Tandem [4 + 2]/[3 + 2] cycloadditions of nitroalkenes. 13. The synthesis of (-)-detoxinine

Article abstract of DOI:10.1021/jo962306n

(-)-Detoxinine, an unusual, highly-functionalized amino acid, is the core residue of many components that comprise the detoxin complex. The synthesis of (-)-detoxinine was accomplished in 10 steps and 13.4% overall yield from commercially available dichlorodiisopropylsilane. The key step is an asymmetric tandem inter [4 + 2]/intra [3 + 2] cycloaddition between silyoxynitroalkene 17 and chiral vinyl ether (-)-24, illustrating the application of a temporary silicon tether in the tandem nitroalkene cycloaddition process.

Full text of DOI:10.1021/jo962306n

1668
J . Org. Chem. 1997, 62, 1668-1674
Ta n d em [4 + 2]/[3 + 2] Cycloa d d ition s of Nitr oa lk en es. 13. Th e
Syn th esis of (-)-Detoxin in e
Scott E. Denmark,* Alexander R. Hurd, and Hubert J . Sacha
Roger Adams Laboratory, Department of Chemistry, University of Illinois, Urbana, Illinois 61801
Received December 10, 1996X
(-)-Detoxinine, an unusual, highly-functionalized amino acid, is the core residue of many
components that comprise the detoxin complex. The synthesis of (-)-detoxinine was accomplished
in 10 steps and 13.4% overall yield from commercially available dichlorodiisopropylsilane. The
key step is an asymmetric tandem inter [4 + 2]/intra [3 + 2] cycloaddition between silyoxyni-
troalkene 17 and chiral vinyl ether (-)-24, illustrating the application of a temporary silicon tether
in the tandem nitroalkene cycloaddition process.
In tr od u ction
The detoxin complex is a collection of 12 unique
depsipeptides (2-13), isolated from Streptomyces cae-
spitosus var. detoxicus 7072 GC1, which displays an
intriguing detoxifying effect against the nucleoside an-
tibiotic blasticidin S, Figure 1.1,2 Coadministration of
blasticidin S and the detoxin complex reduces the cyto-
toxicity of the antibiotic without reducing the curative
effect in the treatment of rice blast disease.3 Ten of the
twelve characterized depsipeptides (2, 5-13) of the
detoxin complex possess the unusual amino acid (-)-
detoxinine as the core scaffold. The absolute configura-
tion of (-)-detoxinine (1) was assigned by chemical
synthesis from D-glucose.4
Although (-)-detoxinine is not known to possess any
particular biological activity of its own, the incorporation
of unusual amino acids into peptide structures can
promote interesting and unusual biological responses.5
Besides the obvious interest in highly functionalized
amino acids, the synthesis of detoxinine has also been
undertaken to illustrate new synthetic methods and has
previously been reported for both enantiomers and the
racemate.6 Most of the syntheses are fundamentally
similar, utilizing an acetate aldol addition as a major
bond-forming event either before pyrrolidine ring for-
mation6b or with the pyrrolidine ring intact.6a,c,e Ad-
ditionally, most syntheses originate with chiral pool
starting material.
F igu r e 1. The detoxin complex.
lighted a general method for the asymmetric synthesis
of pyrrolizidine alkaloids, exemplified by the synthesis
of (-)-hastanecine,11 (-)-rosmarinecine,12 and (+)-cro-
tanecine.13 Additionally, a synthesis of the Sceletium
alkaloid (-)-mesembrine has been completed which
(7) (a) Denmark, S. E.; Dappen, M. S.; Cramer, C. J . J . Am. Chem.
Soc. 1986, 108, 1306. (b) Denmark, S. E.; Cramer, C. J .; Sternberg, J .
A. Helv. Chim. Acta 1986, 69, 1971. (c) Denmark, S. E.; Moon, Y.-C. ;
Cramer, C. J .; Dappen, M. S.; Senanayake, C. B. W. Tetrahedron 1990,
46, 7373. (d) Denmark, S. E.; Kesler, B. S.; Moon, Y.-C. J . Org. Chem.
1992, 57, 4912.
Recent reports from these laboratories have described
the application of asymmetric tandem [4 + 2]/[3 + 2]
nitroalkene cycloaddition reactions7-10 in the synthesis
of natural products, Chart 1. These reports have high-
(8) (a) Denmark, S. E.; Senanayake, C. B. W.; Ho, G.-D. Tetrahedron
1990, 46, 4857. (b) Denmark, S. E.; Schnute, M. E. J . Org. Chem. 1991,
56, 6738. (c) Denmark, S. E.; Schnute, M. E.; Senanayake, C. B. W. J .
Org. Chem. 1993, 58, 1859. (d) Denmark, S. E.; Schnute, M. E.; Marcin,
L. R.; Thorarensen, A. J . Org. Chem. 1995, 60, 3205.
(9) (a) Denmark, S. E.; Moon, Y.-C.; Senanayake, C. B. W. J . Am.
Chem. Soc. 1990, 112, 311. (b) Denmark, S. E.; Thorarensen, A. Chem.
Rev. 1996, 96, 137. (c) Denmark, S. E.; Senanayake, C. B. W. J . Org.
Chem. 1993, 58, 1853. (d) Denmark, S. E.; Senanayake, C. B. W.;
Schnute, M. E.; Moon, Y.-C.; Ho, G.-D.; Middleton, D. S. Proceedings
of the Fifth International Kyoto Conference on New Aspects of Organic
Chemistry; VCH Verlagageselschaft and Kodansha, Ltd. 1992. (f)
Denmark, S. E.; Stolle, A.; Dixon, J . A.; Guagnano, V. J . Am. Chem.
Soc. 1995, 117, 2100.
(10) (a) Denmark, S. E.; Marcin, L. R. J . Org. Chem. 1993, 58, 3857.
(b) Denmark, S. E.; Schnute, M. E. J . Org. Chem. 1994, 59, 4576. (c)
Denmark, S. E.; Marcin, L. R. J . Org. Chem. 1995, 60, 3221.
(11) Denmark, S. E.; Thorarensen, A. J . Org. Chem. 1994, 59, 5672.
(12) Denmark, S. E.; Thorarensen, A.; Middleton, D. S. J . Am. Chem.
Soc. 1996, 118, 8266.
X Abstract published in Advance ACS Abstracts, February 15, 1997.
(1) Yonehara, H.; Seto, H.; Aizawa, S.; Hidaka, T.; Shimazu, A.;
OÄ take, N. J . Antibiot. 1968, 21, 369.
(2) For a review of the detoxin complex, see J oullie, M. M.; Li, W.-
R.; Han, S.-Y. Heterocycles 1993, 36, 359.
(3) Yonehara, H.; Seto, H.; Shimazu, A.; Aizawa, S.; Hidaka, T.;
Kakinuma, K.; Otake, N. Agr. Biol. Chem. 1973, 37, 2771.
(4) Yonehara, H.; Otake, N.; Kakinuma, K. Tetrahedron Lett. 1980,
21, 167.
(5) Wagner, I.; Musso, H. Angew. Chem., Int. Ed. Engl. 1983, 22,
816.
(6) Racemic synthesis: (a) Ha¨usler, J . Liebigs Ann. Chem. 1983, 982.
Natural (-) enantiomer syntheses: (b) Ohfune, Y.; Nishio, H. Tetra-
hedron Lett. 1984, 25, 4133. (c) J oullie, M. M.; Ewing, W. R.; Harris,
B. D.; Bhat, K. L. Tetrahedron 1986, 42, 2421. (d) Kogen, H.;
Kadokawa, H.; Kurabayahi, M. J . Chem. Soc., Chem. Commun. 1990,
1240. Unnatural (+) enantiomer synthesis: (e) Mulzer, J .; Meier, A.;
Buschmann, J .; Luger, P. J . Org. Chem. 1996, 61, 566. Additionally,
there are two formal synthesis of natural (-)-detoxinine on record. (f)
J oullie, M. M.; Ewing, W. R. Heterocycles 1988, 27, 2843. (g) Takahata,
H.; Banba, Y.; Tajima, M.; Momse, T. J . Org. Chem. 1991, 56, 240.
(13) Denmark, S. E.; Thorarensen, A. J . Am. Chem. Soc. 1997, 119,
125.
S0022-3263(96)02306-7 CCC: $14.00 © 1997 American Chemical Society

Tandem [4 + 2]/[3 + 2] Cycloadditions of Nitroalkenes
J . Org. Chem., Vol. 62, No. 6, 1997 1669
Ch a r t 1
Sch em e 1
employed a tandem cycloaddition as the key step.14 For
the syntheses of the rosmarinecine and crotanecine the
critical cis-relationship between HC(1) and HC(7a) was
established by an intramolecular [3 + 2] cycloaddition
through an ester linkage. This also provided the requi-
site functionality for the C(7) hydroxymethyl group
common to these structures. (-)-Detoxinine, however,
bears a simple hydroxyl group in this position and
therefore requires a tether capable of linking the ni-
troalkene and the dipolarophile without carbon atoms.
The synthesis of (-)-detoxinine thus represented a chal-
lenge to extend the scope of the tandem cycloaddition
process in the realm of possibilities for the connecting
tether. The successful resolution of this challenge in the
synthesis of (-)-detoxinine is detailed below.
followed by tether excision using the Tamao-Fleming
oxidation15 of hydroxy lactam 15. Hydroxy lactam 15
would itself be derived from hydrogenolytic unmasking
of nitroso acetal 16, which is the basic subunit created
in the tandem cycloaddition process.
Syn th esis Design
An alternative tether in the tandem nitroalkene cy-
cloaddition process for the synthesis of (-)-detoxinine had
to incorporate the following design features: (1) the
removal of the tether must reveal hydroxyl or equivalent
functional groups at both points of attachment, (2) the
tether must be limited to two atoms to achieve proper
diastereocontrol, and (3) most importantly, the nitroal-
kene must be competent in the tandem cycloaddition
reaction itself. A temporary silicon tether presented an
alternative device which could satisfy these criteria. The
use of a silyloxy linkage between the nitroalkene and the
dipolarophile could provide the appropriate diastereo-
control, yet allow for proper functionalization by the
Tamao-Fleming oxidation.15 A temporary silicon tether
has been employed as a powerful strategy for diastereo-
control in many reactions including cycloadditions.16,17
In designing a synthetic approach to (-)-detoxinine,
we recognized that the natural product would be acces-
sible by hydrolysis of bicyclic lactam 14, Scheme 1. The
lactam 14 was envisioned to arise by R-deoxygenation
Nitroalkene 17 was formulated for the synthesis of
nitroso acetal 16 with consideration of the following
stereocontrol features. First, the C(7) configuration of
nitroso acetal 16 would be established as C(4) in the
intermediate nitronate i by an asymmetric [4 + 2]
cycloaddition using an enantiomerically pure chiral vinyl
ether and an appropriate Lewis acid. This stereocenter
then controls the relative (and conseqently absolute)
configuration of the remaining stereocenters. Second, the
C(7a) configuration in nitroso acetal 16 would be estab-
lished as a result of the intramolecular [3 + 2] cycload-
dition. Third, to establish the correct configuration at
C(4) in nitroso acetal 16, the [3 + 2] cycloaddition must
occur with the tether folded in an endo fashion. Numer-
ous studies have established the strong preference for
an endo orientaion when a “two-atom” tether connects
the nitroalkene and the dipolarophile.9a Nitroalkene 17
should be readily prepared from the coupling of chlorosi-
lane 18 and potassium nitroacetaldehyde (19).18 19 has
been featured in all three pyrrolizidine syntheses and this
silicon-based dipolarophile finds its origins in analogy to
the dienophiles employed by Stork16a and Sieburth.16b
Thus, our first challenge was the selection of appropriate
spectator groups on silicon and a protocol for the new
tandem process.
(14) Denmark, S. E.; Marcin, L. R., J . Org. Chem. 1997, 62, 1675.
(15) (a) Tamao, K.; Ishida, N.; Tanaka, T.; Kumada, M. Organome-
tallics 1983, 2, 1694. (b) Tamao, K.; Ishida, N. J . Organomet. Chem.
1984, 269, C37. (c) Fleming, I.; Henning, R.; Plaut, H. J . Chem. Soc.,
Chem. Commun. 1984, 29. (d) Fleming, I.; Henning, R.; Parker, D. C.;
Plaut, H. E.; Sanderson, P. E. J . J . Chem. Soc., Perkin Trans. 1 1995,
317.
(16) (a) Stork, G.; Chan, T. Y.; Breault, G. A. J . Am. Chem. Soc.
1992, 114, 7578. (b) Sieburth, S. M.; Fensterbank, L. J . Org. Chem.
1992, 57, 5279. (c) Tamao, K.; Kobayashi, K.; Ito, Y. J . Am. Chem.
Soc. 1989, 111, 6478. (d) Bols, M.; Skrydstrup, T. Chem. Rev. 1995,
95, 1253.
Resu lts
Orienting experiments aimed at assaying the stability
of â-(silyloxy)nitroalkenes established that bulky sub-
(17) For a recent example of the use of the temporary silicon tether
in [3 + 2]-dipolar cycloadditions of nitronates see: Righi, P.; Marotta,
E.; Laduzzi, A.; Rossini, G. J . Am. Chem. Soc. 1996, 118, 9446.
(18) Babievskii, K. K.; Belikov, V. M.; Tikhohova, N. A. Bull. Acad.
Sci. USSR (Engl. Transl.) 1970, 1096.

1670 J . Org. Chem., Vol. 62, No. 6, 1997
Denmark et al.
Sch em e 2
Sch em e 3
Ch a r t 2
stituents (e.g. TIPS, TBS) were necessary. The isopropyl
groups were selected to balance product stability with
reactivity of the chlorosilane precursors and ease of
synthesis. The preparation of chlorosilane 18 was ini-
tially envisioned by the addition of silyl anions to carbon-
carbon multiple bonds.19 The reaction of the lower order
(phenyldiisopropylsilyl)cyanocuprate with methyl pro-
piolate provided exclusively phenylsilane (E)-22 albeit in
unsatisfactory (40-68%) yields. As an alternative, the
coupling of silylcopper reagents to haloalkenes was
considered. Although the addition of silyl anions to R,â-
unsaturated esters has been reported by several groups,20
the addition of such cuprates to â-haloacrylates was
unknown. Since both carbocuprates21 and stannyl cu-
prates22 couple with retention of configuration, the ad-
dition of silyl cuprates (specifically phenyldiisopropylsilyl)
to â-iodoacrylates was investigated.
To accomplish this an improved synthesis of phenyl-
diisopropylsilyl chloride (20)23 was first developed by the
monoaddition of phenyllithium to dichlorodiisopropylsi-
lane in 85% yield, Scheme 2. The silyllithium reagent,
prepared by a modified Lambert procedure,24 was treated
with 1 equiv of CuCN at -20 °C for 30 min, and then
the copper reagent was added to a -105 °C solution of
iodoacrylate 2125 affording phenylsilane 22 in 95% yield.26
Our initial plan was to carryout the protodesilylation
of 22 to create a triflate or fluoride. However, the
dearylation of 22 proved to be more challenging than
anticipated. For example, protodesilylation with triflic
acid, tetrafluoroboric acid, or boron trifluoride acetic acid
complex formed the silyl triflate or silyl fluoride but only
in low yield and accompanied by inseparable hydrolysis
products and starting material. Ultimately, an extremely
clean protodesilylation could be affected with excess (30
equiv) dry HCl in chloroform at 80 °C (Carrius tube).
After being heated for 2 h, phenylsilane 22 was converted
to chlorosilane 18 in 98% yield (distilled).
The coupling of chlorosilane 18 with potassium ni-
troacetaldehyde (19) provided nitroalkene 17 (Scheme 3),
which was (despite the isopropyl groups), unfortunately,
not amenable to purification. Attempted distillation (4.5
× 10-5 Torr), chromatography (silica plug, silica Et3N
complex plug, basic alumina III plug), and even cold
chromatography (neutral alumina III, -30 °C) resulted
in isolation of only the hydrolysis product 23 (see Sup-
porting Information). Consequently, 17 had to be used
directly in the next step.
The next and most critical reaction to investigate was
the tandem process itself. The flexibility of the asym-
metric tandem nitroalkene cycloaddition method derives
from the many combinations of Lewis acids and chiral
vinyl ethers8 which are available for optimization studies,
Chart 2. Methylaluminum bis-(2,6-di-tert-butyl-4-meth-
ylphenoxide) (MAD)27 and Ti(Oi-Pr)2Cl2 both proved to
be poorly selective promoters for cycloadditions between
chiral vinyl ether 248c and nitroalkene 17. Methylalu-
minum bis(2,6-diphenylphenoxide) (MAPh)27 was found
to be the Lewis acid of choice, as nitroso acetal 16 was
produced with excellent selectivity (ca 30:1) using either
chiral vinyl ether 24 or 25.8d Ultimately, 24 was chosen
because of the ease of preparation of trans-phenylcyclo-
hexanol (26). The results of previous cycloadditions with
2-substituted nitroalkenes and this vinyl ether promoted
by MAPh suggested that (-)-24 should provide the
correct enantiomer of detoxinine.8c
(19) For a review of silyl anions see: Tamao, K.; Kawachi, A. In
Advances in Organometallic Chemistry; Stone, F. G. A., West R., Eds.;
Academic Press: New York, 1995; Vol. 38, Chapter 1, p 1.
(20) (a) Fleming, I.; Kindon, N. D. J . Chem. Soc., Chem. Commun.
1987, 1177. (b) Fleming, I.; Kindon, N. D. J . Chem. Soc., Perkin Trans.
1 1995, 303. (c) Oppolzer, W.; Mills, R. J ., Pachinger, W.; Stevenson,
T. Helv. Chim. Acta 1986, 69, 1542. (d) Palomo, C.; Aizpurua, J . M.;
Iturburu, M.; Urchegui, R. J . Org. Chem. 1994, 59, 240.
(21) Kozlowski, J . A. In Comprehensive Organic Synthesis, Com-
bining C-C π-Bonds; Paquette, L. A., Ed.; Pergamon: Oxford, 1991;
Vol 4; p 169.
(22) (a) Seitz, D. E.; Lee, S.-H.; Tetrahedron Lett. 1981, 22, 4909.
(b) Piers, E.; Morton, H. E.; Chong, J . M. Can. J . Chem. 1987, 65, 78.
(c) Piers, E.; Morton, H. E. J . Chem. Soc., Chem. Commun. 1978, 1033.
(23) (a) Weidenbruch, M.; Schiffer, W.; Ha¨gele, G.; Peters, W. J .
Organomet. Chem. 1975, 90, 145. (b) Hurd, C. D.; Yarnall, W. A. J .
Am. Chem. Soc. 1949, 71, 755. (c) Metras, F.; Plazanet, J .; Valade, J .
Bull. Soc. Chim. Fr. 1966, 2155.
Unlike most nitroalkene cycloaddition procedures which
display sensitivity to reagent stoichiometry, no discern-
(24) Lambert, J . B.; Urdaneta-Pe´rez, M. J . Am. Chem. Soc. 1978,
100, 157.
(25) Biougne, J .; The´ron, F. C. R. Acad. Sc. Paris 1971, 272, 858.
(26) Higher order silylcyanocuprates provided the â-silyl unsatur-
ated ester in low yield. Additionally, multiple addition products were
suspected as byproducts. Investigations with alternative copper source
also failed.
(27) Maruoka, K.; Itoh, T.; Yamamoto, H. J . Am. Chem. Soc. 1985,
107, 4573. (b) Nonoshita, K.; Banno, H.; Maruoka, K.; Yamamoto, H.
J . Am. Chem. Soc. 1990, 112, 316. (c) Maruoka, K.; Ooi, T.; Yamamoto,
H. J . Am. Chem. Soc. 1990, 112, 9011. (d) Maruoka, K.; Itoh, T.;
Shirasaka, T.; Yamamoto, H. J . Am. Chem. Soc. 1988, 110, 310.

Tandem [4 + 2]/[3 + 2] Cycloadditions of Nitroalkenes
J . Org. Chem., Vol. 62, No. 6, 1997 1671
Ta ble 1. Cycloa d d ition s of 17 a n d 24
entry
T, °C
t, h
yield, %
dra
1
2
3
4
5
6
-80
-85
-85
-85
-55
-30
4
18
18
17
3.5
3.5
44
37
59
53
60
65
>25:1
>25:1
>25:1
>25:1
17:1
F igu r e 2. (a) [4 + 2] Cycloaddition: proposed approach of
vinyl ether (-)-24 to nitroalkene 17. (b) [3 + 2] Cycloaddi-
tion: proposed endo (tether) approach of the dipolarophile to
the same face of the nitronate dipole.
13:1
ation11,31,32 of thionocarbonate (-)-28 afforded lactam (-)-
29 in 84% yield. Tamao-Fleming oxidation15 of the
silyloxy tether was accomplished using tetrabutylammo-
nium fluoride and hydrogen peroxide at 60 °C, providing
lactam diol (-)-14 in 86% yield (Scheme 4). Lactam
hydrolysis was affected with refluxing 10% aqueous
hydrochloric acid for 13 h. Following purification by
cation exchange chromatography and subsequent tritu-
ration with hot ethanol to remove olefinic by-products,
(-)-detoxinine (1) was isolated in 90% yield.33 The
physical properties of synthetic material matched those
The diastereomer ratio was determined by 400 MHz 1H NMR
using a 90° pulse width and 5 s delay time.
a
ible advantage was noticed using 1-3 equiv of both vinyl
ether and Lewis acid, and ultimately, 2 equiv of each
were used. A slight yield and selectivity dependence on
reaction temperature was noticed, Table 1. The diaster-
eomeric ratio was highest at low temperatures, but the
yield was variable and could not be increased with
extended reaction times (entries 1-4).28 As the reaction
temperature was increased (entries 5 and 6), a drop in
selectivity was compensated by a slight boost in yield.
The optimized procedure involved addition of MAPh
to a -78 °C solution of 17 and (-)-24, followed by slow
warming to -15 °C over 14.5 h, Scheme 4. Nitroso acetal
(+)-16 was isolated after workup and chromatography
in a satisfactory 60% yield (two steps) as an inseparable
27/1 mixture of diastereomers. Additionally, 1.16 equiv
of chiral alcohol (-)-26 was recovered. The nitroso acetal
(+)-16 was assigned as arising form an exo mode [4 + 2]
reported. Mp ) 227-229 °C dec, [R]20 ) -4.4° (H2O, c
D
) 0.50); lit.6b mp ) 225-228 °C, [R]D ) -4.8° (H2O, c )
0.50).
Unfortunately, a direct comparison of our synthetic
material with a sample from natural sources was not
possible. Additionally, the only reported physical or
1
spectral property of natural detoxinine is a 100 MHz H
NMR of the hydrochloride salt in pyridine-d5.34 A 500
1
MHz H NMR of synthetic (-)-detoxinine hydrochloride
1
salt in pyridine-d5 is comparable to that in the literature
and is provided (see Supporting Information). Further-
more, 1H NMR, 13C NMR, MS, and IR spectra of our
synthetic material are nearly identical to the spectra of
Ha¨usler’s synthetic racemic detoxinine (see Supporting
Information).
cycloaddition by H NMR analysis of HC(6) and correla-
tion with established precedent.8,9b,13 It could not be
determined whether the minor diastereomer results from
an endo mode [4 + 2] cycloaddition or an exo mode [4 +
2] cycloaddition with the opposite facial selectivity. In
no instances was the intermediate nitronate detected.
The hydrogenolysis of nitroso acetal (+)-16 was de-
pendent upon hydrogen pressure and time. Under
optimal conditions (Raney nickel, 100 psi H2, MeOH, 48
h) a 51% yield of hydroxy lactam (-)-15 and a 93%
recovery of chiral alcohol (-)-26 was realized (Scheme
4).29 Additionally, 27, tentatively assigned as Peterson
olefination product,30 was isolated in 18% yield. Control
experiments established that 27 was a primary product.
Hydroxy lactam (-)-15 could not be converted to 27 either
by heating or exposure to the hydrogenolysis reaction
conditions. Attempts to minimize the Peterson olefina-
tion pathway by solvent change and the use of additives
(HOAc, H2O, CeCl3, and Ti(Oi-Pr)4) as well as buffered
solutions (pH ) 5.5 and 7.8) were to no avail.
Discu ssion
Ta n d em Cycloa d d ition . There are three factors
which influence the stereochemical outcome of the [4 +
2] cycloaddition: (1) the endo/exo approach of the diene-
ophile and (2) the reactive conformation of the vinyl
ether, as either s-trans or s-cis, and (3) the design of the
chiral auxiliary. The stereochemical course of the [4 +
2] cycloaddition of nitroalkene 17 and (-)-24 can be
understood as an exo, s-trans approach of the vinyl ether
to the si face of the nitroalkene, which is analogous to
the previously described MAPh-promoted cycloadditions
using this vinyl ether, Figure 2a.8b,c,9b,d
The selectivity observed in the [3 + 2] cycloaddition is
a direct consequence of the constraints imposed by the
two-atom tether as previously noted.9a Nitroso acetal (+)-
16 results from a [3 + 2] cycloaddition to the same face
of the nitronate dipole to which the tether is attached.
Additionally, the tether is folded endo, and thus, the
carbomethoxy substituent of the dipolarophile is oriented
in an exo fashion, Figure 2b.
Having installed all the stereocenters of (-)-detoxinine,
the remaining transformations necessary were R-deoxy-
genation, silyl oxidation, and hydrolysis. Hydroxy lactam
(-)-15 was converted to thionocarbonate (-)-28 in 85%
yield, which also permitted a convenient determination
of enantiomeric purity (>99% ee by chiral HPLC, Regis,
R,R-Whelk-O1). Barton-McCombie radical deoxygen-
(28) A similar observation was documented for nitroalkene bearing
a 2 methylene tether with MAPh and bornanediol derived chiral vinyl
ether. M. E. Schnute, Ph. D. Thesis, University of Illinois, 1995.
(29) Recovered (-)-26 from all hydrogenolysis reactions was deter-
mined to be of >99% ee by chiral HPLC (R,R-Whelk-O1, Regis).
(30) (a) Peterson, D. J . J . Org. Chem. 1968, 33, 780. (b) Ager, D. J .
Synthesis 1984, 384. (c) Ager, D. J . Org. React. 1990, 38, 1.
(31) Barton, D. H. R.; McCombie, S. W. J . Chem. Soc., Perkin Trans.
1 1975, 1574.
(32) Hartwig, W. Tetrahedron 1983, 39, 2609.
(33) Danishefsky, S.; Dynak, J . J . Org. Chem. 1974, 39, 1979.
(34) Yonehara, H.; Otake, N.; Kakinuma, K. Agr. Biol. Chem. 1974,
38, 2529.

1672 J . Org. Chem., Vol. 62, No. 6, 1997
Denmark et al.
Sch em e 4
Sch em e 5
intermediate(s) the unsaturated lactam 27 originates, but
at any point in the acyclic form, both the syn and anti
elimination processes are available. The use of buffered
solutions (pH ) 5.5, 7.8) as cosolvents and even aqueous
methanol solutions failed to alter the product ratio. It
is also possible that lactamization was slow due to
generation of ring strain in the tricyclic hydroxy lactam
2712,13 and, as a result, greater opportunity for the
Peterson process to occur. However, attempts to acceler-
ate this step with additives (Ti(Oi-Pr)4 or CeCl3) were
unsuccessful.
The enantiomeric enrichments of the intermediates
were accurately determined and deserve brief comment.
After silica gel chromatography, thionocarbonate (-)-28
was determined to be 92% ee by chiral HPLC. Since this
material originated from nitroso acetal (+)-16 of 89% de,
some enrichment occurred, likely due to differential rates
of hydrogenolysis of nitroso acetal diastereomers. The
remaining enrichment to >99% ee occurred during a
single recrystallization of thionocarbonate (-)-28.
The facile oxidation of silane (-)-29 is noteworthy
considering the documented resistance of hindered si-
lanes to oxidation.15a,35 The fact that silane (-)-29 was
readily oxidized suggests this process was facilitated by
release of ring strain.
Syn th etic Su m m a r y
The synthesis of (-)-detoxinine (1) was accomplished
in 10 steps and in a respectable 13.4% overall yield from
commercially available dichlorodiisopropylsilane, making
this the most selective (27/1) and shortest synthesis of
detoxinine to date. All stereocenters were properly
installed in a single tandem cascade event. However,
most noteworthy is the successful demonstration that the
temporary silicon tether provides exceptional diastereo-
selectivity, and that the vestige of the tether can be
completely removed, leaving as its legacy two newly
created hydroxyl stereocenters. The synthesis of more
Hyd r ogen olysis. The hydrogenolysis of nitroso ac-
etals to hydroxy lactams is a remarkable transformation
which deserves comment. It is believed that the process
involves a series of discrete steps as shown Scheme 5.
Double N-O bond cleavage produces hemiacetal ii which
breaks down to amino aldehyde iii. Amino aldehyde iii
undergoes imine formation and saturation to pyrrolidine
iv followed by lactamization. Since the timing of indi-
vidual steps is critical to the overall yield of hydroxy
lactam, hydrogenolysis of each nitroso acetal warrants
individual optimization.
(35) (a) Andry, O.; Landais, Y.; Planchenault, D. Tetrahedron Lett.
1993, 34, 2927. (b) For a useful protocol for the oxidation of hindered
silanes, see Smitrovich, J . H.; Woerpel, K. A. J . Org. Chem. 1996, 61,
6044.
The isolation of only one byproduct of this reaction
provides little insight into the sources of material loss.
This is especially true as it is not clear from which

Tandem [4 + 2]/[3 + 2] Cycloadditions of Nitroalkenes
J . Org. Chem., Vol. 62, No. 6, 1997 1673
δ 7.14 (d, J ) 18.6, 1H), 6.53 (d, J ) 18.6, 1H), 3.79 (s, 3H),
1.23 (sept, J ) 7.3, 2H), 1.08 (d, J ) 7.3, 6H), 1.05 (d, J ) 7.3,
6H); 13C NMR (100.6 MHz, CDCl3) δ 165.54, 140.27, 138.08,
51.83, 16.78, 16.66, 13.65; IR (neat) 1732 (s); MS (CI, CH4)
237 (26), 235 (M+ + 1, 66), 217 (100). Anal. Calcd for
C10H19O2SiCl (234.80): C, 51.15; H, 8.16. Found: C, 51.40;
H, 8.18.
complex natural products using asymmetric, tandem [4
+ 2]/[3 + 2] nitroalkene cycloadditions is currently in
progress.
Exp er im en ta l Section
Gen er a l Exp er im en ta l. See Supporting Information for
details.
Meth yl (E)-3-[Diisop r op yl[(E)-[(2-n itr oeth en yl)oxy]]si-
lyl]p r op en oa te (17). All glassware was flamed-dried under
nitrogen three times. To a cold (-29 °C, bath temperature)
suspension of potassium nitroacetaldehyde18 (19) (503 mg, 3.96
mmol) in 60 mL of CHCl3 (distilled, stored over 4 Å molecular
sieves, and filtered through basic alumina III just prior to use)
and 20 mL of CH3CN in a three-necked, round-bottom flask
equipped with a Schlenk filter leading to an inverted three-
necked round-bottom flask for the cycloaddition was slowly
(over 10 min) added a solution of 22 (933 mg, 3.96 mmol, 1.0
equiv) in 20 mL of CHCl3. The creamy white suspension was
allowed to slowly warm to 5 °C over 2.5 h. The suspension
was stirred for an additional 30 min at rt before Schlenk
filtration into the second flask. For characterization purposes,
a 2.5 mL (2.5%) of the solution of 17 was removed and
concentrated. The remainder was concentrated in vacuo and
used without further purification. Data for 17: 1H NMR (400
MHz, CDCl3) δ 8.11 (d, J ) 10.3, 1H), 7.09 (d, J ) 19.0, 1H),
7.07 (d, J ) 10.5, 1H), 6.46 (d, J ) 19.3, 1H), 3.80 (s, 3H), 1.27
(m, 2H), 1.08 (d, J ) 7.4, 6H), 1.07 (d, J ) 7.4, 6H); 13C NMR
(100.6 MHz, CDCl3) δ 165.03, 156.52, 138.79, 137.54, 128.80,
52.01, 16.45, 11.69.
Ch lor odiisopr opylph en ylsilan e (20). Phenyllithium (34.5
mL, 50.0 mmol, 1.45 M in cyclohexane/Et2O, 1.0 equiv) was
added slowly (45 min) via syringe pump to a cold (-73 °C)
solution of dichlorodiisopropylsilane (9.0 mL, 49.9 mmol, 1.0
equiv) in 50 mL of Et2O. After 30 min, precipitation of salts
was observed. After the addition was complete, the mixture
was allowed to warm to rt over 2.5 h and was stirred for an
additional 2 h. Hexane (50 mL) was added, and the slurry
was Schlenk filtered. The filtrate was concentrated and
fractionally distilled through a 10 cm vacuum-jacketed, Vi-
greaux column to afford 9.59 g (85%) of 20 as a clear, colorless
1
liquid. Data for 20: bp 48-52 °C (0.03 Torr); H NMR (400
MHz, CDCl3) δ 7.6-7.63 (m, 2 H), 7.39-7.44 (m, 3H), 1.43
(sept, J ) 7.3, 2H), 1.10 (d, J ) 7.3, 6H), 1.02 (d, J ) 7.4, 6H);
13C NMR (100.6 MHz, CDCl3) δ 134.18, 132.02, 129.84, 127.69,
16.86, 16.63, 13.6; IR (neat) 2948 (s), 1464 (s), 1428 (s); MS
(70 ev) 226 (M+, 13), 155 (100). Anal. Calcd for C12H19ClSi
(226.82): C, 63.54; H, 8.44; Cl, 15.63. Found: C, 63.45; H,
8.34; Cl, 15.86.
(Diisop r op ylp h en ylsilyl)lith iu m A modification of the
Lambert procedure was followed.24 To a suspension of lithium
shot (1.8 g, 259 mmol, 4.7 equiv) (washed with 3 × 10 mL
hexane, and 10 mL of THF) in 50 mL of THF was added a
solution of chlorodiispropylphenylsilane (20) (12.5 g, 55.1
mmol, 1.0 equiv) in 25 mL of THF. The mixture was sonicated
for 3.5 h and stirred at rt for an additional 13 h. Excess
lithium was removed through a Schlenk filter. The dark green
filtrate was titrated by the method of Gilman36 and determined
to be 0.566 M.
(2S,2a S,4a R,6S,7bS)-3,3-Diisp r op yl-6-[(1S,2R)-(2-p h en -
ylcycloh exyl)oxy]-octah ydr o-7b-m eth yl-3-sila-1,4,7-tr ioxa-
7a -a za bicyclop en t[c,d ]in d en e-2-ca r boxylic Acid Meth yl
Ester ((+)-16). Trimethylaluminum (3.86 mL, 7.72 mmol, 2
equiv) was added rapidly to a solution of 2,6-diphenylphenol
(3.80 g, 15.43 mmol, 4 equiv) in 76 mL of toluene at rt. The
resulting, clear yellow solution was allowed to stir for 30 min.
To a cold (-78 °C) solution of 17 (assumed 1.11g, 3.86 mmol)
in 96 mL of toluene was added (over 3 min) a solution of (-)-
248c (1.57 g, 7.76 mmol, 2 equiv) in 21 mL of toluene. The
resulting solution was stirred for 20 min before the MAPh
solution was added slowly (over 60 min). During the addition
a deep wine-red complex developed and persisted. The reac-
tion was stirred at -78 °C for 4 h, at -70 °C for an additional
8 h, and was then allowed to warm to -15 °C over the 2.5 h.
The reaction was quenched with 20 mL of MeOH and allowed
to warm to rt. The resulting mixture was diluted with CH2-
Cl2 (500 mL) and was washed with water (2 × 250 mL) and
brine (2 × 250 mL). Aqueous washes were back extracted with
CH2Cl2 (3 × 200 mL). Combined organic extracts were dried
(MgSO4), filtered through Celite (3 cm plug), and concentrated.
The resulting, pale-yellow oil was purified by silica gel chro-
matography (pentane/TBME; 9/1, 4/1, 3/1, 2/1, 1/1). Homoge-
neous fractions were concentrated and heated to 50 °C in vacuo
to afford 1.128 g (60%) of 16 as a clear glass. Mixed fractions
containing 2,6-diphenylphenol, (-)-26, and (-)-24 were con-
centrated, and the vinyl ether was hydrolyzed (1/1, MeOH/1
N HCl, 1 h). Purification by silica gel chromatography afforded
3.62 g (95%) of recovered 2,6-diphenylphenol and 787 mg (4.46
mmol, 1.16 equiv) of recovered (-)-26. Data for 16: 1H NMR
(400 MHz, CDCl3) δ 7.15-7.40 (m, 5 H), 5.07 (d, J ) 7.5, 1H),
4.30 (ddd, J ) 6.8, 3.4, 2.2, 1H), 4.23 (dd, J ) 10.7, 6.8, 1H),
4.08 (dd, J ) 7.6, 6.6, 1H), 3.87 (s, 3H), 3.57 (td, J ) 10.3, 4.3,
1H), 2.52 (ddd, J ) 12.5, 10.4, 3.8, 1H), 2.24-2.32 (m, 1H),
2.22 (dd, J ) 10.7, 7.5, 1H), 1.80-1.89 (m, 2H), 1.78 (ddd, J )
14.6, 6.6, 3.4, 1H), 1.70-1.77 (m, 1H), 1.58-1.62 (m, 1H), 1.53
(ddd, J ) 14.6, 7.6, 2.2, 1H), 1.24-1.50 (m, 4H), 1.05-1.25
(m, 2 H), 1.04 (d, J ) 7.5, 3H), 1.0 (d, J ) 7.5, 3H), 0.97 (d, J
) 6.8, 3H), 0.96 (d, J ) 6.9, 3H); 13C NMR (100.6 MHz, CDCl3)
δ 170.37, 144.08, 127.95, 127.69, 126.19, 99.32, 83.11, 81.31,
78.90, 72.59, 52.39, 51.22, 34.50, 32.77, 31.53, 30.00, 25.68,
25.14, 17.21, 17.17, 17.12, 17.00, 12.10, 10.67; IR (CH2Cl2) 1754
Meth yl (E)-3-(Diisop r op ylp h en ylsilyl)p r op en oa te (22).
To a suspension of CuCN (2.46 g, 27.5 mmol, 1.1 equiv) in 100
mL of THF at -20 °C was added (diisopropylphenylsilyl)-
lithium solution (48.5 mL, 27.5 mmol, 1.1 equiv). The result-
ing wine-red solution was stirred for 30 min and was then
cooled to -80 °C. The cuprate solution was added to a solution
of iodoacrylate 2125 (5.30 g, 25 mmol) in 150 mL of THF at
-105 °C. The solution was allowed to warm to 5 °C over 4 h
and was then quenched with 100 mL of NH4Cl solution and
80 mL of H2O. The resulting mixture was extracted with
TBME (4 × 200 mL). The organic extracts were washed with
brine (4 × 15 mL), dried (MgSO4), and concentrated. The
resulting oil was purified by silica gel chromatography (hexane/
EtOAc, 80/1, 50/1, 25/1) and distillation to afford 6.53 g (95%)
of 22 as a clear, colorless oil. Data for 22: bp 150-152 °C
1
(0.25 Torr); H NMR (400 MHz, CDCl3) δ 7.35-7.49 (m, 6H),
6.43 (d, J ) 19.3, 1H), 3.78 (s, 3H), 1.36 (sept, J ) 7.3, 2H),
1.10 (d, J ) 7.3, 6H), 1.00 (d, J ) 7.3, 6H); 13C NMR (100.6
MHz, CDCl3) δ 165.87, 143.79, 137.03, 135.24, 132.44, 129.3,
127.73, 51.71, 17.63, 10.55; IR (neat) 1731 (s); MS (70 eV) 276
(M+, 9), 233 (100); TLC Rf ) 0.59 (hexane/EtOAc, 4/1). Anal.
Calcd for C16H24O2Si (276.45): C, 69.52; H, 8.75. Found: C,
69.87; H, 8.35.
Meth yl (E)-3-(Diisop r op ylch lor osilyl)p r op en oa te (18).
Hydrogen chloride was condensed into a cold (EtOH/N2)
graduated cylinder attached to a Carrius tube. The condensed
HCl (3.4 mL, 111 mmol, 31 equiv) was allowed to transfer upon
the removal of the cooling bath through a glass side arm into
the Carrius tube containing a frozen (N2) solution of 22 (986
mg, 3.57 mmol) in 6.2 mL of CHCl3. The Carrius tube was
then sealed, allowed to thaw, and was then placed in an 80
°C oil bath for 2 h. The tube was then cooled in an ice bath,
and the HCl was carefully released above a concentrated
aqueous solution of KOH. The CHCl3 solution remaining in
the Carrius tube was concentrated to a pale yellow oil and
distilled to afford 837 mg (98%) of 18 as a clear, colorless oil.
Data for 18: bp 100 °C (0.2 Torr); 1H NMR (400 MHz, CDCl3)
(s), 1602 (w); MS (130 ev) 490 (M+, 8), 159 (100); [R]23
)
D
+34.9° (CH2Cl2, c ) 1.09); TLC, Rf ) 0.63 (pentane/TBME,
2/1). Anal. Calcd for C26H39NO6Si (489.69): C, 63.77; H, 8.03;
N, 2.86. Found: C, 63.78; H, 8.18: N, 2.89.
(1S,5a R,7a R,7b S)-7,7-Diisop r op yl-1-h yd r oxy-6-oxa -7-
sila -octa h yd r o-2H-cyclop en ta [g,h ]p yr r olizin -2-on e ((-)-
(36) Gilman, H.; Cartledge, F. K. J . Organomet. Chem. 1964, 2, 447.

1674 J . Org. Chem., Vol. 62, No. 6, 1997
Denmark et al.
15). To a solution of (+)-16 (1.06 g, 2.17 mmol) (dr ) 13:1) in
110 mL of MeOH in a glass-lined steel autoclave was added
Raney Nickel W-2 (washed 5 × 200 mL of MeOH). The
autoclave was sealed, pressurized to 100 psi with H2, and
stirred at room temperature for 45 h. The H2 was carefully
released from steel autoclave, and the reaction mixture was
filtered through Celite (4 cm). The filter cake was washed with
1 L of MeOH, and the filtrate was concentrated. The resulting
pale yellow oil was purified by silica gel chromatography
(pentane/TBME; 2/1; CH2Cl2/MeOH; 50/1, 10/1) to afford 354
mg (93%) of recovered (-)-26 and a mixture of (-)-15 and 27.
The mixture of products was further purified by radial
chromatography (CH2Cl2/MeOH; 50/1) to afford 299 mg (51%)
(-)-15 as a clear oil and 105 mg (18%) 27 as a clear oil. An
analytical sample of (-)-15 was prepared by precipitation from
EtOAc/hexane. Data for (-)-15: mp 104-105 °C (EtOAc/Hex);
1H NMR (400 MHz, CDCl3) δ 4.93 (d, J ) 8.8, 1H), 4.50 (t, J
) 3.7, 1H), 4.11 (dd, J ) 5.9, 3.4, 1H), 3.63 (dt, J ) 11.2, 8.7,
1H), 3.34 (bs, 1H), 3.18 (t, J ) 10.9, 1H), 2.54 (dd, J ) 8.7,
6.0, 1H), 2.18-2.36 (m, 2H), 1.18 (sept, J ) 7.4, 1H), 1.08 (m,
1H), 1.03 (d, J ) 6.6, 3H), 1.01 (d, J ) 7.4, 3H), 0.97 (d, J )
6.8, 3H), 0.96 (d, J ) 6.9, 3H); 13C NMR (100.6 MHz, CDCl3)
δ 176.3, 78.6, 74.9, 65.4, 40.5, 34.2, 32.6, 17.1, 17.0, 16.9, 16.8,
13.1, 12.9; IR (CH2Cl2) 3300-3500 (br), 1702 (s); MS (CI, CH4)
1H), 3.08-3.14 (m, 1H), 3.03-3.10 (m, 1H), 2.49 (d, J ) 8.2,
1H), 2.27-2.32 (m, 2H), 2.02 (dd, J ) 9.4, 7.0, 1H), 0.98-1.17
(m, 14H); 13C NMR (100.6 MHz, CDCl3) δ 175.53, 79.18, 70.42,
40.49, 36.94, 35.41, 20.55, 17.18, 17.11, 17.07, 16.86, 12.61,
11.50; IR (KBr) 1714 (s); MS (CI, CH4) 254 (M+ + 1, 100), [R]20
D
) -26.0° (CHCl3, c ) 0.16); TLC, Rf ) 0.37 (EtOAc); HRMS
for C13H24NO2Si calcd: 254.1576208 actual: 254.156897.
(6R ,6a S ,7S )-6,7-Dih yd r oxy-h e xa h yd r op yr r olizin -2-
on e (14). Potassium fluoride (53.0 mg, 0.912 mmol, 1.7 equiv)
and hydrogen peroxide (0.910 mL, 8.02 mmol, 15 equiv) were
added to a solution of (-)-29 (135 mg, 0.533 mmol) in 7 mL of
THF/MeOH (1/1). The reaction mixture was heated in a 60
°C oil bath for 2 h and then allowed to cool to rt. Silica gel
(1.0 g) was added to the reaction mixture and then concen-
trated. The resulting pale yellow powder was purified by silica
gel chromatography (CH2Cl2/MeOH; 6/1) and recrystallized
twice (acetone/hexane) to afford 71.8 mg (86%) of (-)-14 as a
highly crystalline, white solid. Data for (-)-14: mp 108-109
°C (acetone/hexane); 1H NMR (500 MHz, CDCl3) δ 4.82 (td, J
) 6.8, 2.6, 1H), 4.53 (td, J ) 3.7, 2.2, 1H), 3.92 (br s, 2H) 3.88
(dd, J ) 6.3, 3.4, 1H), 3.84 (dt, J ) 11.5, 8.3, 1H), 3.11 (dddd,
J ) 11.5, 8.9, 3.6, 1.4, 1H), 2.99 (dd, J ) 17.6, 7.3, 1H), 2.45
(dd, J ) 17.6, 2.8, 1H), 2.09-2.22 (ABX3, υa ) 1083.3 Hz, J ax
) 8.9, 8.3, 4.0; υb ) 1062.4 Hz J bx ) 8.3, 4.0, 2.2; J ab ) 13.7);
13C NMR (126 MHz, CDCl3) δ 175.06, 71.03, 68.02, 67.60,
45.13, 40.18, 35.85; IR (KBr) 3395 (s, br), 3257 (s, br), 3123
270 (M+ + H, 100); [R]23 ) -6.66° (CH2Cl2, c ) 0.93); TLC,
D
Rf ) 0.50 (CH2Cl2/MeOH; 20/1). Anal. Calcd for C13H23NO3-
Si (269.42): C, 57.96; H, 8.60; N, 5.20. Found: C, 57.93; H,
8.51: N, 5.18.
(s,br), 1651 (s); MS (70 eV) 157 (m+, 17), 71 (100); [R]23
)
D
-38.8° (CHCl3, c ) 1.02); TLC, Rf ) 0.44 (CH2Cl2/MeOH; 6/1);
thermogravimetric analysis, 3.8% H2O. Anal. Calcd for
C20H27NO4SSi‚0.33H2O (157.17): C, 51.53; H, 7.21; N, 8.58.
Found: C, 51.59; H, 6.94; N, 8.57.
(1S,5aR,7aR,7bS)-7,7-Diisopr opyl-6-oxa-1-[(ph en oxyth io-
ca r bon yl)oxy]-7-sila -octa h yd r o-2H-cyclop en ta [g,h ]p yr -
r olizin -2-on e ((-)-28). To a solution of (-)-15 (198 mg, 0.735
mmol) in 15 mL of CH3CN was added a mixture of DMAP (184
mg, 1.51 mmol, 2.0 equiv in total) and phenyl thionochloro-
formate (156 µL, 1.13 mmol, 1.5 equiv in total) in three equal
portions every 90 min. After 21 h the clear, bright yellow
solution was concentrated, and the resulting oil was purified
by silica gel chromatography (hexane/EtOAc; 3/1, 1/1) to afford
(-)-28 which was determined to be 93% ee by chiral HPLC.
Crude (-)-28 was recrystallized (EtOAc/hexane) to afford 253
mg (85%) of (-)-28 as a highly crystalline, white solid which
was determined to be 99.1% ee by chiral HPLC. Data for (-)-
28: mp 146-147 °C (EtOAc/Hex); 1H NMR (400 MHz, CDCl3)
δ 7.39-7.45 (m, 2H), 7.27-7.32 (m, 1H), 7.07-7.11 (m, 2H),
6.33 (d, J ) 8.8, 1H), 4.57 (t, J ) 3.7, 1H), 4.19 (dd, J ) 6.0,3.5,
1H), 3.74 (dt, J ) 11.5, 8.8, 1H), 3.26 (td, J ) 10.6, 2.2, 1H),
2.95 (dd, J ) 8.8, 6.1, 1H), 2.23-2.41 (ABX3, υa ) 943.9 Hz,
J ax ) 8.7, 2.2; υb ) 915.9 Hz, J bx ) 10.6, 8.7, 3.5; J ab ) 14.6,
2H), 1.25 (sept, J ) 7.5, 1H), 1.09-1.18 (m, 1H), 1.12 (d, J )
7.3, 3H), 1.08 (d, J ) 7.3, 3H), 1.02 (d, J ) 7.1, 3H), 1.01 (d, J
) 7.1, 3H); 13C NMR (100.6 MHz, CDCl3) δ 194.4, 168.8, 153.3,
129.4, 126.6, 121.7, 84.4, 78.8, 65.5, 41.0, 34.2, 30.6, 17.3, 16.9,
16.8, 16.5, 12.8, 12.7; IR (KBr) 1703 (s); MS (70 ev) 405 (M+,
(3R)-3-Hyd r oxy-3-[(2R,3S)-3-h yd r oxy-2-p yr r olid in yl]-
p r op ion ic Acid . (Detoxin in e) ((-)-1). A solution of (-)-
14 (55.2 mg, 0.351 mmol) in 10% aqueous HCl was heated to
reflux for 13 h. The light brown solution was purified by cation
exchange chromatography (Dowex 50 × 8-200), eluting with
1 N NH4Cl. The crude product was triturated twice with
EtOH to afford 55.2 mg (90%) of (-)-1 as a light yellow solid.
Data for (-)-1: mp 227-229 °C dec, lit.6b mp 225-228 °C dec
1
lit.6e mp 224-227 °C; H NMR (500 MHz, D2O) δ 4.51 (t J )
3.3, 1H), 4.33 (ddd, J ) 9.0, 8.0, 4.6, 1H), 3.41-3.56 (m, 3H)
2.64 (dd, J ) 15.7, 4.5, 1H), 2.44 (dd, J ) 15.7, 7.8, 1H), 2.25
(dtd, J ) 14.0, 10.5, 3.3, 1H), 2.13 (ddd, J ) 14.0, 7.5, 2.6,
1H); 13C NMR (100 MHz, D2O) δ 178.71, 69.43, 68.72, 65.90,
42.68, 41.92, 32.78; IR (KBr) 3232 (s), 3212 (s), 3201 (s), 1628
(s); MS (CI, CH4) 176 (M+ + 1, 49), 158 (100); [R]20 ) -4.4°
D
(H2O, c ) 0.50), lit.6b [R]20D ) -4.8° (H2O, c ) 0.50); lit.6c [R]23
D
) -4.1° (H2O, c ) 0.50); lit.6d [R]25D ) -4.7° (H2O,); TLC, Rf )
0.27 (i-PrOH/NH4OH; 7/3). Anal. Calcd for C7H13NO4 (175.19):
C, 47.99; H, 7.48; N, 8.00. Found: C, 47.77; H, 7.52; N, 7.89.
Ack n ow led gm en t. We are grateful to the National
Institutes of Health (GM-30938) for financial support.
A.R.H. thanks the University of Illinois for a Graduate
Fellowship. H.S. thanks the Deutsche Forschungs
Gemeinschoft for a postdoctoral fellowship. We grate-
fully acknowledge the gift of (()-detoxinine from Prof.
M. M. J oullie (University of Pennsylvania).
13), 252 (100); [R]23 ) -110.8° (CH2Cl2, c ) 0.993); HPLC tR
D
(-)-28 9.82 min (99.55%); tR (+)-28 13.8 min (0.45%) (99.1%
ee) (R,R-Whelk-01, 24% EtOAc in hexane, 1.0 mL/min); TLC,
Rf ) 0.29 (EtOAc/Hexane; 1/1). Anal. Calcd for C20H27NO4-
SSi (405.59): C, 59.23; H, 6.71; N, 3.45; S, 7.91 Found: C,
59.16; H, 6.68; N, 3.39; S, 7.78.
(6R,6a S,7S)-7,7-Diisop r op yl-6-oxa -7-sila -octa h yd r o-2H-
cyclop en ta [g,h ]p yr r olizin -2-on e ((-)-29). A solution of
AIBN (43 mg, 0.259 mmol, 0.42 equiv) and n-Bu3SnH (0.250
mL, 0.929 mmol, 1.5 equiv) in 6 mL of toluene was added
slowly (over 15 min) to a solution of (-)-28 (250 mg, 0.616
mmol) in 30 mL of toluene at 100 °C via a Teflon syringe needle
through the top of the reflux condenser. The reaction was
heated at 100 °C for 4.5 h and then concentrated. The
resulting oil was purified by silica gel chromatography (hex-
anes/EtOAc, 3/1, 0/1) and distilled to afford 147 mg (-)-29
(84%) as a clear, colorless oil. Data for (-)-29: bp 155-160
°C (0.03 Torr); 1H NMR (500 MHz, CDCl3) δ 4.45 (td, J ) 3.4,
2.7, 1H), 4.36 (dd, J ) 7.0, 3.7, 1H), 3.65 (dt, J ) 11.4, 8.4,
Su p p or tin g In for m a tion Ava ila ble: A complete listing
1
of H NMR and 13C NMR names with assignments, infrared
absorbances, and mass spectral fragments for all compounds
described, experimental and spectroscopic data for the com-
1
pounds 23 and 27, along with H NMR, 13C NMR, IR, and MS
spectra of synthetic (-)-detoxinine and (()-detoxinine (Ha¨u-
sler) are provided (8 pages). This material is contained in
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
J O962306N