
Tetrahedron 2003, 59, 8613.
(5) For syntheses using the MaitlandÀJapp reaction, see: (a) Japp,
F. R.; Maitland, W. J. Chem. Soc. 1904, 85, 1473. (b) Clarke, P. A.;
Martin, W. H. C. Org. Lett. 2002, 4, 4527. (c) Clarke, P. A.; Martin,
W. H. C.; Hargreaves, J. M.; Wilson, C.; Blake, A. J. Chem. Commun.
2005, 1061. (d) Clarke, P. A.; Martin, W. H. C.; Hargreaves, J. M.;
Wilson, C.; Blake, A. J. Org. Biomol. Chem. 2005, 3, 3551. (e) Clarke,
P. A.; Santos, S.; Martin, W. H. C. Green Chem. 2007, 9, 438. (f) Iqbal,
M.; Mistry, N.; Clarke, P. A. Tetrahedron 2011, 67, 4960.
(6) For general reviews on THP synthesis, see: (a) Larrosa, I.; Romea,
P; Urpı Tetrahedron 2008, 64, 2683. (b) Clarke, P. A.; Santos, S. Eur. J.
Org. Chem. 2006, 2045.
(7) Clarke, P. A.; Santos, S.; Mistry, N.; Burroughs, L.; Humphries,
A. C. Org. Lett. 2011, 13, 624.
(8) Clarke, P. A.; Hargreaves, J. M.; Woollaston, D. J.; Rodriguez
Sarmiento, R. M. Tetrahedron Lett. 2010, 51, 4731.
(9) For reviews on oxy-Michael reactions, see: (a) Nising, C. F.;
€
Brase, S. Chem. Soc. Rev. 2012, 41, 988. (b) Fuwa, H. Heterocycles 2012,
85, 1255.
(10) (a) Bates, R. W.; Song, P. Synthesis 2010, 2935. (b) Uenishi, J.;
ꢀ
Iwamoto, T.; Tanaka, J. Org. Lett. 2009, 11, 3262. (c) Ferrie, L.;
Boulard, L.; Pradaux, F.; Bouzbouz, S.; Reymond, S.; Capdevielle, P.;
Cossy, J. J. Org. Chem. 2008, 73, 1864. (d) Banwell, M. G.; Bui, C. T.;
Pham, H. T. T.; Simpson, G. W. J. Chem. Soc., Perkin Trans. 1 1996,
967.
(11) Fuwa, H.; Noto, K.; Sasaki, M. Org. Lett. 2011, 13, 1820.
(12) (a) White, J. D.; Blakemore, P. R.; Green, N. J.; Hauser, B.;
Holoboski, M. A.; Keown, L. E.; Nylund Kolz, C. S.; Phillips, B. W.
J. Org. Chem. 2002, 67, 7750. (b) Lafontaine, J. A.; Provencal, D. P.;
Gardelli, C.; Leahy, J. W. J. Org. Chem. 2003, 68, 4215.
Org. Lett., Vol. 14, No. 21, 2012
5551

auxiliary with DIBAL-H in CH2Cl2 gave alcohol 9 in 82%
yield. Oxidation with DessÀMartin periodinane gave the
aldehyde in 94% yield. However, it was found that this
aldehyde was unstable, and so it was immediately sub-
jected to FelkinÀAnh controlled addition of (E)-crotyl-
boronic acid pinacol ester13 to generate 10 as a single
diastereomer in 86% yield, which contains the anti, syn-,
anti-, stereochemicaltetradrequired forthesynthesisofthe
C20ÀC32 core.
reported by Lipshutz,16 resulted in 12 being formed in an
improved 79% yield.
Scheme 4. Synthesis of the Cyclization Precursors
Scheme 3. Construction of the Stereochemical Tetrad
With 11 and 12 in hand, we could now consider the
cyclization step. Fuwa conducted his original cyclizations
on the free alcohol using CSA in dichloroethane at 70 °C11
but also reported conditions which delivered the cycliza-
tion in CH2Cl2 at rt.11,17 As CSA has been used to remove
TBS-ethers at rt, we attempted to remove the TBS-ether
with CSA at rt and then, if required, raise the temperature
to 70 °C to affect the cyclization.
Subjecting 11 to Fuwa’s conditions at rt, and at elevated
temperatures, resulted in decomposition with no identifi-
able products being returned (Table 1, entry a). When 12
was treated under these conditions, deprotection was seen
to occur slowly at rt and more rapidly at higher tempera-
tures. At higher temperatures (55 °C), some traces of a
cyclized product 16 could be detected, but the major
product was the deprotected ester 13 (Table 1, entries b
and c). Extended reaction times led to additional products
being formed, which seemed to derive from decomposition
ofthe silyl ether12. As R,β-unsaturated thioesters are more
enone-like than R,β-unsaturated oxoesters, it is perhaps
not surprising that the Brønsted acid catalyzed Michael
addition in 12 generated some cyclized product whereas 11
resulted in decomposition.
With this in mind, an alternative procedure to affect
rapid silyl deprotection, before 11 or 12 could decompose,
was needed. We therefore switched to TBAF buffered with
AcOH. We anticipated that under these conditions fluoride-
mediated deprotection would occur rapidly, and as R,β-
unsaturated thioesters are more enone-like than R,β-un-
saturated oxoesters, cyclization might occur to give the
2,6-cis-THP core. The phenomenon of enones undergoing
Michael cyclization to give 2,6-cis-THPs, while R,β-unsa-
turated oxoesters cyclize to give 2,6-trans-THPs, was
recently reported by Bates.10a When this was attempted
using 11, we found that in situ cyclization did follow
deprotection to give the 2,6-trans-THP 14 as the sole
product (Table 1, entry d). When these conditions were
With 10 in hand, we could now consider converting it to
both the oxoester 11 and the thioester 12. It was decided to
investigate the use of an olefin cross-metathesis reaction to
achieve this transformation.11 In the case of 11, this would
nessecitate the use of cresol acrylate14 as a reagent, and in
the case of the 12, thiocresol acrylate15 would be used. The
HoveydaÀGrubbs II (HG-II) catalyst was used to pro-
mote these reactions (Scheme 4). The formation of ester 11
in 60% occurred without incident when 10 was heated at
55 °C in toluene with cresol acrylate and HG-II catalyst.
However, the cross-metathesis reaction of 10 with thiocre-
sol acrylate was not straightforward. Application of the
same conditions provided 12 in only 28% yield. Increasing
the temperature of the reaction resulted in a reduction
in the yield of 12 to 20% at 70 °C and no product at
90 °C. Changing the solvent to CH2Cl2 and increasing
the catalyst loading did lead to an increase in the yield
of 12 to 50%. It is possible that thiocresol acrylate is
particularly prone to self-dimerization under metathesis
conditions, although we were never able to recover any
of this dimer. It is also possible that the catalyst decom-
posed at elevated temperatures. However, application of
conditions using a CuI additive, based on those recently
(13) Hoffmann, R. W.; Weidmann, U. Chem. Ber. 1985, 118, 3966.
(14) Drewes, S. E.; Emslie, N. D.; Karodia, N. Synth. Commun. 1990,
20, 1915.
(15) Fuwa, H. Personal communication.
(16) Voigtritter, K.; Ghorai, S.; Lipshutz, B. H. J. Org. Chem. 2011,
76, 4697.
(17) Fuwa, H.; Ichinokawa, N.; Noto, K.; Sasaki, M. J. Org. Chem.
2012, 77, 2588.
5552
Org. Lett., Vol. 14, No. 21, 2012

repeated with thioester 12, deprotection and cyclization
occurred rapidly to form 2,6-trans-THP 15 as the only
product, contrary to our hypothesis (Table 1, entry e).
While we were pleased that we had found conditions to
effectdeprotectionand cyclization inthesamepot, wewere
disappointed that the fluoride/Brønsted acid buffered
conditions had led to the formation of the undesired 2,6-
trans-THP 15.18
Table 1. Attempted Cyclization of 11 and 12
We therefore returned to the use of a Brønsted acid
catalyzed deprotection. We chose TFA as it is more acidic
than CSA, and we hoped that it would promote faster
deprotection of the silyl ether and catalyze the cyclization
reaction. When thioester 12 was treated with TFA in
CH2Cl2/H2O, we were pleased to see cyclization occur
and to isolate a 13:1 mixture of the desired 2,6-cis-THP 16
to undesired 2,6-trans-THP 15 (Table 1, entry f). Tetra-
hydropyrans 15 and 16 were generated cleanly as the only
products and could be separated by SiO2 flash column
chromatography to provide 16 in a 71% yield.18 When 11
was subjected to these conditions (Table 1, entry g), 13 was
formed in 68% yield.
The successful formation of 16 constitutes a synthesis
of the C20ÀC32 core of the phorboxazoles, which was
achieved in only seven steps and a 31% overall yield. This
compares favorably with previous syntheses of the frag-
ment.2À4 We were surprised to encounter a Michael cycli-
zation reaction utilizing a thioester electrophile which could
be made to generate either the 2,6-cis- or 2,6-trans-THP
unit with excellent selectivity by initiating the cyclization
with either TFA or TBAF buffered with AcOH, respec-
tively. However, when an oxoester was used as a Michael
acceptor, the AcOH buffered TBAF conditions generated
the 2,6-trans-THP, while Brønsted acid mediated condi-
tions led to silyl deprotection and decomposition. We are
now seeking to examine this switch in the selectivity of the
cyclization process more fully and to take the C20ÀC32
phorboxazole core16forward tocomplete a total synthesis
of phorboxazole B.
entry ester
conditions
yield %
d
a
b
c
d
e
f
11a CSA, 10:1 DCE/MeOH, rt to 70 °C
12a CSA, 3:1 DCE/MeOH, rt to 30 °C 13 (56%),e 16 (trace)
12b CSA, 3:1 DCE/MeOH, rt to 55 °C 16 (20%)
11c TBAF, AcOH, THF, rt
12c TBAF, AcOH, THF, rt
12 TFA/CH2Cl2/H2O, rt
11 TFA/CH2Cl2/H2O, rt
14 (71%)
15 (35%)
16:15 (71%), 13:1 ratio
13 (68%)e
g
a CSA (0.2 equiv) used. b CSA (0.5 equiv) used. c TBAF (1.5 equiv),
AcOH (0.2 equiv) used. d Decomposition of starting material occurred.
e 1H NMR yield.
Acknowledgment. We thank the Wild Fund at the Uni-
versity of York for studentship support (K.E.) and Dr.
John Carey (formerly GSK, Tonbridge) for generous
donations of the oxazole precursor to aldehyde 7.
Supporting Information Available. Full experimental
1
procedures, characterization, and copies of H and 13C
NMR of all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(18) Stereochemistry of the THP rings was determined by 1H NMR
coupling constants and in the case of the 2,6-cis-THP by nOes between
H2 and H6.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 21, 2012
5553