Endo-selective quenching of hexahydropyrrolo[2,3-b]indole-based N-acyliminium ions

Article abstract of DOI:10.1021/jo049757p

Radical decarboxylation of L-tryptophan-derived (2S,3aR,8aS)-8-arylsulfonyl-1,2-di(methoxycarbonyl)-1,2,3,-3a,8, 8a-hexahydro-2H-pyrrolo[2,3-b]indoles 8 and 9 in the presence of diphenyl diselenide results in the endo-selective formation of (2R,3aR,8aS)-8

Full text of DOI:10.1021/jo049757p

derivatives 1 with electrophiles (E+), typically strong
acids or organoselenium-based electrophiles, leads to the
formation of the kinetic exo-substituted hexahydropyr-
rolindole 2, which rapidly equilibrates to the endo-isomer
3.4,18,20,22 Irreversible cyclizations, typically those medi-
ated by oxygen-based electrophiles,8,21,23,24 give mixtures
of exo- and endo-substituted hexahydropyrroloindoles.8,21
Deprotonation of 4, a stabilized form of 3, and alkylation
of the resultant enolate takes place with retention of
configuration, i.e., with high kinetic selectivity for the
exo-face.10 Conjugate additions, and cycloadditions to the
tetrahydro system 5, likewise take place with excellent
exo-face selectivity.13 Generation of radical 6 and quench-
ing with a variety of traps, however, results in prefer-
ential formation of the endo-trapped product, with endo-
selectivity increasing with the size of the trap.19
En d o-Selective Qu en ch in g of
Hexa h yd r op yr r olo[2,3-b]in d ole-Ba sed
N-Acylim in iu m Ion s
Rosa L. Meza-Leo´n,‡ David Crich,† Sylvain Berne`s,‡ and
Leticia Quintero*,‡
Department of Chemistry, University of Illinois at Chicago,
845 West Taylor Street, Chicago, Illinois 60607-7061, and
Centro de Investigacio´n de la Facultad de Ciencias
Quı´micas and Centro de Qu´ımica del Instituto de Ciencias,
Universidad Auto´noma de Puebla, 72570 Puebla,
Puebla, Me´xico
lquinter@siu.buap.mx.
Received February 11, 2004
Abstr a ct: Radical decarboxylation of L-tryptophan-derived
(2S,3aR,8aS)-8-arylsulfonyl-1,2-di(methoxycarbonyl)-1,2,3,-
3a,8,8a-hexahydro-2H-pyrrolo[2,3-b]indoles 8 and 9 in the
presence of diphenyl diselenide results in the endo-selective
formation of (2R,3aR,8aS)-8-arylsulfonyl-1-methoxycarbo-
nyl-2-phenylselenyl-1,2,3,3a,8,8a-hexahydro-2H-pyrrolo[2,3-
b]indoles 10 and 11. These selenides, in conjunction with
Lewis acids, serves as precursors to the corresponding N-acyl
iminium ions, which undego selective endo-face quenching
by allyltributylstannane, allyltrimethylsilane, propargyl-
trimethylsilane, and trimethylsilylcyanide. Stereochemical
assignments rest on NMR data and crystallographic studies.
The endo-selective nature of these reactions is interpreted
in terms of minimization of allylic strain at the transition
state for nucleophilic attack on the N-acyl iminum ion.
This background, together with the generally useful
chemistry of N-acyl iminium ions25 and the current
interest in the endo-face selective quenching of related
fused bicyclic five-membered oxocarbenium ions,26
prompted a study of the generation and trapping of
cationic species 7 at the 2-position of the hexahydropy-
rrolo[2,3-b]indole framework which we report here.
The chemistry of the 2-substituted hexahydropyrrolo-
[2,3-b]indole tautomers 2 and 3 of tryptophan continues
to be of considerable importance owing (i) to the applica-
tion of this system in the synthesis of numerous alkaloids
of diverse biological activity,1-8 (ii) to their use as
templates for the stereocontrolled formation of tryp-
tophan analogues,9-15 and (iii) the intriguing enigma of
the kinetic and thermodymanic stereoselectivity at the
2-position.16-21 With respect to the question of stereo-
selectivity at the 2-position, treatment of tryptophan
† University of Illinois at Chicago.
‡ Universidad Auto´noma de Puebla.
(11) Crich, D.; Pastrana-Grass, I.; Quintero-Cortes, L.; Sandoval-
Ramirez, J . Heterocycles 1994, 38, 719-724.
(1) Bruncko, M.; Crich, D.; Samy, R. J . Org. Chem. 1994, 59, 5543-
5549.
(2) Crich, D.; Pavlovic, A. B.; Samy, R. Tetrahedron 1995, 51, 6379-
6384.
(3) Crich, D.; Fredette, E.; Flosi, W. J . Heterocycles 1998, 48, 545-
547.
(12) Crich, D.; Lim, L. B. L. Heterocycles 1993, 36, 1199-1204.
(13) Bruncko, M.; Crich, D. J . Org. Chem. 1994, 59, 4239-4249.
(14) Dua, R. K.; Phillips, R. S. Tetrahedron Lett. 1992, 33, 29-32.
(15) Burgaud, B. G. M.; Horwell, D. C.; Pritchard, M. C.; Bernad,
N.; Martinez, J . Tetrahedron: Asymmetry 1995, 6, 1081-1084.
(16) Crich, D.; Chan, C.-O.; Davies, J . W.; Natarajan, S.; Vinter, J .
G. J . Chem. Soc., Perkin Trans. 2 1992, 2233-2240.
(17) Crich, D.; Bruncko, M.; Natarajan, S.; Teo, B. K.; Tocher, D. A.
Tetrahedron 1995, 51, 2215-2228.
(4) Depew, K. M.; Marsden, S. P.; Zatorska, D.; Zatorski, A.;
Bornmann, W. G.; Danishefsky, S. J . J . Am. Chem. Soc. 1999, 121,
11953-11963.
(5) Kamanecka, T. A.; Danishefsky, S. J . Angew. Chem., Int. Ed.
1998, 37, 2995-2998.
(18) Crich, D.; Huang, X. J . Org. Chem. 1999, 64, 7218-7223.
(19) Crich, D.; Natarajan, S. J . Org. Chem. 1995, 60, 6237-6241.
(20) Ley, S. V.; Cleator, E.; Hewitt, P. R. Org. Biomol. Chem. 2003,
3492-3494.
(6) Chen, W.-C.; J oullie´, M. M. Tetrahedron Lett. 1998, 39, 8401-
8404.
(7) Schiavi, B. M.; Richard, D. J .; J oullie´, M. M. J . Org. Chem. 2002,
67, 620-624.
(21) Kamenecka, T. M.; Danishefsky, S. J . Chem. Eur. J . 2001, 7,
41-63.
(8) Baran, P. S.; Guerrero, C. A.; Corey, E. J . J . Am. Chem. Soc.
2003, 125, 5628-5629.
(22) Taniguchi, M.; Hino, T. Tetrahedron 1981, 37, 1487-1494.
(23) Hino, T.; Nakagawa, M. In Alkaloids; Brossi, A., Ed.; Academic
Press: New York, 1988; Vol. 34, pp 1-75.
(9) Crich, D.; Davies, J . W. J . Chem. Soc., Chem. Commun. 1989,
1418-1419.
(10) Bourne, G. T.; Crich, D.; Davies, J . W.; Horwell, D. C. J . Chem.
Soc., Perkin Trans. 1 1991, 1693-1699.
(24) Saito, I.; Matsuura, T.; Nakagawa, M.; Hino, T. Acc. Chem. Res.
1977, 10, 346-352.
10.1021/jo049757p CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/24/2004
3976
J . Org. Chem. 2004, 69, 3976-3978

TABLE 1. Dia ster eoselective Rea ction s of 10-en d o w ith
Allyltr im eth ylsila n e a n d Allyltr ibu tyltin in th e P r esen ce
of Lew is Acid s
SCHEME 1
Lewis reaction temp 12/13 yield
entry
nucleophile
acid time (h) (°C) (%) (%)
1
2
3
4
5
6
(CH3)3SiCH2CHdCH2 SnCl4
(CH3)3SiCH2CHdCH2 BF3
(CH3)3SiCH2CHdCH2 TiCl4
nBu3SnCH2CHdCH2 SnCl4
nBu3SnCH2CHdCH2 BF3
nBu3SnCH2CHdCH2 TiCl4
1.5
1.5
5
1.5
1.5
3
-30 46:54 80
75:25 70
25 32:68 55
-30 0:1 90
76:24 78
25 24:76 43
0
0
SCHEME 2
TABLE 2. Dia ster eoselective Rea ction s of 11-en d o w ith
P r op a r gyltr im eth ylsila n e a n d Tr im eth ylsilyl Cya n id e in
th e P r esen ce of Lew is Acid s
Lewis reaction temp
yield
entry
nucleophile
acid time (h) (°C) product (%)
1
2
3
4
5
6
(CH3)3SiCH2CCH SnCl4
(CH3)3SiCH2CCH BF3
(CH3)3SiCH2CCH TiCl4
1.5
1.5
4
1.5
1.5
8
-30
0
25
-30
0
14
14
14
15
16
15
64
53
23
93
57
49
(CH3)3SiCN
(CH3)3SiCN
(CH3)3SiCN
SnCl4
BF3
TiCl4
25
N-acyl iminium ion was poorly selective giving rise to 12
in a mixture with its exo isomer and that, under the
Lewis acid conditions of the reaction, the less stable16 exo-
isomer was undergoing ring opening to the tryptamine
derivative. However, when a purified sample of 12 was
exposed to BF3 at room temperature it was converted to
13 whose specific rotation matched, in sign and magni-
tude, that of the sample isolated from the allylation
reaction. Tryptamine 13 therefore has the absolute
configuration indicated in Scheme 2 and arises from slow
opening of the endo-product 12, with the relative propor-
tions of 12 and 13 (see Table 1) depending on the reaction
conditions.31 This result, of course, leads to the conclusion
that the quenching of cation 7 by allyltrimethylsilane and
allytributylstannane is very highly endo-selective.
Condensation of selenide 11 with propargyltrimethyl-
silane and trimethylsilylcyanide followed the pattern
established with the allylmetal nucleophiles with ex-
tremely high selectivity for the formation of the endo-
product 14 and 15, respectively (Table 2). In the case of
trimethylsilylcyanide the ring-opened product 16 was
isolated when BF3 etherate was employed as Lewis acid.
An appropriate cation precursor, selenide 10, was
obtained in 84% yield as a 19:1 endo/exo mixture by
Barton decarboxylation27,28 of acid 819 in the presence of
diphenyl diselenide.29 The endo-selectivity in this radical
reaction exactly mirrors that observed previously on
quenching of the same radical with diphenyl disulfide.19
The relative stereochemistry of 10-endo was assigned on
the basis of the 3J ) 0 value of the H-2-H-3-endo
coupling, which has been previously shown10,16-18 to be
diagnostic for the 2-endo-substituted series in the pre-
ferred conformation with C-2 puckered in toward the
concave face of the tricyclic system so as to minimize 1,3
A
strain about the N-C carbamate partial double bond.
X-ray crystallographic analysis of 10-endo fully confirms
this assignment.30 As expected the p-toluenesulfonyl
analogue 9 of 8 performed exactly analogously affording
the crystalline endo-selenide 11 (Scheme 1), which was
also X-ray characterized.
With a suitable, stable precursor in hand, attention
was turned to the N-acyl iminium ion chemistry. Reaction
of 10-endo with allyltrimethylsilane in the presence of a
range of Lewis acids resulted in the formation of two
products, the endo-allyl product 12 and the allyl-
tryptamine derivative 13 (Scheme 2) with the yields and
ratios recorded in Table 1. Analogous observations were
made with allyltributylstannane. The configuration of 12,
initially assigned on the basis of a ROESY interaction
between H-2exo and H3a, was unambiguously deter-
mined by single-crystal X-ray crystallographic analysis.
The possibility arose that the initial quenching of the
The relative stereochemistry of the allene 14 was
assigned on the basis of an interaction between H-2exo
and H-3a in the ROESY spectrum whereas the corre-
sponding interaction in the NOESY spectrum of 15
confirmed the endo-nature of the cyano group. The
(25) Speckamp, W. N.; Moolenaar, M. J . Tetrahedron 2000, 56,
3817-3856.
(26) Smith, D. M.; Tran, M. B.; Woerpel, K. A. J . Am. Chem. Soc.
2003, 125, 14149-14152.
(27) Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron
1985, 41, 3901-3924.
(31) The proposed configuration was further confirmed by an X-ray
analysis of 13. It is worth mentioning that 13 crystallizes with two
independent molecules in the asymmetric unit, reflecting a degree of
free rotation for SO2Ph moiety along the N8-S1 bond. On the other
hand, the formation of a double bond during the ring opening is
confirmed by bond lengths C3a-C8a ) 1.333(6) and 1.345(6) Å for the
first and second molecule, respectively.
(28) Crich, D.; Quintero, L. Chem. Rev. 1989, 89, 1413-1432.
(29) Barton, D. H. R.; Bridon, D.; Zard, S. Z. Heterocycles 1987, 25,
449.
(30) The absolute configuration for chiral C-atoms was unambigu-
ously determined as 2R,3aS,8aS and the carbamate partial double
bond is reflected in the short distance N1-C9, 1.369(4) Å.
J . Org. Chem, Vol. 69, No. 11, 2004 3977

processes is dominated by the need to minimize 1,3A
strain between the substituents at C-2 and the carbamate
NdC partial double bond. At the same time torsional
strain around the terminal ring is better minimized when
attack takes place from the endo-face rather than the exo-
face of the system. This argument is best appreciated by
examination of the conformation of the ester 4, which is
well-established to be the thermodynamic isomer. In the
preferred endo-isomer of 4 computational16 and crystal-
lographic studies17,33 show the terminal pyrrolidine ring
to be in an envelope-like conformation with C-2 puckered
in toward the endo face of the molecule. In this confor-
mation the C-2 endo substituent is perpendicular to the
plane of the carbamate, thereby minimizing 1,3A strain,
and torsional interactions around the ring are minimized.
This same conformation is evident in the X-ray crystal-
lographic structures of compounds 10 and 12, as pre-
sented here. It seems entirely reasonable then that the
approach of nucleophiles to cations 7 and 18 as well as
of radical traps to radical 6 will take place from the
direction that engenders minimal 1,3A strain in the
developing bond, i.e., from the endo face. In the alkylation
of the enolate of 4, however, there is a competition for
the favored endo position between the existing C-2-CO2-
Me bond and the incipient bond to the incoming electro-
phile. At the transition state, the need to minimize 1,3A
strain to the existing C-C bond is obviously greater than
that to a partial bond resulting in the observed exo-
selective alkylations.34
SCHEME 3
relative configuration of 15 was also confirmed by single-
crystal X-ray analysis.32 The absolute configuration of the
ring-opened cyanide 16 is assigned by analogy with the
absolute configuration of tryptamine 13.
When 11 was activated with SnCl4, BF3‚OEt2, and
TiCl4 in the presence of vinyltrimethylsilane and tri-
methylsilylacetylene no substitution was observed. In
these cases the only product isolated was the ring-opened
racemic selenide 17, whose structure was also confirmed
crystallographically. Presumably, the formation of 17
arises from ring opening of the initial N-acyl iminium
ion 18 in competition with capture by the less nucleo-
philic vinyl and acetylenyl silanes. The ring-opened
N-acyl iminium ions are then quenched from either face
by the selenide-Lewis acid complex leading to the
formation of racemic 17 (Scheme 3). This sequence is in
line with the X-ray study of 17: in contrast with previous
compounds, 17 crystallizes in a center-symmetric space
group, in agreement with a racemic configuration at C2.
Moreover, as for 13, the double bond formed during the
ring opening is characterized by the expected bond
length, C3a-C8a ) 1.349(8) Å.
Although the nucleophilic capture of 7 and its tosyl
analogue 18 has only been achieved with a limited range
of nucleophiles because of an apparent competition with
the ring opening illustrated in Scheme 3, it is evident
that when capture does take place it occurs with high
selectivity from the endo-face of the tricyclic system.
Thus, the stereoselective nucleophilic capture of C-2
cation parallels that of the quenching of C-2 radicals but
opposes that of the exo-selective alkylation of C-2 eno-
lates. We believe that the stereoselectivity of all three
Ack n ow led gm en t. We thank CONACyT (Project
No. 35102-E and Grant No. 91277) for financial support
of this work.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental details, copies of the 1H and 13C NMR spectra for
compounds 10-17, and X-ray crystallographic details for
compounds 10-13, 15, and 17. This material is available free
of charge via the Internet at http://pubs.acs.org.
J O049757P
(33) Chan, C.-O.; Cooksey, C. J .; Crich, D. J . Chem. Soc., Perkin
Trans. 1 1992, 777-780.
(34) In the possibly related cis-selective attack of nucleophiles on
substituted N-carbamoyl35 and N-sulfonyl36 pyrrolidine derived imi-
nium ions, neighboring group participation from the carbamoyl or
sulfonyl group has been invoked as a possible explanation for the
stereoselectivity. While this may be possible with the N-sufonyl system,
with the tetrahedral sulfur and longer N-S and S-O bonds, it seems
unlikely in N-carbamoyl systems, such as those described here, when
it would require the disruption of the crystallographically well-
established NdC(O-)OMe double bond character.
(32) The cyano group is characterized by a formal triple bond, C13-
N14 ) 1.127 (5) Å; the configuration for C3a and C8a remains
unchanged relative to starting material, while the C2 atom presents
now an S configuration.
(35) Renaud, P.; Seebach, D. Helv. Chim. Acta 1986, 69, 1704-1710.
(36) Ungureanu, I.; Bologa, C.; Chayer, S.; Mann, A. Tetrahedron
Lett. 1999, 40, 5315-5318.
3978 J . Org. Chem., Vol. 69, No. 11, 2004