Enantioselective total synthesis of ( - )-kibdelone C

Article abstract of DOI:10.1021/ja204040k

The kibdelones are aromatic polyketide natural products featuring isoquinolinone and tetrahydroxanthone ring systems. They display potent cytotoxicity toward a range of human cancer cell lines. Here, we present an enantioselective total synthesis of kibdelone C that utilizes a Shi epoxidation to establish the absolute and relative stereochemistry, an acid-catalyzed cyclization to form the tetrahydroxanthone, and a C - H arylation to complete the hexacyclic skeleton.

Full text of DOI:10.1021/ja204040k

COMMUNICATION
pubs.acs.org/JACS
Enantioselective Total Synthesis of (ꢀ)-Kibdelone C
John R. Butler, Chao Wang, Jianwei Bian, and Joseph M. Ready*
Department of Biochemistry, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas,
Texas 75390-9038, United States
S Supporting Information
b
the natural products. We hoped that such a convergent approach
would lend efficiency to the synthesis and ultimately allow access
to diverse derivatives for biological testing.
ABSTRACT: The kibdelones are aromatic polyketide
natural products featuring isoquinolinone and tetrahydrox-
anthone ring systems. They display potent cytotoxicity
toward a range of human cancer cell lines. Here, we present
an enantioselective total synthesis of kibdelone C that
utilizes a Shi epoxidation to establish the absolute and
relative stereochemistry, an acid-catalyzed cyclization to
form the tetrahydroxanthone, and a CꢀH arylation to
complete the hexacyclic skeleton.
The isoquinolinone fragment of the kibdelones was assembled
starting with an amidation of amino alcohol 511 with benzoic acid
612 under SchottenꢀBaumann conditions (Scheme 2). Subse-
quent TEMPO oxidation13 provided aldehyde 9 in good yield.
BCl3 effected cyclization and selective removal of one O-methyl
group;14 dehydration yielded the isoquinolinone ring system 10.
Other Lewis and Bronsted acids proved ineffective, while BBr3
provided mainly doubly demethylated material. Finally, Sonoga-
shira coupling15 and desilylation gave the terminal alkyne 11, a
substrate for a second Sonogashira coupling with the tetrahy-
droxanthone fragment (see below).
he kibdelones (1ꢀ3, Scheme 1) are aromatic polyketides
T
recently isolated from a soil actinomycete by the Capon
group.1 A distinguishing characteristic of the kibdelones is their
propensity to interconvert by way of aerobic oxidation and
disproportionation. Thus, structures 1ꢀ3 differ by virtue of the
oxidation states of their B- and C-rings. Upon standing in MeOH,
purified 2 or 3 evolves to an equilibrium mixture of 1:2:3 in
approximately a 3:1:2 ratio. Kibdelones possess potent nemato-
cidal and antibiotic activity. Additionally, they are impressive
anticancer agents, displaying GI50’s in the low nanomolar range
against a panel of human cancer cell lines. Screening against
the National Cancer Institute's 60-cell panel suggested a novel
mode of action, although their curved, helical, and amphiphilic
character is reminiscent of natural products known to bind
nucleic acids or nucleic acid-binding proteins.2
We wanted to exploit the pseudo-C2 symmetry within the
F-ring that related the two diols to one another. To that end,
diene 12 was subjected to Shi epoxidation conditions16 to yield a
diepoxide that was immediately reduced with borane (Scheme 3).
This sequence, invented by Myers and co-workers,17 generated
doubly protected, C2-symmetric tetraol 14 as a single diaster-
eomer in >95:5 enantiomer ratio. Selective monotosylation
provided secondary alcohol 15. 1H NMR analysis indicated that
the adjacent tosyloxy and silyloxy substituents adopt a trans-
diaxial relationship. Presumably, the steric bulk of the equato-
rially disposed silyloxy group favors monosulfonylation. Swern
oxidation with trifluoroacetic anhydride yielded a ketone that
underwent elimination under the basic reaction conditions.18
Subsequent iodination and Luche reduction secured a vinyl
iodide (16) corresponding to the F-ring of the kibdelones.
Deprotonation of alcohol 16 with CH3Li followed by lithia-
tion with t-BuLi generated a vinyllithium reagent that could
participate in a nucleophilic addition to aldehyde 18. This aldehyde,
prepared as shown in Scheme 4, is both hindered and electron
rich; therefore, it displays minimal electrophilicity. Improved
reactivity was observed with dianionic organometallic reagents
lacking a C-10 protecting group (e.g., 16) compared to O-alkyl or
O-silyl analogues. Similar observations were made by Nicolaou’s
group in the context of their synthesis of diversonol.19 Dessꢀ
Martin oxidation generated an enedione that, upon treatment
with acidic acetone and t-BuOH, shed three protecting groups
and cyclized to construct the tetrahydroxanthone, protected as
an acetonide (21, Scheme 5).19 This transformation may involve
condensation of the D-ring phenol with the F-ring ketone to give
conjugated oxonium 20. Conducting this reaction in the absence
of acetone yielded the unprotected xanthone as a mixture of C-10
epimers, while omitting t-BuOH resulted in the incorporation of
Structurally, the kibdelones present a hexacyclic skeleton
featuring two fully substituted aryl rings and two fully substituted
heteroaryl rings. The F-ring contains three stereogenic centers,
while the A-ring is halogenated. Overall, these natural products
share structural features with other biologically active polyke-
tides, including simaomicin R,3 cervinomycin,4 the kigamicins,5
and SCH 54445.6 Interestingly, among these, only the kibdelones
display significant anticancer activity as single agents.7 Taken togeth-
er, the combination of potent cytotoxicity and structural com-
plexity has engendered excitement in the synthetic community.
Indeed, the Porco group reported an elegant approach to the
ABCD-ring system8 and has recently completed a total synthesis
of (þ)-3.9 Furthermore, other aromatic polyketides have suc-
cumbed to total synthesis through a number of instructive
strategies.10 Here we report our successful efforts to develop a
convergent enantioselective synthesis of kibdelone C (3).
We envisioned assembling the AB-DEF-ring system 4 from an
isoquinolinone and a tetrahydroxanthone fragment (Scheme 1).
In turn, these substructures would both arise from two simpler
precursors (5ꢀ8). We anticipated forming the C-ring through a
CꢀH arylation strategy to construct the full carbon skeleton of
Received: May 2, 2011
Published: June 07, 2011
r
2011 American Chemical Society
9956
dx.doi.org/10.1021/ja204040k J. Am. Chem. Soc. 2011, 133, 9956–9959
|

Journal of the American Chemical Society
COMMUNICATION
Scheme 1. Synthetic Strategy toward the Kibdelones
Scheme 3. Enantioselective Synthesis of the F-Ring
Scheme 4. Synthesis of the D-Ring
The endgame of our synthesis proved complicated. For example,
chlorination with hexachlorodieneone 25 was rapid, although
initially problematic. Thus, reducing unreacted 25 (Na2S2O4)
and extracting the resultant pentachlorophenol prior to concen-
trating the crude reaction mixture proved critical to avoid
decomposition of the chlorinated product.24 Even so, we isolated
the expected chlorinated product 26, along with materials
possessing masses consistent with hydration of the isoquinoli-
none ring.25 Furthermore, these hydrates were formed as dia-
steromeric mixtures and, along with compound 26, exist as
atropisomers at room temperature. Deprotection of this mixture
under acidic conditions removed the Boc group and converted
the aforementioned A-ring hydrates to the isoquinolinone. Next,
oxidation with PhI(OAc)2 gave a mixture of kibdelone B methyl
ether (27) and two derivatives containing either a MOM group
(28) or a methylene acetal (29), presumably arising from
formaldehyde generated during removal of the MOM group. If
these quinones were reduced to the hydroquinone, they under-
went aerobic oxidation back to the quinone in a matter of
minutes. Fortunately, 27ꢀ29 could all be converted to the
natural product by treatment with BCl3 to remove the C-6
methyl ether and the acetals that remained in 28 and 29. Crude
reaction mixtures contained both kibdelones B and C accord-
ing to HPLC/MS analysis. This mixture produced a brown,
sparingly soluble residue upon concentration, possibly as a
result of spontaneous oxidation. Nonetheless, as predicted by
the isolation report,1 reduction of the crude reaction products
with dithionite yielded kibdelone C (3). In our hands, the final
product was subject to oxidation and ill-defined decomposi-
tion. Careful purification by HPLC provided the natural
product in >90% purity.26 The synthetic material displayed
1H and 13C NMR spectra and molecular weight identical to
those reported by Capon and co-workers. The optical rotation
of synthetic 3 was equal in magnitude but opposite in sign to
that reported for the natural product, establishing the absolute
configuration of the kibdelones as enantiomeric to what is
shown in Schemes 1 and 5.
Scheme 2. Synthesis of Isoquinolinone 11
a methylene acetal spanning the C-10 and C-11 alcohols. The
proclivity of this diol to trap formaldehye would re-emerge later
in the synthesis.20
Sonogashira coupling with alkyne 11 required Cu-free con-
ditions and slow addition of the alkyne to avoid oxidative dimer-
ization.21 Next, hydrogenation yielded the pentacyclic structure
22, which contains all the carbons of the kibdelones. A
serendipitously discovered Cu-catalyzed iodination led to a
substrate for CꢀH arylation (23) en route to the C-ring of the
natural products.22 In this context, the coupling of CꢀH bonds
with aryl halides has been studied extensively, although we are
aware of no applications in such a complex setting.18 Extensive
optimization revealed that every parameter was critical for the
success of this cyclization to form the kibdelone skeleton (24):
other phosphines, buffers, solvents, metals, temperatures, or
atmospheric make-ups proved inferior to the conditions shown.
Likewise, the Boc group proved critical for obtaining syntheti-
cally useful yields. The free phenol or alkyl ethers cyclized less
efficiently, and esters were cleaved under the reaction condi-
tions. The most common side products under all conditions
were de-iodinated materials and, at higher temperatures, sub-
stances with aromatized F-rings.
9957
dx.doi.org/10.1021/ja204040k |J. Am. Chem. Soc. 2011, 133, 9956–9959

Journal of the American Chemical Society
COMMUNICATION
Scheme 5. Synthesis of Kibdelone C (3)
In summary, we have developed a convergent, enantioselec-
tive synthesis of the kibdelones. Since the absolute stereochem-
istry was established using asymmetric catalysis, we will have
access to the natural configuration by using the opposite Shi
catalyst. These characteristics should allow us to prepare and
evaluate diverse derivatives of these potent anticancer agents.
Chemother. Pharmacol. 1989, 24, 167. (c) Capranico, G.; Ferri, F.; Fogli,
M. V.; Russo, A.; Lotito, L.; Baranello, L. Biochimie 2007, 89, 482.
(3) Lee, T. M.; Carter, G. T.; Borders, D. B. J. Chem. Soc., Chem.
Commun. 1989, 1771.
(4) Nakagawa, A.; Oura, S. J. Antibiot. 1987, 40, 301.
(5) Someno, T.; Kunimoto, S.; Nakamura, H.; Naganawa, H.; Ikeda,
D. J. Antibiot. 2005, 58, 56.
(6) Chu, M.; Truumees, I.; Mierzwa, R.; Terracciano, J.; Patel, M.;
Loebenberg, D.;Kaminski, J. J.;Das, P.;Puar, M. S.J. Nat. Prod. 1997, 60, 525.
(7) Simaomicin R synergizes with the DNA-damaging agent bleo-
mycin at sub-nanomolar concentrations:Arai, M.; Sato, H.; Kobayashi,
H.; Suganuma, M; Kawabe, T.; Tomoda, H.; Omura, S. Biochem.
Biophys. Res. Commun. 2004, 317, 817.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures and
b
characterization data. This material is available free of charge via
the Internet at http://pubs.acs.org.
(8) Sloman, D. L.; Mitasev, B.; Scully, S. S.; Beutler, J. A.; Porco, J. A.
Angew. Chem., Int. Ed. 2011, 50, 2511.
’ AUTHOR INFORMATION
(9) Sloman, D. L.; Bacon, J. W.; Porco, J. A., Jr. J. Am. Chem. Soc. 2011,
133, No. DOI: 10.1021/ja203642n(preceding paper in this issue).
(10) Cervinomycins:(a) Kelly, T. R.; Jagoe, C. T.; Li, Q. J. Am. Chem.
Soc. 1989, 111, 4522. (b) Rao, A. V. R.; Yadav, J. S.; Reddy, K. K.; Upender, V.
Tetrahedron Lett. 1991, 32, 5199. (c) Mehta, G.; Shah, S. R.; Venkateswarlu,
Y. Tetrahedron 1994, 50, 11729. TMC-66:(d) Hosokawa, S.; Fumiyama, H.;
Fukuda, H.; Fukuda, T.; Seki, M.; Tatsuta, K. Tetrahedron Lett. 2007,
48, 7305. FD-594 Aglycon:(e) Masuo, R.; Ohmori, K.; Hintermann, L.;
Yoshida, S.; Suzuki, K. Angew. Chem. Int. Ed. 2009, 48, 3462.
(11) Johansson, R.; Karlstroem, S.; Kers, A.; Nordvall, G.; Rein, T.;
Slivo, C. Int. Pat. WO 2008/039139A1.
(12) Barfknecht, C. F.; Nichols, D. E. J. Med. Chem. 1971, 14, 370.
(13) De Mico, A.; Margarita, R.; Parlanti, L.; Vescovi, A.; Piancatelli,
G. J. Org. Chem. 1997, 62, 6974.
(14) Walker, E. R.; Leung, S. Y.; Barrett, A. G. M. Tetrahedron Lett.
2005, 46, 6537.
Corresponding Author
joseph.ready@utsouthwestern.edu
’ ACKNOWLEDGMENT
We acknowledge the American Cancer Society, CPRIT
(HIRP101016), and the Welch Foundation (I1612) for Funding.
J.M.R. is a Fellow of the Alfred P. Sloan Foundation. J.R.B. is
supported by the Graduate Programs Initiative from the UT
Board of Regents. We thank Dr. Ranjala Ratnayake for spectra of
the kibdelones and advice regarding handling of 3, and Prof.
Porco for sharing details of his synthesis prior to publication. We
thank Prof. Jef De Brabander for helpful discussions.
(15) Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fu, G. C. Org.
Lett. 2000, 2, 1729.
(16) Shi, Y. Acc. Chem. Res. 2004, 37, 488.
’ REFERENCES
(1) (a) Ratnayake, R.; Lacey, E.; Tennant, S.; Gill, J. H.; Capon, R. J.
Chem. Eur. J. 2007, 13, 1610. (b) Ratnayake, R.; Lacey, E.; Tennant, S.;
Gill, J. H.; Capon, R. J. Org. Lett. 2006, 8, 5267.
(2) (a) Prasad, K. S.; Nayak, R. FEBS Lett. 1976, 71, 171. (b) Latham,
M. D.; King, C. K.; Gorycki, P.; Macdonald, T. L.; Ross, W. E. Cancer
(17) Lim, S. M.; Hill, N.; Myers, A. G. J. Am. Chem. Soc. 2009,
131, 5763.
(18) Tan, L.; Yasuda, N.; Yoshikawa, N.; Hartner, F. W.; Eng, K. K.;
Leonard, W. R.; Tsay, F.-R.; Volante, R. P.; Tillyer, R. D. J. Org. Chem.
2005, 70, 8027.
9958
dx.doi.org/10.1021/ja204040k |J. Am. Chem. Soc. 2011, 133, 9956–9959

Journal of the American Chemical Society
COMMUNICATION
(19) Nicolaou, K. C.; Li, A. Angew. Chem. Int. Ed. 2008, 47, 6579.
(20) Formaldehyde, or its equivalent, arises from deprotection of the
MOM group.
(21) Soheili, A.; Albaneze-Walker, J.; Murry, J. A.; Dormer, P. G.;
Hughes, D. L. Org. Lett. 2003, 5, 4191.
(22) (a) Fortuitously, iodination was originally attempted on phenol
22 recovered from a failed Cu(II)-mediated phenolic coupling. (b) Cu-
catalyzed bromination:Yang, L.; Lu, Z.; Stahl, S. S. Chem. Commun.
2009, 6460. (c) Inadvertent iodination in the presence of Cu:Li, X.;
Hewgley, J. B.; Mulrooney, C. A.; Yang, J.; Kozlowski, M. C. J. Org. Chem.
2003, 68, 5500. (d) Cu(OH)Cl TMEDA:Nakajima, M.; Miyoshi, I.;
3
Kanayama, K.; Hashimoto, S.; Noji, M.; Koga, K. J. Org. Chem. 1999,
64, 2264.
(23) (a) Campeau, L.-C.; Parisien, M.; Jean, A.; Fagnou, K. J. Am.
Chem. Soc. 2005, 128, 581. Reviews:(b) Ackermann, L.; Vicente, R.;
Kapdi, A. R. Angew. Chem. Int. Ed. 2009, 48, 9792. (c) Gutekunst, W. R.;
Baran, P. S. Chem. Soc. Rev. 2011, 40, 1976.
(24) Other chlorination reagents including NCS and PPh3/NCS
performed similarly.
(25) Mass spectrometry and 1H NMR data are consistent with 30.
A similar product was identified during the isolation of 1ꢀ3. See
ref 1.
(26) See Supporting Information for full experimental details.
9959
dx.doi.org/10.1021/ja204040k |J. Am. Chem. Soc. 2011, 133, 9956–9959