Synthetic studies on actinobolin and bactobolin: Synthesis of N-desalanyl-N-[2-(trimethylsilyl)ethanesulfonyl] derivatives from a common intermediate and attempted removal of the SES protecting group

Article abstract of DOI:10.1021/jo960579c

Two closely related syntheses of 5,6-O-(2-propylidene)-N-desalanyl-N-[2-(trimethylsilyl)ethanesulfonyl] bactobolin (9b) from (+)-12, an intermediate previously prepared from D-glucose, are reported. In each case, the key step involves a precedented stereoselective addition of LiCHCl₂ in the presence of CeCl₃ te a suitably protected α-amino ketone. Intermediates from both synthetic routes te 9b can be prepared by degradation of actinobolin (2) thereby establishing a potential method for the transformation of actinobelin into bactebolin. An efficient route to 5,6-O-(2-propylidene)-N-desalanyl-N-[2-(trimethylsilyl)ethanesulfonyl] actinobolin (7b) from (+)-12 involving an unexpected cyclization of 29 was discovered. The 2-(trimethylsilyl)ethanesulfonyl (SES) protecting group in 7b was removed by reaction with Bu₄NF in wet THF. The nature of the Bu₄NF reagent was found te be important to the outcome of the reaction. Several improvements over our previously reported synthesis of actinobolin from D-glucose are noted. Although precedented, the removal of the SES protecting group from 9b could not be achieved thereby preventing completion of a total synthesis of bactobolin.

Full text of DOI:10.1021/jo960579c

5498
J . Org. Chem. 1996, 61, 5498-5505
Syn th etic Stu d ies on Actin obolin a n d Ba ctobolin : Syn th esis of
N-Desa la n yl-N-[2-(tr im eth ylsilyl)eth a n esu lfon yl] Der iva tives fr om
a Com m on In ter m ed ia te a n d Attem p ted Rem ova l of th e SES
P r otectin g Gr ou p
Dale E. Ward,* Yuanzhu Gai, and Brian F. Kaller
Department of Chemistry, University of Saskatchewan, 110 Science Place,
Saskatoon SK, Canada S7N 5C9
Received March 27, 1996X
Two closely related syntheses of 5,6-O-(2-propylidene)-N-desalanyl-N-2-(trimethylsilyl)ethane-
sulfonyl]bactobolin (9b) from (+)-12, an intermediate previously prepared from D-glucose, are
reported. In each case, the key step involves a precedented stereoselective addition of LiCHCl2 in
the presence of CeCl3 to a suitably protected R-amino ketone. Intermediates from both synthetic
routes to 9b can be prepared by degradation of actinobolin (2) thereby establishing a potential
method for the transformation of actinobolin into bactobolin. An efficient route to 5,6-O-(2-
propylidene)-N-desalanyl-N-[[2-(trimethylsilyl)ethanesulfonyl]actinobolin (7b) from (+)-12 involving
an unexpected cyclization of 29 was discovered. The 2-(trimethylsilyl)ethanesulfonyl (SES)
protecting group in 7b was removed by reaction with Bu4NF in wet THF. The nature of the Bu4NF
reagent was found to be important to the outcome of the reaction. Several improvements over our
previously reported synthesis of actinobolin from D-glucose are noted. Although precedented, the
removal of the SES protecting group from 9b could not be achieved thereby preventing completion
of a total synthesis of bactobolin.
The isolation of bactobolin (1) from liquid cultures of
Pseudomonas yoshidomiensis was first reported in 1979.1
Elucidation of the bactobolin structure was facilitated by
its close relationship to actinobolin (2),2 a metabolite of
Streptomyces griseoviridus var atropaciens isolated 20
years earlier. X-ray crystallography confirmed that the
absolute configurations of (-)-11b and (+)-22d were identi-
cal at all equivalent stereogenic centers; the only struc-
tural difference being at the C-3 position where the
CHCl2 group present in 1 has replaced (with inversion
of configuration) the H in 2. Both bactobolin and
actinobolin are broad-spectrum antibiotics and have
antitumor activity but bactobolin is significantly more
potent.3
The synthesis of 1 and 2 has attracted considerable
attention4 and several syntheses of the actinobolin skel-
eton have been published.5,6 The structural similarities
between 1 and 2 suggested that simple modification of a
successful synthetic route to actinobolin would produce
bactobolin.4 Despite several efforts to pursue such a
strategy,7 the only reported synthesis of bactobolin is that
due to Weinreb et al. (Scheme 1).8 A unique feature of
Weinreb’s syntheses of bactobolin and actinobolin is the
introduction of the lactone carbonyl group at a late stage
via an intramolecular acylation reaction (e.g. 8 f 9).
Although Weinreb’s intermediates were racemic, optically
pure final products (-)-1 and (+)-2 were obtained by
separation of the diastereomeric amides formed by acyl-
ation of the amines resulting from deprotection of (()-7
and (()-9, respectively, with L-alanine. We prepared (+)-
12 ([R]D +19; c 0.23, MeOH) by a stereoselective [3+3]
annulation9 of the D-glucose-derived aldehyde 10 with
3-(phenylthio)-2-[(trimethylsilyl)methyl]propene (11) and
followed an intramolecular acylation strategy for the
conversion of (+)-12 into actinobolin (Scheme 1).5g We
now report an improved actinobolin synthesis and the
extension of the synthetic scheme to include the bacto-
bolin series by the efficient transformations of (+)-12 into
(5) (a) Yoshioka, M.; Nakai, H.; Ohno, M. J . Am. Chem. Soc. 1984,
106, 1133; Heterocycles 1984, 21, 121. (b) Rahman, M. A.; Fraser-Reid,
B. J . Am. Chem. Soc. 1985, 107, 5576. (c) Askin, D.; Angst, C.;
Danishefsky, S. J . Org. Chem. 1985, 50, 5005; 1987, 52, 622. (d)
Garigipati, R. S.; Tschaen, D. M.; Weinreb, S. M. J . Am. Chem. Soc.
1985, 107, 7790; 1990, 112, 3475. (e) Kozikowski, A. P.; Konoike, T.;
Nieduzak, T. R. J . Chem. Soc., Chem. Commun. 1986, 1350. (f)
Kozikowski, A. P.; Nieduzak, T. R.; Konoike, T.; Springer, J . P. J . Am.
Chem. Soc. 1987, 109, 5167. (g) Ward, D. E.; Kaller, B. F. Tetrahedron
Lett. 1993, 34, 407; J . Org. Chem. 1994, 59, 4230.
(6) For other synthetic studies on actinobolins see: (a) Rao, M. V.;
Nagarajan, M. Indian J . Chem. 1988, 27B, 718. (b) Kates, S. A. Ph.D.
Thesis, Brandeis University, Waltham, MA, 1988; Diss. Abstr. Int. B
1989, 49, 2651. (c) Marjerle, R. S. K. Ph.D. Thesis, University of
Minnesota, Minneapolis, MN; Diss. Abstr. Int. B 1990, 50, 2932.
(7) (a) Underwood, R.; Fraser-Reid, B. J . Chem. Soc., Perkin Trans.
1 1990, 731. (b) Also see refs 5a and 5c.
X Abstract published in Advance ACS Abstracts, J uly 1, 1996.
(1) (a) Kondo, S.; Horiuchi, Y.; Hamada, M.; Takeuchi, T. Umezawa,
H. J . Antibiot. 1979, 32, 1069. (b) Ueda, I.; Munakata, T.; Sakai, J .
Acta. Crystallogr. 1980, B36, 3128. (c) Munakata, T.; Ikeda, Y.;
Matsuki, H.; Isagai, K. Agric. Biol. Chem. 1983, 47, 929.
(2) (a) Haskell, T. H.; Bartz, Q. R. Antibiot. Annu. 1958-59, 505.
(b) Munk, M. E.; Nelson, D. B.; Antosz, F. J .; Herald, D. L., J r.; Haskell,
T. H. J . Am. Chem. Soc. 1968, 90, 1087. (c) Antosz, F. J .; Nelson, D.
B.; Herald, D. L., J r.; Munk, M. E. J . Am. Chem. Soc. 1970, 92, 4933.
(d) Wetherington, J . B.; Moncrief, J . W. Acta Crystallogr. 1975, B31,
531.
(3) (a) Okumoto, T.; Kontani, M.; Hoshino, H.; Nakanishi, M. J .
Pharmacobio-Dyn. 1980, 3, 177. (b) Munakata, T. Yakugaku Zasshi
1981, 101, 138. (c) Hori, M.; Suzukake, K.; Ishikawa, C.; Asakura, H.;
Umezawa, H. J . Antibiot. 1981, 34, 465. (d) Munakata, T.; Okumotu,
T. Chem. Pharm. Bull. 1981, 29, 891.
(8) (a) Garigipati, R. S.; Weinreb, S. M. J . Org. Chem. 1988, 53, 4143.
(b) Also see ref 5d.
(4) For
a review, see: Fraser-Reid, B.; Lo´pez, J . P. In Recent
Advances in the Chemical Synthesis of Antibiotics; Lukacs, G., Ohno,
M., Eds.; Springer-Verlag: Heidelberg, 1990; pp 285-319.
(9) (a) Ward, D. E.; Kaller, B. F. Tetrahedron Lett. 1991, 32, 843.
(b) Ward, D. E.; Gai, Y.; Kaller, B. F. J . Org. Chem. 1995, 60, 7830.
S0022-3263(96)00579-8 CCC: $12.00 © 1996 American Chemical Society

Synthetic Studies on Actinobolin and Bactobolin
J . Org. Chem., Vol. 61, No. 16, 1996 5499
Sch em e 1
Sch em e 2
the 5,6-O-(2-propylidene)-N-desalanyl-N-[2-(trimethyl-
silyl)ethanesulfonyl] derivatives of bactobolin [(-)-9b;
[R]D -26; c 0.60, CHCl3] and actinobolin [(+)-7b; [R]D
+21.5; c 1.10, CH2Cl2].10
available from 14b (Scheme 1). Treatment of 12 with
BuLi followed by SES-Cl11 gave 13b which was converted
into 14b by reaction with NaOMe in MeOH (88% overall).
Ozonolysis of 14b gave 6b which was transformed into
(+)-7b using Weinreb’s procedure5d (Im2CO, NaH; 80%).12
Because we were unable12 to remove the SES protecting
group from (+)-7b using the reported8,11 conditions (vide
infra) and to conserve synthetic material, we resorted to
the use of the precedented5a,5d PMS group to complete
the synthesis of actinobolin from 12.5g
In contemplating a synthesis of bactobolin from 14b,
the most direct route would involve incorporation of the
dichloromethyl substituent before oxidation of the exo-
cyclic methylene group to the required ketone group.
Collins’ oxidation13 of 14b produced the desired methyl
ketone 15 in 91% yield (Scheme 2). Our initial attempts
to effect addition of LiCHCl2 with8 or without7a added
CeCl3 to the ketone group in 15 were unsuccessful.
Others had observed that this reaction is particularly
sensitive to subtle changes in the substrate structure.7a,8
For example, Weinreb observed8 that the addition of
LiCHCl2/CeCl3 to (()-19 was dramatically more stereo-
selective (>20:1 vs 3:1) than a similar addition to the TBS
ether-protected derivative of (()-19; attempted reactions
without CeCl3 failed to produce the desired product.
Fraser-Reid reported that addition of LiCHCl2 (without
CeCl3) to 21 gave a complex mixture of products whereas
For the synthesis of bactobolin, Weinreb et al. initially
pursued a strategy which exploited 5a , a late stage
intermediate from their elegant actinobolin synthesis
(Scheme 1).8 In that synthesis,5d the choice of the (4-
methylphenyl)methanesulfonyl (PMS) protecting group
was based on earlier work by Ohno et al.5a that estab-
lished an efficient conversion of (+)-7a into 2. They
successfully converted 5a into (()-9a ; however, in con-
trast to 7a , the PMS group could not be removed from
9a without destruction of the substrate. To solve this
problem, the [2-(trimethylsilyl)ethane]sulfonyl (SES) group
was introduced as a new amine protecting group.11
Application of the SES group to the bactobolin synthesis
required substantial retooling since the amine protecting
group was introduced early in the synthesis. The prepa-
ration of (()-9b from (()-4b proceeded in analogy to that
of 9a , and after removal of the SES group with fluoride,
(()-9b was successfully converted into (-)-1 (∼15%
overall yield).8
Considering the above, we selected the SES protecting
group for 12 and reasoned that both 1 and 2 should be
(10) A preliminary account has been reported: Ward, D. E.; Gai,
Y.; Kaller, B. F. Tetrahedron Lett. 1994, 35, 3485.
(11) Weinreb, S. M.; Demko, D. M.; Lessen, T. A.; Demers, J . P.
Tetrahedron Lett. 1986, 27, 2099.
(12) Kaller, B. F. Ph.D. Thesis, University of Saskatchewan, 1993.
(13) Ratcliffe, R. W. Org. Synth. 1976, 55 , 74.

5500 J . Org. Chem., Vol. 61, No. 16, 1996
Ward et al.
Sch em e 3
a similar reaction with 22 gave a single adduct in good
yield.7a Because the amount of synthetic 15 was limited,
we decided to proceed to bactobolin by the known8 route
via 19 (Scheme 2).
The preparation of 19 from 15 was facilitated by using
a route that obviated the need for a protecting group
strategy (Scheme 2). Ozonolysis of 15 gave the dione 18
in excellent yield. Reduction of 18 with NaBH4 according
to our previously developed protocol (50% MeOH in
CH2Cl2, -78 °C, 0.5 h)14 resulted in the chemoselective
and stereoselective15 reduction of the cyclohexanone
carbonyl in the presence of the methyl ketone to give the
desired (-)-19 ([R]D -44; c 0.56, CHCl3) in 72% yield (92%
based on consumed 18) along with recovered 18 (22%).16
The direct conversion of 15 into 19 was attempted by
employing NaBH4 (rather than dimethyl sulfide) for
reductive workup of the ozonolysis reaction; however,
reduction of the putative methoxy hydroperoxide inter-
mediate with NaBH4 under the above conditions to give
18 was not sufficiently rapid at -78 °C to be efficacious.
For example, reaction for 0.5 h gave 19 in <50% yield;
after 1 h, the diol resulting from reduction of both
carbonyl groups in 18 could be detected along with 18
and 19 (1:2, respectively). The spectroscopic properties
of (-)-19 (MS, IR, 1H, and 13C NMR) agreed closely with
those previously reported8 for (()-19. The preparation
of (-)-19 from (+)-12 represented a formal enantio-
specific17 synthesis of bactobolin from D-glucose since the
conversion of (()-19 into (-)-1 had been reported.8
Our attempts to reproduce the reported8 addition of
LiCHCl2 to 19 on small scale (10-20 mg of 19) were, as
with 15, also unsuccessful. To obtain material for further
study of these reactions, we considered the preparation
of 19 (and 15) from actinobolin (2) rather than from
D-glucose.18 In the context of the structure elucidation
of actinobolin, Nelson and Munk had reported19 the
reaction of 5,6-O-(2-propylidene)-N-acetylactinobolin (7;
R ) COCH(Me)NHAc) with aqueous ammonia to produce
N-(acetylalanyl)actinobolone (6; R ) COCH(Me)NHAc)
in 52% yield. The presence of the acetonide protecting
group was shown to be important in preventing elimina-
tion of the C-6 hydroxy group (actinobolin numbering)
and subsequent formation of actinobicyclone and actino-
bolamine derivatives under the reaction conditions.19 We
reasoned that a similar reaction with 7b would produce
6b which could be easily converted into 19. The prepara-
tion of 23 by Edman degradation of 2 has been reported
(Scheme 3).20a In our hands, direct acylation of 23 with
SES-Cl failed to give the desired 24. We prepared 26
from 23 by a slight modification of the known procedure.20
Although 7b could be obtained by acylation of amine 27
obtained by hydrogenolysis of 26, we found that reaction
of 26 with aqueous ammonia to give 28 (57%) presented
a superior route to 19. Treatment of 28 with SES-Cl in
the presence of NaH followed by hydrolysis of the
resulting 29 gave 6b whose spectral data were identical
to those from the 6b obtained previously from D-glucose
via 14b (Scheme 1). Oxidation of 6b produced 18, thus
allowing the preparation of 19 from 2 (Scheme 3) and
establishing a formal route to bactobolin from actinobolin
(cf. Scheme 2).
The preparation of 15 from 2 would require methyl-
enation of the cyclohexanone carbonyl group in one of
the intermediates. Reactions of 29 with Ph3PCH2 gave
the desired 13b in low yields (15-20%); the major
product was 7b (20-30%). Cyclization of 29 was easily
avoided by methylenation under Lombardo’s conditions21
to give 13b in 72% yield. Alternatively, reaction of 28
under these conditions or with Ph3PCH2 gave (+)-12 ([R]D
+20; c 0.50, MeOH) which was identical to that previ-
ously synthesized from D-glucose.5g As shown in Scheme
3, both 15 and 19 were available via nine-step sequences
from actinobolin (2) in ∼19% overall yield for each case.
The cyclization of 29 was unexpected in light of the
earlier speculation by Weinreb et al.8 that an analogous
reaction in their bactobolin synthesis was stereoelec-
tronically disfavored (vide infra). The formation of 7b
clearly indicated that a direct cyclization of 29 under
basic conditions was feasible. Indeed, this process could
be optimized and treatment of 29 with NaH in THF gave
7b in excellent yield. Molecular mechanics calculations22
on 30 as a model for the enol of 29 located a stable
conformation (local minimum) 3.4 kcal/mol above the
global minimum which places the enolic carbon within
3.5 Å of the carbonyl carbon. The relationship between
(14) Ward, D. E.; Rhee, C. K. Can. J . Chem. 1989, 67, 1206.
(15) Stereoisomers of 20 were not detected in the reaction mixture.
(16) Increasing the reaction time to 1 h did not improve the yield of
20; however, diol (∼10%) was detected in the reaction mixture. A larger
scale reaction run at higher concentration on material derived from
degradation of actinobolin gave 19 in 87% yield (see Experimental
Section).
(17) We use this term to describe a diastereoselective synthesis of
a specific enantiomer as opposed to a diastereoselective synthesis of a
racemate or an enantioselective synthesis. For example, see: Ward,
R. S. Chem. Br. 1991, 27, 803.
(18) We thank the Parke-Davis Pharmaceutical Research Divison
of Warner-Lambert Co. for a generous gift of actinobolin sulfate.
(19) Nelson, D. B.; Munk, M. E. J . Org. Chem. 1970, 35, 3832.
(20) Rahman, M. A.; Kelly, D. R.; Ravi, P.; Underwood, R.; Fraser-
Reid, B. Tetrahedron 1986, 42, 2409. (b) We found 23 to be insoluble
in the reported20a solvent (EtOAc) even in the presence of Et3N; thus
a 1.3:1 mixture of EtOH and EtOAc was used as the medium for
acylation with BnOCOCl.
(21) Lombardo, L. Org. Synth. 1987, 65, 81.
(22) Molecular mechanics using the MM2 forcefield included in the
CAChe Worksystem (version 3.7 from CAChe Scientific Inc.). Struc-
tures were minimized with the Newton-Raphson block diagonal method
to 0.001 kcal/mol convergence. Relevant minima were located by
minimizing at least 10 initial conformations generated from driving
the dihedral angles for each rotatable bond.

Synthetic Studies on Actinobolin and Bactobolin
J . Org. Chem., Vol. 61, No. 16, 1996 5501
agreed closely with that previously reported8 for (()-8b.
Ozonolysis of the 85:15 mixture of 16 stereoisomers gave
a separable mixture of the corresponding ketone stereo-
isomers (91%); the major isomer was identical to (-)-8b
in all respects. In our hands, treatment of 8b with MeO2-
CCl gave the enol carbonate 17 in excellent yield.25 The
presence of an enolic CH group in 17 was strongly
implicated by the δH 5.37 (1H, dd, J ) 1, 1 Hz) and δC
112.1 (d) signals in the NMR spectra and the other
spectral data fully supported the assigned structure.
Completion of the bactobolin skeleton was effected by
cyclization of 17 with NaOMe to give (-)-9b ([R]D -26; c
0.60, CHCl3); spectroscopic data (IR, MS, 1H and 13C
NMR) agreed closely with that reported for (()-9b.8 The
syntheses of (-)-9b from (+)-12 (seven steps via 15, 30%;
nine steps via 19, 21%) constitute formal syntheses of
bactobolin from D-glucose.
Sch em e 4
Contrary to expectations but similar to our previous
experience with 7b,12 attempts to remove the SES group
from 9b under the reported conditions (TBAF, THF, 52
°C)8,26 or with CsF (DMF, 95 °C)11 failed to produce the
desired product. With a ready supply of 7b available
from 2 (Scheme 3), we decided to investigate this reaction
further. We were hopeful that identifying suitable reac-
tion conditions for removal of the SES group from 7b
would be facilitated because the desired product (27) was
independently available from 26 (see Scheme 3). Ini-
tially, we established that 27 was stable to TBAF27,28 in
THF solution at 50 °C for several hours; however, heating
under reflux led to slow decomposition of 27 into un-
identified products (<10% of 27 remains after 6 h).
Unfortunately, no reaction was observed after treatment
of 7b with TBAF28a,b in THF at 50 °C for 6 h (64%
recovery of 7b); the presence of 27 was not detected in
the reaction mixture even after prolonged heating under
reflux, conditions which lead to the slow decomposition
of 7b (and 27). Similarly, we were unable to remove the
SES group from 37 with TBAF28b in THF; however,
reaction of 37 with CsF in DMF at 95 °C cleanly produced
38, as previously reported (Scheme 4).11 Treatment of
7b with CsF/DMF led to decomposition without ac-
cumulation of 27.
A recent report by Campbell and Hart29 indicated that
the SES group could be removed from N-acyl SES
sulfonamides under mild conditions. We found that
reaction of 33 with TBAF28b in THF at rt for 15 min gave
36 in excellent yield; however, despite considerable
experimentation,30 we were unable to prepare an N-acyl
derivative of 7b. Considering that 27 may form at a rate
slower than its decomposition, we attempted to trap 27
by reaction of 7b with TBAF in the presence of BnOCOCl
under various conditions; in no case was 26 (or 27)
detected in the reaction mixtures.
the enol and carbonyl groups in this conformation
“loosely” resembles23b the twist-boat TS for addition of
acetaldehyde enolate to formaldehyde;23a a closer rela-
tionship is revealed by examining minimized structures
with the enolic carbon to carbonyl carbon distance
constrained to shorter distances.23c
As a prelude to investigating the addition of LiCHCl2
to 15 and 19, we used 31 as a model compound to
calibrate and optimize the reaction conditions (Scheme
4).24 Reaction of 31 with 3 equiv of LiCHCl2/CeCl3 failed
to give 32 although a similar reaction with 3,4,5-tri-
methoxyacetophenone gave the expected adduct in >90%
yield. The use of 8 equiv of LiCHCl2/CeCl3 gave a 70-
80% yield of 32 as a 5.5:1 mixture of stereoisomers; a
result that could be reproduced on small scale (0.08 mmol
of 31). The stereoselectivity of the reaction presumably
results from a chelation-controlled addition and is con-
sistent with previous observations.7a,8 The yield of 32
was not improved by using up to 20 equiv of LiCHCl2/
CeCl3. Subjecting 19 (0.05-0.08 mmol) to these condi-
tions (20-35 equiv of LiCHCl2/CeCl3) gave 20 as a single
isomer (20-30%) along with recovered 19 (40-60%)
thereby reproducing the results reported by Weinreb et
al.8 (although with lower yield). Similar reactions of 15
(0.05-0.08 mmol) with LiCHCl2/CeCl3 (20-35 equiv)
gave 16 (20-45% yield) as inseparable mixtures of
stereoisomers (∼85:15) along with recovered 15 (30-
45%). The major isomer of 16 was correlated with 20
(vide infra) which confirmed the expected chelation-
controlled stereoselectivity. Considering that the low
conversions in the reactions of LiCHCl2/CeCl3 with 15
and 19 might be due to the acidity of the sulfonamide
NH group, we examined the reaction with 33; however,
a mixture of products was obtained in poor yield (Scheme
4).
(25) Under the same conditions, the formation of the ketone corre-
sponding to 17 (cf. 29) was indicated although spectral data for this
product were not reported.8
(26) We thank Professor Weinreb and Dr. Garigipati for helpful
discussions concerning this reaction.
The oxidation8 of 20 proceeded without incident to give
(-)-8b ([R]D -16; c 0.68, CHCl3) whose spectral data
(27) Similar results were obtained using anhydrous or “wet” TBAF.
(28) An excess of reagent (∼3 equiv) was dispensed from a stock
solution (1 M in THF). (a) Commercial TBAF (1 M in THF) was from
Aldrich Chemical Company (Milwaukee, WI) and was labeled as
containing ∼5 wt % water. This reagent was ∼2 years old and had
been used successfully to remove TBDMS ethers. (b) Anhydrous TBAF
solutions (1 M in THF) prepared from dry TBAF obtained by heating
the hydrate in vacuo: Cox, D. P.; Terpinski, J .; Lawrynowicz, W. J .
Org. Chem. 1984, 49, 3216. (c) Wet TBAF solutions (1 M in THF)
prepared by adding 5% of water (v/v) to an anhydrous solution.
(29) Campbell, J . A.; Hart, D. J . J . Org. Chem. 1993, 58, 2900.
(30) Neustadt, B. R. Tetrahedron Lett. 1994, 35, 379.
(23) (a) Li, Y.; Paddon-Row, M. N.; Houk, K. N. J . Org. Chem. 1990,
55, 481. For the twist-boat TS: torsion 1-2-3-4 ) φ1 ) -57°, torsion
2-3-4-5 ) φ2 ) 25°, angle 2-3-4 ) RN ) 90°, angle 3-4-5 ) RE
104°, distance 3-4 ) 2.35 Å. (b) φ1 ) -66°, φ1 ) 23°, RN ) 143°, RE
)
)
128°. (c) for example, with the 3-4 distance fixed at 2.7 Å: φ1 ) -63°,
φ1 ) 24°, RN ) 127°, RE ) 112° (this conformation was 10.1 kcal/mol
above the global minimum).
(24) We used a slight modification of Weinreb’s protocol8 to accom-
modate a smaller scale. We obtained better reproducibility using MeLi
(as opposed to BuLi) as the base to generate LiCHCl2 and used CeCl3
in a 1.05:1 molar ratio with respect to MeLi (as opposed to 1:1).

5502 J . Org. Chem., Vol. 61, No. 16, 1996
Ward et al.
Sch em e 5
20% yield along with recovered 7b (50%). A superior
procedure involved reaction of the enol benzyl carbonate
derivative 39 under similar conditions (reflux, 30 min)
to give the desired 27 in 58% yield. Unfortunately and
despite considerable experimentation, we were unable to
successfully remove the SES group from 9b or its enol
benzyl carbonate derivative 45 (not shown, cf. 39) under
these or similar conditions.
In conclusion, we have completed two closely related
syntheses of 5,6-O-(2-propylidene)-N-desalanyl-N-[2-(tri-
methylsilyl)ethanesulfonyl]bactobolin (9b) from (+)-12,
an intermediate previously prepared from D-glucose.
Intermediates from both synthetic routes to 9b can be
prepared by degradation of actinobolin (2) thereby es-
tablishing a potential method for converting actinobolin
into bactobolin. Although the conversion of 9b into
bactobolin has been previously reported,8 in our hands
the removal of the SES group from 9b could not be
achieved. By contrast, effective conditions were identi-
fied to remove the SES group from 5,6-O-(2-propylidene)-
N-desalanyl-N-[2-(trimethylsilyl)ethanesulfonyl]actino-
bolin (7b) to give 27, from which actinobolin (2) is readily
prepared (85% yield) according to the published
procedure.5a The recently reported9b enhancements in
the [3+3] annulation of 10 with 11 to produce 12 along
with the efficient synthesis of 7b by direct cyclization of
29 and the successful amine deprotection represent a
considerable improvement over our previously reported
total synthesis of actinobolin.
The diol 24 was readily available by hydrolysis of the
acetonide group in 7b. Reaction of 24 with TBAF28b in
refluxing THF gave 43a which slowly was converted into
a product of undetermined structure31 (Scheme 5). Reac-
tion of 23 under the same conditions produced neither
43a nor its decomposition product. This result indicated
that the lactam in 43a was formed before the loss of the
SES group and suggested the intermediacy of 43b.
Encouraged by a successful (although unproductive)
removal of the SES group, we attempted to decrease the
propensity for intramolecular acylation (cf. 44)32 by
preparation of the enol ether 40 (from 7b and CH2N2
followed by acid). Reaction of 40 with TBAF28b in
refluxing THF cleanly produced a 1.3:1 mixture of 41a
and 42a without evidence of an intermediate. These
results suggest that the facile loss of the SES group from
24 and 40 results from the formation of N-SES lactams
(i.e. 43b and 41b, respectively) which are activated
toward deprotection (cf. 33).29 Presumably, the analogous
reaction of 7b is impeded in favor of other decomposition
pathways by the increased strain in the tetrahedral
intermediate for transacylation (cf. 44a and 44c) due to
the trans fused acetonide.33
Exp er im en ta l Section 35
(3a R,8R,9R,9a R,9b R)-N-[3a ,4,8,9,9a ,9b -H exa h yd r o-5-
h yd r oxy-2,2,8-t r im et h yl-6-oxo-6H -1,3-d ioxolo[4,5-f][2]-
b en zop yr a n -9-yl]-2-(t r im et h ylsilyl)et h a n esu lfon a m id e
(7b). F r om 6b: A solution of 1,1′-carbonyldiimidazole (4.2 mg,
0.026 mmol) and 6b (7.0 mg, 0.017 mmol) in THF (0.5 mL)
was stirred at rt for 16 h. Excess NaH (60% dispersion in oil,
∼10 mg, 0.25 mmol) was added, and the mixture was stirred
at rt for 1 h. The reaction was quenched by addition of
saturated NH4Cl(aq) (∼1 mL), and the mixture was diluted with
water and extraced with EtOAc (×3). The combined organic
layers were dried over Na2SO4, concentrated, and fractionated
by PTLC (5% MeOH in CH2Cl2) to give recovered 6b (3 mg,
43%) and 7b (3 mg, 46%). F r om 29: A solution of 29 (80 mg,
0.18 mmol) in THF (7 mL) was added to NaH (washed with
hexane; 20 mg, 0.8 mmol), and the mixture was stirred at rt
for 5 h. The reaction mixture was diluted with EtOAc
[ca u tion : H2 evolution], washed with brine, dried over
Na2SO4, concentrated, and fractionated by PTLC (5% MeOH
in CH2Cl2) to give 7b (72 mg, 90%): [R]D +21.5 (c 1.10, CH2Cl2);
IR νmax 3274, 2952, 2923, 2853, 1780, 1721, 1644, 1593, 1229,
To determine if the formation of an N-SES lactam (cf.
44a ) was feasible for 7b, we reexamined the reaction with
TBAF in an effort to detect the acetonide derivative of
43a . As before, reaction of 7b with anhydrous TBAF28b
in refluxing THF led only to decomposition without
evidence for formation of a lactam analogous to 43a .
Surprisingly, reaction with “wet” TBAF28c in refluxing
THF gave 27 in low yield (Scheme 3).34 The optimal
result was obtained after reaction for 1 h to give 27 in
1
858 cm-1; H NMR δ 13.56 (1H, s), 4.60 (1H, dq, J ) 1.5, 6.5
Hz), 4.17 (1H, d, J ) 10 Hz), 4.00 (1H, ddd, J ) 1.5, 3.5, 10
(34) We have no convincing explanation for the discrepancy in the
results obtained using commercial28a vs prepared28c “wet” TBAF
solutions.
(35) General procedures are included in the supporting information
and are similar to those recently described.5g Optical rotations were
determined at ambient temperature on a Perkin-Elmer 141 polarimeter
using a 1 mL, 10 dm cell; concentrations (c) are in g/100 mL. Unless
otherwise noted: chemical ionization mass spectra (CIMS) were
recorded at 50 eV with NH3 as the reagent gas; IR spectra were
(31) This compound could not be obtained in pure form. Spectral
data suggested a structure related to actinobolamine2,19 (i.e. elimination
of the C-5 hydroxy group and intramolecular conjugate addition of the
amine nitrogen).
(32) The feasibility of intramolecular acylation via an intermediate
like 44 is supported by the formation of 43a from 24, the cyclization
of 29 to 7b, and by Ohno’s actinobolin synthesis.5a
(33) This can be evaluated qualitatively, for example, by comparing
the difference in the calculated22 total strain energy for 44b and 7 (R
) Ms, keto form) (48.9-27.1 ) 21.8 kcal/mol) with that for the
corresponding diols 44d and 24 (R ) Ms, keto form) (40.5-19.1 ) 21.4
kcal/mol).
recorded on
a Fourier transform interferometer using a diffuse
reflectance cell; NMR spectra were measured in CDCl3 solution at 300
MHz for 1H and 75 MHz for 13C with signals from the solvent used as
internal standards (CHCl3: 7.26 δ; CDCl3 77.0 δ). The 1H NMR
chemical shifts and coupling constants were determined assuming first-
order behavior and multiplicity is indicated by one or more of the
following: s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet),
br (broad), ap (apparent). The multiplicity of 13C NMR signals refers
to the number of attached H’s (i.e., s ) C, d ) CH, t ) CH2, q ) CH3)
and was determined by J modulation.

Synthetic Studies on Actinobolin and Bactobolin
J . Org. Chem., Vol. 61, No. 16, 1996 5503
Hz), 3.79 (1H, ddd, J ) 5.5, 9, 11 Hz), 3.52 (1H, dd, J ) 9, 10
Hz), 3.13-2.96 (2H, m), 2.95 (1H, dd, J ) 5.5, 17 Hz), 2.92
(1H, m), 2.66 (1H, ddd, J ) 3, 11, 17 Hz), 1.48 (3H, d, J ) 6.5
Hz), 1.46 (3H, s), 1.45 (3H, s), 1.12 (2H, m), 0.06 (9H, s); 13C
NMR δ 176.5 (s), 170.6 (s), 112.0 (s), 89.9 (s), 78.3 (d), 76.1
(d), 74.0 (d), 51.1 (t), 50.4 (d, C-4), 41.3 (d), 34.6 (t), 27.3 (q),
26.8 (q), 18.1 (q), 10.5 (t), -1.6 (q); CIMS m/z (relative intensity)
451 ([M + 18]+, 26), 434 ([M + 1]+, 100), 418 (12), 284 (13), 90
(79), 73 (38).
m/z (relative intensity) 537 (3), 535 (1), 533 ([M + 18]+, 15),
520 (7), 518 (28), 516 ([M + 1]+, 34), 90 (100), 73 (18).
(3a R ,4R ,4′R ,5R ,7a R )-4-[H e xa h yd r o-2,2-d im e t h yl-6-
m eth ylen e-1,3-ben zod ioxol-4-yl]-5-m eth yl-3-[2-(tr im eth -
ylsilyl)eth a n esu lfon yl]-2-oxa zolid in on e (13b). F r om 12:
BuLi (2.5 M in hexane) was added to a solution of 12 (12 mg,
0.045 mmol) and 1,10-phenanthroline (trace) in THF (1 mL)
at 0 °C until a red color persisted. After 5 min, SES-Cl11 (0.025
mL, 27 mg, 0.14 mmol) was added, and the mixture was stirred
at 0 °C for 30 min. The mixture was diluted with CH2Cl2;
washed sequentially with NaOH (2 M), brine and water; dried
over Na2SO4; concentrated; and fractionated by FCC (30%
ethyl acetate in hexane) to give 13b as an oil (17 mg, 88%).
F r om 29: The methylenation reagent was prepared from
activated Zn (5.75 g, 88 mmol), CH2Br2 (5.01 g, 28.8 mmol),
and TiCl4 (3.91 g, 20.6 mmol) in THF (50 mL) according to
Lombardo’s procedure21 and stored in a freezer (∼-20 °C). An
aliquot of the above reagent (7.2 mL, ∼2.7 mmol) was added
to a solution of 29 (118 mg, 0.27 mmol) in CH2Cl2 (4 mL) at 0
°C. The mixture was stirred at 0 °C for 2 h and then was
poured onto saturated NaHCO3 and extracted with EtOAc. The
combined organic layers were dried over Na2SO4, concentrated,
and fractionated by FCC (30% EtOAc in hexane) to give 13b
(1′R,2′S,3a R,4R,7a R)-N-[3,3-Dich lor o-1-(h exa h yd r o-2,2-
d im eth yl-6-oxo-1,3-ben zod ioxol-4-yl)-2-h yd r oxy-2-m eth -
ylpr opyl]-2-(tr im eth ylsilyl)eth an esu lfon am ide (8b). Fr om
20: Solid CrO3 (83 mg, 0.83 mmol) was added to a solution of
pyridine (0.23 mL, 0.22 g, 2.8 mmol) in dry CH2Cl2 (2 mL) at
0 °C.13 The mixture was stirred at rt for 10 min, and and then
a solution of 20 (27 mg, 0.055 mmol) in CH2Cl2 (0.5 mL) was
added. After 35 min, the mixture was filtered through a pad
of celite, and the filter cake was washed with EtOAc. The
combined filtrate and washings were concentrated, and the
resulting residue was fractionated by FCC (40% EtOAc in
hexane) to provide 8b (25 mg, 93%):36 [R]D -16 (c 0.68, CHCl3);
IR νmax 3468, 3299, 2986, 1715, 1326, 1141, 1110, 858, 842
1
cm-1; H NMR δ 6.01 (1H, s), 4.62 (1H, d, J ) 10 Hz), 4.26
1
(84 mg, 72%): IR νmax 2994, 2862, 1778, 1137, 842 cm-1; H
(1H, d, J ) 10 Hz), 3.72 (2H, m), 3.11 (2H, m), 2.93 (1H, m),
2.70 (1H, dd, J ) 2, 11 Hz), 2.65 (1H, s br), 2.58 (1H, dd, J )
13, 14.5 Hz), 2.31 (2H, m), 1.53 (3H, s), 1.50 (3H, s), 1.46 (3H,
s), 1.12 (2H, m), 0.08 (9H, s); 13C NMR δ 206.8 (s), 111.9 (s),
78.9 (s), 78.7 (d), 78.1 (d), 76.6 (d), 55.4 (d), 51.0 (t), 45.0 (t),
40.4 (t), 37.2 (d), 27.1 (q), 27.0 (q), 19.3 (q), 10.9 (t), -1.8 (q);
CIMS m/ z (relative intensity) 494 (1), 492 (4), 490 ([M + 1]+,
6), 403 (98), 294 (100), 90 (97), 73 (56). F r om 16: A stream of
ozone in oxygen was bubbled through a solution of 16 (23 mg,
0.047 mmol; a 6:1 mixture of 2′S:2′R isomers) in 20% MeOH
in CH2Cl2 (1 mL) at -78 °C until a blue color was persistent.
A stream of Ar was bubbled through the solution to remove
the excess ozone, and then pyridine (2 drops) and dimethyl
sulfide (0.1 mL) were added, and the solution was allowed to
stand in a refrigerator (3 °C) overnight. The solution was
concentrated to provide an oil which was fractionated by MPC
(2.5% MeOH in CH2Cl2) to provide 8b as a 5:1 mixture of 2′S:
2′R isomers (1H NMR) (21 mg, 91%). The isomers were
separated by MPC (30% EtOAc in hexane) to give 8b (16 mg,
70%) and the corresponding 2′R isomer (2.2 mg, 10%)
(1′R,2′R,3a R,4R,7a R)-N-[3,3-Dich lor o-1-(h exa h yd r o-2,2-
d im eth yl-6-oxo-1,3-ben zod ioxol-4-yl)-2-h yd r oxy-2-m eth -
ylp r op yl]-2-(tr im eth ylsilyl)eth a n esu lfon a m id e: IR νmax
3455, 3342, 2986, 1713, 1650, 1230, 1142, 837 cm-1; 1H NMR
δ 5.76 (1H, s), 4.62 (1H, d, J ) 10 Hz), 4.22 (1H, dd, J ) 10
Hz), 3.75 (1H, ddd, J ) 4.5, 9, 12.5 Hz), 3.66 (1H, dd, J ) 9,
9.5 Hz), 3.08 (1H, m), 2.95 (1H, ddd, J ) 2, 4.5, 14.5 Hz), 2.71
(1H, m), 2.68 (1H, s), 2.56 (1H, dd, J ) 13, 14.5 Hz), 2.35 (1H,
dd, J ) 12.5, 14.5 Hz), 2.28 (1H, m), 1.60 (3H, s), 1.50 (3H, s),
1.45 (3H, s), 1.21 (2H, ddd, J ) 4, 7, 14 Hz), 0.08 (3H, s); CIMS
(isobutane) m/z (relative intensity) 494 (9), 492 (32), 490
([M + 1]+, 40), 342 (66), 340 (94), 101 (100).
(3a R,8S,9R,9a R,9b R)-N-[8-(Dich lor om et h yl)-3a ,4,8,9,
9a ,9b-h exa h yd r o-5-h yd r oxy-2,2,8-tr im eth yl-6-oxo-6H-1,3-
dioxolo[4,5-f][2]ben zopyr an -9-yl]-2-(tr im eth ylsilyl)eth an e-
su lfon a m id e (9b). NaOMe (1 M in MeOH, 0.033 mL, 0.033
mmol) was added to a solution of 17 (5 mg, 0.0087 mmol) in
dry MeOH (1.5 mL), and the mixture was stirred at rt and
monitored by TLC (3% MeOH in CH2Cl2). After 6.5 h,
additional NaOMe (0.020 mL, 0.020 mmol) was added and,
after stirring for 1.5 h, the mixture was diluted with EtOAc,
washed with saturated NH4Cl, dried over Na2SO4, concen-
trated, and fractionated by MPC (3% MeOH in CH2Cl2) to give
9b (3 mg, 67%):36 [R]D -26, (c 0.60, CHCl3); IR νmax 3238, 1743
(sh), 1652, 1585, 1382, 1233, 1103 cm-1; 1H NMR δ 11.90 (1H,
s), 6.04 (1H, s), 4.46 (1H, dd, J ) 3.5, 10.5 Hz), 4.27 (1H, d, J
) 10.5 Hz), 3.82 (1H, ddd, J ) 6, 9, 11 Hz), 3.63 (1H, dd, J )
9, 9 Hz), 3.14 (3H, m), 3.00 (1H, dd, J ) 6, 18 Hz), 2.71 (1H,
ddd, J ) 2.5, 11, 18 Hz), 1.72 (3H, s), 1.50 (6H, s), 1.13 (2H,
m), 0.08 (9H, s); 13C NMR δ 179.6 (s), 169.2 (s), 112.2 (s), 88.2
(s), 86.1 (s), 75.4 (d), 74.4 (d), 73.8 (d), 51.7 (t), 51.6 (d), 37.8
(d), 35.0 (t), 26.9 (q), 26.7 (q), 19.2 (q), 10.7 (t), -2.0 (q); CIMS
NMR δ 4.93 (2H, m), 4.73 (1H, dq, J ) 1.5, 6.5 Hz), 4.03 (1H,
dd, J ) 1.5, 4 Hz), 3.67 (1H, ddd, J ) 4, 14, 14 Hz), 3.34 (3H,
m), 2.74 (1H, dd, J ) 1, 4, 13 Hz), 2.47 (1H, ddd, J ) 1, 4, 13
Hz), 2.28-2.15 (2H, m), 1.88 (1H, br dd, J ) 13, 13 Hz), 1.47
(3H, d, J ) 6.5 Hz), 1.42 (3H, s), 1.39 (3H, s), 1.20 (1H, ddd, J
) 4, 14, 14 Hz), 1.09 (1H, ddd, J ) 4, 14, 14 Hz), 0.06 (9H, s);
13C NMR δ 152.4 (s), 140.7 (s), 115.2 (t), 110.6 (s), 79.5 (d),
79.0 (d), 73.5 (d), 63.9 (d), 50.4 (t), 41.0 (d), 37.3 (t), 33.9 (t),
27.0 (q), 26.4 (q), 20.9 (q), 9.0 (t), -2.0 (q); CIMS m/z (relative
intensity) 449 ([M + 18]+, 23), 432 ([M + 1]+, 100), 416 (16),
282 (20), 268 (22), 210 (11), 172 (26).
(1′R,2′R,3a R,4R,7a R)-N-[1-(Hexa h yd r o-2,2-d im eth yl-6-
m eth ylen e-1,3-ben zod ioxol-4-yl)-2-h yd r oxyp r op yl]-2-(tr i-
m eth ylsilyl)eth a n esu lfon a m id e (14b). Excess NaH (∼5
mg) was added to a stirred solution of 13b (30 mg, 0.069 mmol)
in MeOH (3 mL). After stirring for 3 h, the mixture was
diluted with CH2Cl2, washed with saturated NH4Cl and with
brine, dried over Na2SO4, concentrated, and fractionated by
FCC (40% EtOAc in hexane) to provide 14b (28 mg, 99%): IR
νmax 3512, 3279, 2983, 1648, 1450, 1118, 859 cm-1; 1H NMR δ
4.90 (2H, m), 4.74 (1H, d, J ) 9 Hz), 4.15 (1H, dq, J ) 2, 6.5
Hz), 3.45 (1H, dd, J ) 9, 10 Hz), 3.34 (1H, ddd, J ) 4, 8.5, 12
Hz), 3.26 (1H, ddd, J ) 2, 6.5, 9 Hz), 2.99 (2H, m), 2.72 (1H,
ddd, J ) 1.5, 4.5, 12 Hz), 2.59 (1H, ddd, J ) 3, 13 Hz), 2.22
(1H, dd, J ) 12, 13 Hz), 1.99 (1H, dd, J ) 13, 13 Hz), 1.83
(1H, m), 1.48 (3H, s), 1.44 (3H, s), 1.30 (3H, dd, J ) 6.5 Hz),
1.10 (2H, m), 0.07 (9H, s); 13C NMR δ 141.9 (s), 114.4 (t), 110.2
(s), 81.0 (d), 79.6 (d), 67.3 (d), 61.8 (d), 50.6 (t), 43.5 (d), 37.4
(t), 36.5 (t), 27.0 (q), 26.9 (q), 20.6 (q), 10.8 (t), -1.9 (q); CIMS
m/z (relative intensity) 406 ([M + 1]+, 100), 332 (14), 330 (20),
199 (68), 196 (53), 90 (88).
(1′R ,3a R ,4R ,7a R )-N -[1-(H e xa h yd r o-2,2-d im e t h yl-6-
m e t h yle n e -1,3-b e n zod ioxol-4-yl)-2-oxop r op yl]-2-(t r i-
m eth ylsilyl)eth a n esu lfon a m id e (15). CrO3 oxidation of
14b (76 mg, 0.19 mmol) according to the general procedure
described for the preparation of 8b gave 15 (74 mg, 98%) after
fractionation by FCC (40% EtOAc in hexane): IR νmax 3278,
1
2954, 1720, 1649, 1420, 1147, 835 cm-1; H NMR δ 5.21 (1H,
d, J ) 9.5 Hz), 4.87 (2H, m), 4.32 (1H, dd, J ) 3, 9.5 Hz), 3.41
(2H, m), 2.89 (2H, m), 2.71 (1H, dd, J ) 4, 12.5 Hz), 2.32 (3H,
s), 2.20 (1H, m), 1.99 (3H, m), 1.46 (3H, s), 1.45 (3H, s), 1.04
(2H, m), 0.07 (9H, s); 13C NMR (C6D6) δ 205.6 (s), 142.3 (s),
114.4 (t), 110.4 (s), 79.7 (d), 79.3 (d), 62.6 (d), 49.8 (t), 41.2 (d),
37.7 (t), 33.0 (t), 27.3 (q), 27.2 (q), 27.1 (q), 10.5 (t), -2.1 (q);
CIMS m/z (relative intensity) 421 ([M + 18]+, 27), 404 ([M +
1]+, 100), 199 (49), 90 (64).
(1′R,2′R*,3a R,4R,7a R)-N-[3,3-Dich lor o-1-(h exa h yd r o-
2,2-dim eth yl-6-m eth ylen e-1,3-ben zodioxol-4-yl)-2-h ydr oxy-
2-m eth ylpr opyl]-2-(tr im eth ylsilyl)eth an esu lfon am ide (16).
MeLi (1.0 M in diethyl ether, 1.5 mL, 1.5 mmol) was added to
a solution of CH2Cl2 (0.25 mL; freshly distilled from CaH2) in
THF (0.30 mL) at -100 to -110 °C (bath). After the mixture

5504 J . Org. Chem., Vol. 61, No. 16, 1996
Ward et al.
was stirred for 30 min, diethyl ether (3.2 mL; freshly distilled
from sodium) was added dropwise to the resulting white
supension. Anhydrous CeCl3 (383 mg, 1.55 mmol; prepared37
by heating in vacuo according to Imamoto’s procedure) was
added portionwise via a side arm. The heterogeneous reaction
mixture was stirred at -100 to -110 °C for 1 h, and then a
solution of 15 (19 mg, 0.047 mmol) in CH2Cl2 (0.5 mL) was
added dropwise. The reaction mixture was stirred at -100 to
-110 °C for 2 h and quenched by dropwise addition of MeOH
(1 mL). The mixture was diluted with EtOAc, washed with
saturated NH4Cl, dried over MgSO4, concentrated, and frac-
tionated by MPC (25% EtOAc in hexane) to give recovered 15
(8 mg, 42%) and 16 as a 6:1 mixture of the 2′S:2′R isomers
(10.2 mg, 44%): IR νmax 3456, 3308, 2986, 2954, 1450, 1420,
1142, 837 cm-1; 1H NMR δ for major isomer 6.10 (1H, s), 4.90
(2H, m), 4.65 (1H, d, J ) 10 Hz), 4.12 (1H, br d, J ) 10 Hz),
3.40 (2H, m), 3.08 (2H, m), 2.73 (1H, dd, J ) 4, 12.5 Hz), 2.59
(1H, m), 2.22 (1H, dd, J ) 11.5, 12.5 Hz), 1.99 (2H, m), 1.52
(3H, s), 1.44 (3H, s), 1.41 (3H, s), 1.10 (2H, m), 0.07 (9H, s), δ
for minor isomer 5.81 (1H, s); 13C NMR δ for major isomer
142.4 (s), 114.5 (t), 110.3 (s), 80.2 (d), 79.0 (d), 78.8 (s), 78.4
(d), 56.2 (d), 50.9 (t), 41.0 (d), 37.7 (t), 34.0 (t), 27.2 (q), 26.9
(q), 19.5 (q), 10.9 (t), -1.8 (q); CIMS m/z (relative intensity)
492 (8), 490 (29), 488 ([M + 1]+, 38), 196 (19), 181 (23), 90 (100).
(3a R,4R,4′R,5′S,7a R)-4-[5,5-(Dich lor om eth yl)-5-m eth yl-
3-(2-(tr im eth ylsilyl)eth a n esu lfon yl)-2-oxo-oxa zolid in -4-
yl]-3a ,4,7,7a -tetr a h yd r o-2,2-d im eth yl-1,3-ben zod ioxol-6-
yl Meth yl Ca r bon a te (17). A solution of 8b (28 mg, 0.057
mmol), methyl chloroformate (0.9 mL), and 4-pyrrolidino-
pyridine (2 mg, 0.013 mmol) in Et3N (6 mL) was stirred at rt
for 15 h. The reaction mixture was diluted with EtOAc,
washed with brine, dried over Na2SO4, concentrated, and
fractionated by FCC (20% ethyl actate in hexane) to give 17
as an oil (29 mg, 88%) which contained a small amount (∼15%
by 1H NMR) of the corresponding methyl carbamate alcohol
(i.e. the precursor for oxazolidinone formation):25 IR νmax 2986,
1793, 1761, 1697, 1441, 1359, 1258, 1098, 843 cm-1; 1H NMR
δ 5.80 (1H, s), 5.37 (1H, dd, J ) 1, 1 Hz), 4.90 (1H, d), 3.82
(3H, s), 3.79 (2H, m), 3.57 (2H, m), 2.94 (1H, m), 2.68 (1H, m),
2.55 (1H, m), 1.76 (3H, s), 1.46 (6H, s), 1.13 (2H, m), 0.08 (9H,
s); 13C NMR δ 153.3 (s), 151.5 (s), 149.8 (s), 112.1 (d), 111.5
(s), 86.5 (s), 76.4 (d), 75.5 (d), 74.8 (d), 58.7 (d), 55.5 (q), 51.8
(t), 40.4 (d), 32.5 (t), 27.2 (q), 26.7 (q), 15.7 (q), 9.3 (t), -2.1
(q); CIMS m/z (relative intensity) 595 (4), 593 (14), 591 ([M +
18]+, 18), 160 (65), 138 (100), 132 (52), 90 (58).
mL of solvent) gave recovered 18 (0.9 mg, 22%) and 19 (2.9
mg, 72%):36 [R]D -44 (c 0.56, CHCl3); IR νmax 3488, 3287, 2952,
1
1718, 1370, 1325, 840 cm-1; H NMR δ 5.20 (1H, d, J ) 9.5
Hz), 4.30 (1H, dd, J ) 3.5, 9.5 Hz), 3.86 (1H, m), 3.42 (2H, m),
2.89 (2H, m), 2.46 (1H, m), 2.31 (3H, s), 2.07 (1H, m), 1.80-
1.50 (3H, m), 1.43 (6H, ap s), 1.04 (2H, m), 0.04 (9H, s); 13C
NMR δ 205.9 (s), 111.2 (s), 79.2 (d), 68.3 (d), 62.3 (d), 49.7 (t),
37.8 (d), 37.1 (t), 33.7 (t), 27.9 (q), 27.0 (q), 26.9 (q), 10.2 (t),
-2.0 (q); CIMS m/z (relative intensity) 425 ([M + 18]+, 4), 408
([M + 1]+, 9), 255 (13), 229 (34), 199 (100), 90 (40).
(1′R,2′S,3a R,4R,7a R)-N-[3,3-Dich lor o-1-(h exa h yd r o-6-
h yd r oxy-2,2-d im eth yl-1,3-ben zod ioxol-4-yl)-2-h yd r oxy-2-
m eth ylp r op yl]-2-(tr im eth ylsilyl)eth a n esu lfon a m id e (20).
Reaction of 19 (27 mg, 0.066 mmol) with LiCHCl2/CeCl3 (20
equiv) according to the general procedure described for the
preparation of 16 gave recovered 19 (11 mg, 40%) and 20 (10
mg, 31%) after PTLC (40% EtOAc in hexane):36 IR νmax 3448,
3324, 2986, 2953, 1371, 1252, 1139, 840 cm-1; 1H NMR δ 6.08
(1H, s), 4.78 (1H, d, J ) 10 Hz), 4.11 (1H, dd, J ) 2, 10 Hz),
3.91 (1H, m), 3.43 (1H, ddd, J ) 3.5, 9, 12 Hz), 3.33 (1H, dd,
J ) 9, 9.5 Hz), 3.07 (2H, m), 2.51-2.42 (1H, m), 2.33-2.25
(1H, m), 2.08-1.95 (1H, m), 1.60-1.26 (2H, s), 1.52 (3H, s),
1.43 (3H, s), 1.41 (3H, s), 1.11 (2H, s), 0.07 (9H, s); 13C NMR
δ 110.9 (s), 80.0 (d), 78.9 (s), 78.3 (s), 77.1 (d), 68.7 (d), 55.9
(d), 51.0 (t), 37.8 (d), 37.5 (t), 34.9 (t), 27.1 (q), 26.9 (q), 19.4
(q), 10.8 (t), -1.8 (q); CIMS m/z (relative intensity) 494 (1),
492 (4), 490 ([M + 1]+, 7), 200 (60), 199 (30), 142 (40), 90 (100),
86 (73).
(3aR,8R,9R,9aR,9bR)-9-Am in o-3a,4,8,9,9a,9b-h exah ydr o-
5-h y d r o x y -2,2,8-t r im e t h y l-6H -1,3-d io x o lo [4,5-f][2]-
ben zop yr a n -6-on e (27). F r om 7b: TBAF (1 M in THF
containing 5% H2O (v/v); 0.053 mL, 0.053 mmol)28c was added
to a solution of 7b (7.6 mg, 0.017 mmol) in dry THF (0.5 mL),
and the mixture was heated under reflux for 1 h (preheated
bath). The cooled (rt) reaction mixture was concentrated and
fractionated by FCC (40% EtOAc in hexane; 5% MeOH in
CH2Cl2) to give 7b (3.8 mg, 50%) and 27 (1 mg, 20%). F r om
39: TBAF (1 M in THF containing 5% H2O (v/v); 0.039 mL,
0.039 mmol)28c was added to a solution of 39 (6.4 mg, 0.011
mmol) in dry THF (0.5 mL), and the mixture was heated under
reflux for 30 min (preheated bath). Workup as above gave 27
(1.8 mg, 58%). F r om 26:20 A suspension of 26 (18 mg, 0.045
mmol) and 10% Pd-C (∼5 mg) in EtOH (1.5 mL) was stirred
at rt under H2 (1 atm) for 30 min. The mixture was filtered
with the aid of EtOAc, and the combined filtrate and washings
were concentrated and fractionated by MPC (5% MeOH in
CH2Cl2) to give 27 (9.5 mg, 79%): IR νmax 3325, 2984, 1655,
(1′R,3a R,4R,7a R)-N-[1-(Hexa h yd r o-2,2-d im eth yl-6-oxo-
1,3-b e n zod ioxol-4-yl)-2-oxop r op yl]-2-(t r im e t h ylsilyl)-
eth a n esu lfon a m id e (18). F r om 6b: CrO3 oxidation of 6b
(96 mg, 0.23 mmol) according to the general procedure
described for the preparation of 8b gave 18 (85 mg, 89%) after
fractionation by FCC (40% EtOAc in hexane). F r om 15:
Ozonolysis of 15 (5.0 mg, 0.012 mmol), according to the
procedure described for the preparation of 8b, gave 18 as an
oil (4.4 mg, 91%) after fractionation by FCC (50% EtOAc in
1
1593, 1226, 1094 cm-1; H NMR δ 4.51 (1H, dq, J ) 1.5, 6.5
Hz), 3.77 (1H, ddd, J ) 6, 9.5, 11 Hz), 3.62 (1H, dd, J ) 9.5,
9.5 Hz), 3.21 (1H, dd, J ) 1.5, 3.5 Hz), 2.92 (1H, ddd, J ) 1, 6,
17.5 Hz), 2.81 (1H, ddd, J ) 1, 3, 3.5, 9.5 Hz), 2.62 (1H, ddd,
J ) 3, 11, 17.5 Hz), 1.48 (3H, s), 1.45 (3H, s), 1.41 (3H, d, J )
6.5 Hz); 13C NMR δ 175.1 (s), 171.1 (s), 111.6 (s), 90.6 (s), 78.9
(d), 76.1 (d), 74.4 (d), 47.3 (d), 42.0 (d), 34.8 (t), 27.2 (q), 27.0
(q), 17.5 (q); CIMS m/z (relative intensity) 270 ([M + 1]+, 100),
212 (12), 57 (69).
1
hexane): IR νmax 3273, 2953, 1720, 1120, 858 cm-1; H NMR
δ 5.28 (1H, d, J ) 9 Hz), 4.40 (1H, dd, J ) 2.5, 9 Hz), 3.85
(1H, dd, J ) 9, 10 Hz), 3.70 (1H, ddd, J ) 4.5, 9, 13.5 Hz),
2.91 (3H, m), 2.59 (1H, dd, J ) 13.5, 13.5 Hz), 2.39 (1H, dddd,
J ) 2.5, 5.5, 10, 10 Hz), 2.33 (3H, s), 2.19 (2H, m), 1.52 (3H,
s), 1.50 (3H, s), 1.03 (2H, m), 0.04 (9H, s); 13C NMR (C6D6) δ
204.4 (s), 202.7 (s), 111.6 (s), 78.0 (d), 76.6 (d), 61.7 (d), 49.3
(t), 44.5 (t), 38.5 (t), 37.2 (d), 26.9 (q), 26.8 (q), 26.6 (q), 10.3
(t), -2.4 (q); CIMS m/z (relative intensity) 423 ([M + 18]+, 100),
406 ([M + 1]+, 35), 199 (82), 90 (63).
(3a R,4R,4′R,5R,7a R)-4-[Hexa h yd r o-2,2-d im eth yl-6-oxo-
1,3-ben zodioxol-4-yl]-5-m eth yl-3-[2-(tr im eth ylsilyl)eth an e-
su lfon yl]-2-oxa zolid in on e (29). F r om 28: NaH (80% dis-
persion in oil, 21 mg, 0.70 mmol) was added to a stirred
solution of 28 (63 mg, 0.23 mmol) in THF (4 mL) at 0 °C. After
5 min, SES-Cl11 (180 mg, 0.90 mmol) was added, and the
mixture was stirred at 0 °C for 1 h and at rt for an additonal
2 h. The reaction mixture was diluted with EtOAc, washed
with saturated NH4Cl, dried over Na2SO4, concentrated, and
fractionated by FCC (40% EtOAc in hexane) to give recovered
28 (19 mg, 30%) and 29 as an oil (68 mg, 69%). F r om 13b:
Ozonolysis of 13b (23 mg, 0.053 mmol), according to the
procedure described for the preparation of 8b, gave 29 as an
oil (21 mg, 91%) after fractionation by FCC (50% EtOAc in
hexane): IR νmax 2985, 2849, 1778, 1720, 1354, 1139 cm-1; 1H
(1′R,3a R,4R,6R,7a R)-N-[1-(H exa h yd r o-6-h yd r oxy-2,2-
d im eth yl-1,3-ben zod ioxol-4-yl)-2-oxop r op yl]-2-(tr im eth -
ylsilyl)eth a n esu lfon a m id e (19). Excess NaBH4 (50 mg, 1.3
mmol) was added to a solution of 18 (165 mg, 0.41 mmol) in
1:1 MeOH:CH2Cl2 (10 mL) at -78 °C. After stirring for 30
min, the reaction was quenched by addition of acetaldehyde
(0.2 mL) and stirred for an additional 30 min at -78 °C. The
reaction mixture was allowed to warm to rt and then was
diluted with CH2Cl2, washed with water, dried over Na2SO4,
concentrated, and fractionated by FCC (75% EtOAc in hexane)
to give 19 (145 mg, 87%). A similar reaction on much smaller
scale (18: 4.0 mg, 0.0099 mmol) and lower concentration (1
(36) The spectral data agreed closely with those previously reported5d
for (()-isomer.
(37) Imamoto, T.; Takiyama, N.; Nakamura, K.; Hatajima, T.;
Kamiya, Y. J . Am. Chem. Soc. 1989, 111, 4392.

Synthetic Studies on Actinobolin and Bactobolin
J . Org. Chem., Vol. 61, No. 16, 1996 5505
NMR δ 4.73 (1H, dq, J ) 1.5, 6.5 Hz), 4.08 (1H, dd, J ) 1.5, 3
Hz), 3.78-3.62 (3H, m), 3.38 (1H, ddd, J ) 5, 14, 14 Hz), 2.95
(1H, dd, J ) 1.5, 4.5 Hz), 2.63-2.35 (3H, m), 2.16 (1H, dd, J
) 12, 16 Hz), 1.50 (3H, d, J ) 6.5 Hz), 1.49 (3H, s), 1.45 (3H,
s), 1.19 (1H, ddd, J ) 4, 15, 15 Hz), 1.08 (1H, ddd, J ) 4, 15,
15 Hz), 0.07 (9H, s); 13C NMR δ 203.7 (s), 152.5 (s), 112.5 (s),
78.7 (d), 77.0 (d), 74.3 (d), 63.4 (d), 50.6 (t), 44.8 (t), 40.1 (t),
38.0 (d), 27.1 (q), 26.7 (q), 21.0 (q), 9.3 (t), -1.9 (q); CIMS m/z
(relative intensity) 451 ([M + 18]+, 100), 434 ([M + 1]+, 31),
418 (13), 284 (28), 172 (41).
2.87 (1H, ddd, J ) 2, 5.5, 18 Hz), 2.63 (1H, ddd, J ) 4.5, 10,
18 Hz), 1.47 (3H, s), 1.46 (3H, s), 1.45 (3H, d, J ) 6.5 Hz),
1.12 (2H, m), 0.04 (9H, s); 13C NMR δ 160.4 (s), 159.7 (s), 151.4
(s), 134.2 (s), 128.8 (d), 128.6 (d), 128.5 (d), 112.6 (s), 110.2 (s),
78.6 (d), 75.6 (d), 73.1 (d), 71.1 (d), 51.4 (t), 51.1 (d), 42.8 (d),
35.0 (t), 27.1 (q), 26.8 (q), 18.3 (q), 10.5 (t), -1.9 (q); CIMS m/z
(relative intensity) 585 ([M + 18]+, 21), 568 ([M + 1]+, 9), 524
(16), 434 (62), 358 (17), 418 (25), 91 (100), 90 (46).
Ackn owledgm en t. Financial support from the Natu-
ral Sciences and Engineering Research Council (Canada)
and the University of Saskatchewan is gratefully ac-
knowledged.
(3a R,8R,9R,9a R,9b R)-3a ,4,8,9,9a ,9b -H exa h yd r o-2,2,8-
tr im eth yl-9-[[2-(tr im eth ysilyl)eth a n esu lfon yl]a m in o]-6-
oxo-6H -1,3-d ioxolo[4,5-f][2]b en zop yr a n -5-yl
P h en yl-
m eth yl Ca r bon a te (39). NaH (prewashed with hexane, 3
mg, 0.12 mmol) was added to solution of 7b (22 mg, 0.051
mmol) and benzyl chloroformate (48 mg, 0.28 mmol) in THF
(1.5 mL) at rt under argon. The reaction mixture was stirred
for 1.5 h and then was diluted with EtOAc, washed with
aqueous NaHCO3, dried over Na2SO4, concentrated, and
fractionated by FCC (30% EtOAc in hexane) to provide 39 (27
mg, 96%): IR νmax 3278, 2986, 1762, 1721, 1618, 1188, 842
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedures and spectral data for 6b, 12, 24, 28, 31-36, 40-43,
and 45; 13C NMR spectra for 8b, 9b, 19, and 20; 1H NMR
spectra for all compounds (45 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.
1
cm-1; H NMR δ 7.39 (5H, m), 5.28 (1H, d, J ) 12 Hz), 5.23
(1H, d, J ) 12 Hz), 4.86 (1H, d, J ) 10 Hz), 4.62 (1H, dq, J )
2, 6.5 Hz), 4.08 (1H, ddd, J ) 3, 3, 10 Hz), 3.84 (1H, ddd, J )
5.5, 10, 10 Hz), 3.68 (1H, dd, J ) 10, 10 Hz), 3.03 (3H, m),
J O960579C