On the reaction of tryptophan derivatives with N-phenylselenyl phthalimide: The nature of the kinetic and thermodynamic hexahydropyrrolo[2,3-b]indole products. Alkylation of tryptophan with inversion of configuration

Article abstract of DOI:10.1021/jo991093+

The cyclization of a range of tryptophan derivatives with N- phenylselenyl phthalimide leading to 3a-phenylselenyl-1,2,3,3a,8,8a- hexahydro[2,3-b]pyrroloindoles has been studied. It is established that in each case the kinetic diastereomer has the C-2 substituent on the exo face of the diazabicyclo-[3.3.0]octane skeleton, whereas the thermodynamic isomers have the C-2 substituent endo. A simple protocol for the alkylation of tryptophan, leading to α-substituted tryptophans with clean inversion of configuration, is presented.

Full text of DOI:10.1021/jo991093+

7218
J . Org. Chem. 1999, 64, 7218-7223
On th e Rea ction of Tr yp top h a n Der iva tives w ith N-P h en ylselen yl
P h th a lim id e: Th e Na tu r e of th e Kin etic a n d Th er m od yn a m ic
Hexa h yd r op yr r olo[2,3-b]in d ole P r od u cts. Alk yla tion of
Tr yp top h a n w ith In ver sion of Con figu r a tion
David Crich* and Xianhai Huang
Department of Chemistry, University of Illinois at Chicago, 845 West Taylor Street,
Chicago, Illinois 60607-7061
Received J uly 7, 1999
The cyclization of a range of tryptophan derivatives with N-phenylselenyl phthalimide leading to
3a-phenylselenyl-1,2,3,3a,8,8a-hexahydro[2,3-b]pyrroloindoles has been studied. It is established
that in each case the kinetic diastereomer has the C-2 substituent on the exo face of the diazabicyclo-
[3.3.0]octane skeleton, whereas the thermodynamic isomers have the C-2 substituent endo. A simple
protocol for the alkylation of tryptophan, leading to R-substituted tryptophans with clean inversion
of configuration, is presented.
Over a period of approximately 10 years, this labora-
entry into the synthesis of related indole alkaloids.10 The
success of this chemistry is underpinned by the initial
highly diastereoselective ring closure, which enables
preparation of 2 and so 4 on a significant scale without
recourse to chromatographic purification.1-3 Perhaps, the
most striking observation in all of the above is the
thermodynamically more stable nature of the 2-endo-
carbomethoxy isomer 2,1-3 which prompted us to carry
out a series of computational11 and crystallographic stu-
dies aimed at understanding this phenomenon.12,13 Ul-
timately, we determined that the thermodynamic prefer-
ence for the endo site is a subtle function of torsional
interactions around the bicyclo[3.3.0]octane nucleus,
rather than a minimization of 1,3A-strain due to the Nd
C double implicit in the N1-carbamate or a consequence
of π-stacking between the endo ester and the surface of
the fused benzene ring.13 This is not to say that these
factors do not contribute to the overall stability of the
system but simply that they are not the decisive factors.
This notion is reinforced by the fact that under equili-
brating conditions the 2-CO2Me substituent in the afla-
toxin model 8 also prefers the endo position: seemingly
the same torsional interactions are at work here.14
Similar observations have also been reported for systems
related to 8 in which the 2-substituent was OMe or
OH.15,16 This thermodynamic preference for the endo-
position in 2-substituted bicyclo[3.3.0]octane nuclei is not
limited to heterocyclic systems. As noted previously,13
under equilibrating conditions, 2-alkylbicyclo[3.3.0]octan-
1-ones are 1:1 mixtures of exo and endo isomers, which
tory has exploited the closure of the tryptophan deriva-
tive 1 to a thermodynamic 9:1 mixture of the two dia-
stereomeric hexahydropyrroloindole tautomers 2 and 3
on dissolution in 85% phosphoric acid, as first reported
by Taniguchi and Hino.1 Both 2 and 3 are somewhat
unstable with respect to reversion to 1 in the absence of
strong acid, but sulfonylation of the crude mixture greatly
enhances stability and permits the ready isolation of
crystalline 4 in excellent overall yield and as a single
diastereoisomer.2,3 Alkylations and kinetic aldol conden-
sations of the lithium enolate of 4, and related deriva-
tives, take place exclusively on the exo surface of the
diazabicyclo[3.3.0]octane nucleus so permitting, after
cycloreversion to the tryptophan skeleton, the overall
conversion of L-tryptophan to optically pure R-substituted
derivatives without recourse to the use of an external
chiral auxiliary or catalyst.2-5 The overall concept is
related, but not identical to, Seebach’s self-reproduction
of chirality method:6 the essential difference being that
the key ring formation arises from a simple tautomer-
ization rather than formation of a derivative such as a
cyclic acetal. Dehydrogenation of 4, by the usual sequence
of selenenation and selenoxide syn elimination, provides
5, which is amenable to conjugate additions and cycload-
ditions, again on the exo face, ultimately providing â- and
R,â-substituted tryptophan derivatives in enantiomeri-
cally pure form.7,8 Radical bromination of 4 with NBS in
CCl4 provided 6,9 which, on treatment with allyltribu-
tylstannane and AIBN, afforded 7 and so a convenient
(1) Taniguchi, M.; Hino, T. Tetrahedron 1981, 37, 1487-1494.
(2) Crich, D.; Davies, J . W. J . Chem. Soc., Chem. Commun. 1989,
1418-1419.
(10) Bruncko, M.; Crich, D.; Samy, R. J . Org. Chem. 1994, 59, 5543-
5549.
(11) Crich, D.; Chan, C.-O.; Davies, J . W.; Natarajan, S.; Vinter, J .
G. J . Chem. Soc., Perkin Trans. 2 1992, 2233-2240.
(12) Chan, C.-O.; Cooksey, C. J .; Crich, D. J . Chem. Soc., Perkin
Trans. 1 1992, 777-780.
(13) Crich, D.; Bruncko, M.; Natarajan, S.; Teo, B. K.; Tocher, D. A.
Tetrahedron 1995, 51, 2215-2228.
(14) Civitello, E. R.; Rapoport, H. J . Org. Chem. 1994, 59, 3775-
3782.
(15) Bujons, J .; Sanchez-Baeza, F.; Messeguer, A. Tetrahedron 1994,
50, 7597-7610.
(16) Iyer, R. S.; Coles, B. F.; Raney, K. D.; Thier, R.; Guengerich, F.
P.; Harris, T. M. J . Am. Chem. Soc. 1994, 116, 1603-1609.
(3) Bourne, G. T.; Crich, D.; Davies, J . W.; Horwell, D. C. J . Chem.
Soc., Perkin Trans. 1 1991, 1693-1699.
(4) Crich, D.; Pastrana-Grass, I.; Quintero-Cortes, L.; Sandoval-
Ramirez, J . Heterocycles 1994, 38, 719-724.
(5) Crich, D.; Lim, L. B. L. Heterocycles 1993, 36, 1199-1204.
(6) Seebach, D.; Sting, A. R.; Hoffmann, M. Angew. Chem., Int. Ed.
Engl. 1996, 35, 2708-2748.
(7) Bruncko, M.; Crich, D. J . Org. Chem. 1994, 59, 4239-4249.
(8) Bruncko, M.; Crich, D. Tetrahedron Lett. 1992, 33, 6251-6254.
(9) Bruncko, M.; Crich, D.; Samy, R. Heterocycles 1993, 36, 1735-
1738.
10.1021/jo991093+ CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/18/1999

Alkylation of Tryptophan with Inversion of Configuration
J . Org. Chem., Vol. 64, No. 19, 1999 7219
Ta ble 1. Cycliza tion Rea ction s
suggests that similar factors exist in this carbocyclic
system,17,18 albeit to a lesser extent. Finally, it is neces-
sary to draw attention to two characteristics of the H
total
cyclized
substrate yield (%) product
other
1
endo
exo product endo/exo products
([R]D)
ratio
(%)
NMR spectrum of 4 and all of its 2-endo congenors. These
are the unusual upfield shift of the Me group in the C2
ester and the approximately 90° torsion angle and so
minimal scalar coupling, between the 2-exo-H and the
3-endo-H that is reflected in the simple doublet nature
of the H2 resonance.3,11,13 Together, these phenomena
indicate that in solution the terminal heterocyclic ring
of 4 is in an envelope conformation with the flap (C2)
folded under the endo surface of the fused nucleus, as
found in the crystal. We stress that the same features
are found in the 3a-ramified systems 6 and 7, which
indicates that substitution at this position does not
change the basic conformational preference.10 We also
point out that the same coupling pattern, and so confor-
mation, is found in the dioxabicyclooctane system, which
serves again to illustrate the generality of the phenom-
enon.14,17,18
11
14
17
20
23
26d
27
65
76
83
40
31
0
12
15
18
21
24
13 (-78.4°)a
16 (-74.5°)
19 (-100.6°)
22 (-62.0°)
25 (-35.6°)
1:9b
1:12c
1:11c
<2:98
<2:98
0
28 (85)
a
b
Taken on the 9:1 inseparable mixture. Inseparable by chro-
matography. c Separable by chromatography. No reaction.
d
11 and related tryptophan derivatives with N-phenylse-
lenophthalimide and report here on our findings.
A number of N,N′-diprotected L-tryptophan derivatives
were prepared and treated with N-phenylselenophthal-
imide in dichloromethane with catalysis by p-toluene-
sulfonic acid. The reactions were clean with no significant
byproducts; in each case, the mass balance consisted of
recovered starting material. Examination of the crude
1
reaction mixtures by H NMR spectroscopy revealed the
presence of exo- and endo-cyclized products in almost
every case. As we have discussed numerous times previ-
ously, the spectra taken at room temperature were ill-
resolved owing to a dynamic NMR phenomenon.3,13
However, at 50 °C much sharper spectra were obtained.
Chromatography on silica gel then permitted separation
of the cyclic products from minor, unidentified byprod-
ucts. The various cyclizations, together with their isolated
yields and endo/exo ratios, are presented in Table 1.
With this background, we were considerably surprised
to read a 1994 paper by Danishefsky and co-workers on
the synthesis of the indole alkaloids amauromine (9) and
5-N-acetylardeemin (10).19 In this paper, it was reported
With 11 a 9:1 ratio of cyclized products was obtained,
which, as reported by Danishefsky, we were unable to
separate chromatographically. With the closely analogous
dicarbamates 14 and 17 the ratio was 12:1 and 11:1,
respectively, and the isomers could now be separated
chromatographically without difficulty. When the indole
nitrogen was protected with a sulfonyl group, as in 20
and 23, the yield of cyclized products was lower but the
ratios much higher giving essentially a single isomer. In
the case of 26, with the free indole NH, no cyclized
product was obtained. This result can probably be at-
tributed to the instability of any such cyclic product in
that cyclization of 11 with N-phenylselenophthalimide
led to an initial 1:1 mixture of the endo- and exo-
hexahydropyrroloindoles 12 and 13 and that equilibra-
tion, under unspecified conditions, resulted in conversion
to an inseparable 1:9 mixture favoring the exo isomer
13. Perplexed by this apparent exception to the above
body of work, we have reinvestigated the cyclization of
(17) Kakiuchi, K.; Takeuchi, H.; Tobe, Y.; Odaira, Y. Bull. Chem.
Soc. J pn. 1985, 58, 1613-1614.
(18) Ghera, E. J . Org. Chem. 1968, 33, 1042-1051.
(19) Marsden, S. P.; Depew, K. M.; Danishefsky, S. J . J . Am. Chem.
Soc. 1994, 116, 11143-11144.

7220 J . Org. Chem., Vol. 64, No. 19, 1999
Crich and Huang
the absence of an electron-withdrawing group on both
nitrogen atoms (cf. 2 and 3 above). Finally, with the
N-methyl derivative 27 the only product formed arose
from substitution at the indole C2 position, in accord with
previous work from this laboratory on the reaction of
tryptophan derivatives not bearing an electron-withdraw-
ing group on the indole N with selenium-based electro-
philes.20
With a range of cyclic tautomers in hand, we turned
to the assignment of configuration. In the mixture of 12
and 13, the minor isomer exhibited a singlet at δ 3.07
reminiscent of the ester methyl group in 4. Similar
observations were made for the minor isomers arising
from the cyclizations of 14 and 17. The products derived
from the cyclizations of 20 and 23 had no such upfield
singlets. These observations suggested that the major
isomer, in each case, was the exo isomer in agreement
with the conclusion of Danishefsky and co-workers.
Nevertheless, there remained the possibility, however
small given the fact that 6 fit well the established
pattern, that the 3a-phenylseleno moiety was having an
unexpected effect on the chemical shifts of the methyl
esters. Thus, the mixture of 12 and 13 was reduced with
tributyltin hydride and AIBN in benzene at reflux to give
a 1:9 mixture of 29 and 30. Still, the characteristic upfield
shift of the methyl ester was found only in the minor
isomer. A similar experiment was conducted with both
19 and 25, leading to 31 and 32, respectively, with the
same result. Thus, it was established that the major
isomer arising from each of the cyclizations had the C-2
CO2Me group on the exo surface of the bicyclo[3.3.0]-
octane nucleus. This assignment is borne out by an X-ray
crystallographic study from the J oullie´ group that ap-
peared recently.21
In principle, equilibration between the exo and endo
isomers in any given case can occur through a reversal
of the ring closing and selenation steps or by epimeriza-
tion at C-2. The two processes will give enantiomeric
products. Thus, the cycloreversion/selenation mode of
equilibration will give the products in the configurations
depicted all having the S configuration at C-2. On the
other hand, epimerization at C-2 would lead to the
enantiomers of the products drawn. For example, the
cycloreversion/selenation mode of equilibration of 12
would provide 13, whereas base-catalyzed epimerization
at C-2 would lead to ent-13. As Danishefsky had not
specified the equilibration conditions, it was necessary
to ascertain which, if any, of the two processes was
occurring. This was readily determined from the specific
rotations of the various products. Our extensive experi-
ence with dozens of cyclic tautomers of tryptophan, both
2-exo- and 2-endo-substituted, taught us that the sign
of the specific rotation at the sodium D lines is determined
by the absolute configurations at the two ring junction
stereogenic centers. Other substituents, exo or endo,
make only small contributions to the numerical value of
the specific rotation. Thus, all of the 2-endo products 4-7,
derived from L-tryptophan have strongly dextrorotatory
specific rotations, whereas that of the 2-exo product 32,
also derived from L-tryptophan, was levorotatory. The
major products from each of the cyclizations conducted
here were levorotatory (Table 1), indicating that they had
the absolute configurations indicated.
Next, we turned to the question of the thermodynamic
or kinetic nature of the product mixtures isolated. Dan-
ishefsky had indicated that his initial reaction mixtures
had a 12:13 ratio of 1:1 and that this was improved to
1:9 upon the unspecified equilibration. In our hands, we
never observed the 1:1 mixture in any instance. Neither
did the product ratios change significantly with time
when left exposed to the cyclization reaction conditions.
This suggested that either the equilibration was very
rapid or, alternatively, that the ratios observed were
kinetic and that no equilibration was taking place. To
clarify the situation, the 1:9 mixture of 12 and 13 was
dissolved in perdeuterio-tert-butyl alcohol and treated
with a catalytic quantity potassium tert-butoxide. Within
the space of a few minutes all the C-2 hydrogens were
exchanged for deuterium atoms and the nature of the
minor and major product spectra were reversed. The new
ratio was 7:3 in favor of the endo isomer with its
strikingly upfield methyl ester chemical shift at δ 3.07
(Scheme 1). The fact that this conversion is taking place
under equilibrating conditions, and does not arise from
a kinetic exo protonation of the enolate, is incontrovert-
ibly established by the complete exchange of the C-2
hydrogens in both epimers. Finally, the tert-butyl alcohol
was removed and the reaction mixture taken up in CDCl3
when the same spectral features were noted: the ob-
served changes in the spectra in t-BuOH-d10 therefore
did not arise from the unusual solvent. The specific
rotation of the final equilibrium mixture was strongly
negative (-34°). This indicates that this equilibration has
taken place simply through epimerization at C-2 rather
than via a coupled process including deprotonation at C-2
and cycloreversion, which would have led to complete
racemization. Analogous experiments were conducted
with pure exo-19 and pure exo-25, which were both
converted to fully C-2-deuterated mixtures (80:20 and 86:
14, respectively) in favor of the endo isomers. It is
therefore unambiguously established that, in the pair of
diastereoisomers 12 and 13, the former is the thermo-
dynamically favored isomer. This is fully consistent with
the general picture established previously in this labora-
tory and briefly reviewed in the Introduction. It is also
clear that the exo isomers are the kinetically preferred
products and may be isolated in high yield under the
selenation conditions. Again, this is consistent with
previous work from this laboratory wherein it was
demonstrated that even under the phosphoric acid cy-
clization conditions 1 is converted kinetically to 3, which
very rapidly equilibrates to 2.13 We conclude that the 1:9
ratio of 12 and 13 originally reported by Danishefsky was
a kinetic and not a thermodynamic ratio. Subsequent
(20) Crich, D.; Davies, J . W. Tetrahedron Lett. 1989, 30, 4307-4308.
(21) Chen, W.-C.; J oullie, M. M. Tetrahedron Lett. 1998, 39, 8401-
8404.

Alkylation of Tryptophan with Inversion of Configuration
J . Org. Chem., Vol. 64, No. 19, 1999 7221
Sch em e 1
Sch em e 3
second, more plausible, scenario, the stereodiscriminating
step is the ring closure with B f D being easier than C
f E. This requires the equilibration of B and C, via A,
to be rapid on the time scale of the ring closure. It follows
from this scenario that the transition state for the ring-
closure step is early, as a late, product-like transition
state would result in a kinetic preference for E. In an
early transition state, the attacking carbamate ought still
to be protonated and the nucleophilic lone pair to have
considerable p-character, whereas, after ring closure and
deprotonation, it ends up as a σ bond. Thus, the confor-
mation of the forming ring at the transition state for ring
closure will be rather different from that of the final ring
in the product. There is therefore no reason to expect the
same set of torsional strains in the transition state as in
the product and, consequently, no reason for the kinetic
and thermodynamic stereoselectivity to have the same
bias.
Sch em e 2
Finally, we returned briefly to an old theme in this
laboratory, namely, the alkylation of tryptophan with
inversion of configuration. Our original interest in the
chemistry of cyclic tautomers of tryptophan was in
alkylation at the stereogenic center with clean retention
or inversion without recourse to the use of a chiral
auxiliary. We were very successful with retention using
4 and related products. Here, the enolate is quenched
under kinetic conditions exclusively on the exo face, i.e.,
with clean retention of configuration. Moreover, this
chemistry was very practical as 4 could be obtained
stereochemically pure in excellent yield from L-tryp-
tophan.3,20 We were much less successful with inversion,
which would allow passage from the cheap L-tryptophan
series to the more expensive D series. We were able to
demonstrate that the kinetic product 32 did indeed
undergo alkylation with clean inversion of configura-
tion.13 However, the rapid equilibration of 2 and 3 under
the H3PO4 cyclization conditions meant that 32 could only
be obtained in very low yield, which rendered the process
inefficient and unattractive. We have now taken advan-
tage of the formation of 19 as the major isomer from
closure of 17 under kinetic conditions (Table 1) to briefly
demonstrate this alkylation with inversion. Thus, 19 was
deprotonated with LDA at -78 °C in THF and the
resulting enolate treated with methyl iodide and p-
bromobenzyl bromide (Scheme 3). In each case, alkylation
occurred in moderate yield with clean inversion of con-
figuration as signaled by the now familiar change in
chemical shift of the ester Me group in the 1H NMR
spectra. A byproduct from the reaction with methyl iodide
was the tetrahydropyrroloindole 35. Treatment of both
33 and 34 with neat TFA at room temperature brought
work from the Danishefsky group concurs with this
assessment.22
It is of some interest to ask why the 2-CO2Me exo
isomers are kinetically favored in these ring-closure
reactions. In principle, this kinetic preference might arise
at either of two steps. In the first scenario it is the initial
attack of the electrophile (X+ ) H+ or PhSePhth) on the
indole that shows face selectivity leading preferentially
to indolenium ion B rather than C (Scheme 2). Subse-
quent ring closure then gives the products D and E,
respectively. This seems unlikely simply because the only
stereogenic center in the substrate (A) is two freely
rotating bonds removed from the reaction center and so
is unlikely to confer the type of induction seen. In the
(22) Depew, K. M.; Marsden, S. P.; Zatorska, D.; Zatorski, A.;
Bornmann, W. G.; Danishefsky, S. J . J . Am. Chem. Soc. 1999, 121, in
press.

7222 J . Org. Chem., Vol. 64, No. 19, 1999
Crich and Huang
Hexa h yd r op yr r olin d ole 22, a viscous yellow oil, was
obtained in 40% yield as the pure exo isomer: [R]20D ) -62.0°
about the ring opening and provided the respective
R-alkylated D-tryptophans 36 and 37 in excellent yield.
1
(c ) 0.83, CHCl3); H NMR δ 7.91 (m, 2 H), 7.21-7.52 (m, 10
H), 7.03 (t, J ) 7.3 Hz, 1 H), 6.92 (d, J ) 6.9 Hz, 1 H), 6.17 (br
s, 1 H), 3.86 (dd, J ) 10.0, 6.2 Hz, 1 H), 3.61 (s, 3 H), 3.59 (s,
3 H), 2.74 (dd, J ) 12.7, 6.2 Hz, 1 H), 2.37 (dd, J ) 12.7, 10.0
Hz, 1 H); 13C NMR δ 171.2, 154.3, 140.6, 139.5, 137.6, 133.4,
129.9, 129.6, 129.4, 128.9, 128.6, 127.9, 126.48, 125.4, 124.1,
118.0, 84.9, 59.2, 52.8, 52.5, 39.9. Anal. Calcd for C26H24N2O6-
SSe: C, 54.64; H, 4.23. Found: C, 54.59; H, 4.32.
Exp er im en ta l Section
Gen er a l Meth od s. All solvents were dried and distilled by
standard means. All experiments were conducted under an
atmosphere of dry N2 or Ar. 1H NMR spectra were recorded
at 300 MHz and 13C NMR spectra at 75 MHz in CDCl3 solution
unless otherwise stated. Chemical shifts (δ) are in ppm
downfield from tetramethylsilane. Microanalyses were con-
ducted by Midwest Microlabs, Indianapolis, IN.
Hexa h yd r op yr r olin d oles 12 a n d 13. Gen er a l P r otocol
for Cycliza tion w ith N-Ben zen eselen yl P h th a lim id e. To
a solution of 11 (0.403 g, 0.964 mmol) in dry dichloromethane
(6 mL) were added Na2SO4 (1.5 g), and p-TsOH‚H2O (19 mg,
0.10 mmol, 10 mol %) successively. The resulting mixture was
stirred for 5 min before N-phenylselenyl phthalimide (1.5
equiv, 0.438 g, 1.45 mmol) was added. The system was stirred
at room temperature under Ar for 24-48 h. The reaction was
worked up by addition of EtOAc (30 mL) and H2O (15 mL).
The water layer was extracted with EtOAc (3 × 10 mL). The
combined organic phases were washed with water (10 mL) and
brine (10 mL) and dried over Na2SO4. The solvent was removed
under vacuum and the residue taken up in chloroform and
hexane and filtered. The filtrate was concentrated to dryness
and the residue purified with flash chromatography on silica
gel (hexane/EtOAc 1:1) and afforded the pure title compound
as an amorphous white solid (0.333 g, 65%; 100% conversion
Hexa h yd r op yr r olin d ole 25, a pale yellow solid, was
obtained in 31% yield as the pure exo isomer: mp 75-77 °C;
[R]20 ) -35.6° (c ) 1.45, CHCl3); 1H NMR δ 7.86 (br d, 2 H),
D
7.49 (d, J ) 7.3 Hz, 2 H), 7.31-7.41 (m, 4 H), 7.23 (t, J ) 7.8
Hz, 1 H), 7.02 (t, J ) 7.4 Hz, 1 H), 6.93 (m, 1 H), 6.86 (d, J )
8.7 Hz, 2 H), 6.16 (s, 1 H), 3.86 (dd, J ) 10.2, 6.2 Hz, 1 H),
3.78 (s, 3 H), 3.71 (br s, 3 H), 3.60 (br s, 3 H), 2.73 (dd, J )
12.6, 6.2 Hz, 1 H), 2.36 (dd, J ) 12.6, 10.2 Hz, 1 H); 13C NMR
δ 171.1, 163.3, 140.5, 137.4, 133.2, 130.7, 130.3, 129.7, 129.5,
126.5, 125.1, 123.8, 117.68, 113.8, 84.7, 58.9, 55.5, 55.2, 52.7,
52.3, 52.1, 39.6. Anal. Calcd for C27H26N2O7SSe: C, 53.91; H,
4.36. Found: C, 54.02; H, 4.53.
N
in d -Met h yl-2-p h en ylselen yl-Nr-m et h oxyca r b on yl-L-
tr yp top h a n Meth yl Ester (28). Application of the general
protocol for cyclization to 27 yielded only the title compound,
as a white solid, in 85% yield: [R]20 ) + 25.3° (c ) 0.38,
D
1
CHCl3); H NMR δ 7.65 (d, J ) 7.8 Hz, 1 H), 7.04-7.35 (m, 8
H), 5.20 (d, J ) 7.7 Hz, 1 H), 4.66 (br dd, J ) 13.9, 6.6 Hz, 1
H), 3.73 (s, 3 H), 3.67 (s, 3 H), 3.59 (s, 3 H), 3.44 (ABX system,
J ) 29.4, 14.5, 6.6 Hz, 2 H); 13C NMR δ 172.7, 156.5, 138.7,
132.2, 129.7, 128.8, 127.3, 126.6, 124.7, 123.4, 120.0, 119.4,
117.9, 110.2, 54.9, 52.6, 52.4, 31.8, 29.4. Anal. Calcd for
based on recovered 11) as an exo/endo inseparable mixture
(exo/endo ca. 9:1): [R]20 ) -78.4° (c ) 1.9, CHCl3); H NMR
1
D
δ endo (12 taken from mixture) 7.36-6.94 (m, 9 H), 6.31 (s, 1
H), 4.42 (br d, J ) 7.2 Hz, 1 H), 3.07 (s, 3 H), 2.91 (d, J ) 12.4
Hz, 1 H), 2.60 (dd, J ) 12.4, 9.6 Hz, 1 H), 1.54 (s, 9 H), 1.40
(br s, 9 H); 1H NMR δ exo (13 taken from mixture) 7.36-6.94
(m, 9 H), 6.26 (s, 1 H), 3.89 (dd, J ) 9.8, 6.6 Hz, 1 H), 3.63 (s,
3 H), 2.88 (dd, J ) 12.4, 6.6 Hz, 1 H), 2.36 (dd, J ) 12.4, 6.6
Hz, 1 H), 1.54 (s, 9 H), 1.33 (br s, 9 H); 13C NMR exo/endo
mixture δ 172.4 (br), 152.2, 142.0, 137.6, 132.6, 129.5, 129.3,
129.0, 126.2, 123.5, 117.3 (br), 82.6 (br), 81.7, 80.8 (br), 60.5,
59.2, 54.7 (br), 52.3, 52.0, 38.7, 28.4 (br). Anal. Calcd for
C
21H22N2O4Se: C, 56.63; H, 4.98. Found: C, 56.72; H, 5.01.
Hexa h yd r op yr r olin d ole 31. Gen er a l P r otocol for Re-
d u ction w ith Tr ibu tyltin Hyd r id e. n-Bu3SnH (0.568 mL,
2.04 mmol) was added to a solution of diphenyl diselenide23
(47.7 mg, 0.15 mmol, 15 mol %) in degassed toluene (5.0 mL)
at room temperature under Ar. The yellow solution was stirred
at room temperature for 30 min until the system became
almost colorless. Then a solution of 19 (0.481 g, 1.02 mmol)
and AIBN (25.6 mg, 0.153 mmol, 15 mol %) in degassed toluene
(2.0 mL) was added dropwise. The resulting solution was
refluxed for 12 h before it was cooled to room temperature,
diluted with EtOAc (30 mL) and H2O (15 mL), and extracted
with EtOAc (3 × 15 mL). The combined organic phases were
washed with water (10 mL) and brine (10 mL) and dried over
Na2SO4. The solvent was removed under vacuum, the residue
C
28H34N2O6Se‚3/2H2O: C, 56.00; H, 6.21. Found: C, 56.34; H,
5.94.
Hexa h yd r op yr r olin d oles 15 a n d 16 were obtained in
76% yield as a 12/1 exo/endo mixture from which the pure exo
isomer (16), a glass, was obtained by column chromatography.
1
16: [R]20 ) -74.5° (c ) 1.73, CHCl3); H NMR δ 7.03-7.45
D
(m, 9 H), 6.32 (s, 1 H), 5.24 (br s, 1 H), 5.14 (br s, 1 H), 3.99
(dd, J ) 9.7, 6.7 Hz, 1 H), 3.69 (s, 3 H), 3.46 (br s, 3 H), 2.98
(dd, J ) 12.6, 6.7 Hz, 1 H), 2.44 (dd, J ) 12.6, 9.7 Hz, 1 H);
13C NMR δ 171.9, 152.9, 141.3, 137.5, 136.0, 132.6, 129.5,
128.9, 128.8, 128.5, 125.6, 124.4, 123.6, 117.6, 83.1, 67.9, 59.2,
52.8, 52.6, 38.2. Anal. Calcd for C28H26N2O6Se‚H2O: C, 57.64;
H, 4.84. Found: C, 57.88; H, 4.74. 15 was not isolated pure,
but key data were taken from a mixture: 1H NMR δ 7.60-
6.96 (m, 14 H), 6.37 (s, 1 H), 5.26 (br s, 1 H), 5.17 (br s, 1 H),
4.53 (br d, 1 H), 3.58 (br s, 3 H), 3.09 (s, 3 H), 3.01 (d, J ) 12.4
Hz, 1 H), 2.69 (dd, J ) 12.4, 6.6 Hz, 1 H).
24
dissolved in methanol (3.0 mL), and NaBH4 (77.5 mg, 2.04
mmol) added. After the mixture was stirred for 5-10 min at
room temperature, the methanol was removed in vacuo and
the residue subjected to flash chromatography on silica gel
(hexane/EtOAc 1:2) to afford the title compound as a white
solid (0.31 g, 90%): [R]20D ) -111.2° (c ) 0.17, CHCl3); 1H NMR
δ 7.64 (d, J ) 8.0 Hz, 1 H), 7.24 (t, J ) 7.7 Hz, 1 H), 7.17 (d,
J ) 7.4 Hz, 1 H), 7.05 (dt, J ) 7.4, 0.9 Hz, 1 H), 6.35 (d, J )
5.8 Hz, 1 H), 4.00 (m, 1 H), 3.86 (s, 3 H), 3.72 (s, 3 H), 3.64 (s,
3 H), 2.59 (dd, J ) 12.8, 7.1 Hz, 1 H), 2.31 (ddd, J ) 12.8, 9.6,
7.1 Hz, 1 H); 13C NMR δ 172.8, 154.0, 141.7, 131.5, 128.8,
124.1, 123.6, 117.3, 77.1, 59.1, 53.1, 52.9, 52.5, 45.4, 32.6. Anal.
Calcd for C16H18N2O6‚1/2H2O: C, 55.98; H, 5.58. Found: C,
56.08; H, 5.66.
Hexa h yd r op yr r olin d oles 18 a n d 19 were obtained in
83% yield as an 11/1 exo/endo mixture from which both isomers
were obtained pure by column chromatography. 19: mp 57-
1
59 °C; [R]20 ) -100.6° (c ) 2.76, CHCl3); H NMR δ 7.07-
D
Hexa h yd r op yr r olin d oles 29 a n d 30. Reduction of a 1:9
mixture of 12 and 13 according to the general protocol gave
94% of a 1:9 mixture of 29 and 30, in the form of a white
foam: mp 53-55 °C; [R]20D ) -58.8° (exo/endo c ) 3.0, CHCl3);
1H NMR δ endo (29 taken from mixture) 6.41 (d, J ) 6.7 Hz,
1 H), 4.56 (d, J ) 8.8 Hz, 1 H), 3.13 (s, 3 H), 1.54 (s, 9 H), 1.46
(s, 9 H); exo (30 taken from mixture) 6.37 (d, J ) 5.9 Hz, 1 H),
3.95 (dd, J ) 9.9, 7.1 Hz, 1 H), 3.72 (s, 3 H), 2.54 (dd, J )
12.7, 7.1 Hz, 1 H), 2.28 (ddd, J ) 12.9, 9.9, 7.1 Hz, 1 H), 1.57
(s, 9 H), 1.40 (s, 9 H); exo/endo (overlapping data) 7.53 (br s,
7.23 (m, 9H), 6.30 (s, 1 H), 4.00 (dd, J ) 9.6, 6.8 Hz, 1 H), 3.78
(br s, 3 H), 3.69 (s, 3 H), 3.61 (br s, 3 H), 2.98 (dd, J ) 12.7,
6.8 Hz, 1 H), 2.44 (dd, J ) 12.7, 9.6 Hz, 1 H); 13C NMR δ 171.7,
153.2, 141.1, 137.2, 132.2, 129.3, 128.7, 125.4, 124.1, 123.4,
117.3, 83.1, 58.9, 52.8, 52.4, 38.0. Anal. Calcd for C22H22N2O6-
Se‚1/2H2O: C, 53.02; H, 4.65. Found: C, 52.66; H, 4.76. 18:
[R]20 ) +12.1° (c ) 1.17, CHCl3); 1H NMR δ 7.13-7.33 (m, 8
D
H), 7.03 (t, J ) 7.5 Hz, 1 H), 6.32 (s, 1 H), 4.51 (br d, 1 H),
3.76 (s, 3 H), 3.69 (br s, 3 H), 3.07 (s, 3H), 3.01 (d, J ) 12.9
Hz, 1 H), 2.67 (dd, J ) 12.9, 9.1 Hz, 1 H); 13C NMR δ 170.9,
153.3, 137.3, 129.4, 129.3, 128.8, 125.7, 123.9, 123.7, 117.1,
83.6, 59.3, 52.8, 52.0, 39.0. Anal. Calcd for C22H22N2O6Se‚
1/2H2O: C, 53.02; H, 4.65. Found: C, 53.43; H, 4.66.
(23) Crich, D.; Yao, Q. J . Org. Chem. 1995, 60, 84-88.
(24) Crich, D.; Sun, S. J . Org. Chem. 1996, 61, 7200-7201.

Alkylation of Tryptophan with Inversion of Configuration
J . Org. Chem., Vol. 64, No. 19, 1999 7223
1 H), 7.26-6.96 (m, 3 H), 3.96 (m, 1 H); 13C NMR (exo/endo) δ
173.1, 152.3, 142.3, 132.0, 128.2, 123.4, 117.6, 81.5, 80.7, 76.7,
58.9, 52.1, 44.7, 32.8, 28.3, 28.1. Anal. Calcd for C22H30N2O6‚
0.5H2O: C, 61.80; H, 7.31. Found: C, 61.77; H, 7.36.
phases were washed with water (10 mL) and brine (10 mL)
and dried over Na2SO4. The solvent was removed under
vacuum and the residue purified by flash chromatography on
silica gel (hexane/EtOAc 1:1) to afford 33 (65 mg, 54%) and
Hexa h yd r op yr r olin d ole 32. Reduction of 25 according to
35 (17 mg, 21%) both as colorless oils. 33: [R]20 ) -40.0° (c
D
the general protocol gave 74% of 32:13 [R]20 ) -167.2° (c )
) 2.73, CHCl3); 1H NMR δ 7.11-7.48 (m, 8 H), 7.02 (t, J ) 7.3
Hz, 1 H), 6.35 (s, 1 H), 3.74 (s, 3 H), 3.63 (s, 3 H), 3.18 (d, J )
13.3 Hz, 1 H), 2.97 (s, 3 H), 2.43 (d, J ) 13.3 Hz, 1 H), 1.62 (s,
3 H); 13C NMR δ 173.4, 154.0, 153.8, 142.7, 137.5, 129.6, 129.4,
129.3, 128.9, 125.9, 123.9, 123.8, 123.6, 117.5, 85.1, 66.6, 53.1,
52.5, 52.3, 48.3, 25.7. Anal. Calcd for C23H24N2O6Se: C, 54.88;
H, 4.80. Found: C, 54.55; H, 4.95. 35: [R]20D ) -2.7° (c ) 1.07,
CHCl3); 1H NMR δ 7.94 (m, 1 H), 7.18-7.26 (m, 3 H), 3.95 (s,
3 H), 3.76 (s, 6 H), 3.43 (d, J ) 14.8 Hz, 1 H), 2.92 (d, J ) 14.8
Hz, 1 H), 1.85 (s, 3 H); 13C NMR δ 172.6, 153.3, 151.6, 140.3,
138.8, 125.3, 123.7, 122.7, 117.7, 114.9, 104.7, 75.9, 54.3, 53.3,
53.0, 38.4, 24.5. Anal. Calcd for C17H18N2O6‚1/2H2O: C, 57.46;
H, 5.39. Found: C, 57.95; H, 5.36.
D
1
0.75, CHCl3); H NMR δ 7.60 (d, J ) 8.0 Hz, 2 H), 7.58 (br s,
1 H), 7.28 (t, J ) 7.6 Hz, 1 H), 7.12 (t, J ) 7.2 Hz, 1 H), 7.03
(d, J ) 7.2 Hz, 1 H), 6.80 (d, J ) 8.7 Hz, 2 H), 6.12 (d, J ) 4.8
Hz, 1 H), 3.97 (br t, 1 H), 3.79 (s, 6 H), 3.72 (s, 3 H), 3.40 (br
m, 1 H), 2.37 (ddd, J ) 12.8, 6.9, 2.2 Hz, 1 H), 2.22 (dt, J )
12.8, 7.9 Hz, 1 H).
Equ ilibr a tion of a 1:9 Mixtu r e of 12 a n d 13 in ter t-
Bu tyl Alcoh ol-d 10 w ith Ca ta lytic P ota ssiu m ter t-Bu tox-
id e. Equilibration of a 1:9 mixture of 12 and 13, as described
below for 19, led to a 70/30 endo/exo mixture of 12:13 and their
enantiomers with complete exchange of the enolizable protons
in all products. The specfic rotation of the final mixture was
[R]D ) - 34° (c 3.0, CHCl3): 1H NMR δ endo ((12-d1 taken
from mixture) 7.36-6.93 (m, 9 H), 6.31 (s, 1 H), 3.08 (s, 3 H),
2.92 (d, J ) 12.9 Hz, 1 H), 2.61 (d, J ) 12.9 Hz, 1 H), 1.55 (s,
9 H), 1.41 (br s, 9 H); δ exo ((13-d1 taken from mixture) 7.36-
6.93 (m, 9 H), 6.27 (s, 1 H), 3.65 (s, 3 H), 2.89 (d, J ) 12.4 Hz,
1 H), 2.37 (d, J ) 12.4 Hz, 1H), 1.55 (s, 9 H), 1.34 (s, 1 H).
Equ ilibr a tion of 19 in ter t-Bu tyl Alcoh ol-d 10 w ith
Ca ta lytic P ota ssiu m ter t-Bu toxid e. 19 (11.2 mg, 0.023
mmol) in t-BuOH-d10 (0.55 mL) in an NMR tube was treated
with t-BuOK powder (ca. 20 mol %) under Ar at 40 °C. An 1H
NMR spectrum taken after 15 min showed isomerization. A
further spectrum recorded after 45 min, 22, 60, and 70 h
showed no difference to the one taken after 15 min. The
reaction mixture was filtered through a short silica gel column
(0.5 × 0.5 cm), eluting with CHCl3. The solvent was removed
under vacuum and the mixture reexamined by 1H NMR
spectroscopy in CDCl3 when it was determined that 100% d
substitution had occurred at the enolizable carbon of both
Hexa h yd r op yr r oloin d ole 34. Alkylation of 19 with p-
bromobenzyl bromide according to the general protocol gave
46% yield of 34, as a colorless oil: [R]20 ) -31.0° (c ) 1.16,
D
1
CHCl3); H NMR δ 7.01-7.59 (m, 11 H), 6.83 (d, J ) 7.9 Hz,
2 H), 5.89 (s, 1 H), 3.82 (s, 3 H), 3,72 (s, 3 H), 3.04 (d, J ) 13.4
Hz, 1 H), 3.02 (s, 3 H), 2.93 (d, J ) 14.9 Hz, 2 H), 2.48 (d, J )
13.4 Hz, 1 H); 13C NMR δ 173.3, 154.2, 138.1, 137.5, 134.2,
132.5, 131.8, 129.7, 129.5, 129.3, 125.9, 124.1, 123.6, 121.4,
117.7, 84.1, 70.0, 53.2, 52.6, 52.4, 42.2, 38.3. Anal. Calcd for
C
29H27BrN2O6Se‚3/2H2O: C, 50.82; H, 4.41. Found: C, 50.76;
H, 4.33.
D-Tr yp top h a n 36. 33 (0.047 g, 0.1 mmol) was stirred in
TFA (2.0 mL) at room temperature for 4 h followed by removal
of the solvent in vacuo. The 1H NMR spectrum showed
complete conversion to 36, which could subsequently be
isolated by filtration on silica gel, as a colorless oil: [R]20
)
D
1
-42.9° (c ) 0.68, CHCl3); H NMR δ 8.12 (br d, 1 H), 7.48 (d,
J ) 7.6 Hz, 1 H), 7.21-7.33 (m, 3 H), 5.48 (s, 1 H), 4.02 (s, 3
H), 3.70 (s, 3 H), 3.68 (s, 3 H), 3.54 (d, J ) 14.4 Hz, 1 H), 3,31
(d, J ) 14.4 Hz, 1 H), 1.68 (s, 3 H); 13C NMR δ 174.1, 155.3,
135.0, 130.9, 124.6, 124.0, 122.7, 119.0, 115.9, 115.1, 77.1, 60.3,
53.7, 52.7, 51.8, 31.4, 23.7. Anal. Calcd. for C17H20N2O6‚1/
2H2O: C, 57.14; H, 5.92. Found: C, 56.86; H, 5.86.
products and that the endo/exo ratio was 80:20: [R]20
)
D
1
-37.5° (c ) 0.73, CHCl3); H NMR δ endo (18-d1 taken from
mixture) 7.44-6.98 (m, 9 H), 6.35 (s, 1 H), 3.78 (s, 3 H), 3.70
s, 3 H), 3.10 (s, 3 H), 3.00 (d, J ) 13.0 Hz, 1 H), 2.67 (d, J )
13.0 Hz, 1 H); δ exo (19-d1 taken from mixture) 7.44-6.98 (m,
9 H), 6.31 (s, 1 H), 3.79 (s, 3 H), 3.69 (s, 3 H), 3.61 (s, 3 H),
2.97 (d, J ) 12.9 Hz, 1 H), 2.45 (d, J ) 12.9 Hz, 1 H).
D-Tr yp top h a n 37. Treatment of 34 with TFA, as described
above for 36, gave 94% of 37 in the form of an amorphous white
1
Equ ilibr a tion of 25 in ter t-Bu tyl Alcoh ol-d 10 w ith
Ca ta lytic P ota ssiu m ter t-Bu toxid e. Equilibration of 25, as
described above for 19, led to an 86/14 endo/exo mixture with
complete exchange of the enolizable protons in both products:
solid: mp 115-117 °C; [R]20 ) -15.3° (c ) 0.51, CHCl3); H
D
NMR δ 8.11 (br d, 1 H), 7.50 (d, J ) 7.6 Hz, 1 H), 7.37 (d, J )
8.3 Hz, 2 H), 7.20-7.33 (m, 3 H), 6.93 (d, J ) 8.3 Hz, 2 H),
5.58 (s, 1 H), 4.02 (s, 3 H), 3.95 (d, J ) 13.6 Hz, 1 H), 3.89 (d,
J ) 13.6 Hz, 1 H), 3.73 (s, 3 H), 3.65 (s, 3 H), 3.30 (d, J ) 31.0
Hz, 1 H), 3.26 (d, J ) 31.0 Hz, 1 H); 13C NMR δ 172.2, 155.2,
135.0, 131.2, 130.7, 124.6, 124.0, 122.6, 121.0, 119.0, 115.7,
[R]20 ) +6.7° (c ) 0.46, CHCl3); 1H NMR δ endo (24-d1 taken
D
from mixture) 7.89 (d, J ) 8.9 Hz, 2 H), 7.50 (d, J ) 6.9 Hz, 2
H), 7.38-6.90 (m, 9 H), 6.31 (s, 1 H), 3.82 (s, 3 H), 3.52 (s, 3
H), 3.16 (s, 3 H), 2.89 (d, J ) 12.9 Hz, 1 H), 2.59 (d, J ) 12.9
Hz) δ exo (25-d1 taken from mixture) 7.89 (d, J ) 8.9 Hz, 2
H), 7.50 (d, J ) 6.9 Hz, 2 H), 7.38-6.90 (m, 9 H), 6.17 (s, 1 H),
3.80 (s, 3 H), 3.70 (s, 3 H), 3.61 (s, 3 H), 2.72 (d, J ) 12.9 Hz,
1 H), 2.40 (d, J ) 12.9 Hz, 1 H).
115.1, 65.9, 53.7, 52.6, 51.9, 40.4, 30.9. Anal. Calcd for C23H23
BrN2O6: C, 54.88; H, 4.60; Found: C, 54.60; H, 4.80.
-
Ack n ow led gm en t. We thank the University of
Illinois at Chicago for support of this work and for a
University Fellowship to X.H. We thank Professor S. J .
Danishefsky for helpful and stimulating discussion.
Gen er a l P r oced u r e for Alk yla tion w ith In ver sion of
Con figu r a tion . Hexa h yd r op yr r olin d ole 33 a n d Tetr a h y-
d r op yr r olin d ole 35. To a solution of 19 (0.115 g, 0.24 mmol)
in THF (4 mL) LDA (0.96 mL, 0.48 mmol, 0.5 M in THF) was
added dropwise at -78 °C. After the solution was stirred at
-78 °C for 0.5 h, MeI (0.075 mL, 1.2 mmol, 5 equiv) was
added dropwise. After 3 h, the reaction was quenched with
saturated NH4Cl solution (15 mL) and EtOAc (30 mL) and then
extracted with EtOAc (3 × 15 mL). The combined EtOAc
Su p p or tin g In for m a tion Ava ila ble: Procedures for the
preparation of 11, 14, 17, 20, 23, and 27 together with full
characterization data. This material is available free of charge
via the Internet at http://pubs.acs.org.
J O991093+