Asymmetric formal synthesis of (-)-cryptosporiopsin, a metabolite of Phialophora asterris, and its 3-deschloro congener

Article abstract of DOI:10.1139/v04-145

An asymmetric formal synthesis of the fungal metabolites (-)-cryptosporiopsin and (-)-3-deschlorocryptosporiopsin has been accomplished. The asymmetry was introduced via a diastereoselective [2+2]-cycloaddition between an enoxy-ester, bearing a bulky chir

Full text of DOI:10.1139/v04-145

1686
Asymmetric formal synthesis of (–)-
cryptosporiopsin, a metabolite of Phialophora
asterris, and its 3-deschloro congener
David I. MaGee, Tammy Mallais, and George M. Strunz
Abstract: An asymmetric formal synthesis of the fungal metabolites (–)-cryptosporiopsin and (–)-3-deschlorocrypto-
sporiopsin has been accomplished. The asymmetry was introduced via a diastereoselective [2+2]-cycloaddition between
an enoxy-ester, bearing a bulky chiral auxiliary, and dichloroketene. Diastereomeric ratios of 10:1 were obtained from
the reaction mixture, but this could be increased by simple crystallization. Completion of the synthesis was only ac-
complished after the serendipitous discovery of a base-catalyzed rearrangement to furnish the requisite 3-chloroenone
moiety.
Key words: diastereoselective [2+2]-cycloaddition, rearrangement, chiral auxiliary.
Résumé : On a effectué une synthèse asymétrique formelle des métabolites de champignons, les (–)-cryptosporiopsine
et (–)-3-deschlorocryptosporiopsine. L’asymétrie a été introduite par le biais d’une cycloaddition diastéréosélective
[2+2] entre le dichlorocétène et un ester énoxy portant un auxiliaire chiral auxiliaire volumineux. On a obtenu des rap-
ports de diastéréomères de 10:1 à partir du mélange réactionnel, mais il a été possible de les augmenter par cristallisa-
tion simple. On n’a complété la synthèse qu’après la découverte fortuite d’une transposition catalysée par les bases qui
permet d’obtenir la portion 3-chloroénone requise.
Mots clés : cycloaddition diastéréosélective [2+2], transposition, auxiliaire chiral.
[Traduit par la Rédaction] MaGee et al. 1691
Introduction
tion, was a capricious and low-yield process. A later synthe-
sis involving ring contraction of a pre-formed quinonoid
intermediate provided a somewhat more efficient route to
cryptosporiopsin (9).
Following elucidation of the interesting chemistry associ-
ated with dichloroketene cycloaddition with olefins (10), the
important synthetic potential of this reaction allowing ready
access, inter alia, to a variety of substituted four- and five-
membered ring systems (see, e.g., refs. 10 and 11) was rec-
ognized.
In 1999, a highly efficient racemic synthesis of 3-
deschlorocryptosporiopsin (4), a metabolite of S. affinis (1),
was reported from our laboratories (12), in which the pivotal
step was cycloaddition of dichloroketene with methyl 2-
trimethylsilyloxy-2-hexenoate (5) (Fig. 2). Whereas the orig-
inal biomimetic route to cryptosporiopsin via contraction of
a six-membered ring was not readily amenable to chiral
modification, the dichloroketene cycloaddition process can
be conducted diastereoselectively, e.g., when the olefin bears
a suitable chiral auxiliary (13).
Cryptosporiopsin (1, Fig 1) and chlorinated cyclopenten-
one congeners were isolated in 1969, independently from
Sporormia affinis (1), Cryptosporiopsis sp. (2), and Peri-
conia macrospinosa (3). Several biogenetically related
dihydroisocoumarins from P. macrospinosa (e.g., 2) had
been reported previously (4), and new members of the group,
including the enantiomeric (–)-cryptosporiopsin (3) from
Phialophora asterris (5), were subsequently disclosed (6).
The antibiotic activity of cryptosporiopsin against an array
of plant pathogenic fungi and other microorganisms was de-
scribed (7) and this, together with its intriguing richly
functionalized structure, provided the impetus for develop-
ment of a synthetic route to the metabolite.
Total synthesis of cryptosporiopsin, in racemic form, was
first accomplished in 1973 via a biomimetic pathway (8).
The key step in this synthesis, ring contraction of a phenolic
precursor through the action of alkaline hypochlorite, though
providing access to advanced intermediate in a single opera-
In the present paper we report application of this strategy
in an enantioselective synthesis of (–)-cryptosporiopsin, a
product of Phialophora asterris (5), as well as that of an
enantiomer of the S. affinis metabolite, 3-deschlorocrypto-
sporiopsin (1).
Greene et al. (14) have shown that diastereoselectivities of
around 16:1 can be realized in the cycloaddition of
dichloroketene with a chiral enol ether such as 7 in Fig. 3.
We considered that the most effective chiral induction for
our purposes might be realized if the cycloaddition of
dichloroketene was conducted with a cyclic precursor, such
Received 7 May 2004. Published on the NRC Research Press
Web site at http://canjchem.nrc.ca on 2 February 2005.
D.I. MaGee1 and T. Mallais. Department of Chemistry,
University of New Brunswick, Fredericton, NB E3B 6E2,
Canada.
G.M. Strunz. Department of Chemistry, University of New
Brunswick, Fredericton, NB E3B 6E2, Canada and Canadian
Forest Service–Atlantic Forestry Centre, Natural Resources
Canada, P.O. Box 4000, Fredericton, NB E3B 5P7, Canada.
1Corresponding author (e-mail: dmagee@unb.ca).
Can. J. Chem. 82: 1686–1691 (2004)
doi: 10.1139/V04-145
© 2004 NRC Canada

MaGee et al.
1687
Fig. 1. Various fungal chlorinated metabolites.
sion to produce diastereometrically enriched cyclopentanone
12 in 90% yield.4
This was deemed an appropriate point in the projected
synthetic sequence for removal of the chiral auxiliary, pref-
erably concomitant with direct transesterification. Familiar
synthetic terrain, traversed previously (though with trimethyl-
silyloxy rather than t-butyldimethylsilyloxy protection) (12),
would thus be attained. In any event, attempts at direct trans-
esterification of 12 to the corresponding methyl ester using
conventional methodology (see, e.g., ref. 15) were unsuccess-
ful, undoubtedly as a result of the considerable steric encum-
brance in the vicinity of the ester group. (In many cases the
dimethyl acetal of the ketone resulted.) The chiral auxiliary
could, however, be successfully removed by treatment of 12
with neat trifluoroacetic acid (TFA) (18) under reflux which,
not surprisingly also removed the t-butyldimethylsilyl alco-
hol protecting group. The product, 13, without purification,
was directly esterified with diazomethane to afford the
methyl hydroxy ester 14 in an overall yield of 53%.5
Difficulties were encountered in attempts at resilylation of
the free hydroxyl group of 14 (including formation of an
acyclic product by a retro-aldol process when basic condi-
tions were used), suggesting that the silylation step might be
better conducted following desaturation of the ring. Accord-
ingly, 14 was converted to cyclopentenone 16 (54% yield,
based on consumed starting material) by treatment with
excess selenium dioxide (12). Reaction of 16 with trimethyl-
silyl trifluoromethanesulfonate (TMSOTf) then afforded 17,
identical in all respects, other than optical rotation, with its
racemic counterpart, prepared in the course of the earlier
synthesis (12). Preparation of 17 thus constitutes a formal
total synthesis of the enantiomer of natural 3-deschloro-
cryptosporiopsin (cf. refs. 1 and 12).
as the dioxolanone 8 or lactone 9 (15). In any event, the fail-
ure of both of these olefins to undergo satisfactory cyclo-
addition with dichloroketene led us to explore instead the
diastereoselectivity of dichloroketene cycloaddition with a
series of simple chiral 2-silyloxy-2-hexenoic esters (15), of
which 10 (which has a (t-butyldimethylsilyl (TBS) group)
proved to be functional. Useful enantioselectivity has not
been observed in conjugate addition of nucleophilic reagents
to conjugated chiral esters or during their reaction with diazo-
methane to give pyrazolines (16). It was reasoned, however,
that the presence of a moderately, but not excessively,2
sterically demanding silyloxy group in the 2-position of a
suitable conjugated chiral ester might exert conformational
effects resulting in enhanced asymmetric induction during a
cycloaddition process.
This hypothesis was substantiated, and optimum diastereo-
selectity (>10:1) was, in fact, observed in dichloroketene
cycloadditions with the 2,2-diphenylcyclopentane esters of
2-silyloxy-2-hexenoic acids (15). Molecular mechanics cal-
culations3 suggested that the preferred conformation of the
(–)-2,2-diphenylcyclopentane ester of 2-t-butyldimethylsilyl-
oxy-2-hexenoic acid (10), a candidate synthetic precursor,
should favour a Si approach of the dichloroketene (Fig. 4).
This ester thus emerged as the precursor of choice for the
enantioselective synthesis of the P. asterris metabolite, (–)-
cryptosporiopsin.
In the racemic synthesis, the cyclopentenone was next
transformed to the corresponding trimethylsilyl (TMS)-dienol
ether 20 prior to introduction of the side-chain unsaturation.
Under somewhat modified conditions, 17 with chlorotri-
methylsilane in refluxing triethylamine afforded, to our
pleasant surprise, not the expected TMS-dienol ether but the
rearranged product 18 in modest yield. Further investigation
revealed that the rearrangement occurred simply on heating
17 at reflux in THF with 2 mole equivalents of triethylamine
alone, in the absence of chlorotrimethylsilane!
Cycloaddition of 10 (17) with dichloroketene afforded,
after crystallization from hexanes, cyclobutanone 11 (Scheme 1)
in 69% yield with 11.5:1 diastereoselectivity. Treatment of
11 with diazomethane (12) effected regiospecific ring expan-
A rigorous explanation for the rearrangement of 17 to 18
requires additional mechanistic studies, beyond the scope of
the present synthetic project. It may be noted, however, that
the gem-dichloroketone (17) can be expected to be a rela-
2 Cycloaddition failed when attempted with the excessively bulky 2-t-butyldiphenylsilyloxy-2-hexenoic diphenylcyclopentyl ester.
3 Molecular mechanics calculations were performed using the MMFF94 program. No restrictions on bond rotations were imposed. Although
there were approximately 20 low-energy conformations within 2.5 kcal/mol, 15 of them adopted the s-cis conformation (the lowest-energy
s-cis conformation had a calculated potential energy of 89.1 kcal/mol). Each of these conformations indicated that the Si-face was more ac-
cessible for cycloaddition. The remaining low-energy conformations had the s-trans arrangement (the lowest-energy s-trans conformation
had a calculated potential energy of 91.2 kcal/mol), again with the Si-face appearing more accessible. Of particular note was that a dihedral
angle much less than 180°, and in some instances as low as 139°, between the carbonyl and the carbon–carbon double bond (O=C–C=C)
was observed. This helps to corroborate our rationale as to why [2+2]-cycloadditions with these alkenes proceed while those with 8 and 9
do not.
4 The absolute configurations depicted in 11 and 12 and subsequent structures were, in fact, only rigorously confirmed on completion of the
synthesis.
5 The free acid could, in fact, be obtained without removal of the TBS group by degradation of 12 to 15, through the agency of ruthenium
trichloride in the presence of sodium periodate (19), followed by decarboxylative elimination using 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU). The yield of the desired silyloxy carboxylic acid, however, was not competitive with that from the TFA procedure adopted and de-
scribed in the text.
© 2004 NRC Canada

1688
Can. J. Chem. Vol. 82, 2004
Fig. 2. Attempted Kabanyane route to the synthesis of cryptosporiopsin.
may undergo γ-deprotonation with triethyamine to produce
dienolate i; if this then suffers an SN1 type solvolysis of one
of the chlorine atoms, oxyallyl cation ii is produced.
Trapping of ii with in situ chloride would generate enolate
iii, which after protonation and conjugation of the alkene,
produces 18. Unlike the two aforementioned mechanistic
proposals, the product formed via this pathway is produced
by a nucleophilic chlorine source. This mechanistic rationale
is also attractive since base-induced oxyallyl cation forma-
tions are well documented (20).
Fig. 3. Greene’s enol ether and various other enol ethers investi-
gated.
The formal synthesis of (+)-cryptosporiopsin was com-
pleted by removal of the TMS protecting group using known
methodology (12). The product, 19, obtained in a modest
41% yield from 18 via base-induced (K2CO3) desilylation
was identical in all respects, other than its opposite optical
rotation, with (–)-dihydrocryptosporiopsin, obtained by cata-
lytic hydrogenation of (+)-cryptosporiopsin (2). The optical
rotation of the dihydro derivative obtained from the natural
product was +187° and that of synthetic 19, –160°. This rep-
resents an enantiomeric excess of 85%. As desaturation of
racemic dihydrocryptosporiopsin to racemic cryptosporiop-
sin has been accomplished previously (8), the present work
constitutes a formal total synthesis of the P. asterris metabo-
lite (–)-cryptosporiopsin (5).
tively efficient source of electrophilic chlorine, perhaps
capable of chlorinating triethylamine to a reactive triethyl-
chloroammonium cationic species, as in Scheme 2. The
dienolate ii formed from i via prototropic shifts could then
undergo α-chlorination by the latter reactive species, afford-
ing the observed product 18. The fact that the putative inter-
mediate ii undergoes α-chlorination, in contrast to the gem-
dichlorosilyl enol ether 20 (which reacts with electrophiles
at the γ-position) (12), may be due to steric differences and
the possible existence of ii as a tight ion pair. That said, one
cannot rule out the possibility that enolate ii, in contrast to
Kabanyane’s TMS-dienol ether 20, is simply more suscepti-
ble to α-chlorination owing to the increased electron density,
or higher HOMO, on the α-carbon.
A slight permutation on the above mechanism may also
be considered, in which 17, as its dienolate (or dienol), un-
dergoes direct chlorination by a second molecule of 17, to
furnish trichloro and monochloro intermediates. The mono-
chloro species, as its conjugated dienolate (or dienol, follow-
ing prototropic shifts), could then undergo halogenation at
its free α-position by transfer of one of the geminal chlorine
atoms of the trichloro product. In our experience (12), gem-
dichloroenones such as 17 tend to react with electrophiles at
the γ- rather than the α-position, so one might expect the lat-
ter mechanism to produce a mixture of α- and γ-chlorinated
products. As only the α-chlorination product 18 was ob-
served, we are inclined to favour the former mechanism.
That the chlorine at C5 assumes the correct configuration
relative to the functionality at C1 under the equilibrating
conditions employed has much precedent (6, 12).
Experimental procedures
All reactions, unless stated otherwise, were performed
under inert atmospheres of either argon or nitrogen in flame-
dried glassware (Pyrex). Solvents, such as ether, tetrahydro-
furan, xylenes, benzene, and toluene, were distilled over
sodium, using benzophenone as an indicator, prior to use.
Other solvents such as dichloromethane, triethylamine, and
acetonitrile were freshly distilled over calcium hydride.
Standard techniques were used in handling air-sensitive re-
agents. Silica gel chromatography (TLC) was performed
with DC-Fertigplatten SIL G-25 UV254 glass plates (pur-
chased from Aldrich Chemicals, Mississauga, Ont.) or by
flash chromatography using Silicycle Ultra Pure Silica Gel
1
for chromatography (Silicycle, Quebec City, Que.). All H
and 13C NMR spectra were recorded on a Varian UNITY
400-MHz spectrometer using CDCl3 as solvent. Optical rota-
tions were recorded on a Perkin-Elmer 241 polarimeter us-
ing the Na lamp. All IR spectra were recorded on a Bruker
IFS 25 instrument using NaCl windows. All high-resolution
mass spectra (HR-MS) were run on a Kratos MS50 instru-
ment.
Finally, an alternate pathway to account for the rearrange-
ment of 17 to 18 is shown in Scheme 3.6 Initially, enone 17
6 The authors wish to thank one of the referees for suggesting and pointing out the possibility of this mechanism.
© 2004 NRC Canada

MaGee et al.
1689
Fig. 4. Molecular mechanics (MMFF94) calculated structures for a model diphenylcyclopentyl TMS-enol ester.
Scheme 1. Steps in the enatioselective synthesis of (–)-cryptosporiopsin. a: Cl3CCOCl, Zn, Et2O; b: CH2N2, MeOH, Et2O; c: TFA, ∆,
12 h; d: CH2N2, Et2O; e: SeO2, C6H5CH3, ∆; f: TMSOTf, NEt3, THF; g: NEt3, ∆; h: K2CO3, MeOH.
ether (10 mL) was added followed by dropwise addition
over 4–5 h of a mixture of trichloroacetyl chloride (1.53 mL,
13.7 mmol) and silyl enol ether 10 (3.20 g, 6.9 mmol) (18).
The heterogenous solution was stirred for 16 h at RT, di-
luted with diethyl ether, and filtered through Celite®. The
1-t-Butyldimethylsilyloxy-2,2-dichloro-3-oxo-4-
propylcyclobutanecarboxylic acid (S)-(2,2-
diphenyl)cyclopentyl ester (11)
Activated Zn dust7 (1.12 g, 17.1 mmol) was flame-dried
in a flask under argon. After cooling to room temperature,
7 Zn dust was activated by vigorously stirring with 10% HCl for 1–2 min, filtered, and then washed with H2O (3×) followed by acetone (3×).
Zinc activated in this manner was stored in an oven at 140 °C until it was needed.
© 2004 NRC Canada

1690
Can. J. Chem. Vol. 82, 2004
Scheme 2. A proposed mechanism for the rearrangement of 17
and 18.
Scheme 3. An alternate mechanism for the rearrangement of 17
and 18.
filtrate was concentrated to 1–5 mL by rotary evaporation,
diluted with pentane (15 mL), and then vigorously stirred for
5 min to induce precipitation of Zn salts. After the precipi-
tate had been allowed to settle, the pentane solution was de-
canted from the Zn salts. This decanting/washing protocol
was repeated twice more. The combined pentane extracts
were washed with cold saturated NaHCO3 (5%, 2×), H2O
(1×), and brine (1×) and dried over Na2SO4. The solvent was
evaporated, and the crude residue was purified by filtration
through a short column of SiO2 (20:1 hexane:ethyl acetate)
followed by recrystallization from hexanes to furnish
cycloadduct 11 (2.68 g, 69% yield) as a colorless solid; mp
(CDCl3) δ: –2.8, –2.2, 14.0, 19.4, 20.2, 20.5, 26.1, 26.2,
30.7, 31.6, 34.6, 36.4, 39.9, 59.3, 83.4, 86.8, 126.1, 126.2,
127.3, 128.0, 128.2, 128.4, 128.5, 128.6, 144.3, 144.9,
168.0, and 197.8. IR (cm–1): 3030, 2404, 1776, 1738, 1518,
1424, 1218, and 926. HR-MS: (C32H42Cl2O4Si – C4H9)obs
531.15287; (C32H42Cl2O4Si – C4H9)calcd. = 531.15252. [α]D25.
–42.4° (c = 2.5, CH2Cl2).
=
=
Methyl 2,2-dichloro-1-hydroxy-3-oxo-5-n-
propylcyclopentane carboxylate (14)
1
76–78 °C. H NMR (CDCl3) δ: 0.22 (s, 3H), 0.36 (s, 3H),
Cyclopentanone 12 (2.47 g, 4.2 mmol) and trifluoroacetic
acid (150 mL) were heated at reflux for 15 h. Upon cooling,
the solvent was removed in vacuo to give 13 as a black oil.
The oil was dissolved in ether (100 mL) and cooled to 0 °C,
and diazomethane (0.2 mol/L, 33.5 mL, 5.0 mmol) was
added dropwise. The reaction mixture was allowed to stir at
0 °C for 5 min, then warmed to RT, and stirred for 30 min.
The dark reaction mixture was quenched with glacial acetic
acid (10 drops) and the solvent evaporated. The residue was
purified by SiO2 chromatography (10:1 hexane:ethyl acetate)
to furnish cyclopentanone 14 (0.602 g, 53% yield) as a col-
orless oil. 1H NMR (CDCl3) δ: 0.94 (t, J = 7.2 Hz, 3H), 1.31
(m, 2H), 1.52 (m, 2H), 2.21 (dd, J1 = 9.1, J2 = 18.9 Hz, 1H),
2.85 (dd, J1 = 9.2, J2 = 19.0 Hz, 1H), 3.11 (m, 1H), 3.69 (d,
J = 1.88 Hz, 1H), and 3.94 (s, 3H). 13C NMR (CDCl3)
δ: 14.1, 20.7, 31.5, 36.5, 37.2, 53.4, 83.5, 87.0, 169.0,
and 198.0. IR (cm–1): 3480, 2958, 2874, 1778, 1728, 1450,
1156, and 820. HR-MS: (C10H14Cl2O4)obs. = 268.02769;
(C10H14Cl2O4)calcd. = 268.02692. [α]2D5 = +40° (c = 2.0,
EtOH).
0.73 (t, J = 7.3 Hz, 3H), 0.92 (s, 9H), 1.29 (m, 2H), 1.51 (m,
3H), 1.87 (m, 1H), 2.04 (m, 1H), 2.31 (m, 1H), 2.47 (m,
1H), 2.65 (m, 1H), 2.95 (t, J = 7.7 Hz, 1H), 6.24 (d, J =
4.8 Hz, 1H), and 7.23 (m, 10H). 13C NMR δ: –2.8, 14.1,
19.9, 20.7, 21.2, 26.3, 26.7, 28.8, 30.3, 35.2, 59.9, 64.3,
83.9, 126.3, 126.5, 126.7, 126.8, 127.6, 128.4, 128.5, 128.7,
128.8, 144.5, 145.1, 168.2, and 194.8. IR (cm–1): 3030,
2878, 1818, 1735, 1518, 1460, 1225, and 920. [α]2D5 = –65.2°
(c = 2.5, CH2Cl2).
1-t-Butyldimethylsilyloxy-2,2-dichloro-3-oxo-5-
propylcyclopentanecarboxylic acid (S)-(2,2-
diphenyl)cyclopentyl ester (12)
Dichlorocyclobutanone 11 (2.65 g, 4.69 mmol) was dis-
solved in diethyl ether (20 mL) and cooled to 0 °C, and then
diazomethane in ether (0.2 mol/L, 28 mL, 5.6 mmol) was
carefully added dropwise. Anhydrous methanol (5 mL) was
subsequently added and the reaction mixture was stirred at
0 °C for 5–10 min. The excess diazomethane was quenched
with glacial acetic acid (2 drops) and then the solvent was
removed in vacuo. (Note: The reaction mixture can be left
overnight at RT before removal of the solvent to ensure that
all the diazomethane has been quenched or evaporated.) The
residue was purified by SiO2 chromatography (30:1 hex-
ane:ethyl acetate) to provide dichlorocyclopentanone 12
Methyl 5,5-dichloro-1-hydroxy-4-oxo-2-n-propyl-2-
cyclopentene carboxylate (16)
Selenium dioxide (0.70 g, 6.4 mmol) was added to a solu-
tion of cyclopentanone 14 (0.57 g, 2.13 mmol) in toluene
(20 mL). The solution was heated at reflux for 24 h, cooled
to room temperature, then diluted with hexane and filtered
through Celite®, and the solvent was evaporated. After puri-
fication by SiO2 chromatography (5:1 hexane:ethyl acetate),
cyclopentenone 16 (0.25 g, 63% yield based on consumed
cyclopentanone 14) was obtained as a colorless oil. 1H NMR
1
(2.50 g, 90% yield) as a colorless oil. H NMR (CDCl3) δ:
0.14 (s, 3H), 0.39 (s, 3H), 0.74 (m, 3H), 0.82 (s, 9H), 0.96
(m, 4H), 1.65 (m, 1H), 1.97 (m, 3H), 2.17 (m, 1H), 2.46 (m,
3H), 2.75 (m, 1H), 6.35 (d, J = 4.4 Hz, 1H), 7.15 (m, 2H),
7.26 (m, 6H), and 7.37 (d, J = 7.69 Hz, 2H). 13C NMR
© 2004 NRC Canada

MaGee et al.
1691
1
(CDCl3) δ: 0.98 (t, J = 7.3 Hz, 3H), 1.58 (m, 2H), 2.23 (m,
1H), 2.47 (m, 1H), 3.82 (s, 3H), 4.47 (s, 1H), and 6.28 (s,
1H). 13C NMR (CDCl3) δ: 13.6, 19.7, 30.9, 54.6, 86.7, 87.0,
127.2, 169.8, 173.3, and 189.4. IR (cm–1): 3480, 2952, 1736,
1640, 1464, 1272, 1176, 952, and 840. HR-MS:
rived from the natural compound. H NMR (CDCl3) δ: 0.98
(t, J = 7.4 Hz, 3H), 1.52 (m, 2H), 2.42 (ddd, J1 = 6.6 Hz,
J2 = 9.6 Hz, J3 = 13.3, 1H), 2.55 (ddd, J1 = 6.0 Hz, J2 =
9.3 Hz, J3 = 13.2 Hz, 1H), 3.82 (s, 3H), 4.40 (bs, 1H), and
4.55 (s, 1H). 13C NMR (CDCl3) δ: 14.4, 20.6, 29.4, 54.6,
65.6, 83.7, 133.9, 166.5, 170.8, and 188.0. IR (cm–1): 3504,
3040, 3008, 1736, 1620, 1524, 1420, 1216, and 932. HR-
(C10H12Cl2O4 – Cl)obs. = 231.04094; (C10H12Cl2O4 – Cl)calcd.
=
231.04248. [α]2D5 = –143° (c = 0.65, EtOH).
MS: (C10H12Cl22O54)obs. = 266.01014; (C10H12Cl2O4)calcd.
=
266.01126. [α]D = –160° (c = 0.1, CHCl3). Comparison
Methyl 5,5-dichloro-4-oxo-2-n-propyl-1-
trimethylsilyloxy-2-cyclopentene carboxylate (17)
with spectral data of the authentic sample of (+)-
1
dihydrocryptosporiopsin: H NMR (CDCl3) δ: 0.98 (t, J =
Trimethylsilyl trifluoromethanesulfonate (0.03 mL,
0.14 mmol) was added dropwise to a solution of alcohol 16
(18.8 mg, 0.07 mmol), triethylamine (0.03 mL, 0.17 mmol),
and diethyl ether (1 mL). The reaction mixture was stirred at
RT for 30 min, then quenched with saturated NH4Cl (1 mL).
The organic layer was separated, washed with brine (1×),
and dried over MgSO4, and the solvent was evaporated. Pre-
parative TLC (5:1 hexane:ethyl acetate) provided 17
7.3 Hz, 3H), 1.50 (m, 2H), 2.42 (ddd, J1 = 6.6 Hz, J2 =
9.6 Hz, J3 = 13.3, 1H), 2.55 (ddd, J1 = 6.0 Hz, J2 = 9.3 Hz,
J3 = 13.2 Hz, 1H), 3.82 (s, 3H), 4.40 (s, 1H), and 4.57 (s,
1H). 13C NMR (CDCl3) δ: 14.4, 20.6, 29.4, 54.6, 65.7, 83.7,
133.9, 166.5, 170.7, and 188.0. [α]2D5 = +187° (c = 0.5,
CHCl3).
1
(14.5 mg, 61% yield) as a colorless oil. H NMR (CDCl3) δ:
References
0.35 (s, 9H), 0.99 (t, J = 7.3 Hz, 3H), 1.61 (m, 2H), 2.20 (m,
2H), 2.41 (m, 2H), 3.73 (s, 3H), and 6.19 (s, 1H). 13C NMR
(CDCl3) δ: 2.1, 13.7, 19.8, 30.8, 53.4, 90.6, 125.3, 168.8,
175.5, and 189.9. IR (cm–1): 3000, 1752, 1640, 1448, 1224,
1. W.J. McGahren, J.H. van den Hende, and L.A. Mitscher. J.
Am. Chem. Soc. 91, 157 (1969).
2. G.M. Strunz, A.S. Court, J. Komlossy, and M.A. Stillwell.
Can. J. Chem. 47, 2087 and 3700 (1969).
3. D. Giles and W.B. Turner. J. Chem. Soc. C, 2187 (1969).
4. W.J. McGahren and L.A. Mitscher. J. Org. Chem. 33, 1577
(1968).
894, and 856. HR-MS: (C13H20Cl2O4Si – CH3)obs.
323.02643; (C13H20Cl2O4Si – CH3)calcd. = 323.02731. [α]D25
–109° (c = 0.97, EtOH).
=
=
5. R.J.J. Ch. Lousberg, Y. Tirilly, and M. Moreau. Experientia,
Methyl 3,5-dichloro-4-oxo-2-n-propyl-1-
trimethylsilyloxy-2-cyclopentene carboxylate (18)
32, 331 (1976).
6. G.M. Strunz, P.I. Kazinoti, and M.A. Stillwell. Can. J. Chem.
52, 3623 (1974).
7. M.A. Stillwell, F.A. Wood, and G.M. Strunz. Can. J.
Microbiol. 15, 501 (1969).
8. G.M. Strunz and A.S. Court. J. Am. Chem. Soc. 95, 3000
(1973).
9. G.B. Henderson and R.A. Hill. J. Chem. Soc., Perkin Trans. 1,
2595 (1983).
10. W.T. Brady. Tetrahedron, 37, 2949 (1981).
11. A.E. Greene and J.P. Deprés. J. Am. Chem. Soc. 101, 4003
(1979).
12. S. Kabanyane, A. Decken, C.-M. Yu, and G.M. Strunz. Can. J.
Chem. 78, 270 (2000).
13. P. Nebois and A.E. Greene. J. Org. Chem. 61, 5210 (1996).
14. A.E. Greene, A. Kanazawa, and S. Gillet. Tetrahedron, 63,
4660 (1998).
A mixture of enone 17 (16.9 mg, 0.049 mmol) and
triethylamine (0.014 mL, 0.099 mmol) was heated at reflux
in THF and monitored by TLC. When the starting material
was completely consumed, the solvent was removed in
vacuo and the residue was purified by SiO2 chromatography
(10:1 hexane:ethyl acetate) to furnish 18 (10.5 mg, 62%
1
yield) as a colorless oil. H NMR (CDCl3) δ: 0.27 (s, 9H),
0.97 (t, J = 7.4 Hz, 3H), 1.51 (m, 2H), 2.40 (m, 1H), 2.50
(m, 1H), 3.71 (s, 3H), and 4.44 (s, 1H). 13C NMR (CDCl3) δ:
2.1, 14.2, 20.4, 29.3, 53.2, 66.4, 86.1, 132.5, 168.2, 168.8,
and 188.2. IR (cm–1): 2961, 2877, 1769, 1622, 1436, 1253,
878, and 761. HR-MS: (C13H20Cl2O4Si – CH3)obs.
323.02695; (C13H20Cl2O4Si – CH3)calcd. = 323.02731. [α]D25
–75° (c = 0.6, EtOH).
=
=
15. T.C. Mallais. Ph.D. thesis, Department of Chemistry, Univer-
sity of New Brunswick, Fredericton, N.B. 2003.
16. F.S. Impastato, L. Barash, and H.M. Walborsky. J. Am. Chem.
Soc. 81, 1514 (1959).
(–)-Dihydrocryptosporiopsin (19)
Chloroenone 18 (11.8 mg, 0.03 mmol) was dissolved in
methanol (1 mL) and cooled to 0 °C. Potassium carbonate
(6 mg, 0.04 mmol) was added in one portion, and the reac-
tion mixture was stirred at 0 °C. When TLC showed com-
plete consumption of the starting material, the mixture was
poured into cold H2O and extracted with ether (3×). After
drying with MgSO4, the solvent was removed in vacuo and
the residue was purified by SiO2 chromatography (5:1 hex-
ane:ethyl acetate) to give (–)-dihydrocryptosporiopsin (19)
as a colorless solid (4 mg, 41% yield), which had identical
spectroscopic properties, except rotation, with a sample de-
17. D.I. MaGee, T.C. Mallais, and M. Eic. Tetrahedron: Asymme-
try, 14, 3177 (2003).
18. F.X. Webster, J. Rivas-Enterrios, and R.M. Silverstein. J. Org.
Chem. 52, 689 (1987).
19. P.H.J. Carlsen, T. Katsuki, V.S. Martin, and K.B. Sharpless. J.
Org. Chem. 46, 3936 (1981).
20. For a review on oxyallyl cation formation, see: J. Mann. Tetra-
hedron, 42, 4611 (1986).
© 2004 NRC Canada