Asymmetric reductive cyclization. Total synthesis of (-)-C10-desmethyl arteannuin B

Article abstract of DOI:10.1021/jo9600417

An efficient total synthesis of (-)-C₁₀-desmethyl arteannuin B (5) has been achieved. The sequence features the use of a chiral auxiliary to introduce absolute asymmetry at an early stage and a stereoselective, chelation-controlled reductive cyclization of 8, using samarium diiodide as the reducing agent. The methodology promises to be applicable to the synthesis of a wide range of analogs capable of being converted to potent antimalarial agents related to artemisinin (1).

Full text of DOI:10.1021/jo9600417

3240
J . Org. Chem. 1996, 61, 3240-3244
Articles
Asym m etr ic Red u ctive Cycliza tion . Tota l Syn th esis of
(-)-C10-Desm eth yl Ar tea n n u in B
Michael Schwaebe and R. Daniel Little*
Department of Chemistry, University of California, Santa Barbara, Santa Barbara, California 93106
Received J anuary 4, 1996X
An efficient total synthesis of (-)-C10-desmethyl arteannuin B (5) has been achieved. The sequence
features the use of a chiral auxiliary to introduce absolute asymmetry at an early stage and a
stereoselective, chelation-controlled reductive cyclization of 8, using samarium diiodide as the
reducing agent. The methodology promises to be applicable to the synthesis of a wide range of
analogs capable of being converted to potent antimalarial agents related to artemisinin (1).
Sch em e 1
Malaria continues to be a scourge to millions of people
throughout the world. Unfortunately, drug-resistant
strains have developed. Consequently, several of the
previously used antimalarial agents are no longer effica-
cious.1,2 One of the more effective materials in the
treatment of the disease is artemisinin (qinghaosu, 1), a
tetracyclic sesquiterpene lactone possessing an interest-
ing 1,2,4-trioxane unit.3 Mechanistic studies have shed
light upon the importance of this subunit and its interac-
tion with iron-rich parasites, in the expression of biologi-
cal activity.4
A number of syntheses of artemisinin (1), and its
structural analogs, have been developed.5 Several rely
upon the use of artemisinic acid (6) and arteannuin B
(4) as precursors.6,7 Like the natural product, the analog
10,11-didesmethyl artemisinin (3), displays significant
antimalarial activity against strains of Plasmodium
falciparum.8 We report the total synthesis of (-)-C10-
desmethyl arteannuin B (5), a substance that could be
converted to 2 using known methodology.6 Our approach,
illustrated in Scheme 1, is highlighted by its simplicity
and efficiency (17% overall yield).
Sch em e 2a
The key step in the sequence is the reductive cycliza-
tion designed to convert enone 8 to the bicyclic hydroxy
ester 7. Our previous studies utilizing the electroreduc-
tive cyclization (ERC) reaction en route to natural
products suggested that it would be ideally suited to
achieving this objective.9-12
As shown in Scheme 2, the sequence begins with the
conversion of 3-methylcyclohexenone (9) to the corre-
sponding (S)-(-)-1-amino-2-(methoxymethyl)pyrrolidine
X Abstract published in Advance ACS Abstracts, May 1, 1996.
(1) Klayman, D. L. Science 1985, 228, 1049-1055.
(2) Sharma, R. P.; Zaman, S. S. Heterocycles 1991, 32, 1593-1638.
(3) J ung, M. Curr. Med. Chem. 1994, 1, 35-49.
(4) Posner, G. H. J . Am. Chem. Soc. 1995, 117, 5885-5886.
(5) Zhou, W.-S. Pure Appl. Chem. 1986, 58, 817-824.
(6) Lansbury, P. T.; Nowak, D. M. Tetrahedron Lett. 1992, 33, 1029-
1032.
(SAMP) hydrazone (as a 1.4:1 mixture of E/ Z isomers)
10,13 thereby introducing absolute asymmetry at an early
stage of the sequence. Formation of the carbanion of 10
at -78 °C, followed by treatment with iodide 11 afforded
a mixture of regioisomers resulting from alkylation at
the R- and R′-positions.14 The selectivity and yield
improved significantly when 10 mol % lithium chloride,15
premixed with diisopropylamine in THF, was used in the
(7) Roth, R. J .; Acton, N. J . Nat. Prod. 1989, 52, 1183-1185.
(8) Avery, M. A.; J ennings-White, C.; Chong, W. K. M. J . Org. Chem.
1989, 54, 1792-1795.
(9) Fry, A. J .; Little, R. D.; Leonetti, J . J . Org. Chem. 1994, 59,
5017-5026.
(10) Sowell, C. G.; Wolin, R. L.; Little, R. D. Tetrahedron Lett. 1990,
31, 485-488.
(11) Little, R. D.; Baizer, M. M.; Moe¨ns, L. J . Org. Chem. 1986, 51,
4497-4498.
(13) Enders, D. In Asymmetric Synthesis, 1st ed.; Morrison, J . D.,
Ed.; Academic Press Inc.: Orlando, 1984; Vol. 3, pp 275-340.
(14) Ahlberg, P.; Wu, Y. J . Org. Chem. 1994, 59, 5076-5077.
(15) Lipshutz, B. H.; Wood, M. R.; Lindsley, C. W. Tetrahedron Lett.
1995, 36, 4385-4388.
(12) Little, R. D.; Sowell, C. G. In Electroorganic Synthesis Festschrift
for Manuel M. Baizer; Little, R. D., Weinberg, N. L., Eds.; Marcel
Dekker: New York, 1991; Vol. 1, pp 232-330.
S0022-3263(96)00041-2 CCC: $12.00 © 1996 American Chemical Society

Asymmetric Reductive Cyclization
J . Org. Chem., Vol. 61, No. 10, 1996 3241
stereoisomers 18, none of which corresponded to an
adduct that could be converted to arteannuin B (4).
We were delighted to discover that in contrast to both
of the transformations illustrated above, samarium di-
iodide proved to be a very effective reagent in promoting
the cyclization of enone 8 (see eq 3).21 Careful attention
to several experimental parameters proved critical. For
example, it was important to thoroughly de-gas the
solution containing the substrate prior to use (N2 stream).
In this manner, it was possible to use only 2.5 equiv of
the reducing agent instead of the customary large excess.
To obviate simple reduction of the enone to the corre-
sponding allylic alcohol, and to promote the desired
cyclization required the simultaneous, but separate, addi-
tion of a dilute solution of the proton donor (MeOH) and
samarium diiodide to the substrate. Even then, it proved
important to carry out the reduction at 0 °C, rather than
at -78 °C. At the lower temperature, only reduction to
the allylic alcohol was observed, whereas at 0 °C, cycliza-
tion proved fast enough to compete effectively.
Under the conditions just described, the samarium
diiodide-promoted reductive cyclization of 8 afforded the
bicyclic γ-hydroxy ester 7 in greater than 95% yield.
Within the limits of detection by 200 MHz 1H NMR
spectroscopy, no stereoisomer was produced. Given the
instability of the product to even trace amounts of acid,
a standard workup proved unsatisfactory, leading to low
yields and poor mass balance. The workup was simpli-
fied and the yield improved significantly by using Roch-
elle’s salt in an aqueous medium containing 10% potas-
sium carbonate. Rather than the insoluble salts which
normally accompany a basic workup, the aqueous layer
was homogeneous and clear. To the best of our knowl-
edge, this procedure has not been employed previously.
We recommend its use in the isolation of acid-sensitive
materials.
preparation of LDA. In this manner, the time needed
for carbanion formation was reduced from 8 h at -78 °C
to 4 h at -95 °C. Alkylation then proceeded cleanly to
afford a single diastereomer, 12, in 78% yield. Oxidative
cleavage of the chiral auxiliary and removal of the acetal
was achieved using ozone followed by treatment with
trifluoroacetic acid in methylene chloride/water.16 Ole-
fination of the resulting keto aldehyde 13 proceeded
routinely, to afford enoate 8 in a 95% yield as a 7:1
mixture of E/ Z isomers. Once separated, the stage was
set to attempt reductive cyclization.
Both electrochemical and reagent-induced cyclizations
were explored.10,17 The need to form a σ-bond between
the â-carbon of the enoate and the ketone carbonyl carbon
calls for either the selective 1,4-reduction of the unsatur-
ated ester, or 1,2-reduction of the enone. However, the
R,â-unsaturated ester is more difficult to reduce than the
enone (Ep -2.45 vs -2.2 V),18 and reduction of the enone
promised to lead to saturation of the CdC π-bond which
is needed to permit introduction of the epoxide. On the
other hand, reduction of the epoxy ketone 14 illustrated
in eq 1 (Ep -2.1 V), promised to afford the â-hydroxy
ketone 15 as an intermediate,19 a system that maintains
functionality in the form of the hydroxyl group, that can
be converted to the double bond. This intermediate could
then undergo a second two-electron reduction forming the
cyclized material 16. Consequently, we first elected to
examine the electroreductive cyclization of epoxy ketone
14, rather than enone 8.
While the cyclization proceeded in a respectable 78-
95% yield, it was not a stereoselective process (see eq 1).
This observation is similar to that of Dreiding and co-
workers,20 whose attempts to effect cyclization of 17 using
a zinc-copper couple, culminated in the formation of
Determination of the stereochemical outcome was
facilitated by comparing the coupling constants, J ab and
J
ac, in structure 7 with values reported by Dreiding for
(16) Ellison, R. A.; Lukenbach, E. R.; Chiu, C.-W. Tetrahedron Lett.
1975, 499.
(17) Enholm, E. J . Tetrahedron Lett. 1989, 30, 1063.
(18) Fry, A. J . Synthetic Organic Electrochemistry, 2nd ed.; J ohn
Wiley & Sons: New York, 1989.
the corresponding protons in arteannuin B (4).20 As
illustrated, the agreement is striking, leaving little doubt
concerning the accuracy of the assignment.
(19) Shapiro, E. L.; Gentles, M. J . J . Org. Chem. 1981, 46, 5017-
We attribute the stereoselectivity of the cyclization to
the formation of a transition state similar to 19, one with
5019.
(20) Dreiding, A. S.; Goldberg, O.; Deja, I.; Rey, M. Helv. Chim. Acta
1980, 63, 2455-2468. Epoxide 14 was isolated as a pure substance
from a mixture of diastereomers which were obtained in the epoxida-
tion of the corresponding enone.
(21) Molander, G. A. In Organic Reactions; Wiley: New York, 1995;
Vol. 46, p v.

3242 J . Org. Chem., Vol. 61, No. 10, 1996
Schwaebe and Little
lactone 22 in a 46% yield, over four steps (8 f 22) without
the need to purify labile intermediates. When purifica-
tion was attempted, significant material loss was expe-
rienced.
the side chain oriented axially so that re-face attack on
the enone carbonyl carbon is achieved. There are many
examples suggesting the formation of medium to large
ring transition structures involving samarium che-
lates.22,23 Clearly, the high oxophilicity of samarium is
advantageous in this context to the extent that its
positioning between the two oxygens promotes stereo-
control.
Treatment of 22 with 3 equiv of LDA and formalin
afforded the desired â-hydroxy lactone.27 This material
was most efficiently used directly in the subsequent
mesylation-elimination processes to deliver enantiomeri-
cally pure desmethyl arteannuin B (5) in a 68% yield over
the final three steps. The structure of 5 was confirmed
via X-ray crystallographic analysis; an ORTEP repre-
sentation has been provided as supporting information.29
In principle, one could complete the synthesis of
desmethyl arteannuin B (5) in any of a number of ways.
However, given the acid lability of many of the interme-
diates, the sequence of events proved critical. A satisfac-
tory solution was initiated by reduction of 7 to the diol.
Hydroxyl-directed epoxidation (MCPBA, 20 mol % K2-
CO3) afforded epoxide 2024 (Scheme 3), the first substance
after the cyclization that did not display exceptional acid
lability. At this stage, we assessed the diastereoselec-
tivity of the initial alkylation, cyclization, and epoxidation
events. To do so, we converted 20 to the corresponding
Finally, we note that treatment of the initially formed
cyclized material, 7, with LAH followed by 5 mol % of
TPAP and NMO as the cooxidant affords lactone 23 in
76% yield (two steps).26 This material is the 10,11-
didesmethyl analog of an intermediate which has been
converted to artemisinin (1).6 We anticipate that the
same methodology could be used to convert lactone 23
to 3, the 10,11-didesmethyl analog of artemisinin that
displays potent antimalarial activity against resistant
strains of Plasmodium falciparum.8
1
Mosher’s ester 21. The H NMR spectrum displayed a
27/1 ratio of signals corresponding to the methyl groups
of the diastereomeric methyl ethers, i.e., to a diastereo-
meric excess of >93%.25,28
Sch em e 3
In summary, we have developed an efficient total
synthesis of (-)-C10-desmethyl arteannuin B (5) begin-
ning with readily available starting materials. The route
features a stereoselective, chelation-controlled reductive
cyclization to convert 8 to 7. The methodology appears
to be sufficiently general to allow the construction of a
(27) Schlosser, M.; Coffinet, D. Synthesis 1971, 380.
(28) This value is consistent with that obtained earlier from an
examination of the Mosher ester of alcohol 24, prepared from the
alkylation of SAMP hydrazone 10 with the corresponding silyl-
protected iodide. Within the limits of detection of 200 MHz NMR, no
other diastereomer was observed.
Lactonization was conveniently accomplished using 5
mol % of tetra-N-propylammonium perruthenate (TPAP)
and 4-methylmorpholine N-oxide (NMO) as the cooxi-
dant.26 This exceptionally fine method afforded the epoxy
(22) Enholm, E. J .; J iang, S. J . Tetrahedron Lett. 1992, 33, 313-
316.
(23) Chiara, J . L.; Cabri, W.; Hanessian, S. Tetrahedron Lett. 1991,
32, 1125-1128.
(24) Henbest, H. B.; Wilson, R. A. L. J . Am. Chem. Soc. 1957, 1958.
(25) Mosher, H. S.; Dale, J . A. J . Am. Chem. Soc. 1973, 95, 512.
(26) Ley, S. V.; Norman, J .; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639-666.
(29) The author has deposited atomic coordinates for this structure
with the Cambridge Crystallographic Data Centre. The coordinates
can be obtained, on request, from the Director, Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ, UK.

Asymmetric Reductive Cyclization
J . Org. Chem., Vol. 61, No. 10, 1996 3243
32.31, 29.02, 27.11, 26.47, 24.48, 23.70, 22.25, 22.25; MS m/ z
(relative intensity) 338 (8), 307 (12), 294 (20), 293 (100), 192
(9), 84 (10), 75 (24), 71 (23), 70 (19), 45 (15); HRMS calcd for
C19H34N2O3 338.2569; found 338.2569; σ ) 0.1 ppm.
great many analogs of the potent antimalarial agent,
artemisinin (1).
Exp er im en ta l Section
(R)-(-)-4-(4′-Meth yl-2′-oxocycloh ex-3′-en -1′-yl)bu tyr a l-
d eh yd e Dim eth yl Aceta l (12a ). The alkylated SAMP hy-
drazone 12 (4.82 g, 14.3 mmol) was dissolved in 50 mL of
methylene chloride and cooled to -78 °C. Ozone was bubbled
into the solution until TLC analysis indicated that the reaction
was complete. The mixture was degassed with nitrogen and
allowed to warm to room temperature with continuous degas-
sing. The solvent was removed under vacuum, and the
resulting brown oil was chromatographed on silica gel (hex-
anes/EtOAc, 5:1) yielding the enone as a colorless oil (2.77 g,
12.3 mmol, 86%): [R]25D ) -68.1° (c, 0.010, CH2Cl2); IR (neat)
Gen er a l. Melting points are uncorrected. Specific rota-
tions were measured with the use of a 10 cm cell. 1H NMR
spectra were recorded at 200 or at 500 MHz, and 13C NMR
spectra were recorded at 200 MHz. All spectra were recorded
in CDCl3 as the solvent, and chemical shifts are reported in δ
relative to TMS. IR spectra of solid compounds were obtained
as solutions in CDCl3 (NaCl plates). Solvents were dried
(drying agent in parentheses) and distilled prior to use: Et2O
and THF (Na/benzophenone) CH2Cl2 and Et3N (CaH2). Sa-
marium diiodide was purchased from Aldrich (0.1 M in THF)
and used as is. Thin-layer chromatography (TLC) was per-
formed with precoated glass plates; Kieselgel 60 GF254 (Merck).
Column chromatography was performed using ICN (32-63,
60A) silica gel and the indicated solvents reported by volume
(v/v). Capillary gas chromatography was performed on a
Hewlett-Packard 5890 gas chromatograph. All compounds
were purified. Those which were unstable to purification were
used in subsequent reactions as crude products. All reactions
were conducted under an argon atmosphere
2942, 2871, 2830, 1646, 1455, 1367, 1189 cm-1 1H NMR
;
(CDCl3, 500 MHz) δ 5.80 (q, J ) 1.5 Hz, 1H), 4.34 (dt, J ) 1.5,
6 Hz, 1H), 3.29 (s, 3H), 3.28 (s, 3H), 2.29 (t, J ) 6 Hz, 2H),
2.17 (dddd, J ) 5, 7, 10.5, 12 Hz, 1H), 2.08 (dq, J ) 5, 13.5
Hz, 1H), 1.90 (s, 3H), 1.84 (m, 1H), 1.71 (dddd, J ) 7, 10.5,
13.5, 17.5 Hz, 1H), 1.60 (m, 2H), 1.43 (m, 1H), 1.35 (ddd, J )
3.5, 7.5, 11.5 Hz, 2H); 13C NMR (CDCl3, 200 MHz) δ 201.69,
161.52, 126.43, 104.57, 52.95, 52.86, 45.47, 32.75, 30.32, 29.08,
27.65, 24.30, 22.20; MS m/ z (relative intensity) 196 (13, M+
- CH3O), 195 (100, M - CH3O), 194 (6), 180 (8), 136 (10), 135
(76), 75 (25); HRMS calcd for C12H18O2 (M - CH3OH) 194.1307;
found 194.1302; σ ) 2.7 ppm.
(S )-(+)-[(3-Me t h ylcycloh e x-2-e n -1-yid in e )a m in o]-2-
(m eth oxym eth yl)p yr r olid in e (10). To a solution of 3-me-
thylcyclohex-2-en-1-one (10.23 g, 92.9 mmol) dissolved in 100
mL of benzene was added (S)-(-)-1-amino-2-(methoxymethyl)-
pyrrolidine (SAMP) (12.50 g, 96.0 mmol), and the solution was
refluxed with azeotropic removal of water. After 72 h, the
reaction mixture was allowed to cool, and the solvent was
removed under reduced pressure, affording the crude product
as a yellow oil. Distillation under vacuum afforded 20.05 g
(90.3 mmol, 97%) of the hydrazone in a 1.4:1 ratio of isomers
as a yellow oil (bp 104-106°C, 0.4 mmHg): IR (neat) 2932,
(R)-(-)-4-(4′-Meth yl-2′-oxocycloh ex-3′-en -1′-yl)bu tyr a l-
d eh yd e (13). To 2 mL (26 mmol) of trifluoroacetic acid in 30
mL of water was added the dimethyl acetal 12a (1.35 g, 5.97
mmol) in 30 mL of methylene chloride. The mixture was
stirred at room temperature until TLC analysis indicated that
the reaction was complete (usually 2-4 h). The reaction
mixture was extracted with methylene chloride (3 × 20 mL).
The combined organic layers were washed with a saturated
bicarbonate solution (20 mL) and brine (2 × 20 mL), dried over
magnesium sulfate, and concentrated in vacuo. The resulting
oil was chromatographed on silica gel (hexanes/EtOAc, 5:1)
to give the aldehyde as a colorless oil (1.05 g, 5 83 mmol,
98%): [R]25D ) -32.9° (c, 0.033, CH2Cl2); IR (neat) 3029, 2928,
2865, 2724, 1724, 1673, 1430 cm-1; 1H NMR (CDCl3, 500 MHz)
δ 9.75 (t, J ) 1.5 Hz, 1H), 5.79 (q, J ) 1.5 Hz, 1H), 2.45 (t, J
) 7.5 Hz, 2H), 2.30 (ddd, J ) 4.5, 10 Hz, 1H), 2.17 (dddd, J )
5, 7, 11, 12 Hz, 1H), 2.08 (ddd, J ) 4.5, 9.5, 13 Hz, 1H), 1.92
(s, 3H), 1.82 (dddd, J ) 5.5, 11, 13.5, 16.5 Hz, 1H), 1.61-1.77
(m, 4H), 1.38 (dddd, J ) 5.5, 7.5, 11.5, 13 Hz, 1H); 13C NMR
(CDCl3, 200 MHz) δ 202.67, 201.29, 161.76, 126.37, 45.34,
44.12, 30.42, 28.91, 27.80, 24.30, 19.75; HRMS calcd for
C11H17O2 (P + H)+ 181.1229; found 181.1230; σ ) 0.9 ppm.
1H NMR d a ta for (1′R, 3′R, 4′R)-m eth yl (2E)-6-(3′,4′-
2859, 1720, 1664, 1432 cm-1
;
1H NMR (CDCl3, 500 MHz) δ
6.49 (s, 1H), 5.98 (s, 1H), 3.46 (dd, J ) 2, 7 Hz, 1H), 3.43 (dd,
J ) 3.5, 9 Hz, 1H), 3.34 (s, 3H), 3.33 (s, 3H), 3.16-3.27 (m,
8H), 2.79 (ddd, J ) 4.5, 9, 16 Hz, 1H), 2.54 (dd, J ) 8, 17 Hz,
1H), 2.42 (dd, J ) 8.5, 17 Hz, 1H), 2.34 (m, 1H), 2.1-2.17 (m,
4H), 1.98-2.07 (m, 4H), 1.86 (s, 3H), 1.81 (s, 3H), 1.78 (m, 4H),
1.68 (m, 4H); 13C NMR (CDCl3, 200 MHz) δ 161.83, 161.18,
148.98, 145.76, 123.99, 117.51, 75.35, 75.35, 66.45, 66.33,
59.05, 59.05, 56.01, 54.28, 31.35, 31.13, 30.31, 26.58, 26.58,
26.36, 24.24, 23.97, 22.61, 22.31, 22.05, 22.05; HRMS calcd
for C13H22N2O 222.1732; found 222.1726; σ ) 2.8 ppm.
(2S,5R)-(-)-[[5-(4′,4′-Dim eth oxybu tyl)-3-m eth ylcycloh ex-
2-en -1-ylidin e]am in o]-2-(m eth oxym eth yl)pyr r olidin e (12).
To a solution of diisopropylamine (3.9L mL, 29.8 mmol) and
LiCl (85 mg, 2 mmol) in 20 mL of freshly distilled THF was
added n-butyllithium (2.98 mL of 10M, 29.8 mmol) at -78 °C.
The mixture was allowed to warm to 0 °C over 20 min. The
solution was cooled to -95 °C and the SAMP hydrazone 10
(4.41 g, 19.9 mmol) was added dropwise. The reaction mixture
was stirred at -95 °C for an additional 4 h. After enolate
formation was complete (usually 4 h), 4-iodobutyraldehyde
dimethyl acetal (7.27 g, 29.8 mmol) was added dropwise, and
the reaction mixture, after warming to 0 °C over 30 min, was
quenched with a saturated ammonium chloride solution (30
mL). The aqueous layer was extracted twice with 20 mL
portions of methylene chloride and the combined organic layers
were washed with brine (2 × 30 mL), dried over magnesium
sulfate, and concentrated under reduced pressure to afford a
yellow oil. The crude product was chromatographed on silica
gel (hexane/EtOAc/NEt3, 80:28:2) to yield 12 as a pale yellow
oil (5.29 g, 15.7 mmol, 79%): IR (neat) 2942, 2871, 2830, 1646,
1
ep oxy-4′-m eth yl-2′-oxocycloh ex-1′-yl)h exen oa te (14): H
NMR (CDCl3, 500 MHz) δ 6.95 (dt, J ) 6.5, 15 Hz, 1H), 5.78
(dt, J ) 1.5, 15 Hz, 1H), 3.69 (s, 3H), 3.04 (s, 1H), 2.13 (m,
3H), 1.86 (m, 2H), 1.72 (m, 1H), 1.63 (m, 3H), 1.45 (m, 2H),
1.41 (s, 3H).
(R)-(+)-Meth yl (2E)-6-(4′-Meth yl-2′-oxocycloh ex-3′-en -
1′-yl)h exen oa te (8). To a solution of sodium bis(trimethyl-
silyl)amide (1.53 mL, 1.53 mmol) in 10 mL of THF, was added
dropwise via a precooled (-78 °C) cannula, trimethyl phospho-
noacetate (0.28 g, 1.5 mmol) at -78 °C. The mixture was
allowed to stir for 15 min, and then the aldehyde 13 (0.25 g,
1.39 mmol) was diluted in 5 mL of THF and added dropwise
via the precooled cannula. The reaction mixture was allowed
to warm to 0 °C and was quenched with brine. The aqueous
layer was extracted with ether (2 × 20 mL), and the combined
organic layers were washed with brine (20 mL) and dried over
magnesium sulfate, and the solvent was removed in vacuo.
The resulting oil was chromatographed on silica gel (hexanes/
EtOAc, 5:1) to give the unsaturated ester as a colorless oil (0.27
1455, 1367, 1189, 1123 cm-1 1H NMR (CDCl3, 500 MHz) δ
;
5.90 (s, 1H), 4.38 (t, J ) 5.5 Hz, 1H), 3.43 (dd, J ) 4, 9 Hz,
1H), 3.34 (s, 3H), 3.33 (s, 3H), 3.24 (m, 2H), 3.20 (dd, J ) 8,
16.5 Hz, 1H), 3.14 (ddd, 1H), 3.07 (m, 2H), 2.45 (dd, J ) 8.5,
17 Hz, 1H), 2.20 (m, 1H), 2.05 (ddd, J ) 7.5, 13.5, 19.5 Hz,
1H), 1.97 (dd, J ) 5, 18.5 Hz, 1H) 1.88 (ddd, J ) 2.5, 5, 13.5
Hz, 1H), 1.81 (s, 3H), 1.81 (m, 1H), 1.62 (m, 4H), 1.45 (m, 2H),
1.34 (m, 2H); 13C NMR (CDCl3, 200 MHz) δ 164.98, 142.45,
120.48, 101.76, 73.55, 66.05, 58.87, 55.51, 52.63, 52.40, 33.72,
g, 1.15 mmol, 83%, E isomer): [R]25 ) +40.9° (c, 0.029, CH2-
D
Cl2); IR (neat) 2932, 2859, 1719, 1664, 1432 cm-1
;
1H NMR
(CDCl3, 500 MHz) δ 6.95 (dt, J ) 6.5, 15 Hz, 1H), 5.83 (s, 1H),
5.82 (d, J ) 15 Hz, 1H), 3.71 (s, 3H), 2.30 (m, 2H), 2.14-2.24
(m, 2H), 2.07 (ddd, J ) 5, 10, 13.5 Hz, 1H), 1.93 (s, 3H), 1.83
(m, 1H), 1.71 (m, 1H), 1.51 (m, 3H), 1.37 (m, 1H); 13C NMR
(CDCl3, 200 MHz) δ 201.74, 161.87, 149.68, 126.71, 121.53,

3244 J . Org. Chem., Vol. 61, No. 10, 1996
Schwaebe and Little
51.84, 45.68, 32.74, 30.67, 30.66, 29.30, 28.14, 25.98, 24.58;
HRMS calcd for C14H20O3 236.1412; found 236.1409; σ ) 1.5
ppm.
Cl2); mp 144-146 °C; IR (dissolved in CDCl3) 2936, 1781, 1447
cm-1; 1H NMR (CDCl3, 500 MHz) δ 2.83 (s, 1H), 2.60 (dd, J )
14.5, 16.5 Hz, 1H), 2.49 (dd, J ) 7, 16.5 Hz, 1H), 2.21 (m, 1H),
1.75-1.95 (m, 6H), 1.59 (m, 2H), 1.42 (dddd, J ) 3.5, 7, 8.5,
12 Hz, 1H), 1.38 (s, 3H), 1.34 (m, 1H), 1.23 (m, 1H); 13C NMR
(CDCl3, 200 MHz) δ 175.44, 82.99, 57.87, 56.88, 47.14, 37.64,
32.75, 26.22, 25.23, 24.42, 24.07, 22.98, 19.46; HRMS calcd
for C13H18O3 (M + H)+ 223.1334; found 223.1342; σ ) 3.8 ppm.
Anal. Calcd for C13H18O3: C, 70.25; H, 8.16. Found C, 70.19;
H, 8.11.
Met h yl
(1′S,4a ′R,8a ′S)-2-(8a ′-H yd r oxy-7′-m et h yl-
1′,2′,3′,4′,4a′,5′,6′,8a′-octah ydr o-1′-n aph th yl)acetate (7). The
unsaturated ester 8 (0.51 g, 2.16 mmol) was dissolved in 50
mL of dry THF and degassed with argon for 10 min. The
solution was then chilled to 0 °C and SmI2 (0.1 M in THF, 80
mL, 8 mmol) was added slowly via cannula. While the SmI2
was being introduced, methanol (0.41 mL dissolved in 10 mL
THF, 10 mmol) was also added slowly. When TLC analysis
indicated that the reaction was complete (usually under 15
min, the reaction mixture was still dark blue), a saturated
solution of sodium-potassium tartrate containing 10% potas-
sium carbonate was added. The aqueous layer was extracted
with methylene chloride (3 × 30 mL) and the combined organic
layers were washed with brine (2 × 20 mL) and dried over
magnesium sulfate. The solution was then filtered through a
short pad of silica and concentrated in vacuo to yield the
cyclized product (0.53 g): IR (neat) 3494 br, 2927, 2851, 1733,
(-)-(1′S,4′R,7′R,8′R,8a ′S)-2-(7′,8′-Ep oxy-8a ′-h yd r oxy-7′-
m eth ylp er h yd r o-1′-n a p h th yl)p r op en -8a ′-olid e (5). To a
solution of butyllithium (0.90 mL, 2.25 mmol) in 5 mL of dry
THF at -78 °C was added diisopropylamine (0.30 mL, 2.3
mmol); the reaction mixture was warmed to 0 °C over 15 min.
The solution was cooled to -78 °C and the lactone 22 (0.140
g, 0.63 mmol) in 5 mL of THF was added and stirred for 1 h.
Then, 5 mL of a freshly prepared formaldehyde solution in
THF (generated from 0.20 g of paraformaldehyde, 7 mL of THF
and 0.3 mL of BF3‚OEt, Schlosser procedure) was added. The
reaction mixture, maintained at -78 °C for 15 min, was
allowed to warm to 0 °C over 30 min and quenched with
saturated ammonium chloride (10 mL). The aqueous layer
was extracted with methylene chloride (2 × 10 mL), washed
with 0.1 N HCl (10 mL) and then with a 10% bicarbonate
solution, and brine, and dried over magnesium sulfate. The
solvent was removed in vacuo; the resulting crude mixture was
dissolved in methylene chloride, and cooled to 0 °C. Triethy-
lamine (0.50 mL, 3.6 mmol) was added, followed by a dropwise
addition of mesyl chloride (0.20 mL, 2.6 mmol). The reaction
mixture was allowed to stir for 12 h. The reaction was
quenched with a saturated bicarbonate solution (10 mL) and
the aqueous layer was extracted with methylene chloride (2
× 10 mL). The combined organic layers were washed once
with 0.1 M HCl (10 mL) brine and dried over magnesium
sulfate. The solvent was removed in vacuo and replaced with
benzene to which was added DBU (0.47 mL, 3.15 mmol). After
1 h, the reaction was quenched with a saturated ammonium
chloride solution, dried with magnesium sulfate and chro-
matographed on silica gel (hexanes/EtOAc/NEt3, 78:20:2).
Recrystallization from Et2O yielded C10-desmethyl arteannuin
B as a colorless crystalline solid (0.102g, 0.44 mmol, 69%): mp
147-149 °C; [R]25D ) -68.09° (c, 0.010, CH2Cl2); IR (dissolved
1
1717, 1652, 1434 cm-1; H NMR (CDCl3, 500 MHz) δ 5.36 (s,
1H), 3.66 (s, 3H), 2.87 (dd, J ) 5, 15 Hz, 1H), 2.14 (dd, J ) 9,
15 Hz, 1H), 2.12 (m, 1H), 1.98 (m, 2H), 1.86 (m, 2H), 1.68 (s,
3H), 1.62 (m, 4H), 1.44 (m, 2H), 1.35 (dddd, J ) 4, 8, 13.5, 17
Hz, 1H), 1.24 (ddd, J ) 3.5, 12.5, 15 Hz, 1H), 0.98 (ddd, J )
3.5, 12.5, 16 Hz, 1H).
(1′S,4′R,7′R,8′R,8a ′S)-2-(7′,8′-Ep oxy-8a ′-h yd r oxy-7′-m e-
th ylp er h yd r o-1′-n a p h th yl)eth a n ol (20). Lithium alumi-
num hydride (123 mg, 3.24 mmol) was added to 20 mL of dry
THF at 0 °C. To this was added the γ-hydroxy ester 7 (0.53
g, 2.16 mmol) dissolved in 5 mL of THF; stirring continued
until TLC analysis indicated that the reaction was complete
(usually under 15 min). The reaction was quenched with 0.12
mL of water followed by 0.12 mL of 1 N NaOH and then 0.36
mL of water. The mixture was then filtered over a bed of
sodium sulfate and concentrated in vacuo to yield the crude
product as a colorless oil. To the crude diol (0.5 g, 2.16 mmol)
in 15 mL of methylene chloride and 15 mL of hexanes was
added solid potassium carbonate (0.30 g, 2.17 mmol) and
m-chloroperbenzoic acid (0.75 g, 4.35 mmol) at room temper-
ature. The mixture was stirred until TLC analysis indicated
that the reaction was complete (2-4 h). The white slurry was
diluted with 30 mL of methylene chloride, filtered, and washed
with 10% sodium bisulfite (30 mL), followed by saturated
sodium bicarbonate (20 mL), and brine (20 mL). The aqueous
layers were sequentially back-extracted with methylene chlo-
ride (2 × 20 mL), and the resulting extracts were washed with
saturated sodium bicarbonate solution and brine. The organic
extracts were combined, and the solvent was removed in vacuo
to yield the crude epoxide as a viscous colorless oil: 1H NMR
(CDCl3, 500 MHz) δ 5.29 (s, 1H), 3.79 (ddd, J ) 6, 11, 16.5
Hz, 1H), 3.64 (ddd, J ) 5, 8, 11 Hz, 1H), 2.08 (dddd, J ) 5.5,
8, 11, 13.5 Hz, 1H), 1.88 (dd, J ) 5.5, 15 Hz, 1H), 1.65-1.81
(m, 6H), 1.25-1.50 (m, 2H), 1.36 (s, 3H), 1.06-1.19 (m, 2H).
(-)-(1′S,4′R,7′R,8′R,8a ′S)-2-(7′,8′-Ep oxy-8a ′-h yd r oxy-7′-
m eth ylp er h yd r o-1′-n a p h th yl)eth a n -8a ′-olid e (22). To the
crude epoxide 20 (0.5 g, 2.16 mmol), dissolved in 20 mL of
methylene chloride, was added crushed 4 Å molecular sieves
(1 g), 4-methylmorpholine N-oxide (0.38 g, 3.25 mmol), and
tetrapropylammonium perruthenate(VII) (35 mg, 0.1 mmol)
at 0 °C. The reaction mixture was allowed to warm to room
temperature with constant stirring. When the reaction was
complete, as indicated by TLC analysis, the mixture was
diluted with 20 mL of methylene chloride and filtered through
a short pad of silica. The solvent was removed in vacuo, and
the resulting crude oil was chromatographed on silica gel
(hexanes/EtOAc/NEt3, 88:10:2), yielding a white crystalline
solid. Recrystallization from Et2O yielded the lactone 22 (0.22
g, 1.0 mmol, 46% over 4 steps). [R]25D ) -10.2° (c, 0.024, CH2-
1
in CDCl3) 3099, 2952, 2927, 2863, 1764, 1442 cm-1; H NMR
(CDCl3, 500 MHz) δ 6.16 (d, J ) 4 Hz, 1H), 5.42 (d, J ) 4 Hz,
1H), 2.73 (dddd, J ) 3, 6, 9.5, 12 Hz, 1H), 2.67 (s, 1H), 2.07
(dddd, J ) 1.5, 3, 7, 8.5 Hz, 1H), 1.76-1.95 (m, 4H), 1.64
(ddddd, J ) 1.5, 4.5, 8.5, 12, 13.5 Hz, 1H), 1.49 (dddd, J ) 4,
8, 12.5, 17 Hz, 1H), 1.35, (m, 2H), 1.34 (s, 3H), 1.28 (m, 2H);
13C NMR (CDCl3, 200 MHz) δ 169.93, 138.44, 117.76, 81.26,
58.33, 58.10, 52.40, 37.01, 26.46, 24.72, 24.47, 22.90, 22.14,
19.52; HRMS calcd for C14H18O3 (M + H)+ 235.1334; found
235.1336; σ ) 0.9 ppm. Anal. Calcd for C14H18O3: C, 71.77;
H, 7.74. Found C, 71.72; H, 7.80.
Ack n ow led gm en t. We express our sincere grati-
tude to the Electric Power Research Institute (EPRI)
and the National Science Foundation (NSF) who, in a
jointly sponsored program, have funded this research.
Su p p or tin g In for m a tion Ava ila ble: Spectral data for
compounds 5, 7, 8, 10, 12, 12a , 13, 20, and 22 and Figure 1
(25 pages). This material is contained in libraries on micro-
fiche, immediately follows this article in the microfilm version
of this journal, and can be ordered from the ACS; see any
current masthead page for ordering information.
J O9600417