Diastereoselective intramolecular α-amidoalkylation reactions of L-DOPA derivatives. Asymmetric synthesis of pyrrolo[2,1-a]isoquinolines

Article abstract of DOI:10.1021/jo051584w

Stereocontrolled intramolecular α-amidoalkylation reactions of L-DOPA-derived succinimides have been studied. Addition of MeLi to nonracemic succinimides 9a-d yields oxoamides, which are cyclized upon treatment with Lewis or protic acids to afford (5S,10b

Full text of DOI:10.1021/jo051584w

Diastereoselective Intramolecular r-Amidoalkylation Reactions of
L-DOPA Derivatives. Asymmetric Synthesis of
Pyrrolo[2,1-a]isoquinolines
Eva Garc´ıa, Sonia Arrasate, Esther Lete,* and Nuria Sotomayor
Departamento de Quı´mica Orga´nica II, Facultad de Ciencia y Tecnolog´ıa, Universidad del Pa´ıs Vasco/
Euskal Herriko Unibertsitatea, Apdo 644, 48080 Bilbao, Spain
esther.lete@ehu.es
Received July 29, 2005
Stereocontrolled intramolecular R-amidoalkylation reactions of L-DOPA-derived succinimides have
been studied. Addition of MeLi to nonracemic succinimides 9a-d yields oxoamides, which are
cyclized upon treatment with Lewis or protic acids to afford (5S,10bS)-trans 11 or (5S,10bR)-cis 12
pyrroloisoquinolines with variable diastereoselectivities, but with high enantiomerical purities (ee
99%).
Introduction
N-Acyliminium ions are extremely versatile intermedi-
ates in organic synthesis. In the context of the intramo-
lecular R-amidoalkylation reactions in carbon-carbon
bond-forming processes, considerable work has been done
in the area of reactions of cyclic N-acyliminium ions with
π-nucleophiles, leading to alkaloids and other nitrogen-
containing biologically active compounds.1 Diastereose-
lectivity in these intramolecular R-amidoalkylations can
be achieved by the steric control of the substituents
already present in the cyclic N-acyliminium ion (R1) or
along the chain connecting the π-nucleophiles and the
nitrogen atom (R2) (Figure 1).2
FIGURE 1. Intramolecular reaction of cyclic N-acyliminium
ions with π-nucleophiles.
For example, we have described a tandem organo-
lithium addition, N-acyliminium ion cyclization sequence
(2) For reviews, see ref 1. For some selected examples of diastereo-
selective intramolecular R-amidoalkylation reactions of cyclic N-
acyliminium ions, see the following references. Steric control by the
substituents in the ring: (a) Ostendorf, M.; van der Neut, S.; Rutjes,
F. P. J. T.; Hiemstra, H. Eur. J. Org. Chem. 2000, 65, 115-124. (b)
Consonni, A.; Danieli, B.; Lesma, G.; Passarella, D.; Piacenti, P.;
Silvani, A. Eur. J. Org. Chem. 2001, 66, 1377-1383. (c) Hwang, D. J.;
Kang, S. S.; Lee, J. Y.; Choi, J. H.; Park, H.; Lee, Y. S. Synth. Commun.
2002, 32, 2499-2505. (d) Padwa, A. Pure Appl. Chem. 2003, 75, 47-
62. (e) Kaluza, Z.; Mostowicz, D. Tetrahedron: Asymmetry 2003, 14,
225-232. (f) Mostowicz, D.; Wo´jcik, R.; Dolega, G.; Kahuza, Z.
Tetrahedron Lett. 2004, 45, 6011-6015. (g) Siwicka, A.; Wojasiewicz,
K.; Rosiek, B.; Leniewski, A.; Maurin, J. K.; Czarnocki, Z. Tetrahe-
dron: Asymmetry 2005, 16, 975-993. (h) Hanessian, S.; Tremblay, M.;
Marzi, M.; del Valle, J. R. J. Org. Chem. 2005, 70, 5070-5085. Steric
control by the substituents along the chain connecting the π-nucleo-
philes and the nitrogen atom: (i) Bahajaj, A. A.; Vernon, J. M.; Wilson,
G. D. J. Chem. Soc., Perkin Trans. 1 2001, 1446-1451. (j) Katritzky,
A. R.; Mehta, S.; He, H.-Y. J. Org. Chem. 2001, 66, 148-152. (k)
Nielsen, T. E.; Meldal, M. J. Org. Chem. 2004, 69, 3765-3773. (l)
Ardeo, A.; Garc´ıa, E.; Arrasate, S.; Lete, E.; Sotomayor, N. Tetrahedron
Lett. 2003, 44, 8445-8448.
* Corresponding author. Phone: 34 94 6012576. Fax: 34 94
6012748.
(1) For reviews on N-acyliminium ion chemistry, see: (a) Speckamp,
W. N.; Hiemstra, H. Tetrahedron 1985, 41, 4367-4416. (b) Hiemstra,
H.; Speckamp, W. N. In The Alkaloids; Brossi, A., Ed.; Academic
Press: New York, 1988; Vol. 32, pp 271-339. (c) Hiemstra, H.;
Speckamp, W. N. In Comprehensive Organic Synthesis; Trost, B. M.,
Fleming, I., Eds.; Pergamon Press: Oxford, 1991; Vol. 2, pp 1047-
1082. (d) de Koning, H.; Speckamp, W. N. In Stereoselective Synthesis
[Houben-Weyl]; Helmchen, G., Hoffmann, R. W., Muzler, J., Schau-
mann, E., Eds.; Thieme: Stuttgart, 1996; Workbench ed. E21, Vol. 3,
pp 1952-2010. (e) Speckamp, W. N.; Moolenaar, M. J. Tetrahedron
2000, 56, 3817-3856. (f) Royer, J.; Bonin, M.; Micouin, L. Chem. Rev.
2004, 104, 2311-2352. (g) Maryanoff, B. E.; Zhang, H.-C.; Cohen, J.
H.; Turchi, I. J.; Maryanoff, C. A. Chem. Rev. 2004, 104, 1431-1628.
(h) Dobbs, A. P.; Rossiter, S. In Comprehensive Organic Functional
Group Transformations II; Katritzky, A. R., Taylor, R. J. K., Eds.;
Elsevier: Oxford, 2005; Vol. 3, pp 419-450.
10.1021/jo051584w CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/12/2005
10368
J. Org. Chem. 2005, 70, 10368-10374

Asymmetric Synthesis of Pyrroloisoquinolines
SCHEME 1a
with N-phenethylimides, which constitutes an effective
route to several types of isoquinoline alkaloids.3 Our
approach can be used to generate cyclic N-acyliminium
ions with a substituent adjacent to the iminium carbon,
whose cyclization leads to the diastereoselective synthesis
of 1,10b-cis thiazoloisoquinolines.4 This methodology has
recently been directed toward the enantioselective syn-
thesis of pyrrolo[2,1-a]isoquinolones via stereocontrolled
R-amidoalkylation reactions starting from an enantiopure
N-phenethylnorborn-5-en-2,3-dicarboximide with a 2-exo-
hydroxy-10-bornylsulfinyl group as chiral auxiliary.5
We were also interested in the effect of a stereogenic
center R to the nitrogen atom on this type of N-acylimin-
ium cyclizations. Thus, in a preliminary communication,
we have reported that addition of organolithiums to
chiral nonracemic L-DOPA-derived succinimide yields
oxoamides, which are cyclized diastereoselectively upon
treatment with BF3‚OEt2, to afford 5,10b-trans pyrrolo-
isoquinolones in high ee (99%), but in moderate overall
yields (34-36%).6 Although the resulting pyrroloisoqui-
nolones were isolated as single 5,10b-trans diastereomers,
not detecting the presence of the corresponding cis
diastereomers, the moderate yields obtained did not allow
us to conclude that the cyclization reactions were com-
pletely diastereoselective. Shortly thereafter, Allin and
co-workers7 reported a related stereoselective approach
to the pyrroloisoquinoline ring system based on the
intramolecular R-amidoalkylation reaction of a bicyclic
lactam derived from (S)-phenylalaninol. In this case, the
cyclization with TiCl4 led to a mixture of product dias-
tereoisomers in high yield (87%) in a 2:1 ratio also in
favor of the 5,10b-trans product.
a Reagents: (a) (Boc)2O, 50% dioxane/H2O, Et3N, rt. (b) Me2SO4,
K2CO3, acetone, reflux. (c) LiAlH4, 50% Et2O/THF, reflux. (d)
TBDPSiCl, imidazole, CH2Cl2, rt (for 5a); BnBr, NaH, THF, 0 °C
to rt (for 5b); KOH, DMSO, MeI, rt (for 5c). (e) TFA, CH2Cl2, rt.
(f) Succinic anhydride, Et2O, reflux. Then, Ac2O, NaOAc, 85 °C.
(g) TBAF, THF, 0 °C to rt.
Results and Discussion
With these precedents we decided to reinvestigate the
stereoselectivity of this type of intramolecular R-ami-
doalkylation reaction, using of a series of L-DOPA-derived
N-phenethylsuccinimides with different substituents in
the R-position to the nitrogen atom, and protic and Lewis
acids to carry out the intramolecular R-amidoalkylation
reaction. We disclose here full details of our investiga-
tions that ultimately led to a successful approach toward
the asymmetric synthesis of pyrrolo[2,1-a]isoquinolones.8
Our first task was the synthesis of N-phenethylsuc-
cinimides 9a-d, which were prepared from L-DOPA 1
by standard functional group manipulation, as outlined
in Scheme 1. Thus, protection of the amino group of
L-DOPA 1 with (Boc)2O, followed by methylation with
dimethyl sulfate, yielded methyl ester 3, which was
submitted to LiAlH4 reduction to give the phenylalanilol
4. Derivatization of the primary hydroxyl group was
achieved using TBDPSCl, benzyl bromide, and io-
domethane to give ethers 5a-c.9 Hydrolysis of Boc
amides 5a-c with TFA in dichloromethane at room
temperature afforded the corresponding amines 8a-c.
Treatment of these amines with succinic anhydride led
to the corresponding amido acids, which, without further
purification, were cyclized with Ac2O and NaAcO to give
succinimides 9a-c in good yields. Finally, 9d was
prepared by hydrolysis of the O-silyl ether of imide 9a
with TBAF. Thus, succinimides 9a-d were prepared in
moderate to good overall yields from L-DOPA without
epimerization of the stereogenic center, as imides 9a-d
showed an enantiomeric excess equal to starting amino
acid (99% ee), determined by chiral phase HPLC.
(3) (a) Lete, E.; Egiarte, A.; Sotomayor, N.; Vicente, T.; Villa, M. J.
Synlett 1993, 41-42. (b) Collado, M. I.; Lete, E.; Sotomayor, N.; Villa,
M. J. Tetrahedron 1995, 51, 4701-4710. (c) Collado, M. I.; Sotomayor,
N.; Villa, M. J.; Lete, E. Tetrahedron Lett. 1996, 37, 6193-6196. (d)
Collado, M. I.; Manteca, I.; Sotomayor, N.; Villa, M. J.; Lete, E. J. Org.
Chem. 1997, 62, 2080-2092. (e) Osante, I.; Garc´ıa, E.; Ardeo, A.;
Arrasate, S.; Lete, E.; Sotomayor, N. Recent Res. Dev. Org. Chem. 2002,
6, 103-111.
(4) (a) Osante, I.; Collado, M. I.; Lete, E.; Sotomayor, N. Synlett 2000,
101-103. (b) Osante, I.; Collado, M. I.; Lete, E.; Sotomayor, N. Eur. J.
Org. Chem. 2001, 1267-1277.
(5) (a) Gonza´lez-Temprano, I.; Sotomayor, N.; Lete, E. Synlett 2002,
593-597. (b) Gonza´lez-Temprano, I.; Osante, I.; Sotomayor, N.; Lete,
E. J. Org. Chem. 2004, 69, 3875-3885.
(6) Garc´ıa, E.; Arrasate, S.; Ardeo, A.; Lete, E.; Sotomayor, N.
Tetrahedron Lett. 2001, 42, 1511-1513.
(7) (a) Allin, S. M.; James, S. L.; Martin, W. P.; Smith, T. A. D.
Tetrahedron Lett. 2001, 42, 3943-3946. (b) Allin, S. M.; James, S. L.;
Martin, W. P.; Smith, T. A. D.; Elsegood, M. R. J. J. Chem. Soc., Perkin
Trans. 1 2001, 3029-3036.
(9) O-Silylation of 4 was carried out with imidazole and TBDPSCl
to give 5a in excellent yield (95%). However, the yields of the benzyl
and methyl ethers 5b and 5c were only moderate due to the formation
of side products 6 (35%) and 7 (30%), respectively. Several attempts
to improve these yields failed. See Supporting Information.
(8) Pyrrolo[2,1-a]isoquinolones have attracted considerable interest
because they possess antidepressant, muscarinic agonist, antiplatelet,
and anticancer activity. Moreover, they can be used as PET radiotrac-
ers for imaging serotonin uptake sites. The importance of these
nitrogen heterocycles is further enhanced by their utility as advanced
intermediates for the synthesis of alkaloids. For a general review on
properties, synthesis, and reactivity of pyrrolo[2,1-a]isoquinolines,
see: Mikhailovskii, A. G.; Shklyaev, V. S. Chem. Heterocycl. Compd.
1997, 33, 243-265; Transl. Khim. Geterotsikl. Soedin.
J. Org. Chem, Vol. 70, No. 25, 2005 10369

Garc´ıa et al.
SCHEME 2a
any of the byproducts formed. Optimization experiments
have allowed us to enhance the yield of pyrroloisoquino-
line 11d to 92% (entry 1) and to isolate the minor 5,10b-
cis diastereomer 12d (R2 ) H). The main experimental
differences were the portionwise addition of a larger
excess of the Lewis acid and longer reaction times. Thus,
BF3‚Et2O (12 mmol) was added to a solution of 10a in
CH2Cl2 at 0 °C under argon atmosphere, and after
allowing the reaction mixture to reach room temperature,
it was heated under reflux for 4 days. Then, the reaction
mixture was cooled, treated again with BF3‚Et2O (12
mmol), and refluxed for 3 days, affording the pyrroloiso-
quinolones 11d/12d as a trans/cis diastereomer mixture
in a 2:1 ratio in favor of the trans isomer (see Table 1,
entries 1-3). The yields are given for isolated and
purified products, since the diastereomers were separable
on a preparative scale by flash column chromatography.
Both isomers were stable, and we have not observed any
equilibration under the reaction conditions. Additionally,
the cyclization took place without epimerization since
both diastereomers were enantiopure (99% ee).11
The formation of 11d (R2 ) H) under these reaction
conditions suggested that the silyl ether was hydrolyzed
to alcohol prior to cyclization, since Allin and co-workers7b
obtained the same level of diastereoselectivity in the
cyclization with TiCl4 of a similar substrate that has a
free hydroxyl group. In fact, monitoring of the reaction
by 1H NMR spectroscopy allowed us to detect the
unprotected oxoamide 10d. Moreover, on treating oxoa-
mide 10d (R1 ) H, Table 1, entry 3) with BF3‚Et2O as
Lewis acid, we obtained the diastereomeric mixture of
pyrroloisoquinolines 11d/12d in the same ratio and
comparable yield.
With these results in hand, we pursued the N-acylimin-
ium cyclization of O-alkyl-substituted oxoamide 10b (R1
) Bn). However, under a range of conditions, the reaction
was sluggish, and we could only isolate the trans-
pyrroloisoquinoline 11d (R2 ) H) in low yield (33%).
Because of the low yield obtained, the diastereomeric
ratio indicated in Table 1 is not indicative of a stereose-
lective cyclization. Therefore, we did not apply these
Lewis acid conditions to 10c (R1 ) Me).
Modest yields and same diastereoselectivities were
obtained when oxoamides 10a (R1 ) TBDPS), 10b (R1 )
Bn), and 10d (R1 ) H) were cyclized employing TiCl4 as
Lewis acid (entries 4-6). The best results were observed
when oxoamides were treated with TiCl4 (4 equiv) in CH2-
Cl2 at -78 °C for 1 h, at -10 °C for 3 h, and at room
temperature for 6 days. Despite the long reaction times
and the excess of Lewis acid used, in all cases the
conversion was low and unprotected oxoamide 10d (R1
) H) was isolated (10-60%), obtaining only low yields
of pyrroloisoquinolines 11d/12d (R2 ) H). The observed
diastereoselectivity (2:1) is identical to that obtained with
BF3‚Et2O and the same for all substrates, which indicates
that deprotection occurs prior to cyclization.
a Reagents: (a) MeLi (4 equiv), -78 °C, 6 h.
TABLE 1. Intramolecular r-Amidoalkylation Reaction
of Oxoamides 10a-d
product
yield (%)
dr
entry
subs
acid
R2
11 trans/12 cis
1
2
10a
10b
10d
10a
10b
10d
10a
10b
10c
10d
BF3‚Et2Oa
BF3‚Et2Oa
BF3‚Et2Oa
92
H
2:1
>95:5
2:1
33
H
3
81
H
b
4
TiCl4
13d
15d
48d
67
H
2:1
b
5
TiCl4
H
2:1
b
6
TiCl4
H
2:1
7
TFAc
TFAc
TFAc
TFAc
H
1:1.3
1:1
8
92
Bn
Me
H
9
80
1:2
1:4
10
88
a (a) BF3‚Et2O (10 equiv), 0 °C. (b) Reflux, 4 days. (c) BF3‚Et2O
(10 equiv), reflux, 3 days. b (a) TiCl4 (4 equiv) at -78 °C. (b) f
-10 °C, 3 h. (c) f rt, 6 days. c TFA (34 equiv), reflux, 1 day.
d Variable yields (10-60%) of deprotected oxoamide 10d were
isolated.
The synthesis of key N-acyliminium ion precursors 10
was accomplished by reaction with organolithium re-
agents.3 Thus, succinimides 9a-d were treated with
MeLi to afford the corresponding oxoamides 10a-d
(Scheme 2). An excess of the organometallic (4 equiv) was
required to achieve the reported yields.10 Oxoamide 10d
was also prepared by hydrolysis of the silyl ether of 10a
with TBAF.
The results of the intramolecular R-amidoalkylation
reactions are listed in Table 1. We started studying
intramolecular R-amidoalkylation of oxoamide 10a (R1
) TBDPS) using BF3‚Et2O as Lewis acid to generate the
N-acyliminium ion. We had reported6 that when 10a was
treated with BF3‚OEt2 (12 equiv, reflux), 5,10b-trans
pyrroloisoquinolone 11d (R2 ) H) was isolated in high
ee (99%), but in moderate overall yields (34-36%).
Although complete conversion of oxoamide 10a was
observed, it was not possible to isolate and characterize
Then we decided to test protic acid conditions. Thus,
oxoamide 10a (R1 ) TBDPS) was treated with an excess
of TFA in refluxing CH2Cl2 for 3 h, isolating oxazolone
13 (14%), together with a low yield of pyrroloisoquinoline
(10) The use of 1, 2, or 3 equiv of MeLi led to lower yields of the
oxoamides and starting material. The requirement of more than
stoichiometric amounts of organolithiums has been attributed to
coordination of the organolithium to oxygenated groups. See: Sardina,
F. J.; Blanco, M. J. Org. Chem. 1996, 61, 4748-4755.
(11) All enantiomeric excesses were checked by chiral stationary
phase HPLC by comparison with the corresponding racemates, ob-
tained by analogous routes from racemic D,L-DOPA.
10370 J. Org. Chem., Vol. 70, No. 25, 2005

Asymmetric Synthesis of Pyrroloisoquinolines
SCHEME 3
11d (R2 ) H, 5%), and a mixture of starting material, its
cyclic tautomer, and dehydrated products (Scheme 3).
Because Allin and co-workers7b had used this oxazolone
as an acyliminium ion precursor, we thought that longer
reaction times would favor the final R-amidoalkylation
reaction. Thus, when oxoamide 10a (R1 ) TBDPS) was
treated with TFA for 1 day, a mixture of pyrroloisoquino-
lines 11d/12d (R2 ) H) was obtained in good yield (entry
7). Significantly, reversal of the stereochemistry was
observed and the (5S,10bR)-cis diastereomer 12d was the
major diastereomer obtained, although with poor dias-
tereoselectivity. These conditions were applied to oxoa-
mides 10b (R1 ) Bn) and 10c (R1 ) Me) (entries 8 and
9). Hydrolysis of methyl or benzyl ether did not occur,
and cyclization took place in high yields, but no signifi-
cant improvement in diastereoselectivity was observed.
Only with the unprotected amide 10d (R1 ) H) a
reasonable stereoselectivity was found obtaining the 12d-
cis (R2 ) H) diastereomer as the major product (4:1)
(entry 10).
In summary, the Lewis acid-promoted cyclization of
oxoamides 10a-d is moderately stereoselective, produc-
ing the trans (5S,10bS)-11d (R2 ) H) as the major
diastereomer in a 2:1 diastereomeric ratio. This fact,
together with the isolation of deprotected oxoamides,
indicates that deprotection occurs prior to cyclization, and
thus the protecting group has no relevance for the
stereochemical course of the reaction. However, when a
protic acid (TFA) is used, a reversal of the stereoselec-
tivity is observed, obtaining mainly the (5S,10bR)-cis
products 12. In these cases, the size of the substituent
at the R-position to the nitrogen atom seems to play an
important role. Thus, as the size of the C-5 substituent
decreases, the cis selectivity improves.
The stereochemical outcome of the cyclization may be
explained as a result of conformational factors in the
transition state. To explain these results, we propose the
conformational models and the transition states depicted
in Figure 2. In most examples described in the literature,
the most favored transition state would minimize A(1,3)
strain, according to Hart’s model.12 Thus, for intramo-
lecular cyclizations of π-nucleophiles on N-acyliminium
ions with substituents adjacent to the nitrogen atom, an
axial orientation is preferred to avoid A(1,3) strain between
the substituent and the carbonyl of the N-acyl group.13
This results in the preferred formation of cis diastereo-
mers of, for instance, 5-phenyltetrahydro[2,1-a]isoquino-
lones,14 5-aryl-10b-butyltetrahydro[2,1-a]isoquinolones,15
indolizidines,16 â-carboline derivatives,17 benzo[a]quino-
FIGURE 2. Stereochemical outcome of the R-amidoalkyl-
ation.
lizidines,18 or polycyclic isoindolinone derivatives.19 How-
ever, in all these cases, the iminium carbon is unsubsti-
tuted.
In our case, the iminium carbon bears a methyl group
and, as deprotection occurs prior to cyclization, the Lewis
acid would probably be coordinated to the hydroxyl group,
and thus it can be considered as a large group. The
acyliminium ion could adopt two possible conformations
A and B, which allow the attack of the aromatic ring from
the Re or Si faces, through transition states 1 and 2,
respectively (Figure 2). A balance between A(1,3) strain
in A and severe syn axial 1,3-interactions between both
substituents in the transition state 2 (TS2) would favor
the pseudo-equatorial disposition for C-5 substituent in
the transition state 1 (TS1). Thus, attack of the aromatic
ring onto the Re face of the N-acyliminium ion leads to
the observed stereochemistry, in which the substituent
in 10b (Me) assumes a pseudo-axial position leading to
trans diastereomer 11. Additionally, the formation of a
chelate with a Lewis acid, as TiCl4, cannot be ruled out.
In this context, Allin and co-workers7b had reported the
synthesis of pyrroloisoquinolines through a TiCl4-pro-
moted R-amidoalkylation reaction of oxazolones, which
are likely to proceed through the same N-acyliminium
intermediate depicted in Figure 2, obtaining the same
(12) Hart, D. J. J. Am. Chem. Soc. 1980, 102, 397-398.
(13) Mooiveer, H.; Hiemstra, H.; Speckamp, W. N. Tetrahedron
1989, 45, 4627-4636.
(14) Maryanoff, B. E.; McComsey, D. F.; Duhl-Emswiler, B. A. J.
Org. Chem. 1983, 48, 5062-5074.
(17) Heaney, H.; Taha, M. O. Tetrahedron Lett. 2000, 41, 1993-
1996.
(18) Bassas, O.; Llor, N.; Santos, M. M. M.; Griera, R.; Molins, E.;
Amat, M.; Bosch, J. Org. Lett. 2005, 7, 2817-2820.
(19) Allin, S. M.; Northfield, C. J.; Page, M. I.; Slawin, A. M. Z.
Tetrahedron Lett. 1998, 39, 4905-4908.
(15) Collado, M. I.; Lete, E.; Sotomayor, N.; Villa, M. J. Tetrahedron
1995, 51, 4701-4710.
(16) Chalard, P.; Remuson, R.; Gelas-Mialhe, Y.; Gramain, J.-C.;
Canet, I. Tetrahedron Lett. 1999, 40, 1661-1664.
J. Org. Chem, Vol. 70, No. 25, 2005 10371

Garc´ıa et al.
stereochemistry in each case, and both pyrroloisoquino-
lones were isolated with high enantiomeric purity (99%).
In summary, a diastereodivergent synthesis of enan-
tiomerically pure pyrrolo[2,1-a]isoquinolones via tandem
organolithium addition-N-acyliminium ion cyclization
has been achieved. The stereoselectivity of the intramo-
lecular R-amidoalkylation reaction depends both on the
nature of the substituent on the chain connecting the
nitrogen atom with the nucleophile and on the nature of
the acid used for the generation of the N-acyliminium
ion. Thus, (5S,10bS)-trans 11 and (5S,10bR)-cis 12 could
be obtained from the same succinimide precursor with
modest selectivities but in high enantiomerical purity
using Lewis or protic acids.
FIGURE 3. Selected NOE enhancements.
level of diastereoselectivity. The stereochemical outcome
is explained also by conformational factors in the tran-
sition state. Interestingly, when the acyliminium is un-
substituted (H instead of CH3), the cis diastereomer was
obtained. This result is in agreement with our pro-
posal.
When a protic acid is used, the stereoselectivity is lost,
or it is inverted in favor of the cis isomer. In this case,
there is no metal to coordinate the oxygen atom on the
C-5 substituent, and therefore the balance between A(1,3)
strain and syn axial 1,3 interactions would favor the
conformation that leads to TS2 again, affording the cis
diastereomers 12. Thus, the cis selectivity improves as
the size of the substituent on C-5 decreases from Bn to
Me and H.
Similar effects have been described for related struc-
tures that led to reversal of diastereoselectivity as a
result of competing steric interactions.20 These results
are in agreement with those obtained in the synthesis of
isoindolinone systems, where a variation of the cis/trans
ratio was observed, depending on the nature of the acid
used for the cyclization.19
Nuclear Overhauser effect difference spectroscopy and
1H-1H decoupling experiments confirmed the stereo-
chemistry of all the pyrroloisoquinoline derivatives 11
and 12. The most significant results obtained with
diasteromeric pyrroloisoquinolines (5S,10bS)-11d and
(5S,10bR)-12d are shown in Figure 3. Thus, for 11d, the
J values of the ABX system formed by H-5 and H-6
protons indicate that H-5 is in a pseudo-axial position.
Additionally, NOE difference spectroscopy showed an
enhancement between H-5 and the methyl group in 10b.
These data are consistent with a preferred half-chair
conformation in which the substituents in C-10b and in
C-5 are in pseudo-axial and pseudo-equatorial positions,
respectively. Thus, the configuration was assigned as 5S,-
10bS. On the other hand, for the cis diastereomer 12d,
the J values of the ABX system formed by H-5 and H-6
protons indicate that H-5 is in a pseudo-equatorial
position. In this case, the absence of NOE enhancement
between methyl protons on C-10b and H-5 and the
enhancement observed between methyl protons on C-10b
and methylene on the C-5 substituent confirmed that
both substituents were in a cis disposition, resulting in
a 5S,10bR configuration. The rest of the NOE experi-
ments carried out were fully consistent with the proposed
Experimental Section
Synthesis of Oxoamides 10a-d. General Procedure.
To a solution of succinimides 9a-d (1 mmol) in dry THF (20
mL), MeLi (4 mmol) was added at -78 °C. The resulting
mixture was stirred at this temperature for 6 h, quenched by
the addition of saturated NH4Cl (20 mL), and allowed to warm
to room temperature. The organic layer was separated, and
the aqueous phase was extracted with Et2O (10 mL) and with
CH2Cl2 (3 × 10 mL). The combined organic extracts were dried
(Na2SO4) and concentrated in vacuo to afford oxoamides 10a-
d, which were purified by flash column chromatography.
(S)-(-)-N-[1-(t-Butyldiphenylsilyloxymethyl)-2-(3,4-di-
methoxyphenyl)ethyl]-4-oxopentanamide (10a). Accord-
ing to the general procedure, imide 9a (270 mg, 0.52 mmol)
was treated with MeLi (2 mL of a 1 M solution in pentane,
2.0 mmol), yielding oxoamide 10a that was purified by flash
column chromatography (silica gel, 50% hexane/ethyl acetate)
(229 mg, 76%): [R]20D -34 (0.07, CH2Cl2); mp (n-pentane) 78-
80 °C; IR (CHCl3) 3340, 1720, 1650 cm-1 1H NMR (CDCl3)
;
1.11 (s, 9H), 2.14 (s, 3H), 2.22-2.38 (m, 2H), 2.67-2.74 (m,
2H), 2.86 (d, J ) 7.1 Hz, 2H), 3.60-3.62 (m, 2H), 3.79 (s, 3H),
3.84 (s, 3H), 4.10-4.21 (m, 1H), 5.87 (d, J ) 8.7 Hz, 1H), 6.67-
6.75 (m, 3H), 7.37-7.45 (m, 6H), 7.60-7.65 (m, 4H); 13C NMR
(CDCl3) 19.4, 26.9, 29.8, 29.9, 36.6, 38.5, 51.5, 55.7, 55.8, 63.4,
110.9, 112.2, 121.3, 127.7, 127.8, 129.8, 130.4, 133.0, 133.1,
135.5, 147.4, 148.7, 171.1, 207.5; MS (EI) m/z (relative
intensity) 547 (M+, 2), 490 (90), 432 (36), 375 (24), 319 (8),
256 (29), 240 (9), 199 (93), 177 (76), 151 (100), 135 (33), 91
(37), 71 (20), 57 (47). Anal. Calcd for C32H41NO5Si: C, 70.17;
H, 7.54; N, 2.56. Found: C, 70.03; H, 7.63; N, 2.49.
(S)-(-)-N-[1-(Benzyloxymethyl)-2-(3,4-dimethoxyphen-
yl)ethyl]-4-oxopentanamide (10b). According to the general
procedure, imide 9b (215 mg, 0.56 mmol) was treated with
MeLi (3.75 mL of a 0.6 M solution in pentane, 2.25 mmol),
yielding oxoamide 10b that was purified by flash column
chromatography (silica gel, ethyl acetate) (148 mg, 66%): IR
(CHCl3) 3313, 1653 cm-1; 1H NMR (CDCl3) δ 2.16 (s, 3H), 2.39
(t, J ) 6.3 Hz, 2H), 2.75 (t, J ) 6.3 Hz, 2H), 3.38-3.39 (m,
2H), 3.81 (s, 3H), 3.84 (s, 3H), 4.21-4.24 (m, 1H), 4.46 (d, J )
12.0 Hz, 1H), 4.53 (d, J ) 12.0 Hz, 1H), 5.97 (d, J ) 8.7 Hz,
1H), 6.67-6.78 (m, 3H), 7.29-7.35 (m, 5H); 13C NMR (CDCl3)
29.9, 30.0, 36.9, 38.4, 50.2, 55.8, 69.6, 73.2, 110.9, 112.4, 121.3,
127.7, 127.8, 130.5, 137.9, 147.4, 148.5, 171.2, 207.5; MS (EI)
m/z (relative intensity) 381 (M+ - 18, 2), 366 (28), 290 (13),
260 (100), 244 (20), 204 (9), 91 (69), 77 (8), 65 (16).
(S)-(-)-N-[2-(3,4-Dimethoxyphenyl)-1-methoxymethyl-
ethyl]-4-oxopentanamide (10c). According to the general
procedure, imide 9c (177 mg, 0.58 mmol) was treated with
MeLi (1.65 mL of a 1.4 M solution in pentane, 2.30 mmol),
yielding oxoamide 10c that was purified by flash column
chromatography (silica gel, ethyl acetate) (93 mg, 50%): IR
(CHCl3) 3303, 1717, 1654 cm-1 1H NMR (CDCl3) δ 2.11 (s,
;
3H), 2.36 (t, J ) 6.3 Hz, 2H), 2.71 (t, J ) 6.3 Hz, 2H)*, 2.65-
2.78 (m, 1H)*, 3.22 (d, J ) 3.9 Hz, 2H), 3.34 (s, 3H)*, 3.31-
(20) Maryanoff, B. E.; McComsey, D. F.; Almond, H. R., Jr.; Mutter,
M. S. J. Org. Chem. 1986, 51, 1341-1346.
10372 J. Org. Chem., Vol. 70, No. 25, 2005

Asymmetric Synthesis of Pyrroloisoquinolines
3.37 (m, 1H)*, 3.79 (s, 3H), 3.81 (s, 3H), 4.06-4.18 (m, 1H),
5.99 (d, J ) 8.3 Hz, 1H), 6.64-6.75 (m, 3H) (*designates
partially overlapped signals); 13C NMR (CDCl3) δ 29.8, 30.5,
36.6, 38.3, 50.0, 55.6, 55.8, 71.8, 110.8, 112.2, 121.2, 130.3,
147.3, 148.5, 171.1, 207.5; MS (EI) m/z (relative intensity) 323
(M+, 4), 290 (1), 208 (100), 193 (3), 177 (24), 151 (9), 99 (26),
74 (44).
99% (Chiralcel OD, hexane/2-propanol 85:15, 0.6 mL/min, tr
(5R,10bS) ) 29.1 min < 1%. tr (5S,10bR) ) 30.2 min > 99%).
Procedure B (with TiCl4). To a solution of oxoamide 10a
(512 mg, 0.93 mmol) in CH2Cl2 (20 mL), TiCl4 (0.41 mL, 3.7
mmol) was added at -78 °C, and the resulting solution was
allowed to reach -10 °C and was stirred for 3 h. Then, the
reaction mixture was allowed to reach room temperature and
stirred for 6 days. After workup and purification as described
in procedure A, pyrroloisoquinolines 11d and 12d were
obtained in a 2:1 ratio (34 mg, 13%).
Procedure C (with TFA). To a solution of oxoamide 10a
(208 mg, 0.38 mmol) in CH2Cl2 (15 mL), TFA (1 mL, 12.9
mmol) was added at room temperature, and the mixture was
refluxed for 3 h. After workup as described in procedure A,
the crude reaction mixture was purified by column chroma-
tography (silicagel, ethyl acetate) to afford (3S)-3-(3,4-dimethox-
yphenylmethyl)-7a-methylperhydropyrrolo[2,b][1,3]oxazol-5-
ona (13), whose data were coincidental to those described in
the literature7b (15 mg, 14%): 1H NMR (CDCl3) δ 1.45 (s, 3H),
2.05-2.22 (m, 1H), 2.40-2.52 (m, 1H), 2.71 (dd, J ) 13.9, 9.5
Hz, 1H)*, 2.67-2.78 (m, 1H), 3.09 (dd, J ) 13.9, 5.5 Hz, 1H),
3.79-3.91 (m, 1H)*, 3.85 (s, 3H)*, 3.86 (s, 3H)*, 4.03 (t, J )
7.1 Hz, 1H), 4.15-4.33 (m, 1H), 6.73-6.81 (m, 3H) (*designates
partially overlapped signals); 13C NMR (CDCl3) δ 24.9, 33.3,
34.6, 39.7, 55.6, 55.8, 55.9, 71.3, 100, 111.0, 112.1, 129.5, 147.7,
148.8, 178.3.
(S)-(-)-N-[2-(3,4-Dimethoxyphenyl)-1-hydroxymethyl-
ethyl]-4-oxopentanamide (10d). According to the general
procedure, imide 9d (202 mg, 0.69 mmol) was treated with
MeLi (1.97 mL of a 1.4 M solution in pentane, 2.76 mmol),
yielding oxoamide 10d that was purified by flash column
chromatography (silica gel, ethyl acetate) (137 mg, 64%): [R]20
D
-25 (0.135, CH2Cl2); mp (hexane/AcOEt 50%) 90 °C; IR
1
(CHCl3) 3302, 1700, 1636 cm-1; H NMR (CDCl3) δ 1.67 (bs,
1H), 2.16 (s, 3H), 2.38 (t, J ) 6.3 Hz, 2H), 2.66-2.89 (m, 2H)*,
2.79 (t, J ) 6.0 Hz, 2H)*, 3.56 (dd, J ) 11.1, 5.0 Hz, 1H), 3.69
(dd, J ) 11.1, 3.2 Hz, 1H), 3.86 (s, 3H), 3.87 (s, 3H), 4.08-
4.14 (m, 1H), 5.9 (d, J ) 7.5 Hz, 1H), 6.72-6.81 (m, 3H)
(*designates partially overlapped signals); 13C NMR (CDCl3)
δ 29.7, 29.8, 36.3, 38.4, 52.6, 55.7, 63.3, 110.9, 112.2, 121.1,
130.1, 147.4, 148.7, 172.4, 208.3; MS (EI) m/z (relative
intensity) 309 (M+, 5), 194 (100), 166 (13), 151 (30), 138 (12),
99 (30), 83 (28), 60 (40). Anal. Calcd for C16H23NO5: C, 62.12;
H, 7.49; N, 4.53. Found: C, 61.87; H, 7.21; N, 4.46.
Intramolecular r-Amidoalkylation. Synthesis of (5S,-
10bS)-(-)-5-Hydroxymethyl-8,9-dimethoxy-10b-methyl-
1,5,6,10b-tetrahydropyrrolo[2,1-a]isoquinolin-3-one (11d)
and (5S,10bR)-(+)-5-Hydroxymethyl-8,9-dimethoxy-10b-
methyl-1,5,6,10b-tetrahydropyrrolo[2,1-a]isoquinolin-3-
one (12d). Procedure A (with BF3‚EtO2). To a solution of
oxoamide 10a (410 mg, 0.74 mmol) in CH2Cl2 (20 mL), BF3‚
EtO2 (0.88 mL, 7.4 mmol) was added at 0 °C, and the resulting
solution was allowed to reach room temperature and then
refluxed for 4 days. Then, another portion of BF3‚EtO2 (0.88
mL, 7.4 mmol) was added again, and the resulting mixture
was refluxed for 3 days. The reaction mixture was treated with
saturated aqueous NaHCO3, the organic layer was decanted,
and the aqueous phase was extracted with CH2Cl2 (3 × 10 mL).
The combined organic extracts were washed with brine (2 ×
10 mL), dried (Na2SO4), and concentrated in vacuo. The
resulting crude reaction mixture was purified by column
chromatography (silicagel, hexane/ethyl acetate/methanol 5:3:
2), yielding pyrroloisoquinolines 11d and 12d in a 2:1 ratio.
In successive experiments, the cyclization was carried out
without the isolation of this intermediate, as follows: To a
solution of oxoamide 10a (270 mg, 0.49 mmol) in CH2Cl2 (15
mL), TFA (1.5 mL, 19.5 mmol) was added at room tempera-
ture, and the mixture was refluxed for 1 day. After workup as
described in procedure A, the crude reaction mixture was
purified by column chromatography to afford a mixture of
pyrroloisoquinolines 11d/12d in a 1:1.3 ratio, determined by
1H NMR (96 mg, 67%).
Synthesis of (5S,10bS)-5-Benzyloxymethyl-8,9-di-
methoxy-10b-methyl-1,5,6,10b-tetrahydropyrrolo[2,1-a]-
isoquinolin-3-one (11b) and (5S,10bR)-5-Benzyloxy-
methyl-8,9-dimethoxy-10b-methyl-1,5,6,10b-tetrahydro-
pyrrolo[2,1-a]isoquinolin-3-one (12b). To a solution of
oxoamide 10b (205 mg, 0.52 mmol) in CH2Cl2 (15 mL), TFA
(1 mL, 12.9 mmol) was added at room temperature, and the
mixture was refluxed for 1 day. After workup as described in
procedure A, the crude reaction mixture was purified by
column chromatography to afford a mixture of pyrroloiso-
quinolines 11b/12b in a 1:1 ratio, determined by 1H NMR, that
could not be separated. Data of the mixture are given (180
(5S,10bS)-(-)-5-Hydroxymethyl-8,9-dimethoxy-10b-
methyl-1,5,6,10b-tetrahydropyrrolo[2,1-a]isoquinolin-3-
one (11d) (128 mg, 60%): [R]20 -203 (0.046, CH2Cl2); mp
D
1
(n-pentane) 124-126 °C; IR (KBr) 3392, 1654 cm-1; H NMR
mg, 92%): IR (CHCl3) 1680 cm-1 1H NMR (CDCl3) 1.45 (s,
;
(CDCl3) δ 1.55 (s, 3H), 2.05-2.19 (m, 1H), 2.31-2.47 (m, 2H),
2.59 (dd, J ) 16.2, 3.6 Hz, 1H), 2.59-2.73 (m, 1H), 3.02 (dd, J
) 16.2, 11.2 Hz, 1H), 3.55-3.67 (m, 1H), 3.84 (s, 3H), 3.86 (s,
3H), 3.98-4.03 (m, 2H), 4.92 (t, J ) 7.1 Hz, 1H), 6.52 (s, 1H),
6.69 (s, 1H); 13C NMR (CDCl3) 27.3, 31.1, 31.3, 34.9, 53.9, 55.8,
56.1, 62.4, 64.4, 107.4, 112.2, 124.3, 133.9, 147.8, 148.0, 174.0;
MS (EI) m/z (relative intensity) 291 (M+, 8), 277 (17), 276 (100),
260 (40), 244 (11), 143 (9), 130 (10), 71 (10), 57 (15). Anal. Calcd
for C16H21NO4: C, 65.96; H, 7.26; N, 4.81. Found: C, 65.77;
H, 7.53; N, 4.67. ee > 99% (Chiralcel OJ, hexane/2-propanol
90:10, 1 mL/min, tr (5S,10bS) ) 30.2 min > 99%, tr (5R,10bR)
) 34.8 min < 1%).
3H trans diastereomer), 1.49 (s, 3H cis diastereomer), 2.04-
2.44 (m, 6H both diastereomers), 2.58-2.70 (m, 2H both
diastereomers), 2.89 (dd, J ) 16.4, 4.0 Hz, 1H trans diastere-
omer ), 2.97 (dd, J ) 16.3, 6.73 Hz, 1H cis diastereomer), 3.06-
3.16 (m, 3H both diastereomers), 3.60 (d, J ) 6.7 Hz, 2H cis
diastereomer), 3.87 (s, 12H both diastereomers), 3.87-3.90 (m,
1H trans diastereomer), 4.19-4.25 (m, 1H trans diastereomer),
4.41 (d, J ) 12.0 Hz, 1H trans diastereomer), 4.54 (bs, 2H both
diastereomers), 4.59 (d, J ) 12.0 Hz, 1H cis diastereomer),
4.63-4.69 (m, 1H cis diastereomer); 6.59 (s, 2H cis diastere-
omer), 6.60 (s, 1H trans diastereomer), 6.66 (s, 1H trans
diastereomer); 13C NMR (CDCl3) 27.8 (trans diastereomer),
29.05 (cis diastereomer), 29.5, 30.8 (both diastereomers), 34.3
(trans diastereomer), 36.3 (cis diastereomer), 46.4 (trans
diastereomer), 49.3 (cis diastereomer), 55.8 (trans diastere-
omer), 55.9 (cis diastereomer), 61.0 (trans diastereomer), 62.5
(cis diastereomer), 68.4, 70.2 (both diastereomers), 72.8 (both
diastereomers), 106.4, 111.7 (both diastereomers), 107.2, 112.2
(trans diastereomer), 123.1 (cis diastereomer), 124.8 (trans
diastereomer), 127.5, 128.3 (both diastereomers), 133.7, 135
(both diastereomers), 138.1, 138.3 (both diastereomers), 147.8
(both diastereomers), 173.8 (both diastereomers). The diaster-
eomers were separated by GC-MS, and the MS spectra
obtained separately, confirming a 1:1 ratio: one isomer MS
(5S,10bR)-(+)-5-Hydroxymethyl-8,9-dimethoxy-10b-
methyl-1,5,6,10b-tetrahydropyrrolo[2,1-a]isoquinolin-3-
one (12d) (74 mg, 32%): [R]20 +38 (0.078, CH2Cl2); mp
D
(hexane/AcOEt 50%) 140 °C; IR (KBr) 3392, 1661 cm-1
;
1H
NMR (CDCl3) δ 1.50 (s, 3H), 2.21-2.47 (m, 3H), 2.68-2.82 (m,
1H), 2.75 (dd, J ) 16.2, 7.9 Hz, 1H), 2.98 (dd, J ) 16.2, 7.1
Hz, 1H), 3.58 (bs, 1H), 3.67-3.84 (m, 2H), 3.86 (s, 3H), 3.87
(s, 3H), 4.21-4.32 (m, 1H), 6.62 (s, 1H), 6.64 (s, 1H); 13C NMR
(CDCl3) δ 28.8, 29.2, 30.4, 35.3, 51.8, 55.8, 56.1, 62.4, 66.8,
106.8, 111.3, 123.5, 134.2, 147.9, 148.0, 175.7; MS (EI) m/z
(relative intensity) 292 (M+ + 1, 3), 291 (M+, 12), 276 (100),
260 (28), 244 (8), 204 (4), 85 (20), 83 (35), 71 (4), 57 (3). ee >
J. Org. Chem, Vol. 70, No. 25, 2005 10373

Garc´ıa et al.
(EI) m/z (relative intensity) 381 (M+, 2), 366 (32), 290 (14),
260 (100), 261 (17), 244 (24), 204 (11), 91 (81), 77 (11), 65 (21),
55 (11). Other isomer MS (EI) m/z (relative intensity) 381 (M+,
2), 366 (23), 290 (12), 260 (100), 261 (17), 244 (23), 204 (11),
91 (69), 77 (9), 65 (17), 55 (9).
omer), 45.9 (cis diastereomer), 49.0 (trans diastereomer), 55.7
(both diastereomers), 55.8 (trans diastereomer), 55.9 (cis
diastereomer), 60.7 (cis diastereomer), 62.5 (trans diastere-
omer), 68.4 (trans diastereomer), 72.4 (cis diastereomer), 106.4,
107.2, 111.6, 112.1 (both diastereomers), 122.8, 124.7 (both
diastereomers), 133.5, 134.9 (both diastereomers), 147.7, 147.8,
147.9 (both diastereomers), 173.6 (both diastereomers). The
diastereomers were separated by GC-MS, and the MS spectra
obtained separately, confirming a 1:2 ratio: minor trans
isomer: MS (EI) m/z (relative intensity) 305 (M+, 5), 291 (14),
290 (77), 261 (17), 260 (100), 244 (20), 204 (10), 160 (5), 122
(7), 91 (5), 77 (6), 65 (17), 55 (9). Major cis isomer: MS (EI)
m/z (relative intensity) 305 (M+, 8), 291 (14), 290 (77), 261 (17),
260 (100), 244 (22), 204 (12), 160 (5), 122 (8), 91 (5), 77 (6), 65
(3), 55 (9).
Synthesis of (5S,10bS)-5-Methoxymethyl-8,9-dimeth-
oxy-10b-methyl-1,5,6,10b-tetrahydropyrrolo[2,1-a]iso-
quinolin-3-one (11c) and (5S,10bR)-5-Methoxymethyl-
10b-methyl-8,9-dimethoxy-1,5,6,10b-tetrahydropyrrolo-
[2,1-a]isoquinolin-3-one (12c). To a solution of oxoamide 10c
(87 mg, 0.26 mmol) in CH2Cl2 (15 mL), TFA (1 mL, 12.9 mmol)
was added at room temperature, and the mixture was refluxed
for 1 day. After workup as described in procedure A, the crude
reaction mixture was purified by column chromatography to
afford a mixture of pyrroloisoquinolines 11c/12c in a 1:2 ratio,
1
determined by H NMR, that could not be separated. Data of
the mixture are given (65 mg, 80%): IR (CHCl3) 1678 cm-1
;
Acknowledgment. Financial support from MCYT
(BQU2000-0223), Gobierno Vasco, and Universidad del
Pa´ıs Vasco is gratefully acknowledged. We also thank
Gobierno Vasco for a grant (E.G.).
1H NMR (CDCl3) δ 1.40 (s, 3H trans diastereomer), 1.46 (s,
3H cis diastereomer), 1.95-2.38 (m, 6H both diastereomers),
2.53-2.76 (m, 4H both diastereomers), 2.85-299 (m, 2H both
diastereomers), 3.27 (s, 3H trans diastereomer), 3.29 (s, 3H
cis diastereomer), 3.43 (d, J ) 7.1 Hz, 2H cis diastereomer),
3.72 (dd, J ) 9.1, 3.7 Hz, 1H trans diastereomer), 3.80 (s, 6H
trans diastereomer), 3.82 (s, 6H cis diastereomer), 3.85-3.89
(m, 1H trans diastereomer), 4.04-4.13 (m, 1H trans diaste-
reomer), 4.54-4.63 (m, 1H cis diastereomer), 6.54, 6.57, 6.65
(3s, 6H both diastereomers); 13C NMR (CDCl3) 27.8 (trans
diastereomer), 28.9 (cis diastereomer), 29.3, 30.7, 30.8 (both
diastereomers), 34.2 (trans diastereomer), 36.4 (cis diastere-
Supporting Information Available: Experimental pro-
cedures and full characterization data for compounds 3-9.
Copies of 1H and 13C NMR spectra of compounds described.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO051584W
10374 J. Org. Chem., Vol. 70, No. 25, 2005