Biomimetic synthesis of antimalarial naphthoquinones

Article abstract of DOI:10.1021/ja050092y

The total synthesis of naphthoquinone natural products isolated from the Bignoniaceae plant family is described. Pinnatal, isopinnatal, sterekunthals A and B, pyranokunthones A and B, and anthrakunthone have been prepared along the lines of a biosynthetic

Full text of DOI:10.1021/ja050092y

Published on Web 04/07/2005
Biomimetic Synthesis of Antimalarial Naphthoquinones
Jeremiah P. Malerich, Thomas J. Maimone, Gregory I. Elliott,† and Dirk Trauner*
Contribution from the Department of Chemistry, UniVersity of California-Berkeley,
Berkeley, California 94720-1460
Received January 6, 2005; E-mail: trauner@cchem.berkeley.edu
Abstract: The total synthesis of naphthoquinone natural products isolated from the Bignoniaceae plant
family is described. Pinnatal, isopinnatal, sterekunthals A and B, pyranokunthones A and B, and
anthrakunthone have been prepared along the lines of a biosynthetic proposal involving pericyclic reactions
as key steps. The first case of catalysis in oxa 6π electrocyclizations is reported.
Scheme 1. Oxa 6π-Electrocyclization/Diels-Alder Cascades in
Biosynthesis
Introduction
Pericyclic reaction cascades are very effective in creating
structurally complex molecules from simple precursors. Increas-
ingly they have been recognized to play a role in the biosynthesis
of natural products. Various combinations of electrocyclizations,
sigmatropic rearrangements, and cycloadditions have been
found, for instance 8π-6π electrocyclization cascades,1 [3,3]-
sigmatropic rearrangements followed by Diels-Alder reactions,2
or combinations of electrocyclizations with cycloadditions.1a,3
Oxa 6π electrocyclization followed by [4+2] cycloadditions
are a special case (Scheme 1). A recent example for such a
sequence can be found in the biosynthesis of torreyanic acid
(3).4 As originally proposed by Clardy and subsequently
demonstrated by Porco’s total synthesis, aldehyde 1 undergoes
reversible 6π electrocyclization to afford two diastereomeric
pyrans, 2a and 2b, which subsequently dimerize via [4+2]
cycloaddition to generate the natural product 3.
A related oxa 6π electrocyclization/Diels-Alder cascade was
proposed by Bohlmann to account for the formation of the
complex sesquiterpene xanthipungolide (7) from xanthatin (4).5
Photochemical isomerization of 4 sets the stage for subsequent
oxa 6π electrocyclization (5 f 6). The resulting pyran 6 is then
intercepted by Diels-Alder reaction involving the exo-meth-
ylene lactone moiety as dienophile to afford the pentacyclic
natural product.
† Current address: Department of Chemistry and The Skaggs Institute
for Chemical Biology, The Scripps Research Institute, 10550 North Torrey
Pines Rd., La Jolla, CA 92037.
(1) Nicolaou, K. C.; Sorensen, E. J. Classics in Total Synthesis; VCH:
Weinheim, 1996. (b) Beaudry, C. M.; Trauner, D. Org. Lett. 2002, 4, 2221-
2224. (c) Moses, J. E.; Baldwin, J. E.; Bruckner, S.; Eade, S. J.; Adlington,
R. M. Org. Biomol. Chem. 2003, 1, 3670-3684. (d) Parker, K. A.; Lim,
Y.-H. J. Am. Chem. Soc. 2004, 126, 15968-15969.
(2) Stocking, E. M.; Williams, R. M. Angew. Chem., Int. Ed. 2003, 42, 3078-
3115. (b) Nicolaou, K. C.; Sasmal, P. K.; Xu, H. J. Am. Chem. Soc. 2004,
126, 5493-5501.
(3) Kurdyumov, A. V.; Hsung, R. P.; Ihlen, K.; Wang, J. Org. Lett. 2003, 5,
3935-3938. (b) Paduraru, M. P.; Wilson, P. D. Org. Lett. 2003, 5, 4911-
4913.
(4) Lee, J. C.; Strobel, G. A.; Lobkovsky, E.; Clardy, J. J. Org. Chem. 1996,
61, 3232-3233. (b) Li, C. M.; Johnson, R. P.; Porco, J., J. A. J. Am. Chem.
Soc. 2003, 125, 5095-5106.
(5) Jakupovic, J.; Ganzer, U.; Pritschow, P.; Lehmann, L.; Bohlmann, F.; King,
R. M. Phytochemistry 1992, 31, 863-880.
The elegance and efficiency of these electrocyclization/Diels-
Alder cascades prompted us to investigate how frequently they
occur in nature. Searching for their retron, a 2-oxabicyclo[2.2.2]-
oct-5-ene moiety, in electronic databases, we discovered pinnatal
(8),6 isopinnatal (9),7 and sterekunthal B (10)8 (Chart 1). These
9
6276
J. AM. CHEM. SOC. 2005, 127, 6276-6283
10.1021/ja050092y CCC: $30.25 © 2005 American Chemical Society

Synthesis of Antimalarial Naphthoquinones
A R T I C L E S
Chart 1. Naphthoquinones from Bignoniaceae
A proved to be the most potent against the poW and Dd2 strains
of P. falciparum (IC50: 3.8 and 1.2 µM, respectively), while
related pinnatal and sterekunthal B were somewhat less active
(IC50: 16-97 µM). Isopinnatal was demonstrated to have
activity against Trypsonoma parasites with IC50 values of 0.37
µM against T. brucei brucei and 0.73 µM against T. brucei
rhodesiense.12 Once again, the kigelinols were 10-fold less
effective. In a study of anticancer activity by Houghton,
isopinnatal was shown to have slightly lower IC50 values against
melanoma cell lines (G361: 33 µM, StML11a: 15 µM) than
nonmelanoma cells (C32: 48 µM, CHO: 176 µM).13 The full
evaluation of these compounds was somewhat hampered by their
limited availability, increasing their attractiveness as targets of
total synthesis.
Unified Biosynthetic Proposal. A detailed retrosynthetic and
biosynthetic analysis of the compounds shown in Chart 1 led
us to conclude that they are all linked by a common biosynthetic
pathway featuring an oxa 6π electrocyclization/Diels-Alder
cascade (Scheme 2). According to our proposal, prenylated
hydroxynaphthoquinone 17 undergoes oxidation of the aromatic
nucleus and in the most activated allylic position to yield
hypothetical intermediate 18. Hydroxynaphthoquinone 17, the
prenylated version of the widely distributed natural product
lapachol, has been previously isolated from the roots of
Conospermum teretifolium, an Australian plant only distantly
related to the Bignoniaceae.14 Its oxidation product 18 eliminates
water and undergoes facile double-bond isomerization to yield
19, whose cyclization via intramolecular hetero Diels-Alder
reaction affords pyranokunthone A (15). Alternatively, dehydra-
tion of 18 to afford 20, followed by oxa 6π electrocyclization,
gives pyranokunthone B (16). Selective allylic oxidation of this
natural product then affords unsaturated aldehyde 21. This key
intermediate undergoes intramolecular [4+2] cycloaddition to
form the complex heterocyclic framework of pinnatal (8).
Another pericyclic step, a retro hetero Diels-Alder reaction,
converts pinnatal (8) into sterekunthal A (13). Finally, a
Baeyer-Villiger type oxidation of 13, followed by elimination
of formic acid, affords fully aromatized anthrakunthone (14).
Anthrakunthone could also arise from sterekunthal A via
vinylogous retro Claisen condensation followed by oxidative
aromatization of the resulting cyclohexadiene.
The kigelinols presumably stem from pinnatal and isopinnatal,
respectively. Baeyer-Villiger type oxidation, possibly catalyzed
by an enzyme of the aromatase class,15 could furnish formate
22. Elimination of formic acid with concomitant opening of the
strained heterocycle, which is stereoelectronically set up for an
E2 elimination, yields kigelinol (11). The isomeric natural
products isopinnatal (9), sterekunthal B (10), and isokigelinol
(12) would arise from an analogous pathway, differing only in
the oxidation pattern of the aromatic ring.
Inspired by this biosynthetic hypothesis we have set out to
synthesize the antimalarial naphthoquinones shown in Chart 1.
We now wish to give a full account of our studies, which
culminated in the total synthesis of pinnatal, isopinnatal,
sterekunthal B, sterekunthal A, anthrakunthone, and the
complex natural products belong to a growing class of naph-
thoquinone derivatives isolated from trees of the Bignoniaceae
family. Other members of this series are kigelinol (11) and
isokigelinol (12),7 sterekunthal A (13),8 anthrakunthone (14),8
and pyranokunthones A (15) and B (16).8 All of them share a
common naphthoquinone (or modified naphthoquinone) moiety,
which is fused to heterocyclic or carbocyclic ring systems of
varying complexity.
Biological Activity. The natural products shown in Chart 1
accumulate in the root bark of Bignoniaceae trees, whose
extracts have been used extensively in African tribal medicine.9
The activity of their components has been evaluated in a number
of contexts.10 Houghton and co-workers showed that isopinnatal
has IC50 values of 0.76 and 1.55 µM against chloroquine-
resistant and chloroquine-sensitive strains of Plasmodium fal-
ciparum, respectively.11 The kigelinols were 10 times less active
than isopinnatal toward the strains tested. In a separate study
by Jennett-Siems, the antiplasmodial activity of the natural
products isolated from S. kunthianum was assessed.8 Sterekunthal
(6) Joshi, K. C.; Singh, P.; Taneja, S.; Cox, P. J.; Howie, R. A.; Thomson, R.
H. Tetrahedron 1982, 38, 2703-2708.
(7) Akunyili, D. N.; Houghton, P. J. Phytochemistry 1993, 32, 1015-1018.
(8) Onegi, B.; Kraft, C.; Kohler, I.; Freund, M.; Jenett-Siems, K.; Siems, K.;
Beyer, G.; Melzig, M. F.; Bienzle, U.; Eich, E. Phytochemistry 2002, 60,
39-44.
(12) Moideen, S. V. K.; Houghton, P. J.; Rock, P.; Croft, S. L.; Aboagye-Nyame,
F. Planta Med. 1999, 65, 536-540.
(9) Irvine, F. R. Woody Plants of Ghana; University Press: Oxford, 1961.
(10) Hoet, S.; Opperdoes, F.; Brun, R.; Quetin-Leclereq, J. Nat. Prod. Rep. 2004,
21, 353-364. (b) Schwikkard, S.; van Heerden, F. R. Nat. Prod. Rep. 2002,
19, 675-692.
(13) Jackson, S. J.; Houghton, P. J.; Retsas, S.; Photiou, A. Planta Med. 2000,
66, 758-761.
(14) Cannon, J. R.; Joshi, K. R.; McDonald, I. A.; Retallack, R. W.; Sierakowski,
A. F.; Wong, L. C. H. Tetrahedron Lett. 1975, 32, 2795-2798.
(15) Michal, G. Biochemical Pathways; Wiley: Berlin, 1999.
(11) Weiss, C. R.; Moideen, S. V. K.; Croft, S. L.; Houghton, P. J. J. Nat.
Prod. 2000, 63, 1306-1309.
9
J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005 6277

A R T I C L E S
Malerich et al.
Scheme 2. Biosynthetic Proposal
available 2-hydroxynaphthoquinone (26), catalytic amounts of
â-alanine, and HOAc accomplished two-thirds of our desired
sequence providing pyran 27. Seeking more forceful conditions
to promote the ensuing Diels-Alder reaction, 27 was heated
to 160 °C in a sealed tube. Analysis of the sole product of this
reaction discounted the anticipated pinnatal analogue 28. Instead,
the compound was identified as methyl ketone 29. We quickly
realized that we had performed the desired intramolecular
Diels-Alder reaction, but it was followed by an unexpected
hetero [4+2] cycloreversion. Literature searches at the time of
this result resulted in no hits for natural products corresponding
to 29. Therefore, while we were excited by this result, our next
goal was to prevent the unforeseen retro Diels-Alder reaction.
Only after we completed the synthesis of pinnatal (see below)
did we realize that 29 corresponds to sterekunthal A (13).
To more closely model the pinnatal and isopinnatal systems,
we sought to prepare analogues containing the phenolic hydroxyl
group in protected form (Scheme 4). Reported procedures to
obtain 2,6- and 2,7-dihydroxynaphthoquinone by oxidation of
naphthalene diols were unsuccessful in our hands.17 Therefore,
we considered methoxy naphthoquinones 31, assuming the
phenol could be deprotected at a later stage in our efforts toward
the natural products. These compounds were readily prepared
by autoxidation of commercially available methoxy tetralones
30.18 Condensation/electrocyclization reactions of 31 with
aldehyde 24 produced corresponding oxygenated pyrans 32. In
a fashion similar to 27, compounds 32 were converted to methyl
ketones 33 when subjected to elevated temperatures.
Attempts to promote the intramolecular Diels-Alder reaction
of esters 32 with various Lewis acids gave mixed results. While
we were never able to obtain the corresponding cycloadducts,
we often isolated the products of Lewis-acid-promoted elec-
trocyclic ring opening (see below). This allowed us to eventually
investigate the kinetics of the oxa 6π electrocyclization in detail
and led to the discovery of acid catalysis in these reactions.
Still hoping to catalyze the desired Diels-Alder reaction, we
turned to organocatalysis (Scheme 5).19 This seemed to be a
prudent strategy, given that transformation of the ester function
in 32 to an aldehyde was necessary to complete the synthesis
of pinnatal. However, reduction of this ester proved problematic
in the presence of other reactive carbonyls and proceeded in
very low yield. This was our first indication that a successful
synthetic strategy would have to circumvent hydride reduction
at a late stage. Nevertheless, sufficient amounts of allylic alcohol
34 were procured through DIBAL reduction of 32b. Subsequent
Swern oxidation gave unsaturated aldehyde 35. Unfortunately,
none of our attempts with several organocatalysts (e.g., with
aminoester 36) were successful. It is likely that the R-branching
of the unsaturated aldehyde 35 rendered our substrate incompat-
ible with this methodology.
pyranokunthones. Our investigations on the asymmetric syn-
thesis of pinnatal and sterekunthal A, as well studies on the
kinetics of the central oxa 6π electrocyclization, are described
in detail as well.
Results and Discussion
Next, we explored the viability of a transannular Diels-Alder
strategy (Scheme 6).20 We reasoned that addition of the allylic
alcohol function in 34 to the more electrophilic carbonyl group
would form a macrocyclic acetal 38 (R ) H, Me), setting the
stage for a potentially biomimetic intramolecular cycloaddition.
Model Systems. Our initial attempts toward pinnatal and
isopinnatal sought to rapidly explore the feasibility of a
Knoevenagel condensation/electrocyclization/cycloaddition
cascade to rapidly assemble the molecular skeleton of 8, 9, and
10 (Scheme 3). Thus, allylic alcohol 24, which was obtained in
three steps from geranyl acetate,16 was oxidized with MnO2 to
give unsaturated aldehyde 25. Heating 25 with commercially
(17) De Min, M.; Croux, S.; Tournaire, C.; Hocquaux, M.; Jacquet, B.; Oliveros,
E.; Maurette, M. T. Tetrahedron 1992, 48, 1869-1882.
(18) Buckle, D. R.; Cantello, B. C. C.; Smith, H.; Smith, R. J.; Spicer, B. A. J.
Med. Chem. 1977, 20, 1059-1064.
(19) Fonseca, M. H.; List, B. Curr. Opin. Chem. Biol. 2004, 8, 319-326.
(20) Soucy, P.; L’Heureux, A.; Toro, A.; Deslongchamps, P. J. Org. Chem.
2003, 68, 9983-9987, and references therein.
(16) Marshall, J. A.; Lebreton, J.; Dehoff, B. S.; Jenson, T. M. J. Org. Chem.
1987, 52, 3883-3889.
9
6278 J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005

Synthesis of Antimalarial Naphthoquinones
A R T I C L E S
Scheme 3. Initial Model Studiesa
Scheme 5. Attempted Organocatalysisa
a Reagents and conditions: (a) DIBAL, 17%. (b) Swern reagent, 57%.
a Reagents and conditions: (a) MnO2, 90%; (b) 26, â-alanine, HOAc,
51%; (c) PhH, 160 °C, 82%.
Scheme 6. Proposed Transannular Cycloaddition
Scheme 4. Oxidized Modelsa
Scheme 7. High-Pressure Diels-Alder Reaction
a Reagents and conditions: (a) KOtBu, O2, 31a: 68%, 31b: 62%; (b)
24, â-alanine, HOAc, 32a: 50%, 32b: 52%; (c) PhH, 160 °C, 33a: 81%,
33b: 82%.
failed, and as a result 31a was subjected to brief treatment with
molten AlCl3, followed by careful aqueous workup, to provide
phenol 41.22 On the other hand, aldehyde 46 was synthesized
in straightforward fashion in eight steps from geraniol, as
previously described.23,24 A Knoevenagel condensation/electro-
cyclization sequence involving 41 and 46, followed by depro-
tection of the resulting THP ether under acidic conditions,
provided allylic alcohol 47. Subsequent Swern oxidation gave
aldehyde 21, a constitutional isomer of pinnatal. Given our
previous experience dealing with the analogous Diels-Alder
reaction model systems, we were prepared to subject this
compound to high-pressure conditions. To our surprise however,
cycloaddition of 21 proceeded at ambient temperature and
pressure to provide pinnatal (8) in very good yield. The mildness
of these conditions suggests that this intramolecular cyclo-
addition is indeed biomimetic. Finally, as predicted by our
biosynthetic proposal, heating of pinnatal to elevated temper-
atures gave sterekunthal A (13) via retro hetero Diels-Alder
reaction.24
Unfortunately, we were never able to reduce this attractive idea
to practice despite exploring a variety of acetalization conditions.
With our attempts at catalysis at ambient temperature
unsuccessful, we turned to changing the thermodynamics of the
system. Since cycloadditions typically have negative activation
and reaction volumes, we reasoned that running the reaction at
high pressure could provide a solution.21 To our delight,
pressurizing 32b to 14 kbar (13800 atm) in CH2Cl2 for 50 h
provided isopinnatal analogue 40 in 52% yield together with a
25% yield of recovered 32b (Scheme 7). No product of
cycloreversion was observed under these conditions.
Total Syntheses. With the results of these model studies in
mind we streamlined our synthetic strategy to avoid an endgame
deprotection of the phenol and oxidation state adjustment of
the ester. Our revised plan sought to synthesize aldehyde 21, a
proposed biosynthetic intermediate (cf. Scheme 2), through
oxidation of the corresponding alcohol.
Encouraged by this success, we hoped that the kigelinols and
anthrakunthone could be procured along the lines of our
To this end, naphthoquinone 31a had to be demethylated
(Scheme 8). Milder conditions (BBr3, pyridine‚HCl, NaSEt)
(22) Davies, J. E.; Roberts, J. C. J. Chem. Soc. 1956, 2173-2176.
(23) Muto, S.; Nishimura, Y.; Mori, K. Eur. J. Org. Chem. 1999, 14, 2159-
2165.
(24) Malerich, J. P.; Trauner, D. J. Am. Chem. Soc. 2003, 125, 9554-9555.
(21) Fringuelli, F.; Taticchi, A. In The Diels-Alder Reaction; Fringuelli, F.,
Taticchi, A., Eds.; Wiley: New York, 2002; pp 206-249.
9
J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005 6279

A R T I C L E S
Malerich et al.
Scheme 8. Synthesis of (()-Pinnatal and (()-Sterekunthal Aa
Scheme 9. Synthesis of Anthrakunthonea
a Reagents and conditions: (a) RhCl(PPh3)3, then air, 47%.
Scheme 10. Condensation/Electrocyclization/Cycloaddition
Cascadea
a Reagents and conditions: (a) TBAF, 92%; (b) Dess-Martin periodi-
nane, 83%; (c) EDDA, 10%.
Scheme 11. Synthesis of (()-Isopinnatala
a Reagents and conditions: (a) AlCl3, 53%; (b) (TFEO)2P(O)CH2COOEt,
KHMDS, 18-crown-6 (76%); (c) DIBAL, 94%; (d) DHP, PPTS, 99%; (e)
TBAF, 90%; (f) MnO2, 81%; (g) â-alanine, HOAc, 54%; (h) TsOH, MeOH,
97%; (i) Swern reagent, 87%; (j) neat, rt, 91%; (k) PhH, 160 °C, 92%.
proposed biosynthesis as readily as their precursors. Unfortu-
nately, attempts to elicit selective Baeyer-Villiger oxidation
of pinnatal and sterekunthal A were unsuccessful, presumably
due to the steric hindrance of the formyl group. This largely
ruled out syntheses of the kigelinols; however, while the
elimination of formic acid was not possible, we hoped that
transition metal mediated deformylation of sterekunthal A,
followed by spontaneous oxidative aromatization of the resulting
cyclohexadiene, would be more fruitful. Indeed, treatment of
sterekunthal A (13) with Wilkinson’s catalyst followed by
aerobic workup yielded anthrakunthone (14) (Scheme 9).
To further shorten our synthesis of pinnatal and minimize
protective group manipulations, we attempted a one-pot Kno-
evenagel condensation/electrocyclization/cycloaddition cascade.
To this end, bisaldehyde 49 was prepared by deprotection and
Swern oxidation of 44. We reasoned that the regiochemistry of
the condensation would be biased toward the desired formyl
group based on steric grounds. In the event, heating naphtho-
quinone 41 with 49 in the presence of ethylenediamine diacetate
(EDDA) did give pinnatal in 10% yield. Unfortunately, this low
yield could not be increased, although a variety of conditions
were explored. Isomerization of the C2-C3 double bond
appeared to be a major side reaction. A one-pot synthesis of
sterekunthal A (13) involving a tandem Knoevenagel condensa-
a Reagents and conditions: (a) AlCl3, 61%; (b) EDDA, 82%; (c) TsOH,
MeOH, 95%; (d) Dess-Martin periodinane, 85%; (e) neat, rt, 14 days, 81%.
tion/electrocyclization/cycloaddition/cycloreversion sequence
was attempted as well, but none of the desired natural product
could be isolated.
The total synthesis of isopinnatal and sterekunthal B followed
analogously that of pinnatal (Schemes 11 and 12). Pyrans 51
and 53 were prepared by condensation of naphthoquinones 50
and 26, respectively, with aldehyde 46 under optimized condi-
tions using EDDA in THF, followed by treatment with TsOH
in MeOH. Oxidation with Dess-Martin and Swern reagents
then gave aldehydes 52 and 54, respectively. Cyclization of 52
to isopinnatal proceeded at a rate comparable to the pinnatal
case, achieving full conversion over 10 days at room temper-
ature. On the other hand, 10 was formed at a much slower rate,
requiring 100 days to consume 54. The X-ray structure of
sterekunthal B is shown in Figure 1.
It is interesting to speculate on the differences in rates of the
Diels-Alder reactions leading to model compound 40, pinnatal
(8), isopinnatal (9), and sterekunthal B (10). Both the electronic
9
6280 J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005

Synthesis of Antimalarial Naphthoquinones
A R T I C L E S
Scheme 13. Synthesis of (()-Pyranokunthones A and Ba
Scheme 12. Synthesis of (()-Sterekunthal Ba
a Reagents and conditions: (a) 46, EDDA, 86%; (b) TsOH, MeOH,
100%; (c) Swern reagent, 76%; (d) neat, rt, 100 days, 70%.
a Reagents and conditions: (a) 55, â-alanine, HOAc, 50% 16 and 5%
15.
Scheme 14. Intramolecular Dynamic Kinetic Resolution
Figure 1. X-ray structure of sterekunthal B.
nature of the aromatic ring and the effects of a free phenol need
to be considered. Whereas the slow rate of the cyclization of
54 could be explained based on electronic grounds, the fact that
ester 32b and aldehyde 37 failed to undergo appreciable
cycloaddition at room temperature and ambient pressure suggests
that the cyclization is catalyzed by the acidic phenol functional-
ity.
Pyranokunthones A and B were obtained simultaneously by
â-alanine-catalyzed reaction of 41 with citral (55), a com-
mercially available 2:1 mixture of geranial and neral (Scheme
13). Under these conditions, the product of oxa 6π electro-
cyclization, pyranokunthone B (16), was formed as the major
isomer (50%), whereas the hetero Diels-Alder product pyra-
nokunthone A (15) was isolated only in very low yield (5%).
This result is commensurate with previously reported pyran
syntheses from 1,3-dicarbonyl compounds and citral.25 Only in
rare cases is the hetero Diels-Alder product observed.
Enantioenriched Pinnatal. The first stereocenter in our
synthetic pathway leading to pinnatal (and sterekunthal A) is
set in the course of the oxa 6π electrocyclization. The resulting
pyran 21, however, is expected to undergo facile racemization
via electrocyclic ring opening to give alkylidene trione 56,
followed by electrocyclic ring closure. In fact, 21 is an oxidized
version of pyranokunthone B (16), which has been isolated only
as a racemate from nature. Even though we did not observe 56
directly since the pyrone-dienone equilibrium lies overwhelm-
ingly on the side of 21, we believed it would be kinetically
accessible at room temperature.
Our hope was to selectively intercept one enantiomer of 21
via intramolecular Diels-Alder reaction using a chiral Lewis
acid (Scheme 14). Provided the racemization (21 / ent-21)
proceeded at a much faster rate than the cycloaddition, an
intramolecular dynamic kinetic resolution (intramolecular dy-
namic catalytic asymmetric transformation) could be achieved.26
Unfortunately, this concept met with limited success. While the
addition of chiral scandium triflate-pybox complexes,27 e.g., 57,
to racemic 21 did provide enantioenriched pinnatal (ee’s 20-
55%), the yields of the natural product never surpassed 40%
(Scheme 15). Note that these yields and enantiomeric excesses
could be accounted for by a simple kinetic resolution of 21
lacking the dynamic component.28 Extensive studies involving
other chiral Lewis acids (e.g., Co-Salen, Ti-Binol, or Eu(hfc)3
complexes) gave inferior results. In many cases, only isomer-
ization to the Z-configured unsaturated aldehyde was observed.
(26) Huerta, F. F.; Minidis, A. B. E.; Ba¨ckvall, J. E. Chem. Soc. ReV. 2001, 30,
321-331.
(27) For a review of pybox complexes as ligands in asymmetric catalysis, see:
Desimoni, G.; Faita, G.; Quadrelli, P. Chem. ReV. 2003, 103, 3119-3154.
(28) However, no unreacted 21 could be recovered in these reactions.
(25) Tietze, L. F.; Bachmann, J.; Wichmann, J.; Burkhardt, O. Synthesis 1994,
1185-1194. (b) Tietze, L. F.; Modi, A. Med. Res. ReV. 2000, 20, 304-
322. (c) Tapia, R. A.; Navarro, O.; Alegria, L.; Valderrama, J. A.
Heterocycl. Commun. 1998, 4, 151-154.
9
J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005 6281

A R T I C L E S
Malerich et al.
Scheme 15. Synthesis of Enantioenriched Pinnatala
a
The absolute configuration of pinnatal is arbitrary.
Scheme 16. Isolation and Electrocyclization of 59a
expected, alkylidene trione 59 underwent clean oxa 6π elec-
trocyclization at room temperature to restore 27.
Our strategy for the asymmetric synthesis of pinnatal was
based on the presumption that the racemization of pyrans of
type 27 would proceed rapidly at room temperature via
electrocyclic ring opening/ring closure. With access to pure ring-
opened product 59, we had the unique opportunity to study the
kinetics of the oxa 6π electrocyclization quantitatively.30 To
this end, 59 was heated to 90 °C in toluene-d8. The concentra-
1
tions of 59 and 27 were monitored by H NMR (Figure 2).
Surprisingly, the disappearance of 59 and the formation of 27
followed a second-order rate law, -d[59]/dt ) k2[59]2, with a
rate constant of k2 ) 2.94 × 10-3 ((0.15 × 10-4) M-1 s-1
.
a Reagents and conditions: (a) TiCl4, then H2O, 63% (78% BORSM);
(b) rt, neat, 90 h, 99%.
Scheme 17. Catalysis of Oxa 6π Electrocyclizations
Kinetics of 6π-Electrocyclizations. As mentioned above,
attempts to catalyze the intramolecular Diels-Alder reaction
of model systems 27 and 32a,b with Lewis acids occasionally
resulted in the isolation of products of electrocyclic ring opening
in varying amounts. Believing that a stoichiometric process was
occurring, we added 1 equiv of TiCl4 to 27. Under these
conditions, the surprisingly stable alkylidene trione 59 could
be isolated after aqueous workup (Scheme 16). Presumably, the
electrocyclic ring opening is driven by the formation of a stable
titanium(IV) dione complex 58, which is subsequently hydro-
lyzed.29 To the best of our knowledge, this is the first time that
a compound of type 59 has been isolated and characterized. As
Figure 2. Second-order plot for 59 f 27 (90 °C, toluene). The first-order
plot is shown for comparison.
While the reasons for its unexpected rate law remain to be
determined,31 this surprisingly slow cyclization led to the
discovery of catalysis in oxa 6π electrocyclizations. Since the
(29) Titanium(IV) dione complexes are well precedented. See, for instance:
Quinkert, G.; Becker, H.; Delgrosso, M.; Dambacher, G.; Bats, J. W.;
Du¨rner, G. Tetrahedron Lett. 1993, 34, 6885-6888.
(30) For a discussion of the kinetics of electrocyclizations, see: Marvell, E. N.
Thermal Electrocyclic Reactions; Academic Press: New York, 1980; Vol.
43.
9
6282 J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005

Synthesis of Antimalarial Naphthoquinones
A R T I C L E S
Dissolution of 59 in protic solvents, such as methanol, also
resulted in faster electrocyclization. Attempts to catalyze the
electrocyclization with chiral EDDA analogues such as 61 or
62, or other chiral Broensted acids, afforded only racemic
product. Although 27 could be resolved by analytical chiral
HPLC, its racemization under these conditions is apparently fast
enough to preclude catalytic asymmetric synthesis.
Conclusion
In conclusion, the concise total synthesis of several biologi-
cally active natural products isolated from Bignoniaceae has
been reported, lending credence to a unifying biosynthetic
proposal.33 Racemic pinnatal, isopinnatal, sterekunthals A and
B, pyranokunthones A and B, and anthrakunthone have been
prepared in quantities that allow for further exploration of their
medicinal potential. Attempts toward the asymmetric synthesis
of pinnatal and sterekunthal A have been met with limited
success. In the course of these investigations, however, the
kinetics of oxa 6π electrocyclizations have been elucidated and
the first case of catalysis in these reactions was observed. Future
investigations will be directed toward the generalization of
catalysis in hetero 6π electrocyclizations, the study of its detailed
kinetics, and the elucidations of its origin. The reasons for the
unusual kinetic behavior of the uncatalyzed oxa 6π electro-
cyclization will be elucidated as well. In combination with
detailed studies on the kinetics of the intramolecular Diels-
Alder reaction, these investigations may lead to a satisfactory
asymmetric synthesis of pinnatal and its congeners, such as
sterekunthal A and kigelinol.
Figure 3. Effects of EDDA on oxa 6π electrocyclizations.
half-life for the conversion of 59 to 27 in toluene was determined
to be 2.5 h for 0.2 M 59, while our Knoevenagel condensation/
electrocyclization tandems (e.g., 25 f 27) are typically complete
in 4 h, we suspected that EDDA was a catalyst not only for the
condensation but also for the electrocyclization step. Accord-
ingly, we repeated our kinetic studies in the presence of EDDA.
Indeed, as shown in Figure 3, addition of 20 mol % EDDA led
to a pronounced rate acceleration of the cyclization.
To our knowledge, this is the first clear-cut example of
catalysis in an oxa 6π electrocyclization.29,32 Presumably, it
involves protonation of a carbonyl group to afford 61, which
then undergoes the electrocyclization at an increased rate.
(31) A possible explanation invokes a reversible [1,7]-hydrogen shift generating
vinylogous acid 63. Compound 63 then acts as a catalyst to promote
electrocyclization of 59.
Acknowledgment. This work was supported in part by NSF
grant CHE-0348770. Financial support by Glaxo Smith Kline,
Eli Lilly, Astra Zeneca, Amgen, and Merck & Co. is also
gratefully acknowledged. J.P.M. thanks Bristol-Myers Squibb
for a graduate fellowship.
Supporting Information Available: Complete experimental
procedures and spectral data for previously unreported com-
pounds, 1H NMR spectra for selected compounds, X-ray crystal
structure coordinates for sterekunthal B (10), and kinetics data
for electrocyclizations. This information is available free of
charge via the Internet at http://pubs.acs.org.
JA050092Y
(32) For an example of catalysis in aza 4π-electrocyclic ring openings, see:
Bongini, A.; Panunzio, M.; Tamanini, E.; Martelli, G.; Vicennati, P.;
Monari, M. Tetrahedron: Asymmetry 2003, 14, 993-998.
(33) For a historical account of the development of this program, see: Malerich,
J. P.; Trauner, D. In Strategies and Tactics in Organic Synthesis, Vol. 5;
Harmata, M., Ed.; Elsevier: Amsterdam, 2004; pp 417-436.
9
J. AM. CHEM. SOC. VOL. 127, NO. 17, 2005 6283