A synthetic approach to nomofungin/communesin B

Article abstract of DOI:10.1021/ol034407v

(Matrix presented) A highly stereoselective intramolecular cycloaddition of an indole tethered to an aza-ortho-xylylene intermediate effects the rapid construction of a substantial portion of the ring system of the cytotoxic natural product communesin B.

Full text of DOI:10.1021/ol034407v

ORGANIC
LETTERS
2003
Vol. 5, No. 18
3169-3171
A Synthetic Approach to Nomofungin/
Communesin B
Seth L. Crawley and Raymond L. Funk*
Department of Chemistry, PennsylVania State UniVersity,
UniVersity Park, PennsylVania 16802
rlf@chem.psu.edu
Received March 7, 2003
ABSTRACT
A highly stereoselective intramolecular cycloaddition of an indole tethered to an aza-ortho-xylylene intermediate effects the rapid construction
of a substantial portion of the ring system of the cytotoxic natural product communesin B. An analogous cycloaddition involving an ortho-
quinone methide intermediate provides an adduct that clearly revealed that the structural assignment for nomofungin was in error.
Hemscheidt and co-workers recently reported the discovery
of a structurally intriguing polycyclic natural product from
an endophytic fungus using a bioassay designed to detect
antimicrotubule/antimicrofilament agents.1a The unidentified
fungus was isolated from the bark of Ficus microcarpa L.
and has subsequently been lost. Accordingly, the natural
product was given the name nomofungin (Scheme 1). The
structural assignment was based primarily on NMR spec-
troscopic measurements, and the exiton chirality method was
used to determine the absolute stereochemistry.1a Curiously,
a structurally quite similar natural product, communesin B
(Scheme 2), which has an NH instead of the proposed pyran
oxygen for nomofungin,2a had previously been reported and
gone relatively unnoticed by the synthetic community.
Moreover, communesin B gives proton and carbon spectra
nearly identical to those reported for nomofungin. This
discrepancy was recently clarified by Stolz and co-workers
who proposed a biosynthetic route to communesin B through
the oxidative coupling of tryptamine with the ergot alkaloid
aurantioclavine.3 Preliminary experimental work corroborated
this conjecture and prompted the retraction of the nomofun-
gin structure.1b We describe herein our independent observa-
tions that the natural products nomofungin and communesin
B are the same compound and provide unequivocable
evidence that the structural assignment for communesin B
is correct.
R. A.; Concepcion, G. P.; Ireland, C. M. J. Org. Chem. 2002, 67, 7124. (c)
For an approach to the total synthesis of perophoramidine, see: Artman,
G. D.; Weinreb, S. M. Org. Lett. 2003, 5, 1523.
(1) (a) Ratnayake, A. S.; Yoshida, W. Y.; Mooberry, S. L.; Hemscheidt,
T. K. J. Org. Chem. 2001, 66, 8717. As expected, nomofungin was
subsequently found to be moderately cytotoxic (LoVo, MIC ) 3.9 µM;
KB, MIC ) 8.8 µM) and shown to disrupt microfilaments in cultured
mammalian cells. (b) Erratum: Ratnayake, A. S.; Yoshida, W. Y.;
Mooberry, S. L.; Hemscheidt, T. K. J. Org. Chem. 2003, 68, 1640.
(2) (a) Numata, A.; Takahashi, C.; Ito, Y.; Takada, T.; Kawai, K.; Usami,
Y.; Matsumura, E.; Imachi, M.; Ito, T.; Hasegawa, T. Tetrahedron Lett.
1993, 34, 2355. Communesin A has an acetamide instead of the sorbamide
functionality found in communesin B. (b) For the isolation of a related
alkaloid, perophoramidine, which possesses a bis-amidine rather than the
bis-aminal functionality present in communesin B, lacks the azepine ring
system of communesin B and has a trans-C(7)-C(13) rather than a cis
stereochemical relationship, see: Verbitski, S. M.; Mayne, C. L.; Davis,
(3) May, J. A.; Zeidan, R. K.; Stoltz, B. M. Tetrahedron Lett. 2003, 44,
1203.
10.1021/ol034407v CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/06/2003

Scheme 1
Scheme 2
We were intrigued by the prospect of rapidly assembling
a substantial portion of the core ring system of nomofungin
via an intramolecular cycloaddition reaction involving an
ortho-quinone methide intermediate.4 An abbreviated ret-
rosynthetic plan that features this key step is outlined in
Scheme 1. Thus, the carboxyl functionality of cycloadduct
1 could serve as a suitable handle for the elaboration of the
pyrrolidine ring of nomofungin by one of several conceivable
synthetic sequences and the transformation of the protected
hydroxymethyl group to the epoxide moiety should be
relatively straightforward. The N,O-acetal 1 was expected
to arise from the cycloaddition of the ortho-quinone methide
2 with the tethered indole heterodienophile5 through the
preferred endo-transition state.6 While there are many ways
to thermally produce ortho-quinone methides (e.g., from
2-vinylphenols, ortho-hydroxybenzyl halides, ortho-hydroxy-
benzyl ethers, 1,2-benzooxazines, etc.),4 we elected to
examine the thermolysis of the benzodioxin 3. The basis for
the selection of this particular thermal precursor was
threefold: (1) the benzodioxin functionality is inert, in
comparison to the aforementioned precursors, to reagents
required for its incorporation into the retrocycloaddition-
cycloaddition substrate; (2) to the best of our knowledge,
the generation of ortho-quinone methides by this protocol
has not been previously reported;7 and (3) we were biased
by our successful exploitation of the related retrocycload-
ditions of substituted 4H-1,3-dioxins in a variety of synthetic
endeavors.8
the carbomethoxy and alkoxymethyl substituents present in
benzodioxin 3. To that end, the Lewis acid-catalyzed
condensation product of phenol with ethyl glyoxylate, glycol
4,9 was converted to the corresponding acetonide. The ester
functionality was then saponified, and the resulting carbox-
ylic acid was transformed to the acid chloride 5. The
construction of the indole sector of the benzodioxin 8 was
inititated with the preparation of indole 6 by straightforward
adaptation of the previously described synthesis of the
analogous N-des-methyl indole.10 Hydrogenation of the nitro
group of indole 6 led to concomitant cyclization of the
intermediate amine to the corresponding lactam, which was
then reduced with lithium aluminum hydride to afford the
benzazepine 7. The final reaction sequence further docu-
mented the stability of the benzodioxin functionality, namely,
acylation of the acid chloride 5 with benzazepine 7 followed
by reduction of the intermediate amide to afford the key
retrocycloaddition-cycloaddition substrate 8. We were
pleased to observe that thermolysis of benzodioxin 8 in
decalin (195 °C, 27 h) resulted in a clean transformation to
a mixture (10:1) of N,O-acetal cycloadducts. The relative
stereochemistry for the expected major isomer, endo-9, was
(7) A benzodioxin has been converted to the corresponding diacetate
and then nucleophilically triggered (NaOH) to generate an ortho-quinone
methide; see: Lhomme, J.; Fixler, N.; Salez, H.; Demeunynck, M. J. Chem.
Soc., Perkin Trans. 1 1995, 1649.
(8) (a) Greshock, T. J.; Funk, R. L. J. Am. Chem. Soc. 2002, 124, 754.
(b) Maeng, J.-H.; Funk, R. L. Org. Lett. 2002, 4, 331. (c) Fuchs, J. R.;
Funk, R. L. Org. Lett. 2001, 3, 3923. (d) Aungst, R. A., Jr.; Funk, R. L.
Org. Lett. 2001, 3, 3553. (e) Greshock, T. J.; Funk, R. L. Org. Lett. 2001,
3, 3511. (f) Fuchs, J. R.; Funk, R. L. Org. Lett. 2001, 3, 3349. (g) Aungst,
R. A., Jr.; Funk, R. L. J. Am. Chem. Soc. 2001, 123, 9455. (h) Maeng,
J.-H.; Funk, R. L. Org. Lett. 2001, 3, 1125. (i) Funk, R. L.; Fearnley, S. P.;
Gregg, R. J. Tetrahedron 2000, 56, 10275.
(9) We followed the procedure used for chiral glyoxylates; see: Bigi,
F.; Bocelli, G.; Maggi, R.; Sartori, G. J. Org. Chem. 1999, 64, 5004.
(10) Clark, R. D.; Weinhardt, K. K.; Berger, J.; Fisher, L. E.; Brown, C.
M.; MacKinnon, A. C.; Kilpatrick, A. T.; Spedding, M. J. Med. Chem.
1990, 33, 633.
To quickly validate this general plan, we decided to
prepare the simplified benzodioxin 8 (Scheme 2), which lacks
(4) For a review, see: Pettus, T. R. R.; Van De Walter, R. W.
Tetrahedron 2002, 58, 5367.
(5) (a) For recent examples of cycloaddition reactions of indoles, see:
Lynch, S. M.; Bur, S. K.; Padwa, A. Org. Lett. 2002, 4, 4643 and ref 9
therein. (b) For a review, see: Lee, L.; Snyder, J. K. In AdVances in
Cycloaddition; Harmata, M., Ed.; JAI: Stamford, CT, 1999; Vol. 6, p 119.
(6) Both the ester and silyloxymethyl substituents will prefer to emerge
on the convex face. The analogous exo transition state has a serious peri
interaction between the ester substituent and the proximate aryl ring
[calculated ∆E ) 2.9 kcal/mol, PC Model, transition state bond order
C(7a)--O ) 0.4, transition state bond order C(7)--C(13) ) 0.2].
3170
Org. Lett., Vol. 5, No. 18, 2003

assigned on the basis of the observed NOEs, the most
diagnostic of which are provided in Scheme 2. While at
first glance these conditions may appear to be harsh, they
are actually ideal in the sense that a reactive intermediate,
prone to side reactions,4 is generated slowly and trapped
immediately by a conformationally restricted and reactive
heterodienophile to afford a quite stable product.
Scheme 3
Unfortunately, our gratification was short-lived. Compari-
son of the diagnostic H7a and C7a NMR resonances of N,O-
acetal endo-9 with those reported for nomofungin1a suggested
a serious discrepancy. The chemical shift of the resonance
for the N,O-acetal proton in endo-9 (δ 5.4, CDCl3) is
significantly downfield from the analogous resonance re-
ported for nomofungin (δ 4.7, CDCl3). The carbon resonance
is also in disagreement (δ 101.0 vs 82.4). Moreover, it
seemed highly unlikely that conformational effects were
responsible for the chemical shift differences since N,O-acetal
endo-9 and nomofungin are locked into nearly identical
conformers. Hence, on the basis of our experimental work,
we were forced to conclude that communesin B represented
the correct structure.11
Fortunately, we could readily adapt the general synthetic
plan outlined in Scheme 1, which now required the genera-
tion of an aza-ortho-xylylene intermediate.12 Hopefully, this
reactive intermediate would also undergo the intramolecular
cycloaddition with the indole heterodienophile and thereby
introduce the “southern” aminal substructure of communesin
B (Scheme 3). Pursuant to that goal, ring opening of the
known epoxide 1013 with the benzazepine 7 gave a regio-
isomeric mixture (9:1) of amino alcohols that could be readily
separated upon conversion to the corresponding phenyl
carbonates. Upon thermolysis in dichlorobenzene, the major
carbonate 1114 was found to be an efficient precursor to
N-acyl-aza-ortho-xylylene 12 that gave rise to a single
cycloadduct, aminal endo-13. The stereochemical assignment
was secured upon hydrolysis of the carbamate moiety of
endo-13 to the aminal 14, which exhibited the analogous
NOEs previously observed for the “nomofungin” structure
endo-9. More importantly, the chemical shift of the reso-
nances for the aminal proton and aminal carbon of aminal
14 (δ 4.5 and 84.4, respectively) are quite similar to those
reported for communesin B (δ 4.7 and 82.4).
In conclusion, we have shown that a hexacyclic substruc-
ture of the core heptacyclic ring system of communesin B
can be rapidly constructed by a novel indole, aza-ortho-
xylylene intramolecular hetero Diels-Alder reaction. Syn-
thetic effort directed toward introduction of the remaining
pyrrolidine ring and completion of the natural product
synthesis is currently underway.
(11) We also recognized that biosynthetic considerations favor commu-
nesin B and proposed a pathway involving tryptamine dimerization similar
to the one advanced by Stolz.3 The validation of our hypothesis will be the
subject of a future communication. For seminal examples of proposed
alkaloid biosynthesis (calycanthaceous) originating from tryptamine dimers,
see: (a) Robinson, R.; Teuber, H. J. Chem. Ind. 1954, 783. (b) Woodward,
R. B.; Yang, N. C.; Katz, T. J. Proc. Chem. Soc. 1960, 76.
(12) For a recent review, see: Wojciechoski, K. Eur. J. Org. Chem. 2001,
3587.
(13) Ruano, J. L. G.; Pedrega, C.; Rodriguez, J. H. Tetrahedron 1989,
45, 203.
(14) This relatively mild protocol for generating an aza-ortho-xylylene
intermediate is underutilized. For the first and only examples, see: Kubo,
H.; Nishiyama, K.; Sato, T.; Higashiyama, K.; Ohmiya, S. Heterocycles
1998, 48, 1103.
Acknowledgment. We appreciate the financial support
provided by the National Institutes of Health (GM28553).
Supporting Information Available: Spectroscopic data
and experimental details for the preparation of all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL034407V
Org. Lett., Vol. 5, No. 18, 2003
3171