Asymmetric synthesis of calyculin C. 2. Synthesis of the C26-C37 fragment and model Wittig couplings

Article abstract of DOI:10.1021/jo960315q

We report our synthesis of the C₂₆-C₃₇ fragment of serine/threonine protein phosphatase PP1 and PP2A inhibitor calyculin C (1). Outlined in this paper are synthetic approaches to the two components based on disconnection at the C₃₃-N₃ amide bond. We report the successful synthesis of the C₃₃-C₃₇ aza-sugar derived from D-lyxose which was coupled onto a C₂₆-C₃₂ aminooxazole originating from L-pyroglutamic acid. Elaboration of the resulting amide to a fully deprotected C₂₆-C₃₇ fragment of calyculin C completed our synthesis. This provided an appropriate phosphonium salt for use in a Wittig olefination for joining both halves of the natural product.

Full text of DOI:10.1021/jo960315q

J . Org. Chem. 1996, 61, 6153-6161
6153
Asym m etr ic Syn th esis of Ca lycu lin C. 2. Syn th esis of th e C26-C37
F r a gm en t a n d Mod el Wittig Cou p lin gs
Anthony K. Ogawa, J ohn A. DeMattei,1 Gerard R. Scarlato,2 J ohn E. Tellew,
Lee S. Chong, and Robert W. Armstrong*
Department of Chemistry and Biochemistry, University of California, Los Angeles, 90095
Received February 15, 1996X
We report our synthesis of the C26-C37 fragment of serine/threonine protein phosphatase PP1 and
PP2A inhibitor calyculin C (1). Outlined in this paper are synthetic approaches to the two
components based on disconnection at the C33-N3 amide bond. We report the successful synthesis
of the C33-C37 aza-sugar derived from D-lyxose which was coupled onto a C26-C32 aminooxazole
originating from L-pyroglutamic acid. Elaboration of the resulting amide to a fully deprotected
C26-C37 fragment of calyculin C completed our synthesis. This provided an appropriate phospho-
nium salt for use in a Wittig olefination for joining both halves of the natural product.
Sch em e 1
In tr od u ction
In the previous paper, the synthesis of the C1-C25
fragment applicable to calyculins A3 and C4 was reported.
Herein the focus is on the synthetic efforts directed
toward the C26-C37 fragment and subsequent couplings
to model completion of the carbon skeleton of serine/
threonine phosphatase inhibitor calyculin C.
Retrosynthetic analysis of calyculin revealed several
possibilities for carbon-carbon bond disconnections. Our
strategy centered on an initial disconnection at the C25-
C26 double bond, dividing the natural product into two
fragments of similar functional density. Coupling at this
junction was perceived to afford a high degree of conver-
gence. In addition, this strategy was in accord with other
synthetic efforts.5-7
The introduction of several functionalities within the
C26-C37 fragment 2 provided concern in establishing our
X Abstract published in Advance ACS Abstracts, August 1, 1996.
(1) Taken in part from the Ph.D. Thesis of J ohn A. DeMattei,
University of California, Los Angeles, 1994.
(2) Taken in part from the Ph.D. Thesis of Gerard R. Scarlato,
University of California, Los Angeles, 1990.
(3) Kato, Y.; Fusetani, N.; Matsunaga, S.; Hashimoto, K.; Fujita,
S.; Furuya T. J . Am. Chem. Soc. 1986, 108, 2780.
(4) Kato, Y.; Fusetani, N.; Matsunaga, S.; Hashimoto, K.; Koski, K.
J . Org. Chem. 1988, 53, 3930.
(5) Preliminary communications from this laboratory on this subject,
see: (a) Zhao, Z.; Scarlato, G. R.; Armstrong, R. W. Tetrahedron Lett.
1991, 32, 1609. (b) Armstrong, R. W.; DeMattei, J . A. Tetrahedron Lett.
1991, 32, 5749.
synthetic plan. Formation of the C33-N3 amide bond in
high yield was viewed as critical if efficient use of both
fragments (3 and 4) was to be maintained (Scheme 1).
(6) For completed total syntheses of both enantiomers of calyculin
A, see: (a) Evans, D. A.; Gage, J . R.; Leighton, J . L. J . Am. Chem.
Soc. 1992, 114, 9434. (b) Tanimoto, N.; Gerritz, S. W.; Sawabe, A.;
Noda, T.; Filla, S. A.; Masamune, S. Angew. Chem. 1994, 106, 674.
(7) For other efforts directed toward the synthesis of calyculins,
see: (a) Evans, D. A.; Gage, J . R. Tetrahedron Lett. 1990, 31, 6129. (b)
Evans, D. A.; Gage, J . R. J . Org. Chem. 1992, 57, 1958. (c) Evans, D.
A.; Gage, J . R.; Leighton, J . L.; Kim, A. S. J . Org. Chem. 1992, 57,
1961. (d) Evans, D. A.; Gage, J . R.; Leighton, J . L. J . Org. Chem. 1992,
57, 1964. (e) Duplanter, A. J .; Nantz, M. H.; Roberts, J . C.; Short, R.
P.; Somfai, P.; Masamune, S. Tetrahedron Lett. 1989, 30, 7357. (f)
Vaccaro, H. A.; Levy, D. E.; Sawabe, A.; J aetsch, T.; Masamune, S.
Tetrahedron Lett. 1992, 33, 1937. (g) Hamada, Y.; Tanada, Y.;
Yokokawa, F.; Shiori, T. Tetrahedron Lett. 1991, 32, 5983. (h) Hara,
O.; Hamada, Y.; Shiori, T. Synlett 1991, 283. (i) Yokokawa, F.; Hamada,
Y.; Shiori, T. Synlett 1992, 149. (j) Smith, A. B., III; Duan, J . J .-W.;
Hull, K. G.; Salvatore, B. A. Tetrahedron Lett. 1991, 32, 4855. (k)
Smith, A. B., III; Salvatore, B. A.; Hull, K. G.; Duan, J . J -W.
Tetrahedron Lett. 1991, 32, 4859. (l) Barrett, A. G. M.; Malecha, J . W.
J . Org. Chem. 1991, 56, 5243. (m) Barrett, A. G. M.; Edmunds, J . J .;
Horita, K.; Parkinson, C. J . J . Chem. Soc., Chem. Commun. 1992, 1236.
(n) Barrett, A. G. M.; Edmunds, J . J .; Hendrix, J . A.; Horita, K.;
Parkinson, C. J . J . Chem. Soc., Chem. Commun. 1992, 1238. (o)
Barrett, A. G. M.; Edmunds, J . J .; Hendrix, J . A.; Malecha, J . W.;
Parkinson, C. J . J . Chem. Soc., Chem. Commun. 1992, 1240.
Foremost in our mind was the judicious choice of
methodology for introduction and subsequent protection
of the amine functionalities at C32 and C36. Two possi-
bilites existed and provided a means for stereochemical
control of amine introduction. One approach to ester 3
was derived from the preset stereochemistry found in
aldopentoses. Reliance on a pentose starting point also
precluded any necessity for chain homologation, and
controlled introduction of nitrogen at C36 could conven-
iently arise from displacement of oxygen.
The potential for an amino acid-based strategy pro-
vided a reasonable alternative to the synthesis of the aza-
sugar and offered a stereospecific method for generating
the C26-C32 oxazole fragment. Synthetic routes origi-
nating from amino acids served as the basis for early
S0022-3263(96)00315-5 CCC: $12.00 © 1996 American Chemical Society

6154 J . Org. Chem., Vol. 61, No. 18, 1996
Ogawa et al.
Sch em e 2
efforts toward the synthesis of the C33-C37 aza-sugar 3
and the successful completion of the C26-C32 oxazole 4
fragment.
C37 proved to be less than trivial while a synthesis of 3
originating from a sugar promised a route of equal
efficiency.
The abundant supply of aldopentoses served as the
basis for our earliest synthetic efforts toward the syn-
thesis of C33-C37 fragment 3.2 At the outset of our work
in this area, the absolute stereochemistry of the natural
product had not yet been established. As a consequence,
our early syntheses required double inversion at C4 of
D-ribose for the synthesis of enantiomeric C33-C37 frag-
ment (eq 2). The same stereochemical outcome could be
effected by a single inversion at C4 of L-lyxose. The cost
of the starting sugar, however, precluded its use in our
initial synthetic efforts.
Resu lts a n d Discu ssion
Syn th esis of th e C33-C37 F r a gm en t. Several routes
were considered for the diversely functionalized C33-C37
aza-sugar fragment. As previously indicated, each strat-
egy was founded on decisions affecting the timing for
introduction and elaboration of the C36 nitrogen.
Early efforts were directed toward an approach in
which L-serine was converted to an aziridine species
representative of C35-C37. Subsequent elaboration af-
forded the desired functionalization for all three stereo-
centers in the fragment while maintaining a blocked C36-
nitrogen. It was envisioned that regioselectve acid-
catalyzed ring opening of the aziridine8 followed by in
situ reductive amination would provide functionality to
account for the completed C33-C37 aza-sugar (eq 1). Ring
(1)
(2)
opening under acidic conditions proceeded with the
expected regiochemistry arising from nucleophilic attack
at the less hindered center of an aziridinium intermedi-
ate.8 Reductive amination of the crude ring-opened
products afforded fully functionalized aza-sugar com-
pounds bearing a reasonable handle at C37 for introduc-
tion of the methyl ether moiety (5, e.g.). Elaboration of
Confirmation of the absolute configuration of the
calyculins by Shiori7g suggested that the more efficient
approach to the C33-C37 aza-sugar from D-lyxose was now
a reasonable synthetic possibility. Synthesis of aza-sugar
3 ester proceeded from known alcohol 6, obtained in four
steps from D-lyxose utilizing the elegant work of van
(8) (a) Elderfield, R. C. Heterocyclic Compounds, Vol. 1; J ohn Wiley
and Sons: New York, 1950; 61. (b) Gabriel, S.; Ohle, H. Ber. 1917, 50,
804.

Asymmetric Synthesis of Calyculin C. 2
J . Org. Chem., Vol. 61, No. 18, 1996 6155
Boom.9 The choice of benzyl protecting groups for the
C34,C35-diol was required due to the rigor of proposed
downstream synthetic steps. This need for stability was
demonstrated as 4-methoxybenzyl (PMB) proved labile
under conditions necessary for fragment elaboration
(J ones oxidation, e.g.).
Methylation of 6 under standard conditions afforded
methyl ether 7 in excellent yield. Conversion to diol 8
proceeded via initial acidic hydrolysis of the methyl
furanoside, which required heat to fully generate the
hemiacetal product, followed by mild reduction to give
the desired product in excellent overall yield (Scheme 2).
Silylation of the 1°-hydroxyl using tert-butyldiphenylsilyl
chloride would allow for unambiguous introduction of
nitrogen upon displacement of the C36-hydroxyl. Silyla-
tion to afford alcohol 9 in quantitative yield was followed
by mesylation and azide inversion at high temperatures
to afford azide 10 which possessed the desired heteroatom
substitution and stereochemistry applicable to ester 3.
Azide 10 was reduced to the free amine 11 with lithium
aluminum hydride in good yield. Amine 11 was methyl-
ated under exhaustive reductive amination conditions
which gave dimethylamine 12 in excellent yield. At this
time, the remaining synthetic task was oxidation to the
C33-carboxylate. Desilylation of 12 with TBAF to give
alcohol 13 provided the desired substrate for oxidation
to the completed aza-sugar fragment. Conversion of 13
to target fragment 3 via J ones oxidation10 followed by
esterification proceeded in modest yield. Despite the
apparent shortcomings of the J ones oxidation10 in this
synthesis, other oxidative methods for direct conversion
of alcohols to carboxylates proved no more effective.
Attempts at effecting stepwise oxidation through the
intermediate aldehyde followed by further oxidation to
the acid met with similar yields. This observation was
attributed to the instability of the intermediate aldehyde.
Regardless, the desired ester 3 was in our possession and
could be coupled to an aminooxazole, thereby affording
a fully protected top half of calyculin C.
(3)
A successful route to oxazole fragment5b 4 was estab-
lished at a time when the absolute configuration of the
calyculins was not known. Initial efforts were directed
toward a synthetic route derived from (R)-pyroglutamic
acid via the enantiomer of acetal 15. Establishment of
the absolute configuration suggested that the antipode
of (R)-pyroglutamic acid possessed the correct stereo-
chemistry for use in the synthesis of the C26-C32 portion
of calyculin C. Lithium enolate formation on known
bicyclic N,O-acetal 1512 and subsequent methylation at
-78 °C afforded 16a as the major diastereomer (60% de)
(Scheme 3). An interesting result involving clean conver-
sion of lactam 16a to its C30-epimer 16b by enolate
protonation at -78 °C clearly demonstrated the over-
whelming preference for endo-alkylation of these bicyclic
acetals. Acid hydrolysis of acetal 16a was followed by
mesylation to afford lactam 17 in reasonable yield. Final
elaboration to a fully functionalized C29-C32 precursor
to the oxazole fragment was accomplished by radical
deoxygenation of an in situ-generated iodide to give
pyrrolidinone 18 in excellent yield. This compound was
successfully transformed to an open chain amide inter-
mediate via N-Boc-protection to 19, followed by Weinreb
aluminum-amide opening of the butyrolactam to provide
amide 20.13
Syn th esis of th e C26-C32 F r a gm en t. Work toward
the C26-C32 fragment originated from a plan for sepa-
rately generating aminooxazoles compatible with the
entire family of calyculins.4 The important distinction
between C26-C32 fragments pertaining to calyculins A
and C is the presence of a methyl substituent at C32 in
the latter isomer.
Formation of the oxazole ring presented the next
synthetic challenge in the synthesis of fragment 4. We
felt that 1,3-dichloroacetone afforded an excellent syn-
thon in harnessing previous work from our group5a to
generate the desired oxazole moiety while providing a
reactive handle (C26-chloride) for further elaboration.14
The importance of this functionality was its potential for
facile conversion to the phosphonium salt desired for
coupling of both halves of the natural product.5a,6a The
C26-chloride, therefore, eliminated the need for extensive
manipulation. To this end, condensation of amide 20
with 1,3-dichloroacetone in chloroform at vigorous reflux
successfully yielded the target C26-C32 fragment in good
yield (Scheme 4). The use of other higher boiling solvents
resulted either in generation of inseparable epimers,
presumably at C30, or in no observable product formation.
Pyroglutamic acid could represent a versatile chiral
template as endo-selective methylation of an acetal
intermediate, following the method developed by Mey-
ers,11 would establish the desired C30-methyl functionality
relevant to both calyculins A and C. Differential forma-
tion of bicyclic acetal intermediates would facilitate the
distinction of functionality at C32 for each individual
fragment type (eq 3). We supposed that the oxazole
fragment of calyculin A could arise from radical decar-
boxylation of a pyrrolidinone intermediate (boxed car-
boxylate in the lower path of eq 3). Similarly, endo-
methylation of acetal 1511 followed by radical deoxy-
genation would yield a precursor to aminooxazole frag-
ment 4 (boxed heteroatom in the top path of eq 3).
(9) Veeneman, G. H.; Gomes, L. J . F.; van Boom, J . H. Tetrahedron
1989, 45, 7433.
(10) (a) Bowden, K.; Heilbron, I. M.; J ones, E. R. H.; Weedon, B. C.
L. J . Chem. Soc. 1946, 39. (b) Bowers, A.; Halsall, T. G.; J ones, E. R.
H.; Lernin, A. J . J . Chem. Soc. 1953, 2548.
(11) (a) Meyers, A. I.; Harre, M.; Garland, R. J . Am. Chem. Soc.
1984, 106, 1146. (b) Meyers, A. I.; Lefker, B. A.; Wanner, K. T.; Aitken,
R. A. J . Org. Chem. 1986, 51, 1936.
(12) Thottathil, J . K.; Moniot, J . L.; Mueller, R. H.; Wong, M. K. Y.;
Kissich, T. P. J . Org. Chem. 1986, 51, 3140.
(13) Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977,
4171.
(14) (a) Turchi, I. J .; Dewar, M. J . S. Chem. Rev. 1975, 75, 389. (b)
Lakham, R.; Ternal, B. Adv. Heterocycl. Chem. 1974, 17, 99.

6156 J . Org. Chem., Vol. 61, No. 18, 1996
Ogawa et al.
Sch em e 3
Sch em e 5
Sch em e 6
Sch em e 4
which provided the impetus to attempt amide formation
with more sterically hindered oxazole 4. The reaction
yielded a 2.7:1 separable mixture of diastereomers 23 and
24 (Scheme 5), presumably at C34, in a 62% overall
yield.17 Epimerization at C34 was hypothesized to occur
prior to amide formation, and supporting evidence came
from two separate reactions in which treatment of a
single amide diastereomer with excess trimethylalumi-
num showed no change over extended reaction times (12
hr at 35 °C) while submission of pure ester 3 to similar
conditions was met with epimerization.
The remaining obstacle in our efforts toward the
synthesis of the C26-C37 fragment 1 was the identifica-
tion of the proper sequence for phosphonium salt forma-
tion and debenzylation. Deprotection of the C34,C35-diol
prior to formation of the phosphonium salt at C26 was
perceived as ideal. This synthetic sequence would afford
flexibility with regard to diol protecting groups in the
event that the proposed C25-C26 Wittig olefination war-
ranted such manipulation.
Com p letion of th e C26-C37 F r a gm en t of Ca lycu lin
C. With the two fragments from the C26-C37 portion of
calyculin C in hand, we next sought a viable means of
forming the C33-N3 amide bond to yield a fully protected
top half segment of the natural product. Numerous
unsuccessful attempts were made in an effort to synthe-
size amide 23 through formation of activated acyl inter-
mediates from the carboxylic acid 14. Most notable was
the isolation of an oxazole-phosphorous adduct from
attempted BOP-Cl15 mediated coupling which suggested
an inability of the zwitterionic acid to participate as a
nucleophile in generating an activated carboxylate.
Aluminum-mediated amide formation arising from
coupling of amines to esters offered an alternative
method for the synthesis of the C33-N3 amide bond.13,16
Model studies involved the coupling of ester 3 onto an
achiral oxazole fragment 22 (eq 4) and established
(4)
Experiments confirming the anticipated lability of the
C26-chloride toward hydrogenation suggested the need for
an alternative method for debenzylation. The conditions
for benzyl removal involving boron trifluoride etherate
and ethyl mercaptan18 were also investigated and were
found to be unfruitful, thus relegating our efforts to diol
deprotection after formation of the C26-phosphonium salt.
reaction conditions in which initial formation of an
aluminum-amide complex (3 equiv) from the amine-
hydrochloride salt was followed by addition of the ester
substrate (1 equiv). Heating of the resulting mixture
over extended reaction times provided a reliably good
yield of amide product while allowing for recycling of
unreacted ester starting material. We felt that these
model reactions sufficiently mimicked the parent system
Model studies showed that the C26-phosphonium salt
was inert under hydrogenation conditions which prompted
conversion of amide 23 to phosphonium salt 25 in good
yield (Scheme 6). Successful completion of the C26-C37
fragment proceeded as phosphonium salt 25 was submit-
(15) (a) Diago-Messeguer, J .; Palomo-Coll, A. L.; Fernandez-Lisarbe,
J . R.; Zugaza-Bilbao, A. Synthesis 1980, 547. (b) Cabre, J .; Palomo, A.
L. Synthesis 1984, 413.
(16) (a) Levin, J . I.; Turos, E.; Weinreb, S. M. Synth. Commun. 1982,
12, 989. (b) Poss, M. A.; Reid, J . A. Tetrahedron Lett. 1992, 33, 1411.
(17) The assignment of stereochemistry is consistent with analysis
of 2D-NOESY spectra in conjunction with 1D-1H NMR for each
diastereomer.
(18) Daly, S. M.; Armstrong, R. W. Tetrahedron Lett. 1989, 30, 5713.

Asymmetric Synthesis of Calyculin C. 2
J . Org. Chem., Vol. 61, No. 18, 1996 6157
Sch em e 7
ted to hydrogenation conditions, thereby affording phos-
phonium salt 2 as a white solid.
Mod el Wittig Cou p lin gs. In an effort to test our
hypothesis that a fully deprotected C26-C37 phosphonium
salt could effect coupling to yield a completed calyculin
backbone, we carried out model Wittig condensations on
a series of systems designed to improve our understand-
ing of the reactivity of diol 2. We felt that the stabilized
ylide generated from 2 would be sufficiently reactive to
afford a coupled product with a high degree of stereo-
chemical purity.5a
The successful condensation of a phosphonium salt (26)
derived from our work using achiral oxazole 22 onto an
excess of butyraldehyde (eq 5) provided preliminary
justification for our approach to formation of the C25-
C
26 double bond. The true test of this strategy, however,
of PP1 and PP2A. An inference based on our success in
the previous model studies involving a de-phospho C1-
C25 aldehyde derivative suggested that the free C17-
hydroxyl did not interfere with coupling. The question
remained, however, regarding the viability of such a
substrate, particularly in reference to the potential for
â-elimination. We felt that a conservative approach in
which initial Wittig condensations utilized simple model
phosphonium salt 26 would more accurately test our
initial hypothesis. The successful generation of the
desired olefinated product 28 from condensation onto
model aldehyde 27 suggested graduation to a structurally
more relevant system. Wittig coupling of 2 with the same
model aldehyde proceeded to product 29 (Scheme 7). This
result confirmed our strategy for C25-C26 double bond
formation in the synthesis of calyculin C, as well as other
relevant analogs.
(5)
involved coupling onto more complex aldehyde sub-
strates. Key issues that needed to be addressed included
(1) questions involving the kinetics of coupling for two
large molecules and (2) the subsequent potential for
competitive â-elimination7d of the aldehyde given steric
constraints related to coupling. Condensation onto a
model spiroketal aldehyde (eq 5) yielded the answers to
these questions and lent futher support to our overall
synthetic strategy.
Con clu sion s
We have successfully completed synthesis of the C26-
37 fragment 2 in our route toward the total synthesis of
C
calyculin C. Our synthetic efforts include a lyxose-based
approach to the C33-C37 aza-sugar piece 3, which was
coupled onto aminooxazole 4 derived from pyroglutamic
acid. We believe that our success in modeling the
proposed C25-C26 Wittig olefination using a fully depro-
tected phosphonium salt provides a clear avenue to the
total synthesis of calyculin C. Current studies for effect-
The de-phospho derivative of calyculin C represents a
key analog in our SAR study of phosphatase inhibitors

6158 J . Org. Chem., Vol. 61, No. 18, 1996
Ogawa et al.
g, 3.70 mmol) in DMF (80 mL) was added TBDPS-Cl (1.05 mL,
4.1 mmol) followed by imidazole (529 mg, 7.4 mmol). The clear
reaction mixture was stirred at rt for 12 h. The reaction
mixture was concentrated in vacuo, and purification via
column chromatography on silica gel (5-15% ethyl acetate-
hexane) afforded silyl ether 9 (2.31 g, 100%): [R]D ) -18.7 (c
ing the established Wittig condensation onto the com-
pleted C1-C25 aldehyde19 and subsequent deprotection
to the natural product are ongoing.
Exp er im en ta l Section
10.1, CHCl3); IR (thin film) 3500, 2928, 1454, 1427, 1112 cm-1
;
Gen er a l. Optical rotations were taken at 22 °C. Mass
spectra were obtained from the Mass Spectroscopy Facilities
at UCLA and UC Riverside. Elemental analyses were ob-
tained from Desert Analysis of Tuscon, AZ. For EI, CI, and
FAB mass spectra, 2σ ) 4 ppm.
Solvents and reagents were used as supplied from com-
mercial sources with the following exceptions or specific
notations. DMF was used as received in Aldrich Sure Seal
packing. THF was distilled from sodium benzophenone ketyl.
Toluene was distilled from calcium hydride. Methanol was
distilled from magnesium turnings, and dichloromethane was
distilled from phosphorus pentoxide. All reactions involving
moisture-sensitive reagents were performed under either a
nitrogen or an argon atmosphere.
Meth yl 2,3-Diben zyl-5-m eth yl-r-D-lyxofu r a n osid e (7).
To a solution of 6 (4.9 g, 14.2 mmol) in DMF (100 mL) was
added iodomethane (1.3 mL, 21.4 mmol). Sodium hydride (375
mg, 15.7 mmol) was then added in several portions, and the
opaque yellow reaction mixture was stirred at rt. An ad-
ditional 2.8 mL of iodomethane and 205 mg of sodium hydride
were added in three portions. After 12 h, the reaction was
quenched by addition of methanol. The resulting mixture was
partitioned between CH2Cl2-hexane (150 mL total) and H2O
(70 mL). The layers were separated, and the organic layer
was washed with H2O (2 × 50 mL). The combined organics
were dried over Na2SO4, filtered, and concentrated. Purifica-
tion via column chromatography on silica gel (10-20% ethyl
1H NMR (400 MHz, CDCl3) δ 7.20-7.70 (20H, m), 4.64 (2H,
m), 4.50 (2H, m), 4.02 (1H, m), 3.90 (1H, dd, J ) 10.6, 7.0 Hz),
3.84 (1H, dd, J ) 10.6, 4.7 Hz), 3.70-3.80 (2H, m), 3.48 (1H,
m), 3.40 (1H, dd, J ) 4.9, 1.4 Hz), 3.30 (3H, s), 1.04 (9H, s);
13C NMR (90 MHz, CDCl3) δ 138.1, 138.0, 135.7, 135.6, 134.7,
133.2, 133.0, 129.7, 128.3, 128.0, 127.7, 127.7, 126.6, 126.6,
79.8, 76.6, 73.7, 73.6, 72.7, 69.6, 62.8, 58.9, 26.8, 26.6, 19.1;
HRFABMS calcd for MH+ (C36H45O5Si), 585.3036, found
585.3030 (error 1.0 ppm).
(2R,3S,4S)-4-Azido-2,3-bis(ben zyloxy)-5-m eth oxy-1-(ter t-
bu tyld ip h en yl)siloxy)p en ta n e (10). To a solution of alcohol
9 (2.195 g, 3.71 mmol) in CH2Cl2 (80 mL) was added triethyl-
amine (1.6 mL, 11.1 mmol) followed by methanesulfonyl
chloride (0.35 mL, 4.45 mmol). The reaction mixture was
stirred at rt for 30 min. The reaction mixture was washed
with H2O (2 × 100 mL), dried over Na2SO4, filtered, and
concentrated. The resulting crude mesylate was eluted in
DMF (80 mL). Sodium azide (1.25 g, 18.6 mmol) was added,
and the suspension was heated to 100 °C. After 19 h, the
reaction mixture was concentrated, and the resulting crude
was partitioned between CH2Cl2 (200 mL) and H2O (200 mL).
The layers were separated, and the aqueous layer was
extracted with CH2Cl2 (1 × 200 mL). The combined organics
were dried over Na2SO4, filtered, and concentrated. Purifica-
tion via column chromatography on silica gel (0-15% ethyl
acetate-hexane) afforded azide 10 (10.8 g, 81% two steps):
[R]D ) -18.1 (c 1.01, CHCl3); 1H NMR (400 MHz, CDCl3) δ
7.15-7.70 (20H, m), 4.66 (1H, d, J ) 11.1 Hz), 4.64 (1H, d, J
) 11.7 Hz,), 4.58 (1H, d, J ) 11.1 Hz), 4.45 (1H, d, J ) 11.7
Hz), 3.92 (2H, m), 3.86 (2H, m), 3.63 (1H, m), 3.51 (2H, m),
3.28 (3H, s), 1.06 (9H, s); 13C NMR (90 MHz, CDCl3) δ 138.1,
137.9, 135.7, 135.6, 135.5, 135.1, 134.8, 133.3, 133.1, 129.7,
129.6, 129.6, 128.3, 127.9, 127.8, 127.7, 127.6, 79.1, 78.3, 73.9,
72.2, 72.0, 62.5, 62.0, 58.9, 26.7, 19.2; HRFABMS calcd for
MH+ (C36H44N3O4Si) 610.3101, found 610.3086 (error 2.5 ppm).
(2R,3S,4S)-4-Am in o-2,3-b is(b en zyloxy)-5-m et h oxy-1-
(ter t-bu tyld ip h en ylsiloxy)p en ta n e (11). To a solution of
10 (11.6 g, 19.0 mmol) in THF (600 mL) was added lithium
aluminum hydride (1.08 g, 28.5 mmol) in several portions over
1.5 h. The slurry was stirred at rt for 5.5 h, at which time 0.5
N aqueous NaOH (30 mL) was added dropwise. Upon stirring,
a white precipitate formed, and after 15 min, the mixture was
neutralized via addition of 1.2 N aqueous HCl. The resulting
mixture was filtered, and the precipitate was washed with
EtOAc (400 mL). The combined organics were concentrated,
and purification via column chromatography on silica gel (20-
60% ethyl acetate-hexane) afforded amine 11 (6.42 g, 58%):
[R]D ) -23.7 (c 8.1, CHCl3); IR (thin film) 2930, 1472, 1456,
1428, 1113 cm-1; 1H NMR (400 MHz, CDCl3) δ 7.70-7.20 (20H,
m), 4.69 (1H, d, J ) 11.7 Hz), 4.68 (1H, d, J ) 11.3 Hz), 4.56
(1H, d, J ) 11.3 Hz), 4.54 (1H, d, J ) 11.7 Hz), 3.99 (1H, dd,
J ) 11.2, 3.8 Hz), 3.92 (1H, dd, J ) 11.2, 5.3 Hz), 3.79 (1H,
m), 3.67 (1H, m), 3.46 (1H, dd, J ) 9.0, 3.3 Hz), 3.34 (1H, dd,
J ) 9.0, 7.4 Hz), 3.28 (3H, s), 3.18 (1H, m), 1.08 (9H, s); 13C
NMR (125 MHz, CDCl3) δ 138.5, 138.4, 135.7, 135.6, 133.3,
133.2, 129.6, 129.6, 128.2, 128.2, 127.7, 127.6, 127.6, 127.6,
127.4, 127.3, 80.6, 80.4, 74.8, 73.7, 72.2, 63.3, 58.7, 52.2, 26.8,
19.1; HRFABMS calcd for MH+ (C36H46NO4Si) 584.3196, found
584.3195 (error 0.2 ppm).
acetate-hexane) afforded methyl ether 7 (4.1 g, 81%): [R]D
)
+26.1 (c, 5.2, CHCl3); IR (thin film) 2932, 1496, 1453, 1346,
1
1196, 1149, 1101 cm-1; H NMR (360 MHz, CDCl3) δ 7.25-
7.45 (10H, m), 5.07 (1H, d, J ) 2.0 Hz), 4.71 (2H, m), 4.62
(1H, d, J ) 12.0 Hz), 4.56 (1H, d, J ) 12.0 Hz), 4.38 (1H, ddd,
J ) 7.0, 6.0, 4.0 Hz), 4.25 (1H, dd, J ) 6.0, 5.0 Hz), 3.93 (1H,
dd, J ) 5.0, 2.0 Hz), 3.75 (1H, dd, J ) 10.0, 7.0 Hz), 3.69 (1H,
dd, J ) 10.0, 4.0 Hz), 3.42 (3H, s), 3.40 (3H, s); 13C NMR (90
MHz, CDCl3) δ 137.8, 137.6, 128.1, 128.1, 127.5, 106.0, 81.7,
77.7, 72.9, 72.2, 72.2, 59.0, 55.0; HRFABMS calcd for MH+
(C21H27O5) 359.18585, found 359.1866 (error 2.1 ppm).
(2R ,3S ,4R )-2,3-B is (b e n zy lo x y )-5-m e t h o x y -1,4-p e n -
ta n ed iol (8). To a solution of furanoside 7 (4.1 g, 11.5 mmol)
in AcOH (64 mL) and H2O (10 mL) was added 1.2 N HCl
aqueous (1 mL). The resulting opaque reaction mixture was
heated at 70 °C for 21 h. The reaction mixture was cooled to
rt, and Na2CO3 (approximately 2 equiv) was added to quench
the HCl. The mixture was concentrated in vacuo, and the
resulting crude product was azeotroped with toluene (3 × 30
mL) and eluted in absolute EtOH (100 mL). Sodium borohy-
dride (1.05 g, 27.6 mmol) was added in several small portions
over 30 min, at which time the reaction mixture was parti-
tioned between CH2Cl2 (100 mL) and H2O (50 mL). The layers
were separated, and the aqueous layer was extracted with CH2-
Cl2 (2 × 100 mL). The combined organics were dried over Na2-
SO4, filtered, and concentrated. Purification via column
chromatography (20-50% ethyl acetate-hexane) afforded diol
8 (3.1 g, 79% two steps): [R]D ) -14.6 (c 4.8, CHCl3); IR (thin
film) 3448, 2921, 1494, 1451, 1397, 1211, 1104 cm-1; 1H NMR
(360 MHz, CDCl3) δ 7.25-7.45 (10H, m), 4.78 (1H, d, J ) 11.0
Hz), 4.65 (2H, s), 4.60 (1H, d, J ) 11.0 Hz), 4.02 (1H, m), 3.89
(1H, dd, J ) 12.0, 4.0 Hz), 3.71-3.82 (3H, m), 3.46 (1H, dd, J
) 10.0, 6.0 Hz), 3.40 (1H, dd, J ) 10.0, 6.0 Hz), 3.32 (3H, s);
13C NMR (90 MHz, CDCl3) δ 137.7, 137.6, 128.1, 128.0, 127.9,
127.5, 127.5, 79.3, 76.8, 73.8, 72.0, 69.2, 60.1, 58.6; HRFABMS
calcd for MH+ (C20H27O5) 347.18585, found 347.1856 (error 0.7
ppm).
(2R,3S,4S)-4-(N,N-Dim eth yla m in o)-2,3-bis(ben zyloxy)-
5-m eth oxy-1-(ter t-bu tyld ip h en yl)siloxy-p en ta n e (12). To
a solution of amine 11 (911 mg, 1.56 mmol) in CH3CN (50 mL)
was added formaldehyde (37% wt in H2O, 2 mL), followed by
AcOH (90 µL, 1.56 mmol) and sodium cyanoborohydride (372
mg, 5.92 mmol). The white suspension was stirred at rt for
15 min, at which time saturated aqueous NaHCO3 (0.5 mL)
was added to quench the AcOH. The resulting mixture was
partitioned between CH2Cl2 (200 mL) and saturated aqueous
NaHCO3 (70 mL). The layers were separated, and the aqueous
(2R,3S,4R)-2,3-Bis(ben zyloxy)-5-m eth oxy-1-(ter t-bu tyl-
d ip h en ylsiloxy)-4-p en ta n ol (9). To a solution of 8 (1.285
(19) Scarlato, G. R.; DeMattei, J . A.; Chong, L. S.; Ogawa, A. K.;
Len, M. R.; Armstrong, R. W. J . Org. Chem., previous paper in this
issue.

Asymmetric Synthesis of Calyculin C. 2
J . Org. Chem., Vol. 61, No. 18, 1996 6159
layer was extracted with CH2Cl2 (2 × 50 mL). The combined
organics were dried over Na2SO4, filtered, and concentrated.
Purification via column chromatography on silica gel (20%
ethyl acetate-hexane) yielded amine 12 (788 mg, 83%): [R]D
) -21.2 (c 7.2, CHCl3); IR (thin film) 2930, 2858, 1455, 1113
CDCl3) δ 7.32-7.46 (5H, m), 6.33 (1H, s), 4.23 (1H, dd, J )
8.0, 6.4 Hz), 4.15 (1H, dddd, J ) 8.0, 7.6, 6.4, 5.5 Hz), 3.48
(1H, dd, J ) 8.0, 8.0 Hz), 2.80 (1H, ddd, J ) 14, 10, 9.2 Hz),
2.55 (1H, ddd, J ) 14.0, 10.0, 5.5 Hz), 2.38 (1H, dddd, J )
13.0, 10.0, 7.6, 3.7 Hz), 1.95 (1H, dddd. J ) 13.0, 10.0, 9.2,
5.5 Hz); 13C NMR (90 MHz, CDCl3) δ 178.0, 138.7, 128.5, 128.4,
1
cm-1; H NMR (400 MHz, CDCl3) δ 7.19-7.70 (20H, m), 4.74
(1H, d, J ) 12.0 Hz), 4.70 (1H, d, J ) 11.3 Hz), 4.60 (1H, d, J
) 12.0 Hz), 4.49 (1H, d, J ) 11.2 Hz), 3.82-3.96 (4H, m), 3.52-
3.60 (2H, m), 3.25 (3H, s), 2.83 (1H, m), 2.28 (6H, s), 1.06 (9H,
s); 13C NMR (90 MHz, CDCl3) δ .139.0, 138.8, 135.7, 133.5,
133.5, 129.6, 128.2, 127.6, 127.6, 127.5, 127.3, 127.2, 80.9, 72.8,
72.7, 69.6, 63.6, 63.5, 58.6, 41.8, 26.9, 19.2; HRFABMS calcd
for MH+ (C38H50NO4Si) 612.3509, found 612.3506 (error 0.5
ppm).
125.9, 87.0, 71.6, 58.7, 33.4, 28.0. Anal. Calcd for C12H13-
NO2: C, 70.92; H, 6.45; N, 6.89. Found: C, 70.89; H, 6.45; N,
6.74.
[3R-(3a,6a,7aa)]-Tetr ah ydr o-6-m eth yl-3-ph en yl-(3S-cis)-
3H,5H-p yr r olo[1,2-c]oxa zol-5-on e (16a ). A solution of N,N-
diisopropylethylamine (10.2 mL, 72.8 mmol) in THF (200 mL)
was cooled to -10 °C. n-Butyllithium (36 mL, 1.98 M in
pentane) was added slowly dropwise, and the resulting mixture
was stirred at -10 °C for 15 min. The clear pale yellow
mixture was cooled to -78 °C, and a precooled solution of 15
(13.2 g, 65.0 mmol) in THF (20 mL) was added dropwise down
the side of the flask. The resulting dark brown mixture was
stirred at -78 °C. After 20 min, iodomethane (20 mL, 325.2
mmol) was added, and the reaction mixture was stirred at -78
°C for 1 h. The reaction was quenched via addition of brine
(50 mL) to the cold reaction mixture. The suspension was
partitioned between EtOAc (300 mL) and brine (150 mL). The
layers were separated, and the aqueous phase was extracted
with EtOAc (3 × 300 mL). The combined organics were dried
(2R,3S,4S)-4-(N,N-Dim eth yla m in o)-2,3-bis(ben zyloxy)-
5-m eth oxy-1-p en ta n ol (13). To a solution of 12 (72 mg,
0.118 mmol) in THF (10 mL) was added TBAF (200 µL, 1.0 M
in THF). The clear, pale yellow reaction mixture was stirred
at rt for 8 h. The reaction mixture was concentrated, and
purification via column chromatography on silica gel (1% CH3-
OH-CH2Cl2) afforded alcohol 13 (42 mg, 96%): [R]D ) -25.2
(c 4.2, CHCl3); IR (thin film) 3393, 2926, 2872, 1456, 1113 cm-1
;
1H NMR (500 MHz, CDCl3) δ 7.31-7.44 (10H, m), 4.80 (1H,
d, J ) 12.1 Hz), 4.71 (1H, d, J ) 11.1 Hz), 4.69 (1H, d, J )
12.1 Hz), 4.51 (1H, d, J ) 11.1 Hz), 3.96 (1H, dd, J ) 11.8, 4.9
Hz), 3.92 (1H, m), 3.77-3.74 (2H, m), 3.68-3.64 (2H, m), 3.35
(3H, s), 3.14 (1H, ddd, J ) 9.2, 6.6, 2.6 Hz), 2.40 (6H, s); 13C
NMR (125 MHz, CDCl3) δ 138.5, 138.3, 128.3, 128.2, 127.8,
127.7, 127.6, 127.4, 80.1, 78.6, 73.0, 71.1, 68.9, 61.8, 60.9, 58.6,
41.7; HRFABMS calcd for MH+ (C22H32NO4) 374.2331, found
374.2336 (error 1.3 ppm).
over Na2
SO4, filtered, and concentrated. Purification via
column chromatography on silica gel (0-20% ethyl acetate-
hexane) afforded lactam 16a (10.3 g, 73%): [R]D
) +205 (c
1.00, CHCl3); IR (thin film) 2969, 1703, 1452, 1376, 1266, 1223
1
cm-1; H NMR (500 MHz, CDCl3) δ 7.31-7.46 (5H, m), 6.33
(1H, s), 4.22 (1H, dd, J ) 8.36.4 Hz), 4.07 (1H, dddd, J ) 7.5,
7.2, 6.9, 6.4 Hz), 3.52 (1H, dd, J ) 8.3, 7.2 Hz), 2.94 (1H, ddq,
J ) 11.3, 8.6 Hz, 6.9 Hz), 2.61 (1H, ddd, J ) 12.0, 8.6, 6.9 Hz),
1.54 (1H, ddd, J ) 12.0, 11.0, 7.5 Hz), 1.24 (3H, d, J ) 7.1
Hz); 13C NMR (90 MHz, CDCl3) δ 179.1, 138.6, 128.4, 128.3,
125.9, 86.7, 72.4, 56.5, 40.0, 34.8, 15.5; HRMS (50 eV EI) calcd
for M - H (C13H14NO2) 216.1025, found 216.1017 (error 3.7
ppm). Anal. Calcd for C13H15NO2: C, 71.85; H, 6.96; N, 6.45,
found C, 71.99; H, 6.97; N, 6.70.
(5R,3R)-5-[(Meth an esu lfon yl)oxy)m eth yl]-2-m eth ylpyr -
r olid in -2-on e (17). To a solution of 16a (10.3 g, 47.4 mmol)
in CH3OH (180 mL) and H2O (20 mL) was added pTSA (215
mg, 1.13 mmol), and the reaction mixture was heated to reflux
(bath temperature ∼80 °C) for 11 h. The mixture was cooled
to rt and concentrated to yield a crude white solid which was
eluted in CH2Cl2 (250 mL). To this mixture were added
triethylamine (10 mL, 71.7 mmol) and methanesulfonyl chlo-
ride (4.4 mL, 56.9 mmol). The resulting mixture was stirred
at rt. After 30 min, the reaction mixture was washed with
H2O (200 mL). The layers were separated, and the organic
layer was concentrated to afford a white solid. Multiple
recrystallizations from ethyl acetate afforded mesylate 17 (6.19
(2S,3S,4S)-4-(N,N-Dim eth yla m in o)-5-m eth oxy-2,3-bis-
(ben zyloxy)va ler ic Acid (14). To a solution of 13 (1.41 g,
3.78 mmol) in acetone (200 mL) was added the J ones reagent
(3.65 mL, 2.66 M in CrO3, 4.14 M in H2SO4), and H2SO4 (750
µL) was added dropwise in five portions over 4 h. After 5 h
total reaction time, the reaction mixture was partitioned
between EtOAc (300 mL) and brine (250 mL). The layers were
separated, and the aqueous layer was extracted with EtOAc
(2 × 200 mL). The combined organics were dried over Na2-
SO4, filtered, and concentrated. Purification via column
chromatography (0-15% CH3OH-CH2Cl2) afforded acid 14
(553 mg, 38%): IR (thin film) ∼3400 (br), 2928, 1617, 1474,
1387, 1101 cm-1; 1H NMR (400 MHz, CDCl3) δ 7.20-7.36 (10H,
m), 4.78 (1H, d, J ) 11.5 Hz), 4.62 (1H, d, J ) 11.4 Hz), 4.48
(1H, d, J ) 11.5 Hz), 4.38 (1H, d, J ) 11.3 Hz), 4.35 (1H, s),
3.85 (2H, m), 3.77 (1H, m), 3.58 (1H, dd, J ) 11.3, 8.8 Hz),
3.27 (3H, s), 2.64 (6H, s); 13C NMR (125 MHz, CDCl3) δ 174.2,
137.8, 137.1, 128.4, 128.3, 128.1, 128.0, 127.9, 127.7, 81.2, 75.8,
71.8, 71.7, 68.3, 62.9, 58.9; HRFABMS calcd for MH+ (C22H30
NO5) 388.2124, found 388.2128 (error 1.0 ppm).
-
(2S,3S,4S)-4-(N,N-Dim eth yla m in o)-5-m eth oxy-2,3-bis-
(ben zyloxy)va ler ic Acid Meth yl Ester (3). To a solution
of acid 14 (18 mg, 0.046 mmol) in CH2Cl2 (3 mL) was added 2
N HCl (3 drops), and TMSCHN2 (500 µL total, 2.0 M in
hexanes) was added until a yellow color persisted. The
reaction mixture was partitioned between CH2Cl2 (10 mL) and
saturated aqueous NaHCO3 (10 mL). The layers were sepa-
rated, and the aqueous layer was extracted with CH2Cl2 (1 ×
20 mL) and EtOAc (2 × 20 mL). The combined organics were
dried over Na2SO4, filtered, and concentrated. Purification via
column chromatography on silica gel (1% CH3OH-CH2Cl2)
afforded ester 3 (17 mg, 97%): [R]D ) -49.2 (c 2.3, CDCl3); IR
(thin film) 2930, 1750, 1455, 1281, 1206, 1117 cm-1; 1H NMR
(360 MHz, C6D6) δ 7.26-7.40 (10H, m), 4.87 (1H, d, J ) 12.2
Hz), 4.56 (1H, d, J ) 11.3 Hz), 4.53 (1H, d, J ) 12.2 Hz), 4.43
(1H, d, J ) 11.3 Hz), 4.32 (1H, d, J ) 2.2 Hz), 3.91 (1H, d, J
) 8.8 Hz), 3.71 (3H, s), 3.66 (1H, dd, J ) 10.1, 2.6 Hz), 3.53
(1H, dd, J ) 10.1, 7.6 Hz), 3.29 (3H, s), 3.12 (1H, m), 2.25 (6H,
s); 13C NMR (90 MHz, CD3CN) δ 171.4, 139.9, 139.5, 129.2,
129.2, 128.9, 128.7, 128.5, 80.5, 78.5, 73.4, 73.2, 70.1, 62.6, 58.8,
51.9, 42.1; HRFABMS calcd for MH+ (C23H32NO5) 402.22805,
found 402.2280 (error 0.1 ppm).
g, 63% two steps): [R]D ) +17 (c 1.1, CHCl3); IR (thin film)
1
3400 (br), 1701, 1645, 1343, 1170 cm-1
; H NMR (500 MHz,
CDCl3
) δ 7.40 (1H, br s), 4.24 (1H, dd, J ) 10.0, 3.6 Hz), 4.01
(1H, dd, J ) 10.0, 7.4 Hz), 3.90 (1H, dd, J ) 7.4, 3.6 Hz), 3.07
(3H, s), 2.49 (1H, m), 2.42 (1H, m), 1.37 (1H, ddd, J ) 12.0,
9.4, 8.1 Hz), 1.16 (3H, d, J ) 7.0 Hz); 13C NMR (90 MHz,
CDCl3) δ 180.4, 71.3, 51.0, 37.4, 35.9, 31.4, 16.0.
(5S,3S)-2,5-Dim eth ylp yr r olid in -2-on e (18). A solution
of 17 (1.100 g, 5.31 mmol) in DME (60 mL) was degassed via
a stream of argon. To this solution were added sodium iodide
(1.53g, 10.2 mmol), tributyltin hydride (2.15 mL, 8.0 mmol),
and AIBN (17 mg, 0.11 mmol). The cloudy white reaction
mixture was heated to reflux (bath temperature ∼100 °C) for
5 h. The reaction mixture was cooled to rt, filtered, and
concentrated. Purification via column chromatography on
silica gel (0-5% CH3OH-CH2Cl2) afforded lactam 18 (533 mg,
89%): [R]D ) -18 (c 0.57, CHCl3); IR (thin film) 3238, 2962,
1711, 1654, 1456, 1427, 1307, 1255 cm-1; 1H NMR (500 MHz,
CDCl3) δ 7.19 (1H, br s), 3.59 (1H, m), 2.44-2.33 (2H, m), 1.18
(1H, m), 1.16 (3H, d, J ) 6.1 Hz), 1.12 (3H, d, J ) 5.3 Hz); 13
C
NMR (90 MHz, CDCl3) δ 180.6, 48.0, 38.7, 37.2, 22.0, 15.9;
HRMS (20 eV CI) calcd for M+ (C6H11NO) 113.0841, found
113.0836 (error 4.4 ppm). Anal. Calcd for C6H11NO: C, 63.67;
H, 9.8; N, 12.38. Found: C, 63.42; H, 9.97; N, 12.27.
Tetr a h yd r o-3-p h en yl-(3R-cis)-3H,5H-p yr r olo[1,2-c]ox-
a zol-5-on e (15). F or 15: [R]D ) +247 (c 1.0, CHCl3); IR (thin
1
film) 2943, 1702, 1350, 1222, 1163 cm-1; H NMR (500 MHz,

6160 J . Org. Chem., Vol. 61, No. 18, 1996
Ogawa et al.
(5S,3S)-N-(ter t-Bu toxyca r bon yl)-2,5-d im eth ylp yr r oli-
d in -2-on e (19). To a solution of 18 (54 mg, 0.48 mmol) in
CH2Cl2 (10 mL) were added triethylamine (200 µL, 1.44 mmol),
DMAP (76 mg, 0.62 mmol), and Boc2O (265 µL, 1.15 mmol).
The clear yellow reaction mixture was stirred at rt for 12 h.
The reaction mixture was partitioned against H2O (10 mL).
The layers were separated, and the aqueous layer was
extracted with CH2Cl2 (3 × 10 mL). The combined organics
were dried over Na2SO4, filtered, and concentrated. Purifica-
tion via column chromatography on silica gel (10% ethyl
acetate-hexane) afforded lactam 19 (84 mg, 82%): [R]D ) -51
(c 1.3, CHCl3); IR (thin film) 2975, 1784, 1748, 1752, 1507,
(15 mL). The layers were separated, and the aqueous layer
was extracted with EtOAc (2 × 30 mL). The combined
organics were dried over Na2SO4, filtered, and concentrated.
Purification via column chromatography on silica gel (0-5%
CH3OH-CH2Cl2) afforded amide 23 (30 mg, 46%), 24 (11 mg,
16%), and epimerized ester starting material (8 mg, 17%). For
23: [R]D ) -37.4 (c 1.6, CHCl3); IR (thin film) 3401, 2928, 1671,
1653, 1522, 1456, 1103 cm-1 1H NMR (500 MHz, CDCl3) δ
;
7.55 (1H, t, J ) 1.0 Hz), 7.19-7.35 (10H, m), 6.62 (1H, br d, J
) 11.3 Hz), 4.72 (1H, d, J ) 14.3 Hz), 4.55 (1H, d, J ) 14.8
Hz), 4.52 (1H, d, J ) 14.4 Hz), 4.48 (2H, d, J ) 1.1 Hz), 4.46
(1H, d, J ) 14.3 Hz), 4.44 (1H, d, J ) 2.2 Hz), 4.18 (1H, dd, J
) 12.5, 1.9 Hz), 4.07 (1H, m), 3.65 (1H, dd, J ) 12.7, 2.8 Hz),
3.56 (1H, dd, J ) 12.7, 7.6 Hz), 3.25 (3H, s), 2.96 (1H, m), 2.72
(1H, m), 2.35 (6H, s), 1.79 (1H, ddd, J ) 18.0, 11.7, 6.7 Hz),
1.56 (1H, ddd, J ) 17.9, 11.3, 6.2 Hz), 1.33 (3H, d, J ) 8.6
Hz), 1.02 (3H, d, J ) 8.3 Hz); 13C NMR (125 MHz, CDCl3) δ
171.9, 168.6, 138.5, 137.4, 137.1, 135.7, 128.6, 128.6, 128.2,
128.6, 127.7, 127.4, 79.8, 79.1, 74.9, 74.5, 68.3, 62.8, 58.5, 42.2,
42.0, 41.8, 37.1, 30.5, 21.2, 17.9; HRFABMS calcd for MH+
(C31H43N3O5Cl) 572.2891, found 572.2892 (error 0.2 ppm).
[2S,3S,4S,N(1R,3S)]-4-(N,N-Dim eth ylam in o)-N-[1-m eth -
yl-3-(4-((t r ibu t ylp h osp h in o)m et h yl)-2-oxa zoyl)b u t yl]-5-
m eth oxy-2,3-bis(ben zyloxy)va ler a m id e (25). To a solution
of 23 (15 mg, 0.026 mmol) in distilled DMF (3 mL) was added
tributylphosphine (31 µL, 0.131 mmol). The clear yellow
mixture was heated to 70 °C for 3 h, at which time the reaction
mixture was cooled and concentrated. Purification via column
chromatography on silica gel (1-15% CH3OH-CH2Cl2) af-
forded phosphonium salt 25 (14 mg, 70%): [R]D ) -4.56 (c
1.6, CHCl3); IR (thin film) 3403, 2961, 2934, 1669, 1653, 1522,
1
1304, 1153 cm-1; H NMR (500 MHz, CDCl3) δ 4.01 (1H, m),
2.50 (1H, m), 2.40 (1H, m), 1.51 (9H, s), 1.36 (3H, d, J ) 6.1
Hz), 1.23 (1H, m), 1.22 (3H, d, J ) 7.1 Hz); 13C NMR (90 MHz,
CDCl3) δ 177.1, 150.4, 82.7, 52.2, 37.6, 34.3, 28.8, 22.0, 16.8.
Anal. Calcd for C11H19NO3: C, 61.93; H, 8.98; N, 6.57.
Found: C, 62.08; H, 8.99; N, 6.29.
(2S,4R)-4-[(ter t-Bu t oxyca r b on yl)a m in o]-2,4-d im et h -
ylbu ta n a m id e (20). Through a solution of 19 (668 mg, 3.13
mmol) in CH2Cl2 was bubbled NH3(g) for ∼3 min. Trimethyl-
aluminum (2.35 mL, 2.0 M in toluene) was added rapidly
dropwise, and the reaction mixture was stirred for 2 h at rt.
The reaction was quenched via dropwise addition of 0.1 N
aqueous HCl (3 mL), and the resulting suspension was stirred
for 5 min, filtered, and concentrated. Purification via column
chromatography on silica gel (0-5% CH3OH-CH2Cl2) afforded
amide 20 (449 mg, 62%) and lactam 19 (210 mg, 31%). 20:
[R]D ) +6.7 (c 1.4, CHCl3); IR (thin film) 3384, 3352, 3027,
2971, 1602, 1643, 1618, 1530, 1275, 1166 cm-1; 1H NMR (500
MHz, CDCl3) δ 6.11 (1H, br s), 5.72 (1H, br s), 4.53 (1H, br d),
3.68 (1H, br m), 2.35 (1H, br m), 1.86 (1H, br m), 1.41 (9H, s),
1.18 (3H, d, J ) 6.9 Hz), 1.12 (3H, d, J ) 6.5 Hz); 13C NMR
(90 MHz, CDCl3) δ 179.3, 156.8, 79.2, 45.5, 41.4, 38.4,
28.4,.21.9, 18.8; HRMS (NH3/CI) calcd for MH+ (C11H23N2O3)
231.1709, found 231.1715 (error 2.6 ppm). Anal. Calcd for
C11H22N2O3: C, 57.37; H, 9.63; N, 12.16. Found: C, 57.77; H,
9.54; N, 12.03.
1456, 1277, 1098 cm-1 1H NMR (500 MHz, CD3CN) δ 7.88
;
(1H, d, J ) 4.0 Hz), 7.29-7.44 (10H, m), 4.61-4.66 (2H, m),
4.48-4.55 (2H, m), 4.33 (1H, s), 4.13 (1H, m), 4.00 (1H, m),
3.59-3.73 (2H, m), 3.58 (2H, dd, J ) 14.5, 0.6 Hz), 3.29 (3H,
s), 2.94-3.10 (1H, m), 2.18-2.33 (12H, m), 1.44-1.65 (14H,
m), 1.31 (3H, d, J ) 6.9 Hz), 1.10 (3H, d, J ) 6.5 Hz), 0.95
(9H, t, J ) 7.3 Hz); 13C NMR (125 MHz, CD3CN) δ 169.9, 139.8,
138.9, 138.6, 138.6, 129.7, 129.6, 129.6, 129.4, 129.1, 128.9,
128.6, 128.4, 80.4, 79.8, 75.2, 74.8, 68.9, 63.7, 58.7, 43.1, 42.7,
42.3, 31.6, 24.6, 24.4, 23.9, 23.8, 19.5, 19.1, 18.4, 13.6; HR-
FABMS calcd for M+ (C43H69N3O5P) 738.4975, found 738.4974
(error 0.1 ppm).
[2(1S,3R)]-4-(Ch lor om eth yl)-2-[3-[(ter t-bu toxyca r bon -
yl)a m in o]-1,3-d im eth ylp r op yl]oxa zole (4). To a solution
of 20 (70 mg, 0.304 mmol) in distilled CHCl3 (7.5 mL) were
added K2CO3 (298 mg, 2.1 mmol) and 1,3-dichloroacetone (250
mg, 1.9 mmol). The reaction mixture was heated to reflux at
a bath temperature of >100 °C. After 9 h at reflux, the
reaction mixture was cooled to rt and partitioned between
CHCl3 (20 mL) and brine/H2O (10 mL/10 mL). The layers were
separated, and the aqueous layer was extracted with CHCl3
(3 × 20 mL). The combined organics were dried over Na2SO4,
filtered, and concentrated. Purification via column chroma-
tography on silica gel (5-20% ethyl acetate-hexane) afforded
oxazole 4 (56 mg, 61%), and amide 20 (4 mg, 6%). [R]D ) 21.6
(c 4.0, CHCl3); IR (thin film) 3438 (br), 2978, 1705, 1505, 1367,
[2S,3S,4S,N(1R,3S)]-4-(N,N-Dim eth ylam in o)-N-[1-m eth -
yl-3-(4-((tr ibu tylp h osp h in o)m eth yl)-2-oxa zoyl)bu tyl]-2,3-
d ih yd r oxy-5-m eth oxyva ler a m id e (2). To a solution of 25
(40 mg, 0.052 mmol) in CH3OH (18 mL) was added 3 N HCl
(anhydrous in CH3OH, 0.52 mL), followed by 10% Pd-C (36
mg). Hydrogen gas was bubbled through the suspension for
5 min, and the reaction mixture was stirred at rt under H2.
After 10 h, an additional 42 mg of 10% Pd-C was added. After
15 h, the mixture was filtered through a plug of Celite which
was rinsed with CH3OH (50 mL), EtOAc (50 mL), and CH2Cl2
(50 mL). The combined organics were concentrated, and
purification via column chromatography on silica gel (5-30%
CH3OH-CH2Cl2) afforded diol 2 (21 mg, 68%): [R]D ) -72.9
(c 0.8, CHCl3); IR (thin film) 3223 (br), 2963, 2934, 1653, 1566,
1
1250, 1163 cm-1; H NMR (400 MHz, CDCl3) δ 7.50 (1H, s),
4.43 (2H, s), 3.73 (1H, m), 3.01 (1H, m), 1.89 (1H, m), 1.65
(1H, m), 1.36 (9H, s), 1.31 (3H, d, J ) 6.9 Hz), 1.09 (3H, d, J
) 6.6 Hz); 13C NMR (100 MHz, CDCl3) δ 168.9, 155.2, 136.9,
135.6, 79.0, 44.3, 42.0, 36.9, 30.8, 28.2, 21.9, 18.1 HRFABMS
calcd for MH+, C14H24N2O3Cl: 303.1475, found 303.1490 (error
4.9 ppm).
1464, 1262, 1098 cm-1 1H NMR (500 MHz, CD3CN) δ 7.91
;
(1H, d, J ) 3.8 Hz), 4.50 (1H, s), 4.36 (1H, br s), 4.09 (1H, br
s), 3.68-3.91 (5H, m), 3.39 (3H, s), 3.13 (1H, m), 2.82 (6H, br
s), 2.22 (6H, m), 2.08 (1H, m), 1.73 (1H, m), 1.59 (6H, m), 1.49
(6H, m), 1.32 (3H, d, J ) 6.9 Hz), 1.19 (3H, d, J ) 6.6 Hz),
0.96 (9H, t, J ) 3.8 Hz); 13C NMR (125 MHz, CD3CN) δ 173.5,
170.0, 139.1, 139.0, 129.2, 129.1, 72.5, 69.9, 68.6, 66.7, 59.4,
44.8, 42.1, 32.3, 24.6, 24.5, 23.9, 23.9, 21.8, 19.4, 19.1, 18.8,
18.4, 12.7; HRFABMS calcd for M+ (C29H57N3O5P) 558.4036,
found 558.4041 (error 0.9 ppm).
[2S,3S,4S,N(1R,3S)]-4-(N,N-Dim eth ylam in o)-N-[1-m eth -
yl-3-(4-(ch lor om et h yl)-2-oxa zoyl)b u t yl]-5-m et h oxy-2,3-
bis(ben zyloxy)va ler a m id e (23). A solution of 4 (94 mg,
0.310 mmol) in EtOAc (11 mL) was cooled to 0 °C, and HCl(g)
was bubbled through for 2 min. The cloudy reaction mixture
was stirred at 0 °C for 15 min. The excess HCl was purged
via a stream of argon (5 min), and a white precipitate formed.
The suspension was concentrated and azeotroped with CH2-
Cl2 (2 × 2 mL) to afford a white solid which was eluted in CH2-
Cl2 (6 mL). Trimethylaluminum (260 µL, 2.0 M in toluene)
was added, and the clear yellow-brown mixture was stirred
for 25 min, at which time a solution of 3 (46 mg, 0.115 mmol)
in CH2Cl2 (1.0 mL and 2 × 0.5 mL rinses) was added. The
resulting clear brown reaction mixture was heated to 35 °C
for 9.5 h. The reaction mixture was quenched via addition of
CH3OH (∼1 mL), followed by 0.1 N aqueous HCl (2 mL). The
mixture was partitioned between EtOAc (20 mL) and brine
ter t-Bu tyl [3-[4-[(1E)-3-[(2R,3R,5R,7S,8R,9R)-2-[(1S,3R,
4S)-3-(ter t-bu tyl d im eth ylsiloxy)-1-m eth oxy-4-m eth yl-5-
h exen yl]-9-(b en zoyloxy)-3-h yd r oxy-4,4,8-t r im et h yl-1,6-
d ioxa sp ir o[4.5]d ec-7-yl]p r op en yl]-2-oxa zolyl]p r op yl]ca r -
ba m a te (28). To a solution of phosphonium salt 26 (11 mg,
0.022 mmol) in DMF (0.44 mL) at 0 °C was added LDA (88
µL, 0.25M in THF). The yellow mixture was stirred at 0 °C
for 10 min, at which time a solution of aldehyde 27 (7 mg,
0.011 mmol) in DMF (0.2 mL + 0.2 mL rinse) was added

Asymmetric Synthesis of Calyculin C. 2
J . Org. Chem., Vol. 61, No. 18, 1996 6161
dropwise. The reaction mixture was stirred at 0 °C for 45 min,
at which time the reaction mixture was partitioned between
5% aqueous NaHCO3 (10 mL) and ether (10 mL). The layers
were separated, and the aqueous layer was extracted with
ether (2 × 10 mL). The combined organics were dried over
Na2SO4, filtered, and concentrated. Purification via prepara-
tive thin layer chromatography (35% ethyl acetate-hexane)
afforded coupled product 28 (5.3 mg, 59%): IR (thin film) 3500,
3350 (br), 2957, 2928, 1717, 1472, 1453, 1366, 1277, 1260,
ethyl acetate (5 × 10 mL). The combined organics were dried
over Na2SO4, filtered, and concentrated. Purification via
column chromatography on silica gel (20% ethyl acetate-
hexanes to 10% CH3OH-CH2Cl2) afforded coupled product 29
(5 mg, 35%): IR (thin film) 3500, 3350(br), 2928, 1717, 1645,
1464, 1277, 1100 cm-1; 1H NMR (500 MHz, CDCl3) δ 8.19 (2H,
m), 7.60 (1H, m), 7.48 (2H, m), 7.46 (1H, s), 7.10 (1H, m), 6.29-
6.35 (2H, m), 5.77 (1H, ddd, J ) 17.7, 10.4, 7.8 Hz), 5.21 (1H,
m), 4.94-4.98 (2H, m), 4.68 (1H, m), 4.15-4.30 (2H, m), 4.11
(1H, dd, J ) 8.2, 4.3 Hz), 3.79-3.90 (3H, m), 3.70 (1H, m),
3.59 (1H, dd, J ) 12.1, 4.3 Hz), 3.52 (3H, d, J ) 137.0 Hz),
3.30-3.45 (4H, m), 2.90-3.10 (3H, m), 2.30-2.50 (9H, m),
1.95-2.10 (2H, m), 1.70-1.90 (3H, m), 1.45 (1H, m), 1.38 (3H,
d, J ) 6.9 Hz), 1.26 (3H, d, J ) 6.6 Hz), 1.21 (3H, s), 1.06 (3H,
d, J ) 7.1 Hz), 1.03 (3H, d, J ) 6.9 Hz), 0.95 (3H, s), 0.91 (9H,
s), 0.09 (3H, S), 0.07 (3H, s); HRFABMS DCM/NBA calcd for
MH+ (C5013CH84N3O12Si) 959.5858, found 959.5903 (error 4.7
ppm).
1
1175, 1101 cm-1; H NMR (360 MHz, CDCl3) δ 8.13 (2H, m),
7.54 (1H, m), 7.42 (2H, m), 7.37 (1H, s), 6.26 (2H, m), 5.69
(1H, ddd, J ) 17.6, 10.3, 7.6 Hz), 5.14 (1H, m), 4.83-4.93 (3H,
m), 4.60 (1H, m), 4.03 (1H, dd, J ) 8.5, 4.2 Hz), 3.77 (1H, m),
3.49 (1H, dd, J ) 12.2, 4.2 Hz), 3.44 (3H, d, J ) 134.5 Hz),
3.27 (2H, m), 3.17 (1H, m), 2.74 (2H, m), 2.15-2.40 (3H, m),
1.85-1.95 (3H, m), 1.73 (1H, m), 1.14 (3H, s), 1.02 (3H, d, J )
6.9 Hz), 0.99 (3H, d, J ) 7.0 Hz), 0.88 (3H, s), 0.84 (9H, s),
0.02 (3H, s), 0.00 (3H, s); HRFABMS DCM/NBA calcd for MH+
(C4513CH73N2O10Si) 842.5068, found 842.5071 (error 0.3 ppm).
(2S,3S,4S)-N-[(1R,3S)-3-[4-[(1E)-3-[(2R,3R,5R,7S,8R,9R)-
2-[(1S,3R,4R)-3-(ter t-Bu tyld im eth ylsiloxy)-1-m eth oxy-4-
m et h yl-5-h exen yl]-9-(b en zoyloxy)-3-h yd r oxy-4,4,8-t r i-
m eth yl-4,6-d ioxa sp ir o[4.5]d ec-2-yl]p r op en yl]-2-oxa zolyl]-
1-m eth ylbu tyl]-4-(dim eth ylam in o)-2,3-dih ydr oxy-5-m eth -
oxyva ler a m id e (29). To a solution of phosphonium salt 2
(13 mg, 0.022 mmol) in DMF (0.3 mL) at 0 °C was added LDA
(88 µL, 0.25M in THF). The yellow mixture was stirred at 0
°C for 10 min, at which time a solution of aldehyde 27 (9.1
mg, 0.015 mmol) in DMF (0.2 mL + 0.1 mL rinse) was added.
The reaction mixture was stirred at 0 °C, and portions of LDA
(4 × 44 µL, 2 equiv total) were added over the course of 1.5 h.
The reaction was quenched via addition of saturated aqueous
NaHCO3 (3 mL). The reaulting mixture was extracted with
Ack n ow led gm en t. We thank Mavis Lin and Greg
Mayeur for their synthetic efforts. Support from the
NSF (CHE 88-58059) is graciously acknowledged.
Su ppor tin g In for m ation Available: Procedures and char-
acterization data for 5, 16b, 21, 22, 24, 26, and 27 and NMR
spectra of all new compounds to indicate purity (65 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
journal, and can be ordered from the ACS; see any current
masthead page for ordering information.
J O960315Q