The N-Acyloxyiminium Ion Aza-Prins Route to Octahydroindoles: Total Synthesis and Structural Confirmation of the Antithrombotic Marine Natural Product Oscillarin

Article abstract of DOI:10.1021/ja030669g

The first enantiocontrolled total synthesis of the marine natural product oscillarin is described. The proposed structure and absolute configuration of oscillarin is thus confirmed, and a previously assigned structure of a subunit was shown to be incorrect. The X-ray structure of an oscillarin-thrombin complex was resolved at 2.0 A resolution, which validated its potent inhibitory activity against the enzyme with an IC₅₀ = 28 nM. Methodology was developed for the synthesis of enantiopure octahydroindole-2-carboxylic acids with usable functionality at C-6. The method consists of the halocarbocyclization of N-acyloxyiminium ions containing an olefinic tether in the presence of fin tetrachloride or fin tetrabromide. This N-acyloxyiminium ion aza-Prins carbocyclization proved to be general for the construction of octahydroindole and perhydroquinoline 2-carboxylic acids. Mechanistic rationales are based on an antiperiplanar attack of the terminal alkene on the iminium ion, leading to an incipient secondary carbocation which is trapped by halide via an equatorial attack. X-ray crystal structures of products corroborate the expected stereochemistry.

Full text of DOI:10.1021/ja030669g

Published on Web 04/16/2004
The N-Acyloxyiminium Ion Aza-Prins Route to
Octahydroindoles: Total Synthesis and Structural
Confirmation of the Antithrombotic Marine Natural Product
Oscillarin
Stephen Hanessian,*,† Martin Tremblay,† and Jens F. W. Petersen‡
Contribution from the Department of Chemistry, UniVersite´ de Montre´al, C.P. 6128,
Station Centre-Ville, Montre´al, P.Q., H3C 3J7 Canada, and AstraZeneca Structural Chemistry
Laboratory, AstraZeneca, R&D Mo¨lndal, S-431 83, Mo¨lndal, Sweden
Received December 18, 2003; E-mail: stephen.hanessian@umontreal.ca
Abstract: The first enantiocontrolled total synthesis of the marine natural product oscillarin is described.
The proposed structure and absolute configuration of oscillarin is thus confirmed, and a previously assigned
structure of a subunit was shown to be incorrect. The X-ray structure of an oscillarin-thrombin complex
was resolved at 2.0 Å resolution, which validated its potent inhibitory activity against the enzyme with an
IC50 ) 28 nM. Methodology was developed for the synthesis of enantiopure octahydroindole-2-carboxylic
acids with usable functionality at C-6. The method consists of the halocarbocyclization of N-acyloxyiminium
ions containing an olefinic tether in the presence of tin tetrachloride or tin tetrabromide. This N-acyloxyiminium
ion aza-Prins carbocyclization proved to be general for the construction of octahydroindole and per-
hydroquinoline 2-carboxylic acids. Mechanistic rationales are based on an antiperiplanar attack of the terminal
alkene on the iminium ion, leading to an incipient secondary carbocation which is trapped by halide via an
equatorial attack. X-ray crystal structures of products corroborate the expected stereochemistry.
compounds,7 have been isolated, their structures determined,
and synthesized.8 A third synthesis of aeruginosin 298A was
Introduction
The linear marine peptides generally known as the aerugi-
nosins1 are structurally unique secondary metabolites produced
by various blue-algal species of prokaryotic aquatic microorgan-
isms.2 Originally isolated from the cyanobacterium species
Microcystis aeruginosa, the first member of this structural class
of marine linear peptides possessing serine protease inhibitory
activity was aeruginosin 298A.1a,3 Initially, two total syntheses
provided definitive evidence for its revised structure and absolute
configuration.4,5 Since then, other aeruginosins,6 and related
recently reported.8a
As a class, several aeruginosins share a common 1-aza[4.3.0]-
bicyclic core unit. Thus, the (2S,3aS,6R-hydroxy,7aS) octahy-
droindole 2-carboxylic acid core (L-Choi) is linked to a D-amino
acid in combination with a D-phenyllactic acid or its substituted
variants (Figure 1). Depending on the type of aeruginosin, the
carboxyl end forms an amide bond with a 1-amino-ω-guanidino
alkyl motif. The serine protease inhibitory activity of the
aeruginosins can be related to the presence of three binding
domains that can be recognized by the S1, S2, and S3 sites in
thrombin. X-ray crystallographic studies3,6a of complexes with
trypsin and functional analogies with known inhibitors have
validated this assumption and extrapolated it to thrombin as well.
We recently described the first total synthesis and structural
confirmation of dysinosin A,9 1 isolated from a new genus of
sponge of the family Dysideidae (Lizard Island, Queensland,
Australia).10 In addition to its inhibitory activity against
† Universite´ de Montre´al.
‡ AstraZeneca.
(1) (a) Murakami, M.; Okita, Y.; Matsuda, H.; Okino, H.; Yamaguchi, K.;
Tetrahedron Lett. 1994, 35, 3129. (b) Murakami, M.; Ishida, K.; Okino,
T.; Okita, Y.; Matsuda, H.; Yamaguchi, K. Tetrahedron Lett. 1995, 36,
2785. (c) Matsuda, H.; Okino, T.; Murakami, M.; Yamaguchi, K.
Tetrahedron 1996, 52, 14501.
(2) For recent reviews, see (a) Leusch, H.; Harrigan, G. G.; Goetz, G.; Horgen,
F. D. Curr. Med. Chem. 2002, 9, 179. (b) Burja, A. M.; Banaigs, B.; Abou-
Mansour, E.; Burgess, J.-G.; Wright, P. C. Tetrahedron 2001, 57, 9347.
(c) For a recent review, see Shiori, T.; Hamada, Y. Synlett 2001, 184. (d)
Lewis, J. R. Nat. Prod. Rep. 1998, 417. (e) Namikoshi, M.; Rinehart, K.
L. J. Indust. Microbiol. 1996, 17, 373. (f) Rinehart, K. L.; Namikoshi, M.;
Choi, B. W. J. Appl. Phycol. 1994, 6, 159.
(3) Rios Steiner, J. L.; Murakami, M.; Tulinsky, A. J. Am. Chem. Soc. 1998,
120, 597.
(4) Wipf, P.; Methot, J.-L. Org. Lett. 2000, 2, 4213.
(5) (a) Valls, N.; Lo´pez-Canet, M.; Vallribera, M.; Bonjoch, J. J. Am. Chem.
Soc. 2000, 122, 11248. (b) Valls, N.; Lo´pez-Canet, M.; Vallribera, M.;
Bonjoch, J. Chem. Eur. J. 2001, 7, 3446.
(6) (a) Sandler, B.; Murakami, M.; Clardy, J. J. Am. Chem. Soc. 1998, 120,
595. (b) Kodani, S.; Ishida, K.; Murakami, M. J. Nat. Prod. 1998, 61, 1040.
(c) Ishida, K.; Okita, Y.; Matsuda, H.; Okino, T.; Murakami, M.
Tetrahedron 1999, 55, 10971. (d) Ishida, R.; Murakami, M. J. Org. Chem.
2000, 65, 5898. (e) Ishida, K.; Kato, T.; Murakami, M.; Watanabe, M.;
Watanabe, M. F. Tetrahedron 2000, 56, 8643.
(7) (a) Fujii, K.; Sivonen, K.; Adachi, K.; Nagachi, K.; Shimizo, Y.; Sano, H.;
Hirayama, K.; Suzuki, M.; Harada, K. Tetrahedron Lett. 1997, 38, 5529.
(b) Banker, R.; Carmeli, S. Tetrahedron 1999, 55, 10835.
(8) (a) Aeruginosin 298A: Ohshima, T.; Gnanadesikan, V.; Shibughushi, T.;
Fukuta, Y.; Nemoto, T.; Shibasaki, M. J. Am. Chem. Soc. 2003, 125, 11206.
(b) Aeruginosin E1461: Valls, M.; Vallribera, M.; Carmeli, S.; Bonjoch,
J. Org. Lett. 2003, 5, 447. (c) Valls, N.; Vallribera, M.; Lo´pez-Canet, M.;
Bonjoch, J. J. Org. Chem. 2002, 67, 4945.
(9) Hanessian, S.; Margarita, R.; Hall, A.; Johnstone, S.; Tremblay, M.; Parlanti,
L. J. Am. Chem. Soc. 2002, 124, 13342.
(10) Carroll, A. R.; Pierens, G.; Fechner, G.; de Almeidaleona, P.; Ngo, A.;
Simpson, M.; Hooper, J. N. A.; Bo¨strom, S. L.; Musil, D.; Quinn, R. J. J.
Am. Chem. Soc. 2002, 124, 13340.
9
6064
J. AM. CHEM. SOC. 2004, 126, 6064-6071
10.1021/ja030669g CCC: $27.50 © 2004 American Chemical Society

Synthesis of the Marine Natural Product Oscillarin
A R T I C L E S
which is linked as an amide in all known aeruginosins is
replaced by a 1-amino-2-(N-amidino-∆3-pyrrolinyl)ethyl group
(Figure 1). In this regard, dysinosin A 19,10 and oscillarin 2 are
among a handful of known natural products that contain
hydroxylated octahydroindole cores with a common arginine
mimetic P1 recognition site for serine protease activity, even
though their respective source microorganisms live and flourish
oceans apart.13 Suomilide,7a a glycosidic azabicyclo[3.3.1]
variant of oscillarin and dysinosin A isolated from Nodularia
spumigena HKW, contains the same 1-amino-2-(N-amidino-∆3-
pyrrolinyl)ethyl group as an amide linkage.
The structure of oscillarin 2 was originally proposed with a
six-membered cyclic guanidine subunit, as shown in expression
2a,17 rather than the actual ∆3-pyrroline counterpart. To the best
of our knowledge, this discrepancy has never been corrected
by the same group.14,17
Previous syntheses of the L-Choi subunit in the aeruginosins
have relied on a clever utilization of L-tyrosine as a chiral
progenitor,18,19 as well as on other methods.8a,20 Although the
methodology that was developed for the corresponding unit in
dysinosin A9 could be used to access L-Choi, by performing a
selective deoxygenation at C5, we opted for a new approach
that exploits a seldom used N-acyloxyiminium ion aza-Prins
cyclization,21 which leads directly to 6-halo-octahydroindole
2-carboxylic acids.
Figure 1. Structures of aeruginosin 298A, dysinosin A 1, oscillarin (D-
Pla-D-Phe-L-Choi-Adc) 2 and an originally presumed structure 2a.
thrombin11 and Factor VIIa, which are essential components in
the blood coagulation cascade,12 dysinosin A differs from all
aeruginosins by the presence of a 5S,6S-diol in the octahydro-
indole-2-carboxylic acid subunit, a unique 1-(N-amidino-∆3-
pyrrolino)-ethyl side chain in place of an acyclic guanidine, and
an O-sulfated D-glyceric acid instead of aromatic counterparts
(Figure 1).13
Results
Enolate alkylation of the dianion derived from dimethyl
N-Boc-L-glutamate relying on 1,3-induction22 afforded enan-
tiopure 4 in excellent yield (Scheme 1). Cyclization to the
corresponding pyroglutamate 5, protection, and partial reduction
of the lactam carbonyl gave the hemiaminal acetate 6. Subse-
quent treatment with tin tetrabromide23 effected cyclization
within 5 min at -78 °C to give the desired aza[4.3.0]-bicyclic
motif 7 in 78% yield. The stereochemistry at C6 was suggested
from detailed 1H NMR analysis and was subsequently confirmed
We now report on the first total synthesis and structural
confirmation of oscillarin 2, a hitherto little known new member
of the aeruginosin family, also designated as D-Pla-D-Phe-L-
Choi-Adc. Oscillarin was isolated from the algal cultures of
Oscillatoria agardhii (strain B2 83), at the Institute of Plant
Physiology of the University of Go¨ttingen. Its structure and
absolute configuration were proposed on the basis of NMR data
and a partially resolved cocrystal complex with trypsin.14 Two
biogenetic and structural features concerning oscillarin are
worthy of note. First, cultures of Oscillatoria agardhii obtained
in Japan15 are known to produce aeruginosin 205A, which was
originally proposed to contain a 6R-chloro substitutent in the
octahydroindole-2-carboxylic acid subunit. However, this struc-
tural assignment has been questioned on the basis of definitive
synthesis of the proposed 2-carboxy-6-chlorooctahydro-
indole.16Second, the acyclic 1-amino-4-guanidino butyl subunit
(17) Konetschny-Rapp, S.; Krell, H.-W.; Martin, U. PCT WO 96/11941; Chem.
Abst. 1996, 124, 315175.
(18) (a) Wipf, P.; Kim, Y.; Goustein, D. M. J. Am. Chem. Soc. 1995, 117, 1106.
(b) Wipf, P.; Maresko, D. A. Tetrahedron Lett. 2000, 41, 4723. Wipf, P.;
Li, W. J. Org. Chem. 1999, 64, 4576.
(19) (a) Bonjoch, J.; Catena, J.; Isabal, E.; Lo´pez-Canet, M.; Valls, N.
Tetrahedron: Asymmetry 1996, 7, 1899. (b) Bonjoch, J.; Catena, J.; Valls,
N. J. Org. Chem. 1996, 61, 7106.
(20) (a) Belvisi, L.; Colombo, L.; DiGiacomo, M.; Manzoni, L.; Vodopivec,
B.; Scolastico, C. Tetrahedron 2001, 57, 6463. (b) Toyooka, N.; Okumura,
M.; Himiyama, T.; Nakazawa, A.; Nemoto, H. Synlett 2003, 55.
(21) For excellent general reviews on N-acyliminium ion chemistry, see (a)
Speckamp, W. N.; Moolenaar, M. J. Tetrahedron 2000, 56, 3817. (b)
Hiemstra, H.; Speckamp, W. N. In ComprehensiVe Organic Synthesis; Trost,
B. M., Fleming, I., Heathcock, C. H., Eds.; Pergamon: New York, 1991;
2, 1047. (c) Speckamp, W. N.; Hiemstra, H. Tetrahedron 1985, 41, 4367.
(d) Speckamp, W. N. Rec. TraV. Chim. Pays-Bas 1981, 100, 345. The term
“N-acyloxyiminium ion aza-Prins cyclization” is used in analogy to the
Prins or oxonia-Prins cyclizations (see refs 23, 39, 40). In the present
examples, an unactivated terminal olefin is engaged in intramolecular C-C
bond formation with an endocyclic N-acyloxyiminium ion via a cationic
intermediate that is trapped by halide ion. For the use of aza-Prins in a
different context, see Nicolaou, K.; Snyder, S. A.; Montagnon, T.;
Vassilikogiannakis, G. Angew Chem. Ind. Ed. 2002, 41, 1668 as cited on
p 1680 in relation to the “Diels-Alder/aza-Prins cascade” during the total
synthesis of methyl homosecodaphniphylate, Ruggeri, R. B.; Hansen, M.
M.; Heathcock C. H. J. Am. Chem. Soc. 1988, 110, 8734. Strictly speaking,
the term azonia-Prins rather than aza-Prins should be used (see IUPAC
rules of nomenclature, 1993; Heterocyclic Systems Rule B-6. Cationic
Hetero Atoms). In the present case, however, the mention of N-acyloxy-
iminium ion already implies the involvement of a charged nitrogen.
(22) Hanessian, S.; Margarita, R. Tetrahedron Lett. 1998, 39, 5887.
(23) For the use of SnBr4 in a Prins cyclization of an oxocarbenium ion, see
Marumoto, S.; Saber, J. J.; Vitale, J. P.; Rychnovsky, S. D. Org. Lett. 2002,
4, 3919.
(11) For selected reviews on thrombin inhibitors, see (a) Sanderson, P. E. J.;
Nayler-Olson, A. M. Curr. Med. Chem. 1998, 5, 289. (b) Ripka, W. E.
Curr. Opin. Chem. Biol. 1997, 1, 242. (c) Pfau, R. Curr. Opin. Drug
DiscoVery 2003, 6, 437.
(12) (a) Steinmetz, T.; Hauptman, J.; Stu¨rzbecher, J. Exp. Opin. InVest Drugs
2000, 10, 845. (b) Lolman, R. W.; Marder, V. J.; Salzman, E.; Hirsch, W.
In Hemoslasis and Thrombosis: Basic Principles and Clinical Practice,
3rd ed.; Colman, R. W., Hirsch, W., Marder, V. J., Salzman, E. W., Eds.;
Lippinott, J. B: Philadelphia, PA, 1994; p 3.
(13) A chlorinated dysinosin-like product has been recently isolated from Dysidea
sp. sponges: Goetz, G. H.; Harrigan, G. G.; Likos, J. J.; Kasten, T. P.
PCT WO 03/051831; Chem. Abst. 2003, 139, 47155.
(14) Engh, R.; Konetschny-Rapp, S.; Krell, H.-W.; Martin, U.; Tsaklakidis, C.
PCT WO97/21725; Chem. Abst. 1997, 127, 122002.
(15) Shin, H. J.; Matsuda, M.; Murakami, M.; Yamaguchi, K. J. Org. Chem.
1997, 62, 1810.
(16) Valls, N.; Vallribera, M.; Font-Bard´ıa; Solans, X.; Bonjoch, J. Tetra-
hedron: Asymmetry 2003, 14, 1241.
9
J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004 6065

A R T I C L E S
Hanessian et al.
Scheme 1 a
a a: LiHMDS, THF, -78 °C, then 3-butenol triflate, 85%; b: TFA, CH2Cl2 then toluene reflux, 92%; c: Boc2O, Et3N, DMAP, CH2Cl2, 90%; d: LiBHEt3,
THF, -78 °C; e: Ac2O, Et3N, DMAP, CH2Cl2, 91% (two steps); f: SnBr4, CH2Cl2, -78 °C, 78%; g: Bu4NOAc, toluene, 40-50 °C, 78%; h: TFA, CH2Cl2,
then EDC, HOBt, 9, CH2Cl2, 91% (two steps); i: NaOMe/MeOH; j: MOMCl, iPr2NEt, CH2Cl2, 80% (two steps); k: LiOH, H2O/THF; l: EDC, HOBt, Et3N,
11, 86% (two steps); m: aq HCl/THF 6N, 70%.
Scheme 2 a
by a single-crystal X-ray structure of an advanced intermediate
(Scheme 1). Displacement of the bromide with tetra-n-butyl-
ammonium acetate in toluene at 50 °C afforded the desired
L-Choi subunit 8 in excellent overall yield (36% over nine steps
from L-glutamic acid). Amide formation with a preformed
MOM-protected D-phenyllactyl-D-phenylalanine 9,24 followed
by deacetylation, and protection gave the corresponding bis-
MOM ether 10. A single-crystal X-ray structure analysis
confirmed the absolute configuration of this advanced interme-
diate, clearly showing the axial orientation of the O-MOM
group, the expected cis-ring junction, and confirming the
structure of the originally proposed N-acyloxyiminium aza-Prins
cyclization product. Ester cleavage, followed by coupling with
2-(N,N′-bis-Boc-N-amidino-∆3-pyrrolino)-ethylamine 119 gave
the penultimate precursor 12. Deprotection under acidic condi-
tions and purification by preparative reverse phase HPLC,
afforded oscillarin 2 as the hydrochloride salt, showing spectral
data identical to that reported for the natural product.14,17
Before we became aware of the discrepancy in the structure
of the guanidine-containing motif in oscillarin, we had embarked
on a synthesis route leading to the originally proposed six-
membered guanidine variant (Figure 1, 2a).17 The synthesis
started with the readily available acetal 13, which was subjected
to a Wittig reaction to give 14 (Scheme 2). Mesylation of the
primary alcohol and displacement with azide gave 15, which
was deprotected and the resulting diol was subjected to a double
Mitsunobu reaction with tri-N-Boc guanidine25 leading to 16.
Reduction of the azide group under Staudinger conditions26
afforded a regioisomeric mixture of cyclic bis-N-Boc guanidine
a a: 3-triphenylphosphoniumpropanol bromide, BuLi, THF, -20 °C,
i
TMSCl, then 13; b: TBAF, THF, 75%; c: MsCl, Pr2NEt, CH2Cl2, then
NaN3, DMF, 50 °C, 85%; d: AcOH (80% in water); e: DEAD, PPh3,CH2Cl2,
tris(Boc)guanidine, 50% (two steps); f: PPh3, H2O, then AcOH, then Dowex
1 × 8-50 (Cl-), 60%; g: LiOH, H2O/THF; h: EDC, HOBt, Et3N, 17, 70%
(two steps); i: aq HCl/THF 4N, 60%.
derivative 17. With the precursor ester 10 already in hand, we
proceeded with its coupling to 17 and deprotection as described
for 12 to afford the initially presumed “oscillarin”, 2a. Not
surprisingly, it was found to have no activity when tested against
(24) See Supporting Information for details.
(25) (a) Feichtinger, K.; Sings, H. L.; Baker, T. J.; Matthews, K.; Goodman,
M. J. Org. Chem. 1998, 63, 8432. (b) McManus, J. C.; Genski, T.; Carey,
J. S.; Taylor, R. J. K. Synlett 2003, 369.
(26) Vaultier, M.; Knouzi, N.; Carrie, R. Tetrahedron Lett. 1983, 24, 763.
9
6066 J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004

Synthesis of the Marine Natural Product Oscillarin
A R T I C L E S
Figure 2. Left: Ribbon representation of thrombin-oscillarin complex at 2.0 Å resolution. Right: Connolly surface representation of thrombin-oscillarin
complex.
R-thrombin.27 Although the guanidine subunit in 2a is isomeric
with that in 2, it clearly does not fulfill the requisite alignment
as an arginine mimetic within the S1 subsite of thrombin.
We were successful in obtaining a ternary complex of 2 with
the R-thrombin-hirugen complex,28 that was amenable to an
X-ray structure elucidation at a resolution of 2.0 Å (Figure 2).
Oscillarin forms an extended H-bonding pattern in the complex
similar to the previously observed complexes reported for
aeruginosin 298A3,9 and dysinosin.10 The P1 arginine mimetic
forms a salt bridge with the carboxylate of Asp 229 at the S1
site. The amide NH from the octahydroindole carboxamide
forms a H-bond with the backbone carbonyl oxygen of Ser 256.
The Phe amide carbonyl and the amide NH are H-bonded to
the amidic subunit of Gly 258, in an antiparallel â-strand
fashion. The terminal hydroxyl forms a H-bond with the
backbone amide nitrogen of Gly 260.
The octahydroindole unit delineates the P2/S2 interaction by
stacking toward Trp 86 and Tyr 83. No interactions were
observed with the 6-hydroxyl group of the Choi unit. The phenyl
group of the P3 unit points toward the S3 hydrophobic site of
the enzyme. No interactions were observed between the terminal
phenyl moiety and the enzyme. This might explain why the
side chain was poorly defined in the electron density. The crystal
structure of oscillarin-bound thrombin is only the third for a
member of the aeruginosin family.3,10 In this context, oscillarin
is the most potent inhibitor of thrombin among the aeruginosins,
with an IC50 ) 28 nM.
tronic effects,30 and allylic strain,31 particularly when dealing
with cyclic systems.
There are several examples that involve intramolecular attack
of a nucleophilic olefinic tether onto an endocyclic iminium
cation for the construction of azabicyclic ring systems.21
However, with few exceptions, they involve ω-unsaturated
nucleophilic carbon chains tethered to nitrogen that undergo
endo aza-Prins N-acyliminium ion type cyclizations (Figure 3A).
Such cyclizations lead to azabicycles in which the nitrogen atom
is at the junction of the two rings as in pyrrolizidinones,
indolizidinones, quinolizidinones, and related compounds. The
intermediate carbenium ions are usually solvolytically trapped
in the medium by a Lewis acid or by loss of a hydrogen.32 As
already pointed out by Fisher and Overman in 1990,33 there
are fewer examples of intramolecular cyclizations in which the
olefinic tether is attached to a distal carbon in an endocyclic
iminium ion ring34 (Figure 3B). Alternative pathways for the
construction of monocyclic and bicyclic N-containing com-
pounds utilizing iminium ion intermediates rely on cationic
azonia-Cope rearrangements21,35,36 or activation of the nucleo-
philic tether via vinylsilane or allylsilane chemistry.37 The
Overman and Hart groups in particular have exploited azonia-
Cope and N-acyliminium ion cyclization chemistry in the
synthesis of complex natural products.38
(29) (a) Overman, L. E. Acc. Chem. Res. 1992, 25, 352. (b) Overman, L. E.;
Ricca, D. J. In ComprehensiVe Organic Synthesis; Trost, B. M., Ed.;
Pergamon: London, 1991; 2, 1007. (c) Franklin, A. S.; Overman, L. E.;
Chem. ReV. 1996, 96, 505.
Intramolecular N-Acyloxyiminium Aza-Prins Cyclizations.
The chemistry of N-acyloxy and related N-substituted iminium
ions has a rich legacy in the realm of nitrogen-containing
heterocyclic compounds.21 Indeed, the scholarly contributions
of the Speckamp, Overman, and Hart groups over the past two
decades have laid a strong foundation for exploiting the synthetic
applications of N-acyliminium ions, particularly for alkaloids
and other complex nitrogen-containing natural products.29,30 The
mechanistic basis for reactions involving N-acyliminium ions
has also invoked fundamental principles involving stereoelec-
(30) See, for example, (a) Stevens, R. V.; Lee, A. W. M. J. Am. Chem. Soc.
1979, 101, 7032. (b) Wenkert, E.; Chang, C.-J.; Chawla, H. P. S.; Cochran,
D. W.; Hagaman, E. W.; King, J. C.; Orito, K. J. Am. Chem. Soc. 1976,
98, 3645. (c) Ziegler, F. E.; Spitzner, E. B. J. Am. Chem. Soc. 1973, 95,
7146. (d) Overman, L. E.; Freerks, R. F. J. Org. Chem. 1981, 46, 2833.
(e) Overman, L. E.; Fukaya, C. J. Am. Chem. Soc. 1980, 102, 1454. (f)
Heathcock, C. H.; Kleinman, E. F.; Binkley, E. S. J. Am. Chem. Soc. 1982,
104, 1054.
(31) See, for example, (a) Hart, D. J. J. Am. Chem. Soc. 1980, 102, 397. (b)
Overman, L. E.; Lesuisse, D.; Hashimoto, M. J. Am. Chem. Soc. 1983,
105, 5373. (c) Overman, L. E.; Fukaya, C. J. Am. Chem. Soc. 1980, 102,
1454. (d) Chow, Y. L.; Colo´n, C. J.; Tam, J. N. S. Can. J. Chem. 1968,
46, 2821. (e) Fraser, R. R.; Grindley, T. B. Tetrahedron Lett. 1974, 416.
(f) Scott, J. W.; Durham, L. J.; DeJongh, H. A. P.; Burkhardt, V.; Johnson,
W. S. Tetrahedron Lett. 1967, 2381. (g) Quick, J.; Mondello, C.; Humora,
M.; Brennan, T. J. Org. Chem. 1978, 43, 2705. (h) Brown, J. D.; Foley,
M. A.; Comins, D. L. J. Am. Chem. Soc. 1988, 110, 7445. (i) Beak, P.;
Zajdel, W. J. J. Am. Chem. Soc. 1984, 106, 1010. For reviews, see
Hoffmann, R. W. Chem. ReV. 1989, 89, 1841. Johnson, F. Chem. ReV.
1968, 68, 375.
(27) Nilsson, T.; Sjoling-Ericksson, A.; Deinum, J. J. Enzyme Inhibition 1998,
13, 11.
(28) Skrypczak-Jankum, E.; Carperos, V. E.; Ravichandran, K. G.; Tulinsky,
A.; Westbrook, M.; Maraganore, J. M. J. Mol. Biol 1991, 221, 1379.
9
J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004 6067

A R T I C L E S
Hanessian et al.
Scheme 3. Variation of the Lewis Acid and N-carbamoyl Group in
The Formation of the Octahydroindole Cores
Figure 3. N-Tethered (A) and C-tethered (B) intramolecular aza-Prins-N-
acyliminium ion type cyclizations.
There are ample examples of Lewis acid or protic acid
promoted intramolecular Prins reactions39 proceeding through
oxocarbenium ions for the formation of halo-oxacycles with
excellent stereocontrol.23,40 Seminal contributions by Specka-
mp,41 Overman,42 and more recently Rychnovsky23,43 and their
respective groups have demonstrated the utility of Prins or so-
called oxonia-Cope variants in natural product synthesis.
In contrast, there are no examples of analogous Lewis acid
promoted N-acyloxyiminium aza-Prins halocyclizations, whereby
an ω-olefinic carbon tether attached as an integral part of the
ring undergoes cyclization onto an incipient endocyclic N-
acyliminium ion with incorporation of a halogen in a 1-azabi-
cyclic ring system related to the octahydroindoles or decahy-
droquinolines (Figure 3B, X ) Cl, Br).44
As already shown in the synthesis of oscillarin, the method
is a novel way to prepare 6-halo-octahydroindole-2-carboxylic
acids.16 The alternative solvolytic method with formic acid34
would not be compatible because of the presence of the acid-
labile N-Boc group.
(32) For the synthesis of Lewis acid catalyzed halogenated azacyclic systems,
see (a) Udding, J. H.; Tuijp, C. J. M.; Hiemstra, H.; Speckamp, W. N. J.
Chem. Soc., Perkin Trans. 2 1992, 857. (b) Rutjes, F. P. J. T.; Hiemstra,
H.; Pirrung, F. O. H.; Speckamp, W. N. Tetrahedron 1993, 49, 10027. (c)
Li, W.; Hanau, C. E.; d′Avignon, A.; Moeller, K. D. J. Org. Chem. 1995,
60, 8155. For solvolytic reactions with formic acid, see (d) Demailly, G.;
Solladie´, G. J. Org. Chem. 1981, 36, 3102. (e) Hart, D. J.; Kanai, K.-I. J.
Am. Chem. Soc. 1983, 105, 1255. (f) Hart, D. J.; Hong, W. P. J. Org.
Chem. 1985, 50, 3670. (f) Cannizzo, L. F.; Grubbs, R. H. J. Org. Chem.
1985, 50, 2316. (g) Brosius, A. D.; Overman, L. E. J. Org. Chem. 1997,
62, 440. (h) Frank, K. E.; Aube´, J. Tetrahedron Lett. 1998, 39, 7239. (i)
Schoemaker, H. E.; Dijkink, J.; Speckamp, W. N. Tetrahedron 1978, 34,
163. (j) Dijkink, J.; Speckamp, W. N. Tetrahedron 1978, 34, 173. For ene-
type reactions, see (k) Ruggeri, R. B.; Hansen, H. M.; Heathcock, C. H. J.
Am. Chem. Soc. 1988, 110, 8734. (l) Tanner, D.; Hagberg, L. Tetrahedron
1998, 54, 7907. For an example of ester participation, see (m) Endoma,
M. A.; Butora, G.; Claeboe, C. D.; Hudlicky, T.; Abboud, K. A. Tetrahedron
Lett. 1997, 38, 8833.
(33) Fisher, M. J.; Overman, L. E. J. Org. Chem. 1990, 55, 1447.
(34) For examples involving solvolysis with formic acid, see (a) Oostveen, A.
R. C.; de Boer, J. J. J.; Speckamp, W. N. Heterocycles 1977, 7, 171. (b)
Schoemaker, H. E.; Kruk, C.; Speckamp, W. N. Tetrahedron Lett. 1979,
2437. For Lewis acids and alkene tethers, see (c) Melching, K. H.; Hiemstra,
H.; Klaver, W. J.; Speckamp, W. N. Tetrahedron Lett. 1986, 27, 4799. (d)
Esch, P. M.; Boska, I. M.; Hiemstra, H.; de Boer, R. F.; Speckamp, W. N.
Tetrahedron, 1991, 47, 4039. For solvolysis of allenic or acetylenic tethers,
see (e) Beyersbergen van Henegouwen, W. G.; Hiemstra, H. J. Org. Chem.
1997, 62, 8862. For applications to â-lactams, see (f) Metais, E.; Overman,
L. E.; Rodriguez, M. I.; Stearns, B. A. J. Org. Chem. 1997, 62, 9210.
(35) (a) Blechert, S. Synthesis 1989, 71. (b) Heimgartner, H.; Hansen, H.-J.;
Schmid, H. In Iminium Salts in Organic Chemistry, Part 2; Bo¨hme, H.,
Viche, J. J., Eds.; Wiley: New York, 1979; p 655-732. (c) Winterfeldt,
E. Curr. Topics Org. Chem. 1971, 16, 1971.
The Scope of the N-Acyloxyiminium Aza-Prins Halocar-
bocyclization. Having demonstrated the utility of an intramo-
lecular aza-Prins carbocyclization as a novel stereocontrolled
route to the 6-halo(or hydroxy)-2-carboxy-octahydroindole core
of oscillarin and related aeruginosins, we studied the scope and
limitations of related reactions.
First, we wished to study the influence of the carbamoyl group
as well as the nature of the Lewis acid on the stereoselectivity
of the halocarbocyclization. We were intrigued to find that in
contrast to the two Lewis acids used, the nature of the
N-carbamoyl group had little influence on the yield or stereo-
selection (Scheme 3). The single-crystal structure of the tert-
butyl ester analogue24 of the major product 18 is shown as an
Ortep diagram (Scheme 3). Clearly, SnBr4 was superior to SnCl4
(36) See, for example, (a) Overman, L. E.; Kakimoto, M. A. J. Am. Chem. Soc.
1979, 101, 1310. (b) Overman, L. E.; Jacobsen, E. J.; Doedens, R. J. J.
Org. Chem. 1983, 48, 3393. (c) Hart, D. J.; Yang, T. K. J. Org. Chem.
1985, 50, 235. (d) Ent, H.; de Koning, H.; Speckamp, W. N. J. Org. Chem.
1986, 51, 1687.
(37) For selected references, see (a) Dobbs, A. P.; Guesne´, S. J. J.; Hursthouse,
M. B.; Coles, S. J. Synlett 2003, 1740. (b) Sun, P.; Sun, C.; Weinreb, S. N.
J. Org. Chem. 2002, 67, 4337. (c) Wo¨lfling, J.; Frank, E.; Schneider, G.;
Bes, M. T.; Tietze, L. F. Synlett 1998, 1205. (d) Kercher, T.; Livinghouse,
T. J. Am. Chem. Soc. 1996, 118, 4200. (e) Gelas-Mialhe, Y.; Gramain,
J.-C.; Hajouji, H.; Remuson, R. Heterocycles 1992, 34, 37. (f) Tietze, L.
F.; Wu¨nsch, J. R. Angew. Chem., Int. Ed. Engl. 1991, 30, 1697. (g)
Overman, L. E.; Sharp, M. J. Tetrahedron Lett. 1988, 29, 901. (h) Kano,
S.; Yokomatsu, T.; Yuasav, Y.; Shibuya, S. Heterocycles 1986, 24, 621.
(i) Shono, T.; Matsumura, Y.; Uchida, K.; Kobayashi, H. J. Org. Chem.
1985, 50, 3243. (j) Hiemstra, H.; Fortgens, H. P.; Speckamp, W. N.
Tetrahedron Lett. 1985, 26, 3155. (k) For reviews see Fleming, I.; Barbero,
A.; Walter, D. Chem. ReV. 1997, 97, 2063. Langkopf, E.; Schinzer, D.
Chem. ReV. 1995, 95, 1375.
(40) For recent examples of halooxacycles via Prins cyclizations, see (a) Hart,
D. J.; Bennet, C. E. Org. Lett. 2003, 5, 1499. (b) Jaber, J. J.; Mitsui, K.;
Rychnovsky, S. D. J. Org. Chem. 2001, 66, 4679. (c) Al-Mutairi, E. H.;
Crosby, S. R.; Darzi, J.; Harding, J. R.; Hughes, R. A.; King, C. D.;
Simpson, T. J.; Smith, R. W.; Wills, C. L. J. Chem. Soc., Chem. Commun.
2001, 835. (d) Cloninger, M. J.; Overman, L. E. J. Am. Chem. Soc. 1999,
121, 1092. (e) Yang, X.-F.; Li, C. J. Tetrahedron Lett. 2000, 41, 1321. (f)
Marko´, I. E.; Chelle´, F. Tetrahedron Lett. 1997, 38, 2895. (g) Perron-Sierra,
F.; Promo, M. A.; Martin, V. A.; Albizati, K. F. J. Org. Chem. 1991, 56,
6188. (h) Coppi, L.; Ricci, A.; Taddei, M. J. Org. Chem. 1988, 53, 911.
(i) Winstead, R. C.; Simpson, T. H.; Lock, G. A.; Schiawelli, M. D.;
Thompson, D. M. J. Org. Chem. 1986, 51, 257 and references therein.
(41) Lolkema, L. D. M.; Hiemstra, H.; Semeyn, C.; Speckamp, W. N.
Tetrahedron 1994, 50, 7115.
(38) For early examples, see: (a) Overman, L. E.; Mendelson, L. T. J. Am.
Chem. Soc. 1981, 103, 5579. See also ref 32c,e.
(42) Blumenkopf, T. A.; Overman, L. E. Chem. ReV. 1986, 86, 857.
(43) Jaber, J. J.; Mitsui, K.; Rychnovsky, S. D. J. Org. Chem. 2001, 66, 4679.
(44) For related examples, see Wo¨lfling, J.; Frank, E.; Schneider, G.; Bes, M.
T.; Tietze, L. Synlett 1998, 1205. See also refs 34d, 34f, 37i.
(39) Snider, B. B. In ComprehensiVe Organic Chemistry; Trost, B. M., Fleming,
I., Heathcock, C. H., Eds.; Pergamon: New York, 1991; Z, p 523.
9
6068 J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004

Synthesis of the Marine Natural Product Oscillarin
A R T I C L E S
Scheme 4 a
Scheme 6 a
aa: LiHMDS, THF, -78 °C, then 3-butenol triflate, 70% (1:1.8 anti/
syn); b: LiHBEt3, THF, -78 °C; c: Ac2O, Et3N, DMAP, CH2Cl2, 87%
(two steps); d: SnCl4, CH2Cl2, -78 °C, 64%.
Scheme 5 a
a a: NaBH4, EtOH, 87%; b: CBr4, PPh3, cyclohexene, MeCN; c: Bu3SnH,
AIBN, toluene, 80 °C; d: Boc2O, Et3N, DMAP, CH2Cl2, 35% (three steps);
e: LiBHEt3, THF, -78 °C; f: Ac2O, Et3N, DMAP,CH2Cl2, (80% two steps);
g: SnCl4, CH2Cl2, -78 °C, 60%.
to promote the cyclization. Interestingly, a small amount (4-
5%) of â-acetoxy group was incorporated instead of Cl at the
six position in the case of SnCl4.24 We next studied the influence
of substrate concentration and the nature of the Lewis acid on
the carbocylization. Using concentrations ranging from 0.01 to
0.2 M of 6 did not appreciably alter the yield of the major
product or the accompanying formation of enecarbamate 19
(15-20%). The proportion of the latter increased substantially
when SbCl5, TMSCl, or BF3‚Et2O were used. While there was
no reaction in the presence of ZnCl2, a complex mixture resulted
in the case of TiCl4. The halocarbocyclizations with SnCl4 or
SnBr4 were essentially complete in a few minutes at -78 °C.34d
The analogous cationic carbocyclization was also successfully
accomplished with the 4-epimeric series (Scheme 4). Thus,
enolate alkylation of N-Boc-methyl-L-pyroglutamate 20 with
1-butenyl triflate gave a mixture of syn- and anti-products which
could be readily separated into their 4-epimeric analogues 5
and 21. Transformation of 21 to the corresponding N-Boc
carbinolamine acetate 22 as previously described for 6, and
carbocyclization with SnCl4 afforded the 6-chloro-octahydro-
indole analogue 23 in 64% yield. To eliminate any involvement
of the electron-rich methoxylcarbonyl (ester) group in coordi-
natively participating in the delivery of halide ion, we carried
out the carbocylization with the corresponding C-methyl lactam
25 via its activated carbinolamine derivative 26 (Scheme 5).
The 6-chloro analogue 27 was formed in 60% yield, ac-
companied by 12% of enecarbamate.
aa: LiHMDS, THF, -78 °C, then 3-butenol triflate, 90% 1.3:1(anti/
syn); b: TFA, CH2Cl2; c: Boc2O, Et3N, DMAP, CH2Cl2, 78% (two steps);
d: LiBHEt3, THF, -78 °C; e: Ac2O, Et3N, DMAP, CH2Cl2; f: SnX4, CH2Cl2,
-78 °C.
and anti-products 30 and 31, respectively. Conversion to the
acetoxy derivative 32, followed by cyclization in the presence
of SnCl4, gave the corresponding bicyclic 6-S-chloro analogue
34 and 6R-epimer 35 in 63% and 8% yields, respectively,
accompanied by ∼10-15% of the enecarbamate 40. Their
structures and absolute configuration were ascertained by single-
crystal X-ray analysis (see below). With SnBr4, only the major
cyclization product 36 was isolated with the enecarbamate as a
byproduct (∼10%). In contrast, cyclization of the trans-
intermediate 33 under the same conditions led to only a modest
yield of the corresponding chloride 37 or bromide 39 (Scheme
6). A minor epichloro isomer 38 was also isolated and
characterized by single-crystal X-ray analysis. Considerable
amounts of the enecarbamate 40 (30-35%) were isolated in
these reactions.
Extension of the aza-Prins halocarbocyclization to generate
bicyclo 1-aza[3.3.0] or bicyclo 1-aza-[3.4.0] systems from chain-
shortened, that is, 4-(1-propenyl), or chain-elongated, that is,
4-(1-pentenyl)-analogues of 6, resulted in the formation of
enecarbamates with no trace of cyclization. Variations in the
We next studied the N-acyloxyiminium aza-Prins halocar-
bocyclizations of C-5 epimeric 6-acetoxy-L-pipecolic acid
analogues, readily available from 6-oxo-N-Boc L-pipecolic
methyl ester 28, via enolate alkylation and separation of syn-
9
J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004 6069

A R T I C L E S
Hanessian et al.
chairlike intermediate in which A1,2 strain is minimized.31 The
halide ion would attack the incipient carbocation preferentially
from an equatorial trajectory to give the major product 7 (Figure
4A). The alternative reactive conformer involving a synclinal
attack (Figure 4B) in which A1,2 strain can also be avoided
would be less favored. Evidently, the minor â-acetoxy byproduct
must arise from attack by the released acetate group. The higher
yield when SnBr4 was used may reflect the better nucleophilicity
nature of bromide versus chloride ion. The same rationales apply
to the formation of the major product 23 in the C4-epimeric
series (Scheme 4).
The results in the perhydroquinoline series derived from the
C2-C5 cis-isomer 32 follow the same line of reasoning (Scheme
5). An antiperiplanar olefin approach leads to the 1-aza cis-
decalinoid carbenium ion which is trapped by halide ion in an
equatorial mode (Figure 5A). The single-crystal X-ray structure
of the major product 34 shows a chair-chair conformation with
a pseudoaxial methoxycarbonyl (ester) group in accord with
minimization of A1,2 strain. The crystal structure of the minor
epimer 35 (not shown) also reveals a chair-chair conformation,
a pseudoaxial methoxycarbonyl (ester) group, and an axial
chlorine substituent. Thus, in this series, favorable stereoelec-
tronic effects and minimization of A1,2 strain are cooperative
features. Although none of the epi-6-bromo isomer could be
found in the case of SnBr4, the difference in yield was not as
important as in the octahydroindole series.
Figure 4. Proposed reactive intermediates in the N-acyloxyiminiumion aza-
Prins halocarbocyclization to octahydroindoles.
stereochemical outcome of related cyclizations depend on the
length of the carbon tether.36
Mechanistic Rationales. As previously mentioned, stereo-
electronic control30 during the intramolecular N-acyloxyiminium
ion cyclizations plays an important role in deciding preferred
trajectories of approach of the tethered alkenic nucleophile. It
is generally agreed that maintaining maximum orbital overlap
of the alkenyl π systems with the developing lone pair on the
nitrogen in five- and six-membered N-acyloxyiminium ions is
an important requirement.30a,45 Added to this crucial control
element are nonbonded interactions that could affect the
conformations of reactive intermediates when bicyclic systems
are involved.29b Thus, excluding intramolecular delivery of
halide ions from carbonyl-coordinated species, a plausible
mechanism for the stereocontrolled formation of 6-halo-octahy-
droindoles such as 7 and 23 (Schemes 1, 4, 5) is to assume an
antiperiplanar alignment of the olefin and the iminium ion in a
In sharp contrast, the C5-epimeric series starting with 33 led
to only modest yields of cyclized products regardless of the
nature of the halide in the Lewis acid. The crystal structure of
the minor cyclization product 38 depicts a chair-chair confor-
mation in which the methoxycarbonyl (ester) group now
occupies a pseudoequatorial orientation (Figure 5B). Assuming
the prevalence of an iminium ion conformation in which the
methoxycarbonyl (ester) group is indeed pseudoequatorial, we
can project a favorable antiperiplanar approach leading to a
Figure 5. Proposed reactive intermediates in the N-acyloxyiminium ion aza-Prins halocarbocyclizations to perhydroquinolines.
9
6070 J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004

Synthesis of the Marine Natural Product Oscillarin
A R T I C L E S
carbocation that is trapped by halide attack in an equatorial or
axial mode to give 37 or 38, respectively. Despite such a
favorable stereoelectronic alignment, the energetic penalty for
A1,2 strain in the reactive conformation (and product) is
presumably too high. Thus, elimination to the enecarbamate
competes with cyclization. The alternative conformer in which
A1,2 strain could be avoided (Figure 5C) must adopt an
unfavorable synclinal approach of the olefin toward the iminium
ion. Thus, both pathways from 33 are disfavored to different
extents, hence the higher propensity for elimination to enecar-
bamate compared to 32. No dihydrooxazinones33 were formed
in these cyclizations and the stereochemistry of the ring junction
was invariably cis.
developed in this work should provide easy access to these two
classes of bicyclic nitrogen heterocycles harboring useful and
diverse functionality for further manipulation to natural and
unnatural products.
Acknowledgment. This paper is dedicated to Professors Nico
Speckamp, Larry Overman, and David Hart for their seminal
contributions to N-acyliminium chemistry. We thank NSERC
for generous financial assistance through the Medicinal Chem-
istry Chair program. We are grateful to Dr. Kenneth Granberg
and co-workers at AstraZeneca, Mo¨lndal, Sweden for the
enzyme inhibitory assays. We thank Michel Simard for the
X-ray analysis of compounds shown in Schemes 1, 3, 6, and
Figure 5, and Dalbir Sekhon for LC/MS assistance. M.T. is
grateful for NSERC and FRNTQ fellowships.
Conclusion
Note Added after ASAP. In the version posted 4/16/04, there
were errors in compound numbers in Scheme 4, in the 3rd
paragraph of the Introduction, and in the 12th and 17th
paragraphs of the Results. The version posted 4/20/04 and the
print version are correct.
We have described the first enantiocontrolled total synthesis
of the marine natural product oscillarin thereby confirming its
revised structure and absolute configuration. A previously
proposed structure containing a cyclic guanidine-containing
subunit disclosed as 2a in a patent was also synthesized and
found to be incorrect. Oscillarin 2 is a potent inhibitor of the
enzyme thrombin whereas the previously presumed product 2a
is inactive. A ternary complex of oscillarin and R-thrombin-
huridin was resolved at a resolution of 2.0 Å.
Supporting Information Available: Experimental procedures
of key reactions, NMR, and X-ray and other data (PDF and
CIF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA030669G
A key reaction in the synthesis of the octahydroindole-2-
carboxylic acid core relied on a seldom used N-acyloxyiminium
ion aza-Prins halocarbocyclization. The scope of this reaction
was general for diastereomeric octahydroindole-2-carboxylic
acid precursors, and its versatility was shown for the synthesis
of enantiopure 7-haloperhydroquinoline 2-carboxylic acids.
(46) For relevant reviews, see, for example, (a) Jeffs, P. W. Sceletium alkaloids.
In The Alkaloids; Rodrigo, R. G. G., Ed.; Academic: New York, 1981V.
19, p 1. (b) Martin, S. F. Amaryllidaceae alkaloids. In The Alkaloids; Brossi,
A., Ed.; Academic: New York, 1987; Vol. 30, p 251. (c) Hoshino, O. In
The Alkaloids; Cordell, G. A., Ed.; Academic: New York, 1987; Vol. 51,
p 323. (e) Pilli, R. A.; Ferreira de Oliveira, M. C. Stemona alkaloids. Nat.
Prod. Rep. 2000, 17, 117.
(47) See, for example, (a) Xu, R.; Dwoskin, L. P.; Grinevich, V. P.; Deaciuc,
G.; Crooks, P. A. Nicotine analogues. Bioorg. Med. Chem. Lett. 2001, 11,
1245. (b) Hanessian, S.; Papeo, G.; Angiolini, M.; Fettis, K.; Berretta, M.;
Munro, A. Constrained proline analogues. J. Org. Chem. 2003, 68, 7204.
(c) Portevin, B.; Longchampt, M.; Canet, E.; De Nanteuil, G. Human
elastase inhibitors. J. Med. Chem. 1997, 40, 1906. (d) Blankley, C. J.;
Kaltenbronn, J. S.; DeJohn, D. E.; Werner, A.; Bennett, L. R.; Bobowski,
G.; Krolls, U.; Johnson, D. R.; Pearlman, W. R.; Hoefle, M. L.; Essenburg,
A. D.; Cohen, D. M.; Kaplan, H. R. ACE inhibitors. J. Med. Chem. 1987,
30, 992.
Octahydroindole and perhydroquinoline cores are integral
subunits of biologically relevant nitrogen-containing natural
products.31,46 Important classes of therapeutic agents contain an
octahydroindole or related bicyclic system.47 The methodology
(45) Deslongchamps, P. Stereoelectronic Effects in Organic Chemistry; Perga-
mon: New York, 1983; Chapter 6.
9
J. AM. CHEM. SOC. VOL. 126, NO. 19, 2004 6071