Synthesis of pterocellin A

Article abstract of DOI:10.1021/ol061002c

The first total synthesis of pterocellin A (1) was achieved in 10 linear steps from commercially available kojic acid (6) and 2-bromo-3-pyridinol (11) in a convergent sequence. The key constructive steps are a directed lithiation to couple two pyridines a

Full text of DOI:10.1021/ol061002c

ORGANIC
LETTERS
2006
Vol. 8, No. 12
2651-2652
Synthesis of Pterocellin A
Meaghan M. O’Malley,† Fehmi Damkaci, and T. Ross Kelly*
E. F. Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467
ross.kelly@bc.edu
Received April 26, 2006
ABSTRACT
The first total synthesis of pterocellin A (1) was achieved in 10 linear steps from commercially available kojic acid (6) and 2-bromo-3-pyridinol
(11) in a convergent sequence. The key constructive steps are a directed lithiation to couple two pyridines and an intramolecular nucleophilic
aromatic substitution to form 1.
In 2003, New Zealand scientists reported the isolation and
characterization of two red natural products, pterocellin A
(1) and B (2).1,2 The pterocellins exhibit in vitro anticancer
activity against a variety of cancer cell lines. Since the
pterocellins are the only known representatives of the
tricyclic pyrido[4,3-b]indolizine ring system, and since no
independent confirmation of the structure determination has
been recorded, the synthesis of pterocellin A was undertaken.
We now report the first synthesis of pterocellin A and the
certification of the original structure assignment.3
in a convergent fashion from 4, 5 and a carbonyl synthon.
In principle, the carbonyl unit could be incorporated initially
attached to either pyridine 4 or 5. Both possibilities were
examined, but in practice initial attachment to pyridine 4
provided the solution.
The synthesis began with the methylation of commercially
available kojic acid (6) to yield pyrone 7,4 which was then
heated with concentrated ammonium hydroxide according
to the procedure of Armit and Nolan to produce the known
pyridone 8 (Scheme 1).5 Pyridone 8 was protected with
p-methoxybenzyl (PMB) chloride to afford pyridine ether 9
in 32% overall yield from 7; the low yield is attributed in
part to difficulty in purifying pyridone 8. Primary alcohol 9
was then oxidized using o-iodoxybenzoic acid (IBX) to
aldehyde 10.
Retrosynthetic analysis suggested (eq 1) that 1 might be
achieved by cyclization of 3. The latter should be available
† Undergraduate research participant; recipient of a Pfizer Summer
Undergraduate Research Fellowship.
(1) Yao, B.; Prinsep, M. R.; Nicholson, B. K.; Gordon, D. P. J. Nat.
Prod. 2003, 66, 1074-1077.
(2) Prinsep, M. R.; Yao, B.; Nicholson, B. K.; Gordon, D. P. Phytochem.
ReV. 2004, 3, 325-331.
(3) For model studies leading to a monoaza counterpart of the ring system
of 1 and 2, see: Kende, A. S.; Henry, O.; Chen, Z. Tetrahedron Lett. 2004,
45, 7809-7812.
(4) Campbell, K. N.; Ackerman, J. F.; Campbell, B. K. J. Org. Chem.
1950, 15, 221-226.
(5) Armit, J. W.; Nolan, T. J. J. Chem. Soc. 1931, 3023-3031.
10.1021/ol061002c CCC: $33.50
© 2006 American Chemical Society
Published on Web 05/19/2006

Scheme 1
Scheme 3
Preparation of the unit corresponding to 5 (Scheme 2)
began with the methoxymethyl (MOM) ether protection of
commercially available 2-bromo-3-pyridinol (11) to produce
pyridine 12. A Kumada cross-coupling6 between bromopy-
ridine 12 and isobutylmagnesium bromide using nickel
catalysis7 gave 13.
in dichloromethane yielded triflate 16 in essentially quantita-
tive yield. The facile removal of the PMB group using 10%
TFA/DCM yielded pyridone 3. As predicted, cyclization of
3 with potassium carbonate and 18-crown-6 in N,N-dimeth-
ylformamide afforded the desired 1 as a red solid in a 57%
yield.
Scheme 2
1
The H NMR, 13C NMR, and UV-vis spectra (see the
Supporting Information) of synthetic 1 are in excellent
agreement with those reported for the natural material. Direct
1
comparison of synthetic and natural 1 (TLC cospotting, H
Selective lithiation of pyridine 13 at the 4-position was
achieved by using t-BuLi and enlisting the coordinating
ability of the MOM group;8 the lithiated pyridine was then
coupled to aldehyde 10 (Scheme 3). Since the purification
of the resulting doubly benzylic alcohol proved problematic,
it was oxidized without purification to dibenzylic ketone 14
using IBX. The MOM group was then removed from bicycle
14 using dilute hydrochloric acid to generate pyridinol 15.
The yield of this reaction is time dependent due to the
competitive, but slower, removal of the PMB group. Esteri-
fication of pyridinol 15 using triflic anhydride and pyridine
NMR spiking experiments, and undepressed mixture melting
point) verifies their identity.
In conclusion, we report the first synthesis of pterocellin
A, a synthesis that confirms the initially reported structure.
Acknowledgment. M.M.O. thanks Pfizer, Inc., for a
Summer Undergraduate Research Fellowship. We are grate-
ful to Dr. Michele Prinsep1,2 (University of Waikato, New
Zealand) for a sample of natural pterocellin A. We also thank
Mr. Alex Scopton (Boston College) for his aid in the
execution of this project.
Supporting Information Available: Experimental pro-
cedures and characterization data for selected compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(6) For a leading reference, see: Ku¨rti, L.; Czako´, B. Strategic Applica-
tions of Named Reactions in Organic Synthesis; Elsevier: Amsterdam, 2005;
pp 258-259.
(7) Ohta, A.; Takahashi, N.; Yuasa, K. Heterocycles 1990, 30, 875-
884.
(8) Winkle, M. R.; Ronald, R. C. J. Org. Chem. 1982, 47, 7, 2101-
2108.
OL061002C
2652
Org. Lett., Vol. 8, No. 12, 2006