Indole diterpene synthetic studies: Development of a second-generation synthetic strategy for (+)-nodulisporic acids A and B

Article abstract of DOI:10.1021/jo062423a

(Chemical Equation Presented) A second-generation strategy for construction of (+)-nodulisporic acids A and B based on the development of a new, effective modular indole synthesis exploiting a sequential Stille cross-coupling/Buchwald- Hartwig union/cyclization tactic is disclosed. This strategy evolved due to the considerable acid instability of the C(24) hydroxyl group observed in several advanced intermediates during our first-generation approach.

Full text of DOI:10.1021/jo062423a

Indole Diterpene Synthetic Studies: Development of a
Second-Generation Synthetic Strategy for (+)-Nodulisporic
Acids A and B
Amos B. Smith, III,* La´szlo´ Ku¨rti, Akin H. Davulcu, Young Shin Cho, and
Kazuyuki Ohmoto
Department of Chemistry, Monell Chemical Senses Center and Laboratory for Research on the Structure
of Matter, UniVersity of PennsylVania, Philadelphia, PennsylVania 19104
smithab@sas.upenn.edu
ReceiVed NoVember 24, 2006
A second-generation strategy for construction of (+)-nodulisporic acids A and B based on the development
of a new, effective modular indole synthesis exploiting a sequential Stille cross-coupling/Buchwald-
Hartwig union/cyclization tactic is disclosed. This strategy evolved due to the considerable acid instability
of the C(24) hydroxyl group observed in several advanced intermediates during our first-generation
approach.
Introduction
generation strategy (Scheme 1) for (+)-nodulisporic acids A
(1) and B (2) based on the indole construction tactic (6+7,
Scheme 1) that we developed in the mid 1980s6,7 and subse-
quently exploited for the total syntheses of several related indole
diterpenoid natural products, including (-)-21-isopentenylpax-
illine8 and (-)-penitrem D.9,10 Specifically, we envisioned late-
stage attachment of the highly strained nodulisporic acid D-ring
The nodulisporanes comprise a new class of insecticidal
indole diterpenes (Figure 1). Structurally the most complex
member, (+)-nodulisporic acid A (1), was isolated by Ondeyka
and co-workers at Merck Research Laboratories in 1997.1,2
Given the architectural novelty of the nodulisporanes, in
conjunction with their potential use as commercial insecticides,
we initiated a synthetic program in 2000 to devise a modular
strategy that would permit construction, not only of the naturally
occurring nodulisporic acids, but also of the congeners not
readily available by chemical modification of the natural
products.3,4 In the preceding paper,5 we outlined a first-
(4) Berger, R.; Shoop, W. L.; Pivnichny, J. V.; Warmke, L. M.; Zakson-
Aiken, M.; Owens, K. A.; deMontigny, P.; Schmatz, D. M.; Wyvratt, M.
J.; Fisher, M. H.; Meinke, P. T.; Colletti, S. L. Org. Lett. 2001, 3, 3715-
3718.
(5) Smith, A. B., III; Davulcu, A. H.; Cho, Y. S.; Ohmoto, K.; Ku¨rti,
L.; Ishiyama, H. J. Org. Chem. 2007, 72, 4596-4610.
(6) Smith, A. B., III; Visnick, M.; Haseltine, J. N.; Sprengeler, P. A.
Tetrahedron 1986, 42, 2957-2969.
(7) Smith, A. B., III; Visnick, M. Tetrahedron Lett. 1985, 26, 3757-
3760.
(8) Smith, A. B., III; Cui, H. Org. Lett. 2003, 5, 587-590.
(9) Smith, A. B., III; Kanoh, N.; Ishiyama, H.; Hartz, R. A. J. Am. Chem.
Soc. 2000, 122, 11254-11255.
(10) Smith, A. B.; Kanoh, N.; Ishiyama, H.; Minakawa, N.; Rainier, J.
D.; Hartz, R. A.; Cho, Y. S.; Cui, H.; Moser, W. H. J. Am. Chem. Soc.
2003, 125, 8228-8237.
(1) Ostlind, D. A.; Felcetto, T.; Misura, A.; Ondeyka, J.; Smith, S.; Goetz,
M.; Shoop, W.; Mickle, W. Med. Vet. Entomol. 1997, 11, 407-408.
(2) Ondeyka, J. G.; Helms, G. L.; Hensens, O. D.; Goetz, M. A.; Zink,
D. L.; Tsipouras, A.; Shoop, W. L.; Slayton, L.; Dombrowski, A. W.;
Polishook, J. D.; Ostlind, D. A.; Tsou, N. N.; Ball, R. G.; Singh, S. B. J.
Am. Chem. Soc. 1997, 119, 8809-8816.
(3) Chakravarty, P. K.; Tyagarajan, S.; Shih, T. L.; Salva, S.; Snedden,
C.; Wyvratt, M. J.; Fisher, M. H.; Meinke, P. T. Org. Lett. 2002, 4, 1291-
1294.
10.1021/jo062423a CCC: $37.00 © 2007 American Chemical Society
Published on Web 05/19/2007
J. Org. Chem. 2007, 72, 4611-4620
4611

Smith et al.
SCHEME 2. Union of Aniline (-)-6 with Lactone (+)-7 and
Indole Formation Studies with Keto Aniline 9
FIGURE 1. Representative nodulisporic acid congeners.
onto an advanced heptacyclic intermediate (cf. 5) employing
the acylation conditions developed by Hendrickson.11 The
requisite dienoic side chain would then be installed via a Stille
cross-coupling tactic.12
SCHEME 1. First-Generation Retrosynthetic Analysis of
Nodulisporic Acid A (1)
Results and Discussion
A First-Generation Strategy for Nodulisporic Acid A:
Challenges and Frustrations. Pleasingly, our indole synthetic
tactic proved successful for the union of the requisite structurally
complex western and eastern hemisphere tricyclic aniline (-)-6
and lactone (+)-7 to furnish aniline 9 (Scheme 2).5 We
anticipated that the Boc group could be readily removed under
mild acidic or basic conditions and the resulting amino ketone
in turn induced to undergo facile cyclodehydration to complete
construction of the desired hexacyclic indole 11. Unfortunately,
selective removal of the Boc group, even under extremely mild
acidic conditions (e.g., SiO2, PPTS, etc.), could not be achieved
without effecting C(24) hydroxyl dehydration to furnish the
hexacyclic indole 10 (Scheme 2). Ultimately, we did achieve
removal of the Boc group, but only after acetylation of the two
(11) Hendrickson, J. B.; Hussoin, M. S. J. Org. Chem. 1989, 54, 1144-
1149.
(12) Farina, V.; Krishnamurthy, V.; Scott, W. J. Org. React. 1997, 50,
1-652.
4612 J. Org. Chem., Vol. 72, No. 13, 2007

Synthetic Strategy for (+)-Nodulisporic Acids A and B
hydroxyl groups, followed by exposure to p-toluenesulfonic acid
(Scheme 2). Removal of the acetates (K2CO3/MeOH) then led
to the unmasked amino ketone (+)-12, albeit in modest yield
(three steps, ca. 25-47%). Indole formation was then achieved
under neutral conditions by heating in benzene at reflux for 6
days; the long exposure time, however, permitted isolation of
(+)-11 in only 29% yield. The modest efficiency of the overall
indole ring construction, in conjunction with the considerable
lability of the C(24) hydroxyl clearly foreshadowed difficulty
with the proposed D-ring installation exploiting the earlier
proposed (vide infra) acidic Hendrickson acylation protocol. We
therefore developed a second-generation strategy (Scheme 3);
two strategic alternatives were envisioned to construct ring D.
The first, as with the Hendrickson scenario, would entail use
of a pre-existing indole moiety, onto which the C(7)-N(1) two-
system of the nodulisporic acids. In this regard, only two reports
exist wherein a two-carbon bridge has been successfully installed
between the C(7) and N(1) (nodulisporic acid numbering) on a
substituted indole or indoline ring. The first entailed construction
of the highly strained 1,7-annulated system 16 (Figure 2).14 Not
surprisingly, indole 16 was reported to be quite unstable, readily
reacting with various nucleophiles (e.g., alcohols and amines)
FIGURE 2. Literature examples of 1,7-annulated indoles.14,15,17
SCHEME 3. Possible Approaches to the Highly Strained
Indole 13
to relieve the ring strain (i.e., relatively weak C-N bond). The
second example was recorded as part of a study by Dankwardt
et al.15 on the structural requirements for a 5-exo-trig versus
6-endo-trig Pd-catalyzed Heck reaction (cf. 24 and 25).16-18
Notwithstanding the foreboding literature precedent, we briefly
explored several seemingly straightforward approaches to install
the C(7)-N(1) bridge. The first involved intramolecular nu-
cleophilic substitution to form the C-N bond (bond b, Scheme
3). Toward this end, R-bromo ketone 27 was prepared from
R-diazoketone 26 via treatment with HBr.19 We envisioned
either heating 27 in a nonpolar solvent or exposure to K2CO3
or LiHMDS might permit the indolyl nitrogen to displace the
R-bromine atom. Unfortunately, both reactions proved unsuc-
cessful. Presumably, the ring strain in conjunction with stereo-
electronic alignment prevents the nitrogen anion from devel-
oping sufficient overlap with the σ* orbital of the C-Br bond.
(15) Dankwardt, J. W.; Flippin, L. A. J. Org. Chem. 1995, 60, 2312-
2313.
(16) Dankwardt and co-workers also examined the effect of the ring size
on the mode of ring closure and therefore prepared an N-acryloyl-8-
bromotetrahydroquinoline derivative upon Heck cyclization, and the ratio
of the 5-exo-trig versus the 6-endo-trig products was reversed, favoring
the former mode of cyclization. The authors concluded that the 6-endo-trig
cyclization mode is only favored in systems (i.e., indoles and indolines)
wherein the strain of the transition state would severely suppress the 5-exo-
trig mode of ring closure.
(17) For a few examples of radical or Pd-catalyzed cyclization of N-allyl-
7-bromoindoles (Scheme 4), see: (a) Dobbs, A. P.; Jones, K.; Veal, K. T.
Tetrahedron Lett. 1997, 38, 5379-5382. (b) Black, D. S. C.; Keller, P. A.;
Kumar, N. Tetrahedron 1992, 48, 7601-7608.
(18) Intramolecular reactions between carbon-centered radicals and
olefins prefer to proceed via a kinetic 5-exo-trig cyclization for stereoelec-
tronic reasons. However, in the case of compound 17, only the corresponding
6-endo-trig cyclization product 19 was obtained along with some reduced
product 20. Similar observations were obtained with the Pd-catalyzed
cyclization of indole 21. The authors14 attributed the observed selectivity
favoring the 6-endo-trig cyclization to the considerable strain and distortion
in the transition state leading to the five-membered 1,7-annulated products.
carbon bridge would be installed. Mild conditions would be
paramount to avoid elimination of the C(24) hydroxyl. The
second alternative would require the development of a de novo
indole synthesis, again employing mild reaction conditions.
Central to the latter scenario would be preservation of the
convergency of the overall synthetic sequence, thereby permit-
ting the union of two modestly re-engineered advanced hemi-
spheres (e.g., 14 and 15).13 To elaborate the D-ring employing
an already existing indole nucleus, three bonds (a, b, and c)
were considered for disconnection. We were of course cognizant
of the considerable strain inherent in the central CDE ring
(13) Smith, A. B., III; Ku¨rti, L.; Davulcu, A. H. Org. Lett. 2006, 8, 2167-
2170.
(14) Black, D. S. C.; Bowyer, M. C.; Catalano, M. M.; Ivory, A. J.;
Keller, P. A.; Kumar, N.; Nugent, S. J. Tetrahedron 1994, 50, 10497-
10508.
(19) See Supporting Information.
J. Org. Chem, Vol. 72, No. 13, 2007 4613

Smith et al.
a fully elaborated D-ring during to the union of an eastern and
western fragment was clearly appealing. To test the viability
of this approach, we turned to the preparation of tetracyclic
indole 33, bearing an isopropenyl substituent on the D-ring as
found in nodulisporic acid B (Scheme 4). The requisite
hydrazine, efficiently constructed from isopropenyl indoline29
in two steps, was condensed with 1,3-cyclohexanedione.30 In
addition to the expected hydrazone 32 (50%), we obtained a
very modest yield (ca. 9%) of the desired tetracyclic indole 33.
However, when the pure hydrazone 32 was subjected to the
same or related Fischer indole protocols, the yield of 33
improved only slightly (ca. 10-20%), presumably due to the
instability of 33 to the reaction conditions. The structure of 33
was confirmed via single-crystal X-ray crystallography (Figure
4). Examination of the bond angles in 33 not surprisingly
revealed significant bond angle distortions. Particularly note-
FIGURE 3. Unsuccessful approaches for the construction of ring D.
We next considered an intramolecular carbenoid insertion
promoted by rhodium.20-23 Required here was 26 (Figure 3).
Two Rh carboxylate catalysts were explored. The common
dirhodium tetraacetate [Rh2(OCOCH3)4] catalyst24 led rapidly
to decomposition. The less common catalyst, dirhodium tet-
raoctanoate,24 also led at room temperature to no reaction, while
at the reflux temperature of dichloroethane, complete decom-
position occurred. Again, ring strain in conjunction with the
stereoelectronics presumably precluded bond formation.
We turned next to the possibility of bond c construction
(Scheme 3). Two approaches were explored: a Dieckmann
condensation25 and a Heck carbonylation.26,27 The requisite
Dieckmann substrate was diester 28.19 No products were
observed employing a variety of bases (e.g., NaH, NaOMe,
LiHMDS). Similarly, treatment of 29 with a series of palladium
catalysts, solvents, and temperature regiments failed to generate
the D-ring. Again the ring strain was presumably too high to
permit cyclization. Taken together, these observations compelled
us to abandon installation of the C(7)-N(1) two-carbon bridge
(D-ring) onto a pre-existing indole nucleus. This decision had
two important consequences: first, a new convergent indole
synthesis was mandated to assemble the highly strained CDE
tricycle; second, the new union tactic would require at least
modest re-engineering of the original eastern and western
hemispheres (6 and 7).
FIGURE 4. Bond angles of tetracyclic indole 33 as determined from
the single-crystal X-ray structure (Jmol Viewer version 10).
worthy, one of the bond angles in ring C deviated by 12.3°
from the expected angle of 120° for benzene. Similar transfor-
mations with the hydrazones derived from either cyclopen-
tanedione or cyclopentanone however led only to decomposition.
Again, the issue of ring strain under the thermodynamic
conditions prevented indole ring formation.
Turning to the literature for alternative construction of the
CDE tricycle, we noted the work of Wijngaarden and co-
workers28 who subjected hydrazone 30 to the classic Fischer
indole synthetic protocol [HCl/AcOH] (Scheme 4). Tetracyclic
At this juncture, we reasoned that access to the γ-amino
ketone that serves as the cornerstone of the Fischer indole tactic
SCHEME 4. The Fischer Indole Synthesis Approach
(20) Doyle, M. P. Top. Organomet. Chem. 2004, 13, 203-222.
(21) Ye, T.; McKervey, M. A. Chem. ReV. 1994, 94, 1091-1160.
(22) Padwa, A.; Krumpe, K. E. Tetrahedron 1992, 48, 5385-5453.
(23) Burke, S. D.; Grieco, P. A. Org. React. 1979, 26, 361-475.
(24) Adams, J.; Spero, D. M. Tetrahedron 1991, 47, 1765-1808.
(25) Davis, D. R.; Garratt, P. J. In ComprehensiVe Organic Syntheses;
Trost, B. M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 2, pp 795-
863.
(26) Trzeciak, A. M.; Ziolkowski, J. J. Coord. Chem. ReV. 2005, 249,
2308-2322.
(27) Skoda-Foldes, R.; Kollar, L. Curr. Org. Chem. 2002, 6, 1097-
1119.
(28) van Wijngaarden, I.; Hamminga, D.; van Hes, R.; Standaar, P. J.;
Tipker, J.; Tulp, M. T. M.; Mol, F.; Olivier, B.; de Jonge, A. J. Med. Chem.
1993, 36, 3693-3699.
(29) O’Connor, J. M.; Stallman, B. J.; Clark, W. G.; Shu, A. Y. L.; Spada,
R. E.; Stevenson, T. M.; Dieck, H. A. J. Org. Chem. 1983, 48, 807-809.
(30) Generally, absolute ethanol at reflux is sufficient for the preparation
of hydrazones; however, even after 12 h at reflux, we observed less than
5% of hydrazone 32, accompanied by significant decomposition of the
hydrazine. Acetic acid was therefore added to catalyze the transformation.
indole 31 was formed in excellent yield. We reasoned that a
similar tactic might provide a convergent synthetic strategy for
the core of the nodulisporic acids. The prospect of generating
4614 J. Org. Chem., Vol. 72, No. 13, 2007

Synthetic Strategy for (+)-Nodulisporic Acids A and B
SCHEME 6. The Stille Cross-Coupling Approach
might permit the development of a new, modular indole
synthetic protocol. This line of analysis suggested that R-ary-
lation of ketones, as introduced by Buchwald and Hartwig,31,32
might be well-suited to permit access to the requisite γ-amino
ketones. Although we were able to reproduce successfully
several R-ketone arylations as reported by Buchwald et al.,31
treatment of 7-bromoindoline or the N-Boc derivative with either
cyclohexanone or 1,3-cyclohexanedione led only to recovery
of starting materials. We suspect that the necessary oxidative
addition most likely fails to take place, again due to steric
reasons. Indeed a careful literature search revealed the lack of
o-haloanilines participating in the Buchwald-Hartwig ketone
arylation process. We therefore halted further efforts to obtain
the required γ-amino ketone intermediates via the Buchwald-
Hartwig protocol.
Still intrigued by the possibility of R-ketone arylations, we
turned to other cross-coupling reactions (i.e., Suzuki or Stille
cross-coupling).12 Overman and co-workers,33,34 en route to the
total synthesis of (-)-idiospermuline, reported the union of a
complex 7-iodo-N-Boc indoline with an aliphatic vinylstannane.
To explore this possibility, we constructed model substrates (35,
37, and 39), albeit reversing the functional groups compared to
the Overman example (Scheme 5). Specifically N-Boc indoline
34 was ortho-lithiated, and the resulting aryllithium reacted with
tributylstannyl chloride to furnish arylstannane 35.35 The
SCHEME 5. Preparation of Substrates for the Stille
Cross-Coupling
requisite R-iodoenones 37 and 39 were prepared using the
procedure developed by Johnson et al.36 We began with the
Stille cross-coupling reaction of enone 37 with arylstannane 35
(Scheme 6). No reaction was observed at room temperature
employing the Overman conditions. However, as the reaction
temperature was raised (ca. >80 °C), the yield of 41 progres-
sively increased. Optimum conditions entailed 1.5 equiv of 35
at ca. 100 °C. When less than 1.5 equiv of 35 was employed,
a sharp drop in the yield was observed.37 In similar fashion,
the Stille cross-coupling of 2-iodocyclohexenone 39 with 35
furnished 40; however, to obtain a high yield, the catalyst and
ligand loadings had to be increased (Scheme 6). Reduction of
enones 40 and 41 with L-Selectride in 1,4 fashion furnished the
desired R-arylcycloalkanones 42 and 43, respectively.38 Again
cognizant of the core CDE ring strain, we turned to the critical
cyclodehydration. Gratifyingly, exposure of 42 to methanolic
HCl at reflux furnished the desired tetracyclic indole 44 in 70%
yield. Treatment of 43, however, with a variety of Brønsted
and Lewis acids (e.g., HCl/MeOH, TFA, TMSI, MgBr2,
p-TsOH) even under forcing conditions, led only to decomposi-
tion. Presumably, the ring strain in indole 45 is again sufficiently
high to prevent indole formation under thermodynamic condi-
tions. This result suggests that even if we had succeeded in
(31) Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem.
Soc. 2000, 122, 1360-1370.
(32) Culkin, D. A.; Hartwig, J. F. Acc. Chem. Res. 2003, 36, 234-245.
(33) Overman, L. E.; Peterson, E. A. Tetrahedron 2003, 59, 6905-6919.
(34) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483.
(35) Diep, V.; Dannenberg, J. J.; Franck, R. W. J. Org. Chem. 2003,
68, 7907-7910.
(36) Johnson, C. R.; Adams, J. P.; Braun, M. P.; Senanayake, C. B. W.
Tetrahedron Lett. 1992, 33, 917-918.
(37) The reaction was also found to be very sensitive to the anhydrous
state of the NMP; that is, not sufficiently dried NMP caused the arylstannane
35 to undergo significant protodestannylation at the reaction temperature.
The protodestannylation was found to be much faster than the cross-
coupling; in some cases, no coupling product (40) was observed.
(38) A survey of the literature revealed that this transformation might
prove quite challenging since a mixture of 1,2- and 1,4-reduction products
had been reported for many combination of enones and reducing agents:
(a) Fortunato, J. M.; Ganem, B. J. Org. Chem. 1976, 41, 2194-2200. (b)
Kowalski, C. J.; Weber, A. E.; Fields, K. W. J. Org. Chem. 1982, 47, 5088-
5093. (c) Lee, H.-Y.; An, M. Tetrahedron Lett. 2003, 44, 2775-2778.
J. Org. Chem, Vol. 72, No. 13, 2007 4615

Smith et al.
facilitating the [3,3]-sigmatropic rearrangement (i.e., in the
proposed Fischer indole synthesis) employing the cyclopen-
tanone and 1,3-cyclopentanedione-derived hydrazones, the
resulting γ-amino ketone intermediates would not likely have
undergone cyclization to furnish tetracyclic indoles.
cycloalkenones 40 and 41 were reduced with L-Selectride and
the resulting lithium alkoxides captured as the enol triflates with
the Comins reagent.45 Enol triflates 46 and 47 were produced
in excellent yields. Removal of the Boc protecting groups was
then achieved with TMSI46 to furnish free amines 48 and 49.
We next screened the most commonly used catalysts [i.e., Pd-
(OAc)2 and Pd2(dba)3], ligands [i.e., BINAP and xantphos],
bases [i.e., NaOt-Bu, LiHMDS, Cs2CO3], and solvent systems
[i.e., toluene, THF] to achieve the Buchwald-Hartwig amina-
tion. We quickly recognized that the use of strong bases, such
as NaOt-Bu and LiHMDS, would not be compatible with the
enol triflate moiety. Rapid decomposition occurred within
minutes even at room temperature. Eventually we discovered
that the combination of a mild base (Cs2CO3) and THF as
solvent effected cyclization of enol triflate 48 to furnish the
corresponding tetracyclic indole 44 in 58% yield (Scheme 7).
However, the identical reaction conditions [2.5 mol % of Pd2-
(dba)3, 7.5 mol % of xantphos, 2 equiv of Cs2CO3 in THF at
reflux] failed to effect the cyclization of enol triflate 49.
Formation of tetracyclic indole 45 did occur, albeit in low yield
(ca. 5%) when the reaction mixture was heated to 100-130 °C
in a sealed tube. This observation again attests to the fact that
the activation energy for cyclization of 49 to 45 is considerably
higher than for the cyclization of 48. We next reasoned that, if
we were to prepare the kinetically derived enol triflate 50 from
ketone 43, the presence of an sp3-hybridized carbon at the
benzylic position might permit a less strained transition state
for the conversion of 50 to 45 (Scheme 8). Pleasingly, treatment
of enol triflate 50, derived from 43, under the optimum
The Buchwald-Hartwig Amination: A Kinetically Con-
trolled Protocol. At this juncture, we recognized that construc-
tion of the requisite C-N bond would require a kinetically
controlled process. We selected the transition-metal-catalyzed
C-N bond formation known as the Buchwald-Hartwig ami-
nation.39 A thorough literature search revealed that examples
of the amination between enol triflates and aliphatic or aromatic
amines were not well precedented.40-42 Only one report by
Willis et al.41 featured secondary amines as the coupling partner.
All of the examples, however, were intermolecular reactions.43,44
Nonetheless, we proceeded with the expectation that the right
combination of catalyst, ligand, base, and solvent would permit
formation of the C-N bond under kinetically controlled
conditions (Scheme 7). Given the availability of the Stille cross-
SCHEME 7. Initial Buchwald-Hartwig Cyclization Studies
SCHEME 8. Buchwald-Hartwig Coupling of Kinetic Enol
Triflate 50
cyclization conditions developed for the conversion of 48 to
44 furnished the highly strained tetracyclic indole 45 in 72%
yield. From the perspective of the nodulisporic acids, this result
clearly demonstrated that an indole such as 45 could indeed be
formed in high yield and, importantly, was stable under the
reaction conditions. However, still determined to define condi-
(39) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002, 219, 131-
209.
(40) Hicks, F. A.; Brookhart, M. Org. Lett. 2000, 2, 219-221.
(41) Willis, M. C.; Brace, G. N. Tetrahedron Lett. 2002, 43, 9085-
9088.
(42) Willis, M. C.; Brace, G. N.; Holmes, I. P. Angew. Chem., Int. Ed.
2005, 44, 403-406.
(43) Barluenga, J.; Valdes, C. Chem. Commun. 2005, 4891-4901.
(44) Dehli, J. R.; Legros, J.; Bolm, C. Chem. Commun. 2005, 973-986.
(45) Comins, D. L.; Dehghani, A. Tetrahedron Lett. 1992, 33, 6299-
6302.
coupling products 40 and 41, formation of the requisite enol
triflates 46 and 47 appeared straightforward. To this end, R-aryl
4616 J. Org. Chem., Vol. 72, No. 13, 2007

Synthetic Strategy for (+)-Nodulisporic Acids A and B
tions that would convert the thermodynamically derived enol
triflate 49 to indole 45, we turned to increasing the catalyst load,
a tactic known in some cases to bring about difficult transition-
metal-catalyzed transformations, given the issue of catalyst
decomposition. Gratifyingly, doubling the catalyst load resulted
in product formation in THF at reflux. Optimum conditions
required a 4-fold increase in both catalyst and ligand (Scheme
9).47 Several hundred milligrams of the tetracyclic indoles 44
to access a more complex indole that would more closely
resemble the substitution pattern found in (+)-nodulisporic acid
A (1), we searched for an appropriate model ketone possessing
a quaternary stereocenter at the R-position. We eventually
selected (+)-estrone methyl ether48 which possesses the same
substitution pattern as (+)-nodulisporic acid A (1) at C(3) and
C(12), except having the opposite absolute stereogenicity
(Scheme 10). Construction of the requisite vinyl iodide (-)-52
was achieved in three steps, involving Saegusa oxidation of the
enol ether derived from (+)-estrone methyl ester to install the
requisite R,â-unsaturation in enone moiety,49,50 followed by
iodination employing the method of Johnson et al.36
SCHEME 9. Buchwald-Hartwig Cyclization of
Thermodynamic Enol Triflate 49
SCHEME 10. Design of a Heptacyclic Strained Indole
[(+)-51] That Resembled Nodulisporic Acid A (+)-1 in
Substitution Pattern at C(3) and C(12)
and 45 were readily prepared. Structure confirmation was
achieved by single-crystal X-ray crystal analysis. As with indole
33, significant bond angle distortion was observed in 44 and
45 (Figure 5).
Pleasingly, the Stille cross-coupling between arylstannane 35
and (-)-52 proved successful; the best conditions entailed 10
mol % of catalyst, 40 mol % of ligand, with 2 equiv of
arylstannane. Ketone (-)-53 was obtained in excellent yield
(ca. 89%). L-Selectride reduction then furnished (+)-56 in 80%
yield. Given our earlier model studies, we were not surprised
to find that all attempts to achieve cyclization of (+)-56 to
heptacyclic indole (+)-51 under thermodynamic conditions (e.g.,
acid) failed; only decomposition occured. The Buchwald-
Hartwig amination protocol, however, proved quite successful.
Treatment of (-)-53 with L-Selectride followed by lithium
enolate capture with the Comins reagent45 furnished the N-Boc
enol triflate (+)-54. Removal of the Boc group with TMSI,
followed by application of the optimized amination cyclization
conditions (ca. 10 mol % of catalyst, 30 mol % of ligand, Cs2-
FIGURE 5. Bond angles of tetracyclic indoles 44 and 45 as determined
from the single-crystal X-ray structures (Jmol Viewer version 10).
To demonstrate that the sequential Stille cross-coupling/
Buchwald-Hartwig union/cyclization tactic could be employed
(48) Xie, W.; Peng, H.; Kim, D. I.; Kunkel, M.; Powis, G.; Zalkow, L.
H. Bioorg. Med. Chem. 2001, 9, 1073-1083.
(49) Ito, Y.; Hirao, T.; Saegusa, T. J. Org. Chem. 1978, 43, 1011-1013.
(50) Egner, U.; Fritzemeier, K.-H.; Halfbrodt, W.; Heinrich, N.; Kuhnke,
J.; Muller-Fahrnow, A.; Neef, G.; Schollkopf, K.; Schwede, W. Tetrahedron
1999, 55, 11267-11274.
(46) Lott, R. S.; Chauhan, V. S.; Stammer, C. H. J. Chem. Soc., Chem.
Commun. 1979, 495-496.
(47) Further increase in catalyst and ligand loadings did not result in
higher yields of tetracyclic indole 45.
J. Org. Chem, Vol. 72, No. 13, 2007 4617

Smith et al.
SCHEME 11. Preparation of the Highly Strained
Heptacyclic Indole (+)-51
FIGURE 6. Single-crystal X-ray structure of indole (+)-51 (Jmol
Viewer version 10).
nodulisporic acids A (1) and B (2) to involve arylstannane 59
and vinyl iodide 58 (Scheme 12). Importantly, the proposed
second-generation synthetic strategy not only preserves the
convergency of the original approach, but also holds the
potential for increased convergency by permitting the union of
more elaborate fragments.
SCHEME 12. Second-Generation Retrosynthetic Analysis of
(+)-Nodulisporic Acid A (1)
CO3 (2 equiv) in dioxane), completed construction of (+)-51
in 55% yield (Scheme 11).51 Single-crystal X-ray crystal-
lography confirmed the structure (Figure 6). We again note
considerable bond angle distortions were observed in the
aromatic ring (see Supporting Information).
Having demonstrated the efficiency of the sequential Stille
cross-coupling/Buchwald-Hartwig union/cyclization tactic, we
have now revised our first-generation synthetic analysis of (+)-
4618 J. Org. Chem., Vol. 72, No. 13, 2007

Synthetic Strategy for (+)-Nodulisporic Acids A and B
Preparation of (-)-53. To a 50 mL round-bottom Schlenk flask,
equipped with a PTFE-coated stirbar, were charged 7-tributylstan-
nyl-N-Boc indoline 359 (1.02 g, 2.00 mmol, 2 equiv), NMP (20
mL), Pd2(dba)3‚CHCl3 (104 mg, 0.100 mmol, 10 mol %), and P(2-
furyl)3 (93 mg, 0.400 mmol, 40 mol %), and the resulting solution
was purged with deoxygenated Ar for 15 min. Afterward, a solution
of (-)-52 (409 mg, 1.00 mmol, 1 equiv) in NMP (5 mL) was added
via syringe, followed by flame-dried CuI (191 mg, 1.00 mmol, 1
equiv). The reaction vessel was then fitted with a reflux condenser
and heated at 100 °C for 1 h to give a dark green mixture. The
reaction mixture was then slowly poured into saturated NH4OH
(75 mL) and Et2O (75 mL) with vigorous stirring. The aqueous
layer was extracted with Et2O (3 × 30 mL) and EtOAc (1 × 30
mL), and the combined organic layers were washed with brine (4
× 25 mL), dried over MgSO4, concentrated in vacuo, and adsorbed
onto silica gel (10 g). Flash chromatographic purification (hexanes/
EtOAc ) 6:1) furnished (-)-53 as an amorphous white solid (445
Summary
A second-generation strategy for construction of (+)-nod-
ulisporic acids A and B, based on the development of a new,
highly efficient, modular indole synthesis exploiting a sequential
Stille cross-coupling/Buchwald-Hartwig union/cyclization tac-
tic, has been developed. Construction of the requisite redesigned
eastern and western hemispheres 58 and 59 for (+)-nodulisporic
acids A and B is currently underway. Progress concerning their
union and future elaboration to (+)-nodulisporic acids A and
B will be reported in due course.
Experimental Section
Preparation of Tetracyclic Indole (44). To a 10 mL round-
bottom flask, equipped with a PTFE-coated stirbar and a reflux
condenser, were charged ketone 42 (70 mg, 0.222 mmol, 1 equiv)
and deoxygenated HPLC grade MeOH (2 mL). In the meantime,
acetyl chloride (174 mg, 2.22 mmol, 10 equiv) was carefully added
to deoxygenated MeOH (2 mL) at 0 °C to produce anhydrous HCl
in MeOH. After 5 min, this freshly prepared MeOH/HCl solution
was added in one portion to a solution of 42 in MeOH at room
temperature, and the resulting mixture was heated at reflux for 15
min. The solvent and excess HCl were removed in vacuo, and the
crude product was purified by preparative TLC (1000 µm plate)
eluted with hexanes/EtOAc ) 3:1, affording 44 as a white solid
after the removal of the solvents in vacuo (31 mg, 70%): mp )
149-150 °C; Rf ) 0.7 (hexanes/EtOAc ) 3:1); 1H NMR (500 MHz,
CDCl3) δ 7.20 (d, J ) 7.7 Hz, 1H), 6.94 (t, J ) 7.3 Hz, 1H), 6.85
(d, J ) 6.7 Hz, 1H), 4.35 (t, J ) 7.1 Hz, 2H), 3.74 (t, J ) 7.1 Hz,
2H), 2.75 (m, 4H), 1.91 (m, 2H), 1.86 (m, 2H); 13C NMR (125
MHz, CDCl3) δ 147.9, 134.8, 123.9, 120.7, 119.6, 115.6, 114.2,
113.4, 47.9, 34.0, 23.3, 23.0, 22.9, 22.6; IR (neat) 3046 (w), 2937
(s), 2912 (s), 2845 (s), 1651 (w), 1507 (w), 1418 (m), 1336 (w),
1309 (m), 1292 (s), 1193 (w), 743 (s), 582 (w). HRMS (CI-MS)
calcd for C14H15N [M+] 197.1204, found 197.1201.
mg, 89%): mp ) 164-165 °C; [R]D20 ) -39.25 (c 0.53, CHCl3);
Rf ) 0.52 (hexanes/EtOAc ) 3:1); 1H NMR (500 MHz, CDCl3) δ
7.49 (s, 1H), 7.24 (dd, J ) 7.9, 5.7 Hz, 2H), 7.15 (d, J ) 7.3 Hz,
1H), 7.03 (t, J ) 7.5 Hz, 1H), 6.75 (dd, J ) 8.6, 2.3 Hz, 1H), 6.69
(d, J ) 2.0 Hz, 1H), 4.33-3.95 (m, 2H), 3.80 (s, 3H), 3.03 (t, J )
7.9 Hz, 2H), 3.00-2.94 (m, 2H), 2.69 (d, J ) 11.5 Hz, 1H), 2.51-
2.36 (m, 2H), 2.31-2.19 (m, 1H), 2.07 (d, J ) 12.8 Hz, 1H), 1.93-
1.68 (m, 3H), 1.66-1.55 (m, 1H), 1.46 (s, 9H), 1.21 (s, 3H); 13C
NMR (125 MHz, CDCl3) δ 209.5, 157.6, 152.5, 149.6, 143.1, 140.6,
137.6, 134.4, 132.2, 128.4, 126.0, 124.1, 123.7, 121.9, 113.9, 111.5,
80.1, 55.2, 54.6, 51.5, 49.8, 45.2, 35.8, 29.8, 29.4, 29.0, 28.4, 26.8,
25.6, 21.1; IR (neat) 2929 (br s), 2858 (m), 1711 (s), 1609 (m),
1576 (w), 1500 (s), 1447 (s), 1434 (s), 1389 (s), 1368 (s), 1336
(s), 1280 (m), 1244 (s), 1163 (s), 1135 (m), 1104 (w), 1075 (w),
1050 (m), 1007 (m), 983 (m), 910 (m), 869 (w), 821 (w), 799 (w),
767 (m), 734 (m), 647 (w), 592 (w). HRMS (ESI-MS) calcd for
C32H37NO4Na [(M + Na)+] 522.2620, found 522.2591.
Preparation of (+)-56. To a 25 mL round-bottom flask,
equipped with a PTFE-coated stirbar, were charged enone (-)-53
(122 mg, 0.240 mmol, 1 equiv) and THF (10 mL), and the resulting
solution was cooled to -78 °C. Next, a 1.0 M solution of
L-Selectride in THF (0.27 mL, 0.27 mmol, 1.1 equiv) was
introduced over 1 min. The cold bath was then removed, and the
reaction mixture was warmed to 0 °C and stirred at that temperature
for 1 h. The reaction mixture was then quenched with 2 N NaOH
(10 mL) and subsequently extracted with Et2O (15 mL). The
resulting organic layer was then washed with water (20 mL)
followed by brine (20 mL). The aqueous phase was extracted with
EtOAc (2 × 15 mL), and the combined organic layers were dried
over MgSO4, concentrated in vacuo, and adsorbed onto silica gel
(3 g). Flash column chromatography (hexanes/EtOAc ) 5:1)
afforded (+)-56 as a white solid (98 mg, 80%): mp ) 193-195
Preparation of Tetracyclic Indole (45). To a 25 mL round-
bottom flask, equipped with a PTFE-coated stirbar and a reflux
condenser, were charged Cs2CO3 (451 mg, 1.386 mmol, 2 equiv),
THF (10 mL), Pd2(dba)3‚CHCl3 (71 mg, 0.07 mmol, 10 mol %),
and xantphos (121 mg, 0.208 mmol, 30 mol %). The resulting
solution was stirred for 5 min and then treated with a solution of
enol triflate 49 (231 mg, 0.693 mmol, 1 equiv) in THF (2 mL),
and the resulting mixture was heated at reflux for 2 h. The reaction
mixture was then cooled to room temperature, quenched with
saturated aqueous NH4Cl (15 mL), and extracted with Et2O (3 ×
30 mL), and the combined organic layers were dried over MgSO4
and concentrated in vacuo. The crude material was adsorbed onto
silica gel (3 g) and subjected to column chromatography (hexanes/
EtOAc ) 30:1), affording 45 as a white solid (69 mg, 55%): mp
°C (dec); [R]D20) +170.00 (c 0.23, CHCl3); Rf ) 0.58 (hexanes/
1
1
) 109-110 °C; Rf ) 0.7 (hexanes/EtOAc ) 2:1); H NMR (500
EtOAc ) 3:1); H NMR (500 MHz, CDCl3) δ 7.21 (d, J ) 8.7
MHz, CDCl3) δ 7.19 (d, J ) 7.8 Hz, 1H), 6.97 (dd, J ) 7.8, 6.8
Hz, 1H), 6.86 (d, J ) 6.8 Hz, 1H), 4.42-4.38 (m, 2H), 3.75 (t, J
) 7.1 Hz, 2H), 2.84 (m, 4H), 2.54 (m, 2H); 13C NMR (125 MHz,
CDCl3) δ 153.1, 143.3, 124.6, 122.3, 121.0, 117.3, 115.8, 113.8,
48.6, 34.1, 28.8, 24.9, 24.7; IR (KBr pellet) 3047 (s), 3015 (w),
2956 (m), 2929 (m), 2901 (m) 2852 (s), 1646 (s), 1496 (s), 1478
(m), 1465 (s), 1456 (m), 1437 (s), 1379 (m), 1368 (m), 1338 (s),
1294 (m), 1269 (s), 1215 (w), 1174 (m), 1156 (w), 1096 (w), 1040
(w), 1012 (m), 951 (w), 822 (w), 752 (s), 743 (s), 617 (w), 588
(w), 517 (w). HRMS (CI-MS) calcd for C13H13N [M+] 183.1048,
found 183.1051.
Hz, 1H), 7.08 (d, J ) 6.9 Hz, 1H), 7.02 (td, J ) 14.8, 7.3 Hz, 2H),
6.73 (dd, J ) 8.6, 2.5 Hz, 1H), 6.65 (d, J ) 2.2 Hz, 1H), 4.23 (t,
J ) 8.6 Hz, 1H), 4.09 (dtd, J ) 18.5, 11.2, 7.6 Hz, 2H), 3.79 (s,
3H), 3.05-2.93 (m, 2H), 2.91 (dd, J ) 8.7, 3.9 Hz, 2H), 2.64 (dd,
J ) 13.1, 5.0 Hz, 1H), 2.47-2.37 (m, 1H), 2.33 (s, 1H), 2.18-
1.91 (m, 2H), 1.71-1.51 (m, 6H), 1.53 (s, 9H), 0.94 (s, 3H); 13C
NMR (125 MHz, CDCl3) δ 219.4, 157.6, 153.6, 142.0, 137.8, 135.2,
132.2, 129.6, 126.2, 126.0, 124.8, 122.9, 113.9, 111.5, 80.7, 55.2,
52.8, 51.0, 48.9, 48.7, 44.1, 37.9, 32.1, 29.8, 29.6, 29.5, 28.4, 26.8,
26.0, 13.7; IR (neat) 2930 (br s), 1736 (s), 1702 (s), 1641 (w),
1609 (m), 1549 (w), 1529 (w), 1500 (m), 1479 (w), 1449 (m), 1433
(m), 1369 (s), 1335 (m), 1281 (w), 1243 (m), 1160 (s), 1125 (m),
1053 (m), 1010 (w), 908 (m), 849 (w), 822 (w), 767 (w), 732 (s),
647 (w). HRMS (ESI-MS) calcd for C32H40NO4 [(M + H)+]
502.2957, found 502.2953.
(51) When THF was used as the solvent, we isolated a side product in
10% yield in which the triflate functionality was reduced and the indoline
ring was oxidized to the corresponding indole. This oxidation most likely
occurred via air oxidation during the purification stage. However, we later
found that, when THF was replaced with dioxane as the solvent, only the
desired heptacyclic indole (+)-51 was obtained.
Preparation of (+)-54. To a 25 mL round-bottom flask,
equipped with a PTFE-coated stirbar, were charged enone (-)-53
J. Org. Chem, Vol. 72, No. 13, 2007 4619

Smith et al.
(375 mg, 0.750 mmol, 1 equiv) and THF (10 mL), and the resulting
solution was cooled to -78 °C. Next, a 1.0 M solution of
L-Selectride in THF (0.83 mL, 0.830 mmol, 1.1 equiv) was added
at -78 °C over 3 min, and the resulting yellow solution was stirred
at -78 °C for 15 min. Next, a solution of Comins’ reagent (442
mg, 1.125 mmol, 1.5 equiv) in THF (4 mL) was introduced over 1
min, and the resulting mixture was stirred for 20 min at -78 °C.
The -78 °C reaction mixture was then poured into saturated
aqueous NH4Cl (20 mL) and extracted with Et2O (4 × 10 mL) and
EtOAc (1 × 10 mL). The combined organic layers were washed
with brine (2 × 20 mL), dried over MgSO4, and concentrated in
vacuo. The crude material was adsorbed onto silica gel (3 g) and
subjected to flash chromatography (hexanes/EtOAc ) 20:1) af-
fording (+)-54 as a thick, colorless oil which solidified upon
standing to a white solid (320 mg, 67%): mp ) 79-80 °C; [R]D
(m), 650 (w), 608 (m). HRMS (ESI-MS) calcd for C28H31F3NO4S
[(M + H)+] 534.1926, found 534.1908.
Preparation of Heptacyclic Indole (+)-51 Derived from
Estrone. To a 25 mL round-bottom Schlenk flask, equipped with
a PTFE-coated stirbar, were charged Cs2CO3 (202 mg, 0.62 mmol,
2 equiv), dioxane (8 mL), xantphos (54 mg, 0.09 mmol, 30 mol
%), and Pd2(dba)3‚CHCl3 (32 mg, 0.031 mmol, 10 mol %) to give
a deep brownish red solution. The resulting solution was stirred
for 5 min, then treated with a solution of the enol triflate (+)-55
(165 mg, 0.309 mmol, 1 equiv) in THF (3 mL). Next, the Schlenk
flask was fitted with a reflux condenser, and the reaction mixture
was brought to reflux, yielding a dark brownish yellow solution.
After 50 min, the reaction was cooled to room temperature and
poured into a vigorously stirred mixture of Et2O (20 mL) and
saturated aqueous NH4Cl (20 mL). The aqueous layer was then
extracted with Et2O (2 × 15 mL), and the combined organic layers
were dried over MgSO4, concentrated in vacuo, and adsorbed onto
silica gel (2 g). Flash chromatography (hexanes/EtOAc ) 12:1)
20) +46.59 (c 0.43, CHCl3); Rf ) 0.60 (hexanes/EtOAc ) 2:1);
1H NMR (500 MHz, CDCl3) δ 7.20 (d, J ) 8.6 Hz, 1H), 7.14 (d,
J ) 7.0 Hz, 1H), 7.04 (td, J ) 14.5, 7.08 Hz, 2H), 6.73 (dd, J )
8.5, 2.8 Hz, 1H), 6.66 (d, J ) 2.7 Hz, 1H), 4.13 (ddd, J ) 11.0,
9.3, 6.0 Hz, 1H), 3.99 (ddd, J ) 11.0, 9.3, 8.0 Hz, 1H), 3.79 (s,
3H), 3.20-2.83 (m, 5H), 2.78-2.58 (m, 2H), 2.37 (td, J ) 10.8,
10.4 Hz, 2H), 2.11 (dt, J ) 11.2, 7.06 Hz, 1H), 1.99 (m 1H), 1.93-
1.86 (m, 1H), 1.78 (dt, J ) 12.7, 3.84 Hz, 1H), 1.72-1.57 (m,
2H), 1.50 (s, 9H), 1.15 (s, 3H); 13C NMR (125 MHz, CDCl3) δ
157.6, 153.0, 151.0, 140.3, 137.8, 134.5, 132.4, 131.4, 127.4, 125.9,
124.2, 123.6, 122.6, 122.1, 119.6, 117.0, 114.5, 113.9, 111.5, 111.5,
80.2, 55.2, 52.9, 49.6, 46.1, 44.2, 36.8, 33.5, 29.6, 29.0, 28.5, 26.8,
26.0, 15.3 (the CF3 group shows up as a quartet and all four peaks
are reported); IR (neat) 2976 (m), 2935 (br s), 1709 (s), 1609 (m),
1575 (w), 1500 (m), 1478 (m), 1447 (m), 1435 (m), 1412 (s), 1378
(s), 1338 (m), 1281 (w), 1241 (s), 1210 (s), 1161 (s), 1140 (s),
1107 (w), 1059 (m), 1036 9 m), 1010 (w), 957 (w), 911 (m), 858
(m), 846 (m), 813 (w), 768 (w), 735 9m0, 653 (w), 608 (m), 567
(w), 504 (w). HRMS (ESI-MS) calcd for C33H38F3NO6SNa [(M +
Na)+] 656.2270, found 656.2281.
Preparation of Enol Triflate (+)-55. To a 10 mL round-bottom
flask, equipped with a PTFE-coated stirbar, were charged (+)-54
(203 mg, 0.321 mmol, 1 equiv) and CH2Cl2 (5 mL), and the
resulting solution was cooled to 0 °C. Next, TMSI (129 mg, 91
µL, 2 equiv) was added at 0 °C over 1 min, and the resulting mixture
was stirred for 45 min at 0 °C. The reaction was then quenched
with saturated aqueous NH4Cl (5 mL) and extracted with CH2Cl2
(3 × 5 mL). The combined organic layers were dried over MgSO4
and concentrated in vacuo, affording analytically pure (+)-55 as a
furnished (+)-51 as a white crystalline solid (65.4 mg, 55%): mp
20
1
) 204-205 °C (dec);[R]D ) +64.20 (c 0.26, CHCl3); H NMR
(500 MHz, CDCl3) δ 7.24 (d, J ) 8.5 Hz, 1H), 7.19 (d, J ) 7.8
Hz, 1H), 6.96 (dd, J ) 7.8, 6.8 Hz, 1H), 6.85 (d, J ) 6.8 Hz, 1H),
6.74 (dd, J ) 8.5, 2.8 Hz, 1H), 6.68 (d, J ) 2.6 Hz, 1H), 4.47 (dt,
J ) 9.3, 4.6 Hz, 1H), 4.39 (dt, J ) 9.4, 5.9 Hz, 1H), 3.80 (s, 3H),
3.79 (m, 1H), 3.73 (dq, J ) 8.9, 4.4 Hz, 1H), 3.09-2.87 (m, 2H),
2.81 (dd, J ) 13.4, 6.2 Hz, 1H), 2.43 (m, 3H), 2.26 (ddd, J )
10.6, 8.3, 4.7 Hz, 1H), 2.11-2.01 (m, 2H), 1.90 (dt, J ) 12.7, 3.7
Hz, 1H), 1.77 (dtd, J ) 16.2, 11.9, 3.4 Hz, 2H), 1.54 (dq, J )
12.0, 6.6 Hz, 1H), 1.04 (s, 3H); 13C NMR (125 MHz, CDCl3) δ
157.5, 152.3, 151.6, 138.0, 132.7, 126.0, 124.6, 121.2, 119.7, 117.7,
115.9, 113.9, 113.7, 111.5, 60.1, 55.2, 48.3, 44.5, 41.7, 37.4, 34.7,
34.3, 29.8, 27.4, 26.3, 26.0, 17.9; IR (neat) 2947 (br m), 1499 (m),
1497 (m), 1456 (m), 1447 (m), 1372 (w), 1362 (w), 1329 (w) 1280
(w), 1245 (m), 1182 (w), 1143 (w), 1046 (w), 909 (w), 732 (m).
HRMS (ESI-MS) calcd for C27H30NO [(M + H)+] 384.2327, found
384.2326.
Acknowledgment. Financial support was provided by the
National Institutes of Health (National Institute of General
Medical Sciences) through Grant GM-29028 and by the Bristol-
Myers Squibb Pharmaceutical Research Institute (for stipend
support of A.H.D.). We are also grateful to Eli Lilly and
Company for the graduate fellowship to L.K., the Merck
Research Laboratories for generous financial support of this
research program, and to Dr. P. T. Meinke for providing
generous samples of natural (+)-nodulisporic acid F and (-)-
nodulisporic acid D. Finally, we thank Dr. G. Furst, Dr. R. K.
Kohli, and Dr. P. Carroll for assistance in obtaining NMR
spectra, high-resolution mass spectra, and X-ray crystallographic
data, respectively.
yellow oil (165 mg, 96%): [R]D20) +27.72 (c 0.29, CHCl3); Rf )
0.35 (hexanes/EtOAc ) 2:1); 1H NMR (500 MHz, CDCl3) δ 7.20
(d, J ) 8.6 Hz, 1H), 7.09 (dd, J ) 7.2, 0.8 Hz, 1H), 6.90 (dd, J )
7.7, 0.5 Hz, 1H), 6.73 (td, J ) 15.0, 5.1 Hz, 2H), 6.66 (d, J ) 2.6
Hz, 1H), 3.79 (s, 3H), 3.59 (dt, J ) 8.5, 1.25 Hz, 2H), 3.06 (t, J )
8.4 Hz, 1H), 2.92 (dd, J ) 9.3, 6.5 Hz, 2H), 2.67 (dd, J ) 14.9,
6.4 Hz, 1H), 2.48-2.40 (m, 3H), 2.37 (dd, J ) 11.0, 4.3 Hz, 1H),
2.08-1.99 (m, 1H), 1.99-1.91 (m, 3H), 1.82 (dt, J ) 12.7, 4.2
Hz, 1H), 1.63 (m, 2H), 1.48 (ddd, J ) 23.2, 11.2, 7.4 Hz, 1H),
1.16 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 157.6, 153.0, 149.0,
137.7, 132.1, 130.0, 129.9, 126.7, 125.9, 124.6, 122.1, 119.6, 118.6,
117.0, 116.1, 114.6, 113.9, 111.5, 55.2, 52.6, 47.1, 46.1, 44.1, 36.7,
33.4, 33.3, 29.8, 29.4, 26.7, 25.9, 15.8 (the CF3 group shows up as
a quartet and all four peaks are reported); IR (neat) 3328 (br w),
2934 (br s), 1608 (m), 1500 (m), 1452 (m), 1414 (s), 1282 (m),
1210 (s), 1139 (s), 1059 (m), 1036 (m), 911 (m), 847 (m), 735
Supporting Information Available: Experimental procedures
and spectroscopic and analytical data for all other new compounds.
Copies of 1H and 13C NMR spectra for all new compounds.
Crystallographic information files (CIF) of compounds 33, 44, 45,
and 51. This material is available free of charge via the Internet at
http://pubs.acs.org.
JO062423A
4620 J. Org. Chem., Vol. 72, No. 13, 2007