Wharton-fragmentation-based approach to the carbocyclic core of the phomoidrides

Article abstract of DOI:10.1021/ol302011b

The carbocyclic core of the phomoidrides has been synthesized efficiently and in high yield. Key steps include a phenolic oxidation/intramolecular Diels-Alder sequence, tandem radical cyclization, and a late-stage Wharton fragmentation of a densely functionalized isotwistane skeleton.

Full text of DOI:10.1021/ol302011b

ORGANIC
LETTERS
2012
Vol. 14, No. 17
4544–4547
Wharton-Fragmentation-Based Approach
to the Carbocyclic Core of the
Phomoidrides
Graham K. Murphy,† Naoto Hama,† Aaron Bedermann,† Ping Dong,†
Chris M. Schneider,† Travis C. McMahon,† Ran N. Tao,† Barry M. Twenter,‡
David A. Spiegel,‡ and John L. Wood*,†
Department of Chemistry, Colorado Sate University, Fort Collins, Colorado 80523,
United States and Department of Chemistry, Yale University, New Haven,
Connecticut 06520, United States
john.l.wood@colostate.edu
Received July 19, 2012
ABSTRACT
The carbocyclic core of the phomoidrides has been synthesized efficiently and in high yield. Key steps include a phenolic oxidation/intramolecular
DielsꢀAlder sequence, tandem radical cyclization, and a late-stage Wharton fragmentation of a densely functionalized isotwistane skeleton.
In their quest for novel cholesterol-lowering and antic-
ancer agents, researchers at Pfizer reported the isolation
and structure elucidation of two novel fungal metabolites,
phomoidrides A and B (1 and 2, respectively).1 This report
spawned a flurry of activity from synthetic chemists who
were intrigued by the molecules’ novel structural features.2
These efforts have not only culminated in several com-
pleted syntheses3 but they have also led to the isolation and
structural elucidation of two novel congeners, phomoi-
drides C and D (3 and 4, respectively, Figure 1).3a,4
† Colorado Sate University.
‡ Yale University.
(1) (a) Dabrah, T. T.; Harwood, H. J.; Huang, L. H.; Jankovich,
N. D.; Kaneko, T.; Li, J. C.; Lindsey, S.; Moshier, P. M.; Subashi, T. A.;
Therrien, M.; Watts, P. C. J. Antibiot. 1997, 50 (1), 1. (b) Dabrah, T. T.;
Kaneko, T.; Massefski, W.; Whipple, E. B. J. Am. Chem. Soc. 1997, 119
(7), 1594.
Almost without fail, orchestrating a synthesis of a com-
plex natural product proves to be an arduous road that
winds around many obstacles arising from the interplay of
densely packed functional groups. Overcoming these ob-
stacles leads to the evolution of a synthetic strategy that is
accompanied with, and often guided by, unanticipated
discoveries. In previous reports on our efforts in the
phomoidride area, we described a synthetic strategy that
evolved from an oxy-Cope based approach to one that
hinged upon a Wharton-type fragmentation.5 Efforts to
(2) For a review of early efforts, see (a) Spiegel, D. A.; Njardarson,
J. T.; McDonald, I. M.; Wood, J. L. Chem. Rev. 2003, 103 (7), 2691. For
subsequent efforts, see (b) Hayashi, Y.; Itoh, T.; Fukuyama, T. Org.
Lett. 2003, 5 (13), 2235. (c) Clive, D. L. J.; Cheng, H.; Gangopadhyay, P.;
Huang, X.; Prabhudas, B. Tetrahedron 2004, 60 (19), 4205. (d) Yoshimitsu,
T.; Sasaki, S.; Arano, Y.; Nagaoka, H. J. Org. Chem. 2004, 69 (26), 9262.
(e) Ashenhurst, J. A.; Gleason, J. L. Tetrahedron Lett. 2008, 49 (3), 504.
(f) Ashenhurst, J. A.; Isakovic, L.; Gleason, J. L. Tetrahedron 2010,
66 (1), 368.
(3) (a) Meng, D. F.; Tan, Q.; Danishefsky, S. J. Angew. Chem., Int.
Ed. 1999, 38 (21), 3197. (b) Nicolaou, K. C.; Baran, P. S.; Zhong, Y. L.;
Choi, H. S.; Yoon, W. H.; He, Y.; Fong, K. C. Angew. Chem., Int. Ed.
1999, 38 (11), 1669. (c) Nicolaou, K. C.; Baran, P. S.; Zhong, Y. L.;
Fong, K. C.; He, Y.; Yoon, W. H.; Choi, H. S. Angew. Chem., Int. Ed.
1999, 38 (11), 1676. (d) Chen, C.; Layton, M. E.; Sheehan, S. M.; Shair,
M. D. J. Am. Chem. Soc. 2000, 122 (30), 7424. (e) Nicolaou, K. C.; Jung,
J. K.; Yoon, W. H.; He, Y.; Zhong, Y. L.; Baran, P. S. Angew. Chem.,
Int. Ed. 2000, 39 (10), 1829. (f) Waizumi, N.; Itoh, T.; Fukuyama, T.
J. Am. Chem. Soc. 2000, 122 (32), 7825.
(4) Spencer, P.; Agnelli, F.; Sulikowski, G. A. Org. Lett. 2001, 3 (10),
1443.
(5) Njardarson, J. T.; Wood, J. L. Org. Lett. 2001, 3 (16), 2431.
r
10.1021/ol302011b
Published on Web 08/23/2012
2012 American Chemical Society

strategy, 9 was seen as arising from a Wharton-type
fragmentation reaction of mesylate 10, which would, in
turn, be prepared from mixed acetal 11.10 As in our
previous studies, assembly of the isotwistane core struc-
ture in 11 was seen as proceeding from exo-methylene
lactone 12 via a radical cyclization cascade that draws upon
the pioneering studies of Ueno and Stork.11 The requisite
[2.2.2]-bicycle in lactone 12 would arise by way of a tandem
phenolic oxidation/inverse electron-demand DielsꢀAlder
sequence applied to phenol 13, a compound envisioned
available from the etherification of 14 with alkyl iodide 15
followed by further manipulation.
Figure 1. Phomoidrides AꢀD.
Scheme 2. Retrosynthetic Analysis
implement the latter approach on substantially functiona-
lized substrates were successful in producing 6, an inter-
mediate from which access to the complete phomoidride
core (7) appeared trivially accessible (Scheme 1).6,7 In yet
anotheraffirmation of the axiom that there isnosuchthing
as a trivial reaction, we were unable to advance 6 to the
corresponding maleic anhydride, a transformation that
also eluded us in efforts to advance substrates possessing
the fully functionalized phomoidride core structure.8
Scheme 1. Wharton-type Fragmentation Approach to the
Phomoidride Core
In the forward sense, the synthesis began by subjecting
phenol 14 (which could be purchased, or prepared from 2,4-
dihydroxybenzaldehyde) to a known three-step allylation/
BaeyerꢀVilliger oxidation/formate ester cleavage sequence
to give 16 in excellent yield (Scheme 3).12 Regioselective
bromination of 16 with NBS was followed by alkylation
with iodide 15 to afford 17 which, upon conversion to the
methyl ester and deallylation, furnished phenol 13a. Pheno-
lic oxidation of 13a, employing Pb(OAc)4, led to a mixture of
bicyclic products 19a and 20a via the intermediacy and
IMDA reaction of o-quinone acetal 18a, which was also
isolated in small quantities. Exposure of 19a to silica com-
pleted the acetate cleavage, and the resulting hemiacetal 20a
was protected employing TMSOTf to give 21a.
Turning toward installation of the exo-methylene lac-
tone, the lynchpin in our planned radical cascade cycliza-
tion, ketone 21a was combined with propionate 22 in a
two-step aldol reaction/Cope elimination sequence that
furnished enoate 23a (Scheme 4).13 Exposure of 23a to
TBAF/AcOH afforded exomethylene lactone 12a which,
Given our inability to advance 6, we considered alter-
native strategies and eventually turned toward an ap-
proach that called for a late-stage carbonylation, a
transformation akin to that employed by Shair on a similar
but regioisomeric substrate.9 Thus, as outlined in retro-
synthetic fashion (Scheme 2), we envisioned accessing the
phomoidride core structure (i.e., 8) via a sequence that
employs keto-ester 9 as the carbonylation substrate. Inter-
estingly, at least from the evolutionary standpoint of this
(6) Njardarson, J. T.; McDonald, I. M.; Spiegel, D. A.; Inoue, M.;
Wood, J. L. Org. Lett. 2001, 3 (16), 2435.
(7) Spiegel, D. A.; Njardarson, J. T.; Wood, J. L. Tetrahedron 2002,
58 (32), 6545.
(8) These efforts will be reported elsewhere.
(10) Although in our initial fragmentation studies we had employed a
Wharton fragmentation, our ability to rapidly generate advanced inter-
mediates containing both the C11 and C12 esters led us to favor this
more convergent approach. Unfortunately, the early introduction of
these esters proved to be the Achilles heal of these efforts.
(11) (a) Ueno, Y.; Chino, K.; Watanabe, M.; Moriya, O.; Okawara,
M. J. Am. Chem. Soc. 1982, 104, 5564. (b) Stork, G.; Mook, R.; Biller,
S. A.; Rychnovsky, S. D. J. Am. Chem. Soc. 1983, 105 (11), 3741.
(12) Stack, G. P.; Gao, H.; Gildersleeve, E. S. US Patent 6,552,049
B2, Apr. 23, 2003.
(9) The late-stage palladium-catalyzed carbonylation reaction at C11
employed by Shair3d to install the maleate and complete the total
synthesis is illustrated below and differs regiochemically from that
illustrated in Scheme 2 for the conversion of 9 to 8.
(13) Yu, L. C.; Helquist, P. J. Org. Chem. 1981, 46 (22), 4536.
Org. Lett., Vol. 14, No. 17, 2012
4545

fragmentation precursor 10a via mesylation of the tertiary
alcohol and benzyl-deprotection (Scheme 5).15 To our sur-
prise, the key Wharton fragmentation reaction of 10a failed
to provide 9a in all attempts using a host of bases and Lewis
acids in various solvents over a wide temperature range.
Scheme 3. Phenolic Oxidation/DielsꢀAlder Sequence
Scheme 5. Attempted Wharton Fragmentation of an Isotwis-
tane
Confounded by the failure of the Wharton fragmenta-
tion, we decided to confirm the integrity of our structural
assignments and obtained single crystal X-ray analyses on
benzyl ether 28a and the product of Bn-deprotection, 10a
(Figure 2). Interestingly, these analyses revealed that the
stereogenic center adjacent to the methyl ester had epimer-
ized during the deprotection, placing the ester away from
the congested polycyclic side of the ring system, giving
C11-epi-10a as the major product.
under conditions developed by Stork, was converted to the
corresponding bromoacetal 25a.11 Although in previous
investigations we had only employed electron deficient
maleate systems in the radical cascade sequence, we were
delighted to find that even the more electron rich vinylo-
gous carbonate moiety in 25a smoothly undergoes a
tandem 5-exo/5-exo cyclization to give 11a in 66% yield
upon exposure to Bu3SnH/AIBN. A second product (26),
resulting from an undesired 6-endo cyclization, was also
recovered in 33% yield.14
Scheme 4. Lactone Formation and Radical Cyclization
Figure 2. X-ray crystal structures of 28a and C11-epi-10a.
Upon re-examining the deprotection reaction we dis-
covered that the desired product 10a could be isolated
(on occasion) in up to 4% yield, and its exposure to base
rapidly induced epimerization to afford C11-epi-10a. At this
point we postulated that 10a and C11-epi-10a were inter-
converting via a retro-aldol fragmentation (Scheme 6), a
reasonable mode of reactivity that would also explain the
failed Wharton fragmentation for substrate 10a.
Having accessed the isotwistane core, mixed acetal 11a was
converted to thioacetal 27a which was, in turn, converted to
(14) The intervention of the 6-endo cyclization had been seen in our
previous cyclization studies of exomethylene lactone substrates; how-
ever with these former maleate containing substrates the 6-endo event
was followed by a 4-exo ring closure. Guided by our previous efforts we
attempted to circumvent the 6-endo pathway via a top-down cyclization
initiated by electron transfer to the vinylogous ester using SmI2.
Although this tranformation furnshed only 11a, isolated yields were
poor. In an effort to neutralize the Michael-acceptor character of the
exo-methylene moiety in 25a, we attempted to reduce the lactone to the
lactol; however in all cases this was found to proceed via preferential
methyl ester reduction.
Suspecting that 10a was engaging in retro-aldol chem-
istry instead of the desired fragmentation we considered
ways to avert the undesired pathway and decided to target
(15) In rare cases where the dithiane was also removed, re-exposure
of the crude product mixture to the dithiane formation conditions
resulted in regeneration of 10a.
4546
Org. Lett., Vol. 14, No. 17, 2012

delighted to find that LiAlH4 reduces the methyl ester in
deference to the lactone, and furnishes diol 31 in good,
though variable yield (Scheme 8). Diol 31 was found to
fragment upon exposure to several bases (e.g., NaH,
t-BuOK, KHMDS) and with KOH in THF/H2O affords
the desired product 32 in 67% yield. Single crystal X-ray
analysis confirmed the identity of the fragmentation pro-
duct 32, which was found to also exist as the hydrated
hemiketal isomer 320.
Scheme 6. Conversion of 10a to C11-epi-10a by a Retro-aldol
Process
Scheme 8. Wharton Fragmentation of Diol 31
β-hydroxyacid 29 as a fragmentation substrate, presuming
that formation of an intermediate carboxylate dianion via
retro-aldol would be less favored thanthe desired Wharton
fragmentation. Unfortunately all attempts to convert 28a
to 29 failed and we had to resort to its de novo preparation.
As illustrated in Scheme 7, we simply altered the alkylation
of 17 so as to deliver benzyl ester 13b and then followed the
same route that had been employed for the methyl ester
series.16 Deprotection of 28b with BBr3 afforded an in-
separable 2:1 mixture of β-hydroxyacid epimers 29 in
excellent yield.17 Upon treating the derived acids with a
variety of bases, none ofthe desired fragmentation product
30 was observed. The reactions afforded either recovered
starting material or decomposition products.
Having established that an appropriately functionalized
isotwistane core is capable of undergoing Wharton frag-
mentation to the phomoidride ring system, we have in-
itiated efforts to advance fully functionalized material
suited for conversion to the natural products. Clearly,
efforts to minimize the late stage oxidation state changes
that have been requiredfor the successful preparation of 32
will be incorporated in our efforts to develop a streamlined
approach to the phomoidrides. These latter efforts are
underway and will be reported in due course.
Scheme 7. Attempt at Wharton Fragmentation on 29
Acknowledgment. J.L.W. thanks Amgen for financial
support. G.K.M. and N.H. acknowledge NSERC of Canada
and the Uehara Foundation, respectively, for postdoctoral
fellowships. Dr. Chris Incarvito (Yale), Dr. Chris Rithner
(CSU), Don Heyse (CSU) and Don Dick (CSU) are
acknowledged for their assistance in obtaining crystallo-
graphic and spectroscopic data.
Although initial attempts to eliminate the undesired
fragmentation pathway had failed, we eventually decided
toreduce the methylester tothe corresponding alcohol and
attempt to fragment the resulting diol. To this end, we were
Supporting Information Available. Spectroscopic and
experimental data for all numbered compounds and
isolated intermediates. Single crystal X-ray analysis data
for indicated compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
(16) Yields, reactivities, and product distributions were comparable
to the methyl ester series throughout the conversion of 17 to 29. See the
Supporting Information for details.
(17) Though these were not characterized, exposing them to ethereal
CH2N2 afforded epimers 10a in quantitative yield.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 17, 2012
4547