Total synthesis of the novel angiogenesis inhibitors epoxyquinols A and B

Article abstract of DOI:10.1016/S0040-4039(03)00621-X

The synthesis of the recently discovered angiogenesis inhibitors epoxyquinols A and B, having novel polyketide derived dimeric structures, has been accomplished from the readily available Diels-Alder adduct of cyclopentadiene and p-benzoquinone through a short, simple and flexible strategy that is diversity oriented and can be adapted to an asymmetric variant.

Full text of DOI:10.1016/S0040-4039(03)00621-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3569–3572
Total synthesis of the novel angiogenesis inhibitors epoxyquinols
A and B
Goverdhan Mehta* and Kabirul Islam
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, India
Received 28 January 2003; revised 20 February 2003; accepted 28 February 2003
Abstract—The synthesis of the recently discovered angiogenesis inhibitors epoxyquinols A and B, having novel polyketide derived
dimeric structures, has been accomplished from the readily available Diels–Alder adduct of cyclopentadiene and p-benzoquinone
through a short, simple and flexible strategy that is diversity oriented and can be adapted to an asymmetric variant. © 2003
Elsevier Science Ltd. All rights reserved.
Angiogenesis or formation of new blood vessels is a
complex, multi-step, biological event and abnormalities
associated with this fundamental process have been
implicated in cancer, rheumatoid arthritis and many
other chronic inflammatory disorders. Thus, inhibition
of angiogenesis has emerged as a promising approach for
many diseases particularly cancer and many compounds
including natural products are being explored as leads for
clinical development.1 In this context, a report by
Japanese scientists in 2002 that novel pentaketide derived
dimers epoxyquinol A 1 and epoxyquinol B 2, isolated
from a soil based unknown fungus, exhibit potent
anti-angiogenic profiles has generated a great deal of
interest.2 An interesting aspect of this finding is that 1 and
2 differ from the known angiogenesis inhibitors in both
structural features and biological activity. The complex
heptacyclic structures of 1 and its diastereomeric sibling
2 were elucidated on the basis of extensive high-field
NMR studies and X-ray crystal structure determination
of epoxyquinol A 1.2a It was hypothesized that biosyn-
thetically 1 and 2 could arise through an endo-selective
Diels–Alder reaction between the diastereomeric 2H-
pyran monomers 3a and 3b through hetero and homo
dimerization, respectively (Scheme 1).2a
Scheme 1. Biogenetic hypothesis for the natural products.
* Corresponding author. E-mail: gm@orgchem.iisc.ernet.in
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00621-X

3570
G. Mehta, K. Islam / Tetrahedron Letters 44 (2003) 3569–3572
The origin of the 2H-pyran monomers 3a,b can be
traced to the precursor dienal 4 through a 6p electrocy-
clization and further to the more robust monocyclic
diol epoxide 5 (Scheme 1). Isolation of 5 from the same
fungus along with 1 and 2 provides strong support to
the biosynthetic proposal.2a Interestingly, 5 bears a
close structural resemblance to a family of epoxyquinol
and epoxyquinone antibiotics that have aroused consid-
erable biological and synthetic interest.3
The unusual structure and anti-angiogenic activity
profile of epoxyquinols A 1 and B 2 and the quest for
new analogues of these natural products has drawn the
immediate attention of the synthetic community.4 In
the past few months, enantioselective syntheses of
epoxyquinols A and B by the Hayashi4a,b and Porco4c
groups have appeared and quite understandably their
endeavors have followed the proposed biosynthetic
pathway. In addition, Porco et al.4c have also reported
the preparation of new diastereomers related to 1 and 2
through thermally induced equilibration and studied
their inhibitory activity towards the transcription factor
NF-kB. We too were enticed by the structures of
epoxyquinols A and B as they surfaced2a in the litera-
ture and wish to report here our accomplishment of the
total synthesis of natural products 1 and 2 in racemic
form. Our approach is also patterned along the biosyn-
thetic pathway2,3d,4 shown in Scheme 1 but follows a
different synthetic strategy to access the key monomeric
precursors 4 and 5.
Scheme 2. Reagents and conditions: (a) 30% H2O2, 10%
Na2CO3, acetone, 0°C, 95%; (b) DBU (2.1 equiv.), 40%
formalin (excess), 0°C, quant.; (c) diphenyl ether, 240°C, 5
min, 97%; (d) TBSCl, imidazole, DMAP, DCM, 0°C, 72%; (e)
DIBAL-H (2 equiv.), THF, −78°C, 82%; (f) 2,2-
dimethoxypropane, PPTS, acetone, rt 97%; (g) i. DIBAL-H,
THF, −50°C, 86%; ii. Ac2O, pyridine, DMAP, DCM, 0°C,
quant.; iii. HF–pyridine, THF, rt 87%.
The monocyclic epoxyquinol natural product 5 was
identified as the initial target and its synthesis emanated
from the readily available Diels–Alder adduct 65 of
cyclopentadiene and p-benzoquinone. Base-mediated
epoxidation to 7 and exhaustive hydroxymethylation in
the presence of DBU delivered 8 and established two
key CꢀC bonds in a single operation and in quantitative
yield.6 Next, the epoxyquinone moiety with two strate-
gically functionalized side arms in place was to be
extracted from 8. The retro-Diels–Alder reaction of 8
proceeded smoothly in nearly quantitative yield and led
to the symmetrical epoxyquinone 9 in which the two
hydroxymethyl arms now needed to be differentiated
(Scheme 2). One of the hydroxyl groups in 9 was
selectively protected as its bulky TBS–ether 10 to facili-
tate regio- and stereoselective DIBAL-H reduction7 to
furnish the diol 11 (Scheme 2).6 The diol moiety in 11
was protected as the acetonide 12. DIBAL-H reduction
of 12 then led to a 2:1 mixture of epimeric alcohols
which were directly acetylated and further deprotection
of the TBS-group led to 13 and 14 (Scheme 2). The
stereochemistry of the major hydroxy epimer 13 was
secured through an X-ray crystal structure
determination.8
The allylic primary hydroxy group in 13 was oxidized
with MnO2 and the resulting aldehyde 15 was subjected
to Wittig olefination to furnish a 1:1 mixture of E:Z-
isomers which on base hydrolysis furnished 16 and 17
(Scheme 3).6,9 The stereochemistry of 16, although
Scheme 3. Reagents and conditions: (a) MnO2, DCM, 0°C,
84%; (b) i. Ph3P(CH2CH3)I, nBuLi, THF, 0°C, 50%; ii.
K2CO3, MeOH, 0°C, quant.; (c) PDC, DCM, rt 81%; (d)
Amberlyst 15, MeOH, rt 85%; (e) MnO2, DCM, 0°C–rt 1
(40%), 2 (25%).

G. Mehta, K. Islam / Tetrahedron Letters 44 (2003) 3569–3572
3571
1
revealed by H NMR spectral analysis, was unambigu-
K.; Ogaswara, K. Tetrahedron Lett. 1996, 37, 499; (d) Li,
C.; Lobskovsky, E.; Porco, J. A. Am. Chem. Soc. 2000,
122, 10484; (e) Genski, T.; Taylor, R. J. K. Tetrahedron
Lett. 2002, 43, 3573.
ously secured through an X-ray crystal structure deter-
mination.8 The E-isomer 16, required for elaboration to
the natural products 5, 1 and 2, was subjected to PDC
oxidation to deliver the dienone 18 which was found to
be identical with the advanced intermediate of Hayashi
et al.4a in their synthesis of epoxyquinol A. Indeed,
deprotection of the acetonide group in 18 yielded 5,
which was found to be identical with the natural
product.4a,9 In a sequence similar to that described for
13, the a-epimer 14 was also transformed to
epoxyquinol 5, thereby making both the diastereomers
obtained from 12 serviceable, and neutralizing the lack
of stereoselectivity in the DIBAL-H reduction of 12
and so enhancing the efficacy of our approach. Finally,
MnO2 oxidation of the exocyclic allylic hydroxy group
4. (a) Shoji, M.; Yamaguchi, J.; Kakeya, H.; Osada, H.;
Hayashi, Y. Angew. Chem., Int. Ed. 2002, 41, 3192; (b)
Shoji, M.; Kishida, S.; Takeda, M.; Kakeya, H.; Osada,
H.; Hayashi, Y. Tetrahedron Lett. 2002, 43, 9155; (c) Li,
C.; Bardhan, S.; Pace, E. A.; Liang, M.-C.; Gilmore, T.
D.; Porco, J. A. Org. Lett. 2002, 4, 3267.
5. (a) Cookson, R. C.; Crundwell, E.; Hill, R. R.; Hudec, J.
J. Chem. Soc. 1964, 3062; (b) Mehta, G.; Srikrishna, A.;
Reddy, A. V.; Nair, M. S. Tetrahedron 1981, 37, 4543.
6. All new compounds were fully characterized on the basis
1
of spectral data (IR, H and 13C NMR, Mass).
7. Kiyooka, S.-I.; Kuroda, H.; Shimasaki, Y. Tetrahedron
1
in 5 was complete within minutes at 0°C and the H
Lett. 1986, 27, 3009.
NMR spectrum of the crude reaction mixture indicated
the presence of 4 along with 3a,b in a ratio of ꢀ60:40.
On allowing the reaction product to warm-up to ambi-
ent temperature (ꢀ28°C) and leaving aside for a few
hours to complete the reaction,10 epoxyquinols A and B
could be isolated along with an as yet uncharacterized
but related dimeric product by column chromatography
as envisaged in the biosynthetic proposal in Scheme
1.2a,4c Our synthetic 1 and 2 were found to be spectro-
scopically identical2,4c with the natural products
(Scheme 3).
8. Crystal data: X-ray data were collected at 293 K on a
SMART CCD–BRUKER diffractometer with graphite
,
monochromated MoKa radiation (u=0.7107 A). The
structure was solved by direct methods (SIR92). Refine-
ment was by full-matrix least-squares procedures on F2
using SHELXL-97. The non-hydrogen atoms were
refined anisotropically whereas hydrogen atoms were
refined isotropically. Compound 13: C13H18O6, MW=
270.27, colorless crystal, crystal system: triclinic, space
group: P-1, cell parameters: a=8.4101(8), b=14.604(1),
3
,
,
c=10.737(1) A, V=1315.24 (6) A , Z=4, Dcalcd=1.365 g
cm−3, F(000)=576.0, v=0.11 mm−1. Total number of l.s.
parameters=487, R1=0.0407 for 4148 Fo>4|(Fo) and
In summary, we have devised a simple and effective
approach to the novel natural products, epoxyquinols
A and B from the readily available Diels–Alder adduct
of cyclopentadiene and p-benzoquinone. Our approach
can be readily adapted towards the generation of struc-
tural diversity and an asymmetric version is currently
under development.
Acknowledgements
We would like to thank Professor Hayashi for compari-
son spectra. K.I. thanks CSIR for the award of a
research fellowship. This research was supported by the
Chemical Biology Unit of the Jawaharlal Nehru Center
for Advanced Scientific Research. We would like to
thank the CCD facility at IISc for providing the X-ray
data.
References
1. (a) Gasparini, G. Drugs 1998, 58, 17; (b) Paper, D. H.
Planta Med. 1998, 64, 686.
2. (a) Kakeya, H.; Onose, R.; Koshino, H.; Yoshida, A.;
Kobayashi, K.; Kageyama, S.-I.; Osada, H. J. Am. Chem.
Soc. 2002, 124, 3496; (b) Kakeya, H.; Onose, R.;
Yoshida, A.; Koshino, H.; Osada, H. J. Antibiot. 2002,
55, 829.
0.0521 for all 5232 data. wR2=0.1129, GOF=1.024,
Restrained GOF=1.024 for all data. Crystallographic
data (without structure factors) have been deposited with
the Cambridge Crystallographic Data Center and the
depository number is CCDC 199401. Compound 16:
C13H18O4, MW=238.27, colorless crystal, Crystal system:
3. Selected references: (a) Taylor, R. J. K.; Alkaraz, L.;
Kapfer-Eyer, I.; Macdonald, G.; Wei, X.; Lewis, N.
Synthesis 1998, 775; (b) Kamikubo, T.; Ogaswara, K.
Chem. Commun. 1996, 1679; (c) Kamikubo, T.; Hiroya,
triclinic, space group: P-1, cell parameters: a=7.333(2),
3
,
,
b=8.465(2), c=10.775(3) A, V=629.45 A , Z=2,
D
calcd=1.257 g cm−3, F(000)=256.0, v=0.09 mm−1
Total number of l.s. parameters=226, R1=0.0437 for
.

3572
G. Mehta, K. Islam / Tetrahedron Letters 44 (2003) 3569–3572
2093 Fo>4|(Fo) and 0.0517 for all 2476 data. wR2=
9. We have elaborated the Z-isomer 17 to a new Z-
epoxyquinol corresponding to 5 and its chemistry is
being investigated.
0.1129, GOF=1.042. Restrained GOF=1.042 for all
data. Crystallographic data (without structure factors)
have been deposited with the Cambridge Crystallo-
graphic Data Center and the depository number is
CCDC 199400. ORTEP diagrams of 13 and 16 are
shown above.
10. We and others4 have found that transformation of 4 to
furnish 1 and 2 via 3a,b is quite sensitive to time, tem-
perature and solvent regimes and efforts are underway
to optimize the conditions for this process.