Total synthesis of stipiamide and designed polyenes as new agents for the reversal of multidrug resistance

Article abstract of DOI:10.1021/ja972603p

The synthesis of (-)-stipiamide (1) is reported together with the designed enynes 2 (6,7-dehydrostipiamide) and 3 that are now shown to reverse the multidrug resistance (MDR) of human breast cancer cells (MCF-7adrR). Stipiamide was assembled using a Still

Full text of DOI:10.1021/ja972603p

J. Am. Chem. Soc. 1997, 119, 12159-12169
12159
Total Synthesis of Stipiamide and Designed Polyenes as New
Agents for the Reversal of Multidrug Resistance
Merritt B. Andrus,*,1 Salvatore D. Lepore, and Timothy M. Turner
Contribution from the Department of Chemistry, Purdue UniVersity,
West Lafayette, Indiana 47907-1393
ReceiVed July 30, 1997X
Abstract: The synthesis of (-)-stipiamide (1) is reported together with the designed enynes 2 (6,7-dehydrostipiamide)
and 3 that are now shown to reverse the multidrug resistance (MDR) of human breast cancer cells (MCF-7adrR).
Stipiamide was assembled using a Stille coupling with (E)-vinyl iodide 17 and (Z)-stannyl amide 16 in 78% yield.
(E)-Vinyl iodide 17 was made using a Takai reaction and a selective dihydroxylation of the terminal olefin of
nonconjugated diene 7 using the Sharpless AD-mix reagent. The precursor to 16, (E,Z)-stannyl diene ester 13, was
assembled with high selectivity in a single operation using a tandem syn-addition of tributyltin cuprate to acetylene
followed by conjugate addition to ethyl propiolate. Structural variants 2 and 3 were assembled using palladium-
catalyzed Sonogashira couplings with vinyl iodides 17 and 35 and acetylenes 22 and 26 in high yield at near 1:1
stoichiometry. Compound 2 was found to be far less toxic than stipiamide and performed much better as an MDR
reversal agent. Compound 3 was better still due to even lower toxicity.
Introduction
Chart 1
(-)-Stipiamide (1) (Chart 1) was discovered by Seto from
the soil bacterium Myxocaccus stipiatus using a colchicine-
resistant screen.2 This resistant KB cell line is known to express
Pgp (P-glycoprotein),3 the well-known membrane-bound mul-
tidrug transporter, a critical factor in the development of
multidrug resistance displayed by transformed cells.4 Ho¨fle also
independently reported the isolation of this compound, calling
it phenalamide A1 using an anti-HIV screen.5 While the related
polyenes, the myxalamides differing only slightly at the left-
hand terminus, are potent antifungal and antibacterial agents,6
stipiamide was found to be only slightly effective as an
antimicrobial agent. Further experiments showed that myxala-
mide B (Chart 2) potently inhibits NADH oxidation by complex
I in isolated beef heart submitochondrial particles (IC50 170 pm/
mg of protein)7 with the same level and manner of electron
transfer inhibition exhibited by the powerful insecticides pier-
cidin A and rotenone.8
Pgp is a heavily glycosylated membrane bound small-
molecule efflux pump that is greatly overexpressed in most
cancer cells found to be cross resistant to therapeutic agents.4
X Abstract published in AdVance ACS Abstracts, December 1, 1997.
(1) Current address: Department of Chemistry and Biochemistry,
Brigham Young University, Provo, UT 84602. E-mail: mbandrus@
chemgate.byu.edu.
Chart 2
(2) Kim, Y. J.; Furihata, K.; Yamanaka, S.; Fudo, R.; Seto, H. J. Antibiot.
1991, 44, 553-556.
(3) Higgins, C. F.; Gottesman, M. M. Trends Biol. Sci. 1992, 17, 18-
21.
(4) Lehnert, M. Eur. J. Cancer 1996, 32A, 912-920. Germann, U. A.
Eur. J. Cancer 1996, 32A, 927-944.
(5) Trowitz-Kiensat. W.; Forche, E.; Wray, V.; Reichenbach, H.;
Jurkiewicz, E.; Hunsmann, G.; Ho¨fle, G. Liebigs Ann. Chem. 1992, 659-
664.
(6) (a) Ho¨fle, G.; Bedorf, N.; Jansen, R.; Steinmetz, H.; Trowitz-Kiensat.
W.; Reichenbach, H.; Forche, E.; Gerth, K.; Irschik, H. Kunze, B. Sasse,
F.; Hunsmann, G.; Jurkiewicz, E. Ger. Patent DE 4041686A1, 1990. (b)
Hunsmann, G.; Jurkiewicz, E.; Reichenbach, H.; Forche, E.; Gerth, K.;
Irschik, H.; Kunze, B.; Sasse, F.; Ho¨fle, G.; Bedorf, N.; Jansen, R.;
Steinmetz, H.; Trowitz-Kiensat. Ger. Patent DE 4041688A1, 1990. (c)
Jansen, R.; Reifenstahl, G.; Gerth, K.; Reichenbach, H.; Ho¨fle, G. Leibigs
Ann. Chem. 1983, 1081-1095.
This condition is commonly developed following exposure to
a single anticancer agent. Its action, the transport of multiple
classes of compounds, is believed to be responsible in most
cases for the failure of therapy, the clinical expression of
multidrug resistance (MDR).9 The ability to bind and transport
(7) Gerth, K.; Reifenstahl, G.; Ho¨fle, G.: Irschik, H.; Kunze, B.;
Reichenbach, H.; Thierbach G. J. Antibiot. 1983, 36, 1150-1156.
(8) Jansen, R.; Ho¨fle, G. Tetrahedron Lett. 1983, 24, 5485.
(9) Ferry, D. R.; Traunecker, H.; Kerr, D. J. Eur. J. Cancer 1996, 32A,
1070-1081.
S0002-7863(97)02603-6 CCC: $14.00 © 1997 American Chemical Society

12160 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
Andrus et al.
Chart 3
Chart 4
a range of structurally dissimilar compounds is truly remarkable
and without precedent. These include the tetracycline antibiotics
daunorubricin and adriamycin and the vinca alkaloids vinblastine
and vincristine, along with others such as colchicine, mitomycin
C, actinomycin D, topotecan, mitoxanantrone, etoposide, teni-
poside, paclitaxel, emetine, ethidium bromide, puromycin, and
mithramycin, to name a few. Bleomycin, methotrexate, and
cisplatinum are known not to be transported.
Scheme 1
Of the numerous compounds that have been found to reverse
MDR through inhibition of Pgp, the most studied compound,
verapamil (Chart 3), a well-known calcium ion channel blocker
used to treat hypertension, is still the standard by which all others
are compared.10 It has been shown to interact directly with Pgp
using photolabeling and competitive displacement studies (600
nM).11 Clinical trials, however, have proven disappointing in
that the concentration required to reverse MDR (10 µM) leads
to hypotension and heart failure.9 Other compounds, also based
on known channel blockers, dexniguldipine,12 S9788,13 and GF
120918,14 have also suffered from hypotension problems.
Hapalosin, a cyclic depsipeptide discovered recently, has
demonstrated MDR reversal at 2.5 µM with MCF-7adrR cells
with added adriamycin (15 µM).15
We now report the full details of our recent synthesis of (-)-
stipiamide (1),16 together with the design and synthesis of new
less toxic polyenes for the reversal of MDR. 6,7-Dehydrostipi-
amide (2), the result of the transposition of the C-6,7 (Z)-alkene
of stipiamide to an alkyne, was made and is shown to be far
less toxic and more potent as an MDR reversal agent. In
addition a less toxic and further simplified enyne (3) designed
lacking the extended π conjugation was produced.
The final step for the synthesis of stipiamide employs a Stille
coupling with an (E)-vinyl iodide and a (Z)-stannyl triene amide
(Chart 4). The vinyl iodide was produced from an enal using
a highly selective Takai reaction. The enal precursor was
generated by oxidative cleavage of a diol that was produced
using a new application of the Sharpless osmium tetroxide AD-
mix reaction that is selective for the terminal olefin of the
nonconjugated diene substrate.17 A Brown crotylborane addition
was used to create the vicinal 1-hydroxy-2-methyl functionality.
The stannyl triene coupling partner was derived from the product
of a new single-flask stannyl cuprate-acetylene-alkyne ester
tandem addition sequence to give the key (E,Z)-stannyl diene
ester in high yield and selectivity. Structural variants 2 and 3
were assembled using improved conditions for the palladium-
catalyzed Sonogashira coupling with (E)-vinyl iodides and
terminal acetylenes in high yield at near 1:1 stoichiometry.
Results and Discussion
The synthesis of the left-hand portion of stipiamide (Scheme
1) began with acyloxazolidinone 4 methylated with greater than
25:1 selectivity according to the procedure of Evans.18 The
product was reacted with LAH in 60% yield and the resultant
alcohol was oxidized to the aldehyde with tetrapropylammonium
perruthenate (TPAP) and N-methylmorpholine N-oxide (NMP).19
The crude aldehyde was treated with (carbethoxyethylidene)-
triphenylphosphorane to give 5 as an 8:1 diastereomeric mixture
of unsaturated esters which were not separated.
(10) Tsuruo, T.; Iida, H.; Tsukagoshi, S.; Sakurai, Y. Cancer Res. 1981,
41, 1967-1972.
(11) Safa, A. R. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 7187-7191.
(12) Hofmann, J.; Gekeler, V.; Ise, W. Biochem. Pharmacol. 1995, 49,
603-609.
(13) Huet, S.; Chapey, B.; Robert, J. Eur. J. Cancer 1993, 29A, 1377-
1388.
(14) Hyafil, F.; Vergely, C.; Du, V. P. Grand, P. T. Cancer Res. 1993,
53, 4595-4602.
Fragment 5 was reduced to the alcohol using 2.5 equiv of
DIBAL in THF at -40 °C. This intermediate was oxidized
(15) Stratmann, K.; Burgoyne, D. L.; Moore, R. E.; Patterson, G. M. L.
J. Org. Chem. 1994, 59, 7219-7226.
(16) Andrus, M. B.; Lepore, S. D. J. Am. Chem. Soc. 1997, 119, 2327.
(17) Andrus, M. B.; Lepore, S. D.; Sclafani, J. A. Tetrahedron Lett. 1997,
38, 4043.
(18) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737-1738.
(19) Griffith, W. P; Ley, S. V. Aldrichim. Acta 1990, 23, 13.

Synthesis of Stipiamide and Designed Polyenes
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12161
Scheme 2
Scheme 3
as noted by others, we have found that the yield is lowered
upon scale up above 1 g.25
using TPAP to give aldehyde 6 which was subsequently reacted
with diisopinocampheyl-(E)-crotylborane derived from (-)-R-
pinene according to the procedure of Brown.20 This gave the
key anti-homoallylic alcohol 7 in 60% yield for the three
combined steps. In order to continue the synthetic sequence,
oxidative cleavage of the terminal olefin in the presence of the
internal trisubstituted olefin was required. Initially it was
considered that this would be a limitation of the crotylboration
method. Initial attempts to selectively oxidize the terminal
olefin of TBS-protected diene 7 were made using standard
osmium tetroxide-NMO conditions with added amines in
various solvents.21 These reactions were found to be sluggish,
giving mixtures of internal and terminal diol products. The
attempt was then made using the commercially available
asymmetric osmium tetroxide dihydroxylation reagent, AD-mix
of Sharpless.22 The terminal olefin of 7 was smoothly converted
to the 1,2-diol 8 in 87% yield with no internal dihydroxylation
or tetrahydroxylation products observed. Many of the factors
contributing to the selectivity of this useful transformation, in
particular the effect of added base, along with various substrates
are included in a recent report.17
With the nonconjugated diene problem solved, diol 8 was
treated with sodium periodate, producing the corresponding
aldehyde which was reacted with (carbethoxyethylidene)-
triphenylphosphorane to give ester 9 in 75% yield as an 8:1
E/Z mixture (Scheme 2). Conversion of the ester to the alcohol
was achieved using DIBAL followed by TPAP and NMO to
provide aldehyde 10. Treatment of enal 10 with iodoform and
chromous chloride in THF gave vinyl iodide 11 in 70% yield
with 20:1 E/Z selectivity.23 Several features of the Takai
reaction deserve some comment in that the initial attempts
following reported procedures gave a wide range of yields (10-
50%) and low selectivity. A recent investigation concerning
the effect of solvent found an optimal solvent system of 6:1
dioxane/THF that gave a 69% yield after 8 h with good
selectivity (13:1 E/Z).24 With enal 10, dioxane as cosolvent
produced an extremely sluggish reaction and provided only
slightly enhanced selectivity at the expense of yield compared
to THF alone. After exploring numerous variations, the
remarkably high 20:1 E/Z selectivity was finally achieved only
after the newly purchased chromous chloride was gently flame
dried under vacuum immediately prior to use. Following
addition of THF to the cooled material, care was taken to
mechanically agitate the chromous chloride to ensure that it had
not fused. This precaution greatly decreased the reaction time
to 3 h at 0 °C, at the same time increasing the yield. In addition,
Inspiration for the new tandem acetylene-propiolate-cuprate
addition reaction developed to construct the right-hand portion
came from the work of Corey and others who found that
alkylcopper reagents add to acetylenic esters to form R,â-
unsaturated esters in high yield and selectivity.26 This useful
cis-stannylvinyl cuprate intermediate, first made by Westmijze,
is the product of a stannylcopper(I) bimetallic reagent reacted
with acetylene in syn-fashion.27 Marino reported that 5 equiv
of cis-(tributylstannyl)vinyl cuprate added in a 1,4-fashion to
cyclohexenonegave the cis-vinylstannane ketone adduct in 80%
yield.28 Subsequent studies by Marino and others have shown
that the reactivity of the (stannylvinyl)copper(I) intermediate
can be increased by starting with copper(I) cyanide.29 Following
this precedence, ethyl propiolate (12) was developed as a
substrate for a new tandem stannylcopper-acetylene-alkyne
ester addition reaction (Scheme 3).
Stannyl cuprate was first generated from hexabutylditin,
butyllithium, methyllithium, and copper(I) cyanide added
sequentially in THF at -78 °C.29 Excess acetylene was then
added directly to the cold solution followed by ethyl propiolate
(12). After quenching with methanol, diene ester 13 was
obtained in 82% yield, based on ethyl propiolate using 6 equiv
of the cuprate, with greater than 25:1 Z,E/Z,Z selectivity. Use
of only 1 equiv of cuprate gave 13 in only 17% yield, and 3
equiv improved the yield to 53%. The kinetic quench conditions
of Piers, with 12 added together with methanol to the cuprate
reagent, also gave 13 with high selectivity.30 As found
previously with the alkylcopper reagents, the stereochemistry
of intermediate copper-enoate adduct was maintained upon
protonation at low temperature.
To continue, DIBAL reduction of ester 13 gave the corre-
sponding allyl alcohol which was oxidized using TPAP and
NMO to give aldehyde 14 (Scheme 4). Treatment with the
Horner-Emmons reagent then gave ester 15 in 73% overall
yield. Only the (E,E,Z)-product shown was detected for this
step. For the hydrolysis of ester 15 to the corresponding acid,
several conditions were explored using sodium and lithium
hydroxide in various concentrations and combinations of water,
methanol, and THF. All gave extensive destannylation products
with yields ranging from only 25% to 35%. Finally it was found
(25) Stamos, D. P.; Kishi, Y. Tetrahedron Lett. 1996, 37, 8643-8647.
(26) Corey, E. J.; Katzenellenbogen, J. A. J. Am. Chem. Soc. 1969, 91,
1851-1853. Siddall, J. B.; Biskup, M.; Fried, J. H. J. Am. Chem. Soc. 1969,
91, 1853-1855. Anderson, R. J.; Corbin, V. L.; Cotterrell, G.; Cox, G. R.;
Henrick, C. A.; Schaub, F.; Siddall, J. B. J. Am. Chem. Soc. 1975, 97,
1197-1205. Piers, E.; Chong, J. M.; Morton, H. E. Tetrahedron Lett. 1981,
22, 4905-4908.
(20) Brown, H. C.; Bhat, K. S. J. Am. Chem. Soc. 1986, 108, 293.
(21) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
(27) Westmijze, H.; Ruitenbert, K.; Meijer, J.; Vermeer, P. Tetrahedron
Lett. 1982, 23, 2797-2800.
1973-1976.
(22) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem ReV.
(28) Marino, J. P.; Long, J. K. J. Am. Chem. Soc. 1988, 110, 7916.
(29) Marino, J. P.; Emonds, M. V. M. ; Stengel, P. J.; Oliveira, A. R.
M.; Simonelli, F.; Ferreira, J. T. B. Tetrahedron Lett. 1992, 33, 49-52.
Barbero, A.; Cuadrado, P.; Fleming, I.; Gonzales, A. M.; Pulido, F.; Rubio,
R. J. Chem. Soc., Perkin Trans. 1 1993, 1657-1663.
1994, 94, 2483.
(23) Takai, K.; Nitta, K.; Utimoto, K. J. Am. Chem. Soc. 1986, 108,
7408-7410.
(24) Evans, D. A.; Black, W. C. J. Am. Chem. Soc. 1993, 115, 4497-
4513.
(30) Piers, E.; Morton, H. E. J. Org. Chem. 1980, 45, 4264-4266.

12162 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
Andrus et al.
Scheme 4
Chart 5
Scheme 5
that slow addition of saturated aqueous lithium hydroxide to a
tert-butyl alcohol solution of ester 15 at 0 °C produced the
desired acid with no destannylation.24 The crude acid was then
coupled with (S)-alaninol using bromotris(pyrrolidino)phospho-
nium hexafluorophosphate (PyBroP) to give amide 16 in 59%
yield for the two combined steps.31 With routes to the right-
and left-hand fragments, the key coupling to form stipiamide
was undertaken.
material clearly revealed the presence of stipiamide, the all-
trans-isomer, and the 4-cis-isomer in a similar ratio. In addition,
the optical rotation of the synthetic material ([R]D -100°)
approximated the weighted average of the rotations reported
for the three natural isomers (-112°).
Vinyl iodide 11 was deprotected in 86% yield using tetrabu-
tylammonium fluoride (TBAF) to give 17 (Scheme ). In a
recent synthesis, a key Stille coupling between vinyltin and vinyl
iodide fragments of similar size and comparable functionality
to this system using Pd2(dba)3CHCl3 as catalyst was reported
where a number of problems with low yields and byproduct
formation were encountered.32 Further review of the literature
provided a number of additional cases where Pd(II) catalysts
worked well, prompting us to investigate the more active
(MeCN)2PdCl2 catalyst.33 It was found that at room temperature
in 15 min the reaction of 16 and 17, at 1:1.1 stoichiometry,
gave the desired product in 78% yield. NMR analysis of this
material showed a 4:2:1 mixture of (-)-stipiamide and the all-
trans- and the (4Z)-isomers, respectively. The coupling experi-
ment was repeated with the exclusion of light followed by rapid
filtration through neutral alumina. The same isomeric ratio was
observed. Due to the stereochemical fidelity of the Stille
reaction, it can be reasoned that the isomerization did not occur
during the brief palladium coupling conditions, but rather during
the workup and isolation steps. Indeed, a similar ratio of
isomers was also observed in a sample of authentic material.
In the case of the heptaene dimethylcrocetin visible light readily
converts the cis-form to the all-trans-form in the time needed
to perform the spectroscopy.34 The process is also reminiscent
of the conversion of all-trans-â-carotene to the mono-cis-
isomers.35 In similar fashion, the pentaene-amide functionality
of stipiamide is readily isomerized. Attempts were made to
separate the stipiamide isomers using reversed-phase HPLC
following the report of Ho¨fle.5 Although complete separation
was not achieved, direct comparison by 1H NMR to the authentic
Design and Synthesis of Simplified Variants. Due to the
structural similarity with the myxalamides, it was reasoned that
the toxicity of stipiamide was due to its ability to function as
an electron acceptor inhibiting mitochondrial NADH oxidation.7
In the design of DHS (2), the goal was to create a molecule of
similar geometry but with different electronic characteristics.
Thus, the conversion of the C-6,7 cis-alkene to a triple bond
served two purposes, a significant synthetic simplification
eliminating the isomerization problem and an electronic varia-
tion. The next goal was to further simplify 2 by interrupting
the extended polyene conjugation in an effort to probe the
connection between toxicity and electron transport inhibition.
This was accomplished by saturating the C-4,5 alkene and
removing the C-16 carbon with its attendant stereocenter along
with the C-10,11 ethylene methyl functionality to give com-
pound 3 (Chart 5).
The synthesis of DHS (2) was accomplished by using a
Sonogashira reaction with the same vinyl iodide 17 used for
the synthesis of stipiamide and a new terminal alkyne amide.
Alcohol 1836 was oxidized to the corresponding aldehyde which
was treated with the phosphonate ester anion to give 19 in 55%
yield for the two steps (Scheme 6).32 Stereospecific bromine-
tin exchange was brought about using bromine in methylene
chloride to provide the vinyl bromide in 92% yield.37
The terminal alkyne was installed by coupling with TMS-
acetylene to give 20 in 91% yield. Removal of the TMS group
was accomplished using potassium carbonate in ethanol. Treat-
ment with lithium hydroxide then provided 21 in 76% yield.
This yellow-orange acid was taken on, without purification, to
the amide using (S)-alaninol and PyBrop to give 22. The
purification of fragment 22 proved difficult due to stability
problems. Thus, it was carried on without delay to the next
reaction.
(31) Coste, J.; Ferot, E.; Jouin, P.; Castro, B. Tetrahedron Lett. 1991,
32, 1967.
(32) Evans, D. A.; Gage, J. R.; Leighton, J. L. J. Am. Chem. Soc. 1992,
114, 9434-9453.
(33) Nicolaou, K. C.; Piscopio, A. D.; Bertinato, P.; Chakraborty, T.
K.; Minowa, N.; Koide, K. Chem. Eur. J. 1995, 1, 318-321. Stille, J. K.;
Groh, B. L. J. Am. Chem. Soc. 1987, 109, 813-814.
(34) Zechmeister, L. Chem. ReV. 1943, 44, 267.
(35) Doering, W. v. E.; Sotiriou-Leventis, C. ; Roth, W. R. J. Am. Chem.
Sco. 1995, 117, 2747.
(36) Jung, M. E.; Light, L. A. Tetrahedron Lett. 1982, 23, 3851-3854.
(37) Nicolaou, K. C.; Veale, C. A.; Webber, S. E.; Katerinopoulos, H.
J. J. Am. Chem. Soc. 1985, 107, 7515.

Synthesis of Stipiamide and Designed Polyenes
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12163
Scheme 6
Scheme 9
to diene 7 in the synthesis of stipiamide, it was found the
protected diene 33 gave poor yields (35-45%) when reacted
with AD-mix-R. Interestingly, use of the opposite enantiomeric
antipode AD-mix-â gave the desired diol 34 in 86% yield with
no internal dihydroxylation.17 Diol 34 was then reacted with
sodium periodate, and the resulting crude aldehyde was treated
under the optimized Takai reaction conditions to give the vinyl
iodide in 57% overall yield for the two steps. TBAF depro-
tection provided the simplified left-hand fragment 35 in 89%
yield.
DHS (2) was formed by a Sonogashira coupling of vinyl
iodide 17 and alkyne 22 (Scheme 9).38 The optimized condi-
tions included (PPh3)2PdCl2 as catalyst with CuI as cocatalyst
with addition of diisopropylamine to an ethyl acetate solution
at -20 °C of the coupling partners at 1:1.7 stoichiometry to
give DHS (2) in 59% yield. Acetylenes 26, 28, and 30 were
coupled to vinyl iodide 35 to give the desired enyne products
3, 36, and 37, as shown in Chart 6. Selected conditions and
reagents explored are shown in Table 1.
Scheme 7
While there are many synthetic applications of the coupling
of a vinyl iodide with a terminal alkyne under palladium-copper
catalysis,39 there have been few systematic studies reported that
investigate each feature of the reaction. Initially (entries 1-3)
the widely used protocol was employed using catalytic Pd-
(PPh3)4 with stoichiometric n-butylamine in benzene.40 The use
of 1 equiv of acetylene 22 gave a low yield of 15% for 2. Even
with 3 equiv of alkyne, 2 was obtained in only 50% yield.
Numerous variations were performed to determine the effect
of the hydroxyl and amide functionalities on the yield (not
shown). These findings suggested that an alkynyl amide is
inherently a poorer coupling partner compared to an analogous
alkynyl ester, giving yields 15-20% lower. The few reported
couplings with amide substrates also occur with low yields in
parallel with our findings.41 Other attempts to remedy the low
yield problem that also proved unsuccessful included slow
addition, changing the mole percent of the copper cocatalyst,
and colder initial temperatures.42 Only with 22 added rapidly
was 2 obtained in 70% yield (entry 3) using this catalyst system.
Alternative forms of palladium were explored in an effort to
increase the yield closer to the ideal 1:1 substrate stoichiometry.
With enyne 3, Pd(PPh3)4 gave yields of 10-12% (not shown)
whereas use of PdCl2(MeCN)2 significantly improved the yield
to 31% (entry 4). The optimal results were finally achieved
using PdCl2(PPh3)2 as catalyst (entries 6-9). Indeed, in the
case of DHS (2) PdCl2(PPh3)2 (entry 5) gave a much improved
59% yield. Most importantly, in this example only 1.7 equiv
of 22 was used.
Scheme 8
The terminal alkynes used to form the simplified stipiamide
variants were prepared as shown in Scheme 7. 4-Butyn-1-ol
(23) was oxidized with PCC, providing the volatile aldehyde
which was carefully concentrated and treated with (carbethoxy-
ethylidene)triphenylphosphorane to give ester 24 in 54% yield.
Hydrolysis of ester 24 with lithium hydroxide gave acid 25
which was coupled with (S)-alaninol in 92% overall yield.
Similar amidation reactions were used with 27 and 29 to access
the corresponding amides 28 and 30.
The synthesis of a simplified left-hand fragment began with
3-phenylpropanal (31) olefinated using 2-(triphenylphosphora-
nylidene) propionaldehyde in 69% yield (Scheme 8). The
resultant enal 32 was treated with diisopinocampheyl (E)-
crotylborane derived from (-)-R-pinene to give anti-homoallylic
alcohol 33 in 80% yield followed by TBS protection. In contrast
Following the lead of Johnson, further improvement could
be realized by starting the reaction at -20 °C with warming to
(38) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975,
4467-4470.
(39) Rossi, R.; Carpita, A.; Bellina, F. Org. Prep. Proc. 1995, 27, 129.
Sonogashira, K. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming, I., Eds., Pergamon Press: Oxford, 1991; Vol. 3, pp 521-549.
(40) Nicolaou, K. C.; Ladduwahetty, T.; Taffer, I. M.; Zipkin, R. E.
Synthesis 1986, 344. Magnus, P.; Carter, P. A. J. Am. Chem. Soc. 1988,
110, 6921. Nicolaou, K. C.; Hwang, C. -K.; Smith, A. L.; Wendeborn, S.
V. J. Am. Chem. Soc. 1990, 112, 7416. Porco, J. A. Jr.; Schoene, F. J.;
Stout, T. J.; Clardy, J.; Schreiber, S. L. J. Am. Chem. Soc. 1990, 112, 7410.
(41) Rossi, R.; Carpita, A.; Ciofalo, M.; Lippolis, V. Tetrahedron 1991,
47, 8443.
(42) Reddy, R. S.; Iguchi, S.; Kobayashi, S.; Hirama, M. Tetrahedron
Lett. 1996, 37, 9335.

12164 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
Andrus et al.
Chart 6
Table 1. Sonogashira Coupling for the Formation of Enynes 2, 3, 36, and 37
entry vinyl iodide alkyne equiv of alkyne product palladium catalyst (%)a CuI (%)a solvent temp (°C)
baseb
% yield
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
17
17
17
35
17
35
35
35
35
35
35
35
35
35
35
22
22
22
26
22
26
26
26
26
28
28
28
30
30
30
1
3
10
2
2
2
3
2
3
3
3
3
36
36
36
37
37
37
Pd(PPh3)4 (10)
Pd(PPh3)4 (10)
Pd(PPh3)4 (10)
5
5
5
benzene rt
benzene rt
benzene rt
n-BuNH2
n-BuNH2
n-BuNH2
Et3N
i-Pr2NH
Et3N
i-Pr2NH
i-Pr2NH
i-Pr2NH
Et2NH
i-Pr2NH
i-Pr2NH
Et2NH
15
50
70
31
59
54
61
72
87
61c
70
85
31
54
95
1.5
(MeCN)2PdCl2 (5)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (10)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (8)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (6)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (5)
(PPh3)2PdCl2 (5)
10
35
10
7
30
10
10
15
32
10
33
18
DMF
rt
1.7
1.5
1.3
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
-20 to rt
rt
0
0
-20 to rt
rt
0
-20 to rt
rt
0 to rt
-20 to rt
i-Pr2NH
i-Pr2NH
a mol %. b The reactions were 0.2 M in base which was added last. c The yield was obtained with Et2NH added last. Slow addition of alkyne 28
gave 36 in 30% yield.
Table 2. MDR Reversal Activity of Stipiamide with
room temperature.43 Ethyl acetate was used for these reactions
Chemotherapeutic Drugs Using MDR-7adrR (Adriamycin Resistant)
Breast Cancer Cells
since the polar alkynyl amides were found to be poorly solvated
by benzene at lower temperatures. The palladium(II) chloride
catalyst again proved optimal in the synthesis of the simplified
variants, giving excellent yields of 85-95% (entries 9, 12, and
15). Diminished yields of 10-15% were obtained using the
less active catalyst Pd(PPh3)4 (not shown). Temperature clearly
plays an important role in the formation of the enynes especially
in the case of 37. This coupling reaction when performed at 0
°C gave only a 54% yield (entry 14), but when started at -20
°C (entry 15) with warming to room temperature, the product
was obtained in 95% yield. The fact that coupling occurs with
high efficiency is significant in this case in that others have
recently reported that alkynes conjugated to electron-withdraw-
ing groups react with very poor yields.44
ED50 of anticancer drug
in MCF-7adrR cell
+ stipiamidea (nM)
anticancer ED50 of anticancer drug
entry
drug
in wild type cell (nM)
1
2
3
4
adriamycin
vinblastine
vincristine
colchicine
0.5
2500
12
400
3
4.4 × 10-6
0.01
0.02
a Stipiamide was present at 50% of its ED50 (4 × 10-12 M).
also found to be incredibly toxic (ED50, 0.01 nM). This activity
can be attributed to inhibition of mitochondrial ATP synthesis
as exhibited by other members of the polyene class.6c
In contrast, the same assay found that the ED50 of DHS (2)
was 4.4 µM, 10 000 times less toxic than stipiamide. Remark-
ably, the simple modification of the C-6,7 (Z)-alkene to alkyne
greatly lowers the toxicity displayed in the normal, wild type
cells. In order to then determine the concentration for MDR
reversal, the standard protocol developed by Chang was used
with 2.46 MCF-7adrR cells were treated with 2 followed by
the addition of adriamycin at its ED50 concentration for the wild
type cells (4 nM). The percentage of cells remaining viable at
a given concentration of 2 was then determined for each assay
performed in quadruplicate at each concentration. The assays
were performed using consistent concentrations both above and
below the ED50 of the compound, allowing for comparisons of
2 and the other simplified variants. The average of the percent
viable cells for the assays was then plotted as a function of
DHS (2) concentration (Figure 1).
Biological Evaluation. Seto found that stipiamide, in the
MDR assay with colchicine resistant KB cells, was able to
restore cytotoxicity to within a factor of six to the wild type
response.2 However, with other anticancer agents, MDR
reversal was not observed using stipiamide. Now with synthetic
stipiamide, MDR reversal using human breast cancer cells
(MCF-7adrR) resistant to multiple anticancer compounds was
investigated.45 Again with several added agents, exclusive MDR
reversal activity for colchicine was observed (Table 2). Only
in the case of colchicine (entry 5) was stipiamide shown to
reduce resistance to within a factor of 100 of the wild type
response, whereas in all other cases the difference between
transformed and wild type response was over 5 orders of
magnitude. Besides the exclusive MDR effect, stipiamide was
(43) Johnson, C. R.; Miller, M. W. J. Org. Chem. 1997, 62, 1582.
(44) Yamanaka, H.; Sakamotot, T. Synthesis 1992, 746.
(45) Cowan, K. H.; Baptist, G.; Tulpule, A.; Biranda, K. S.; Myers, C.
E. Proc. Natl. Acad. Sci. U.S.A. 1986, 83, 9328.
(46) Hall, A. M.; Croy, V.; Chan, T.; Ruff, D.; Kuczek, T.; Chang, C.
J. Nat. Prod. 1996, 59, 35.

Synthesis of Stipiamide and Designed Polyenes
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12165
conjugation impairs the compound’s ability to function as a one-
electron acceptor, leading to lower toxicity. Continuing the
trend, 36 (entry 3) further lacking the R,â-unsaturation of 3 is
the least toxic of the simplified variants. Again the loss of
conjugation appears to play a role in lowering the toxicity.
However, the lack of rigidity and reduced size relative to 3
diminished the ability of 36 to reverse MDR. These findings
together with the optimized coupling conditions will allow for
the design and synthesis of new compounds that will be used
to further define the structural requirements for effective Pgp
transport inhibition and improved MDR reversal.
Figure 1. Fraction of viable MCF-7adrR cells versus the concentration
of DHS with adriamycin present at 4 nM.
Experimental Section
General Information. Air-sensitive reactions were performed under
a positive pressure of argon unless otherwise indicated. Air- and
moisture-sensitive reagents were introduced Via syringe or cannula
through rubber septa. All solvents were distilled prior to use from an
appropriate drying agent. Methylene chloride was distilled from CaH2.
THF and diethyl ether were distilled from sodium-benzophenone ketyl.
Reagents were purchased and used without further purification.
Purification by flash column chromatography was carried out in the
indicated solvent system using 70-230 mesh silica gel. Purification
by radial chromatography was performed using 1, 2, and 4 mm plates
loaded with 230-400 mesh PF-254 gypsum-bound silica. Chro-
mophore-containing compounds were visualized during the radial
chromatography using 254 nm UV light. TLC analysis was conducted
on silica gel F254, 0.25 mm precoated glass plates. The developed plates
were visualized by UV light (254 nm) and/or by dipping into staining
Figure 2. Fraction of viable MCF-7adrR cells versus the concentration
of enynes 3, 36, and 37 with adriamycin present at 4 nM.
1
solutions followed by charring. All H NMR spectra were obtained
using either a 500 or 200 MHz spectrometer employing chloroform
(7.26 ppm) as an internal reference. Signals are reported as m
(multiplet), s (singlet), d (doublet), t (triplet), q (quartet), etc.; coupling
constants (J) are reported in hertz (Hz). Carbon spectra were obtained
at 50 MHz. Infrared spectra were obtained using an FTIR spectrometer.
Only several characteristic absorbencies are reported (in units of
wavenumbers, cm-1). Mass spectra were run by the Purdue campus-
wide mass spectrometry facility. Low-resolution electron impact (EI)
or chemical ionization (CI) spectra were obtained using a mass
spectrometer. Prominent mass fragments are reported along with their
relative intensities (in parentheses). Optical rotations were obtained
using a polarimeter at room temperature (rt) employing the sodium D
line. In all cases, the length of the rotation cell was 10 cm (1 dm),
and solvents employed were of spectroscopic grade. Optical rotation
concentrations are given in grams per milliliter. Microanalyses were
performed by the Purdue Chemistry Department Microanalytical
Laboratory. Finally, bicinchoninic acid MDR assays were performed
by the Purdue Cancer Research Center.
Preparation of Ethyl (2E,4S)-2,4-Dimethyl-6-phenyl-2-hexenoate
(5). To a stirring solution of (4S)-3-[(2S)-2-methyl-4-phenyl-1-
oxobutyl]-4-(phenylmethyl)-2-oxazolidinone (10.41 g, 30.9 mmol) in
ether (0.15 M, 200 mL) at 0 °C was added lithium aluminum hydride
(1 M, 46.3 mL, 46.3 mmol) as a THF solution via syringe pump over
1 h. The mixture was allowed to stir for 2 h and was quenched by the
slow addition of 9:1 THF/water (20 mL) followed by aqueous NaOH
(2 N, 3 mL).
The THF and water layers were separated, and the aqueous layer
was extracted with methlyene chloride (2 × 100 mL). The organic
phases were combined, dried over MgSO4, and distilled down to a crude
oil which was passed through a plug of silica gel (7 × 4 cm) and
distilled down to give 4.3 g of essentially pure alcohol (by TLC).
To a stirring solution of the crude alcohol (4.3 g, 26 mmol) in 50
mL of methylene chloride were added oven-dried, crushed 3 Å
molecular sieves (13 g), NMO (4.6 g, 39 mmol), and TPAP (350 mg,
1 mmol). The solution was stirred for 15 min and passed through a 7
cm diameter course fritted funnel containing sand (0.5 cm) on top, filter
paper in between, and silica gel (3 cm) on the bottom. The material
was rinsed through with ether. The filtered crude was concentrated,
giving 4.18 g of crude aldehyde (one spot by TLC) which was dissolved
in 125 mL of toluene (0.2 M). To this was added (carbethoxyeth-
ylidene)triphenylphosphorane (14 g, 38 mmol), and the solution was
Table 3. Comparison of the ED50 and TIs for Enynes 2, 3, 36, and
37 in MDR-7adrR Adriamycin Resistant Breast Cancer Cells
entry
compd
ED50 (µM)
reversal concn (µM)
TIa
1
2
3
4
2
3
36
37
4.4
14
18
5
6.5
4
14
4
1.48
0.29
0.78
0.8
a TI ) concentration to completely reverse MDR/the ED50.
From the graph it was determined that the concentration for
MDR reversal by 2 required for adriamycin to kill half of the
resistant cells was 6.5 µM. This value is slightly greater than
its own ED50 suggesting that 2, at the MDR reversal concentra-
tion, may not be acting exclusively as a reversal agent but also
may be functioning as a cytotoxin. More importantly 2, in
contrast to stipiamide, now shows MDR reversal activity with
a broad range of cancer agents, not being limited to colchicine.
MDR reversal analysis was then performed with the simpli-
fied enynes 3, 36, and 37. It was found that MDR reversal
now occurred at concentrations lower than the toxicity ED50
(Figure 2). The therapeutic index (TI) was then used as a helpful
convention to rank the success of the new reversal compounds,
being defined as the ratio of the reversal concentration to the
toxicity ED50 as shown in Table 3.
By this definition, smaller TI values (<1) correspond to better
reversal agents. This range indicates that the compounds are
able to reverse MDR at concentrations below their toxicity level
(ED50). DHS (2), with a TI of 1.48 (entry 1), demonstrates
very good MDR reversal when directly compared to verapamil
(TI > 5), the clinical standard. Clearly the lower toxicity
generated by the change to the 6,7-acetylene is a step in the
right direction. More remarkable is the low TI of 0.29 observed
for 3 (entry 2). Although enynes 2, 3, and 37 have similar MDR
reversal concentrations, 3 is less toxic than the other two (14
vs 4.4 and 5 µM). This finding for 3 supports the reasoning of
the design criteria where the interruption of the polyene

12166 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
Andrus et al.
warmed to 90 °C for 15 h. After cooling, the toluene was removed,
and the oil was dissolved in hexanes, causing triphenylphosphine oxide
and excess ylide to precipitate. This suspension was passed through a
plug of silica gel (7 × 3 cm) and divided into three equal portions (1H
NMR showed E:Z ) 8:1). Each portion was radial chromatographed
on a 4 mm plate with 5% EtAOc/Hex as the eluent, giving all together
4.2 g (55%, three steps) of the pure E isomer. Close eluting higher
fractions (the (Z)-isomer) were discarded along with some overlapping
fractions: Rf ) 0.56 in 35% EtOAc/Hex; 1H NMR (200 MHz, CDCl3)
δ 7.36-7.12 (m, 5H), 6.62 (d, J ) 11.6 Hz, 1H), 4.22 (q, J ) 7.0 Hz,
2H), 2.73-2.46 (m, 3H), 1.82 (s, 3H), 1.82-1.58 (m, 2H), 1.32 (t, J
) 7.0 Hz, 3H), 1.05 (d, J ) 7.4 Hz, 3H); 13C NMR (50 MHz, CDCl3)
δ 13.1, 14.8, 20.5, 22.2, 33.3, 34.2, 39.0, 61.0, 126.3, 127.4, 128.8,
142.6, 147.9, 168.9; IR (film) 1712, 1256, 1086. Anal. Calcd for
C16H22O2: C,78.01; H, 9.00. Found: C, 77.76; H, 9.29.
Preparation of (3S,4R,5E,7S)-4-Hydroxy-9-phenyl-3,5,7-trimeth-
ylnona-1,5-diene (7). To a stirring solution of ethyl (2E,4S)-2,4-
dimethyl-6-phenyl-2-hexenoate (5) (6.0 g, 24.4 mmol) in THF (0.2 M,
120 mL) at -40 °C was added DIBAL (1.5 M, 49 mL, 73.1 mmol)
rapidly as a THF solution. The mixture was allowed to stir for 1 h
and was quenched by the slow addition of methanol (2 mL) followed
by saturated aqueous Rochelle salts (15 mL). The solution was allowed
to stir until the aqueous and organic layers separated into distinct phases
(approximately 3 h). The THF and water layers were separated, and
the aqueous layer was extracted with methlyene chloride (2 × 50 mL).
The organic phases were combined, dried over MgSO4, and distilled
down to a crude oil which was passed through a plug of silica gel (7
× 4 cm) and distilled down to give 4.38 g of essentially pure alcohol
(by TLC).
at rt. The reaction was stirred overnight, diluted with 50 mL of
methylene chloride, and extracted with water (50 mL). The separated
water layer was washed with methylene chloride (2 × 50 mL).
Combined extracts were dried over MgSO4. The solution was
concentrated and flashed chromatographed through silica gel (4 × 10
cm) eluting with 250 mL of 1%, 2%, 5% EtOAc/hexanes. Product-
containing fractions were collected and concentrated to give 4.64 g
(95%) of a clear oil as an inseparable mixture of diastereomers (∼7:1
determined by 1H and 13C NMR): [R]D + 2.72° (c ) 0.054, CH2Cl2);
1
Rf ) 0.68 in 5% EtOAc/Hex; H NMR (200 MHz, CDCl3) δ 7.33-
7.13 (m, 5H), 5.97-5.77 (m, 1H), 5.16-4.93 (m, 3H), 3.68 (d, J ) 8
Hz, 1H), 2.72-2.21 (m, 4H), 1.73-1.47 (m, 5H), 0.98-0.83 (m, 15H),
0.04 (s, 3H), 0.02 (s, 3H); 13C NMR (50 MHz, CDCl3) δ 143.4, 142.7,
135.8, 133.6, 128.8, 128.7, 126.0, 114.2, 83.3, 42.7, 40.0, 34.5, 32.3,
26.4, 21.6, 18.8, 17.3, 12.3, -3.8, -4.4. Anal. Calcd for C24H40OSi:
C, 77.35; H, 10.82. Found: C, 77.19; H, 10.86.
Preparation of Ethyl (2E,4S,5R,6E,8S)-5-[(tert-Butylsilyl)oxy]-
10-phenyl-2,4,6,8-tetramethyldodeca-2,6-dienoate (9). To a stirring
solution of (3S,4R,5E,7S)-4-[(dimethyl-tert-butylsilyl)oxy]-9-phenyl-
3,5,7-trimethylnona-1,5-diene (1.25 g, 3.35 mmol) in a mixture of tert-
butyl alcohol (15 mL) and water (15 mL) was added AD-mix-R
(Aldrich, 5 g, 1.5 g/mmol) at rt. The reaction was allowed to proceed
for 30 h, giving exclusively terminal alkene dihydroxylation (Rf ) 0.41
in 35% EtOAc/Hex) and a faint starting material spot. The reaction
mixture was then passed through an 8 cm diameter course fritted funnel
containing sand (0.5 cm) on top, filter paper in between, and silica gel
(4 cm) on the bottom. The material was rinsed through with ether.
The filtered crude was concentrated, giving 1.40 g of crude diol which
was dissolved in a mixture of THF (10 mL) and water (10 mL). To
this solution was added sodium periodate (1.1 g, 5.2 mmol), and the
mixture was allowed to stir for 30 min, giving aldehyde (Rf ) 0.38 in
10% EtOAc/Hex). The reaction mixture was then passed through an
8 cm diameter course fritted funnel containing sand (0.5 cm) on top,
filter paper in between, and silica gel (4 cm) on the bottom. The
material was rinsed through with ether and concentrated, giving 1.13
g of crude aldehyde (one spot by TLC) which was dissolved in 30 mL
of toluene (0.1 M). To this was added (carbethoxyethylidene)-
triphenylphosphorane (2.1 g, 6.0 mmol), and the solution was warmed
to 90 °C for 12 h. After cooling, the toluene was removed, and the oil
was dissolved in hexanes, causing triphenylphosphine oxide and excess
ylide to precipitate. This suspension was passed through a plug of
silica gel (7 × 3 cm) rinsing with 1:1 EtAOc/hexanes. The solution
was concentrated and radial chromatographed on a 4 mm plate using
10% EtAOc/hexanes to give 0.96 g (62% from terminal alkene) of
product (a clear oil) as a mixture of diastereomers (∼7:1 by NMR).
Close eluting higher fractions (the (Z)-isomer) was discarded along with
some overlapping fractions: [R]D + 6.08° (c ) 0.36, CH2Cl2); Rf )
To a stirring solution of the crude alcohol (4.38 g, 21.4 mmol) in
45 mL of methylene chloride were added oven-dried, crushed 3 Å
molecular sieves (10 g), NMO (3.8 g, 32 mmol), and TPAP (300 mg,
0.8 mmol). The solution was stirred for 15 min and passed through
an 8 cm diameter course fritted funnel containing sand (0.5 cm) on
top, filter paper in between, and silica gel (3 cm) on the bottom. The
material was rinsed through with ether. The filtered crude was
concentrated, giving 3.32 g of crude aldehyde (one spot by TLC).
To a stirring -78 °C solution of potassium tert-butoxide (3.13 g,
27.9 mmol) in 15 mL of THF was added via cannula trans-2-butene
(6 mL, ∼54 mmol, condensed into a graduated column at -78 °C).
To the solution was added via syringe n-butyllithium (2.5 M, 11.2 mL,
27.9 mmol) as a hexane solution. The mixture was allowed to stir for
l
15 min, and Ipc2BOMe (diisopinocampheylmethoxyborane derived
from (-)-R-pinene) was slowly added by syringe (2 M, 16.4 mL, 32.8
mmol) as a solution in ethyl ether. The solution was stirred for 30
min, and boron trifluoride etherate was slowly added (4.64 mL, 37.7
mmol) by syringe. The mixture was stirred for 5 min followed by the
addition of the above formed crude aldehyde as an ether solution (5
mL) with two 1 mL ether washings. After 15 h at -78 °C, 10 mL of
saturated NaOAc was added followed by 10 mL of 30% H2O2. The
biphasic mixture was then allowed to warm to rt and stir for 8 h. The
mixture was diluted with ether (10 mL) and washed with brine. The
separated aqueous layer was then washed twice with methylene chloride
(2 × 25 mL), and the combined organic layers were dried over MgSO4.
The solution was concentrated and flashed chromatographed through
silica gel (4 × 10 cm) eluting with 250 mL of 7%, 9%, 11%, and 15%
EtOAc/hexanes. The product-containing fractions were concentrated
to give 3.71 g (60% from the ester) of product as clear oil. 1H NMR
showed the product to be an approximately 7:1 inseparable mixture of
1
0.56 in 35% EtOAc/Hex; H NMR (500 MHz, CDCl3) δ 7.30-7.24
(m, 2H), 7.21-7.13 (m, 3H), 6.64 (d, J ) 11.6 Hz, 1H), 5.35 (d, J )
11.1 Hz, 1H), 4.22-4.12 (m, 2H), 3.76 (d, J ) 8.9 Hz, 1H), 2.70-
2.47 (m, 3H), 2.46-2.36 (m, 1H), 1.85 (s, 3H), 1.68-1.60 (m, 1H),
1.6-1.52 (m, 1H), 1.57 (s, 3H), 1.26 (t, J ) 8.3 Hz, 3H), 0.95 (d, J )
7.4 Hz, 3H), 0.87 (s, 3H), 0.84 (s, 9H), 0.01 (s, 3H), -0.01 (s, 3H);
13C NMR (50 MHz, CDCl3) δ 168.8, 146.8, 143.3, 135.8, 135.4, 128.6,
127.8, 126.1, 83.4, 60.7, 39.9, 38.7, 38.6, 34.4, 32.3, 26.3, 21.7, 18.6,
14.8, 13.2, 12.0; IR (film) 1712, 1184, 1086. Anal. Calcd for
C28H46O3: C,73.31; H, 10.11. Found: C, 73.06; H, 10.39.
Preparation of (1E,3E,4S,5R,7E,8S)-5-[(tert-Butylsilyl)oxy]-1-
iodo-10-phenyl-3,5,7,9-tetramethylundeca-1,3,7-triene (11). To a
stirring solution of ethyl (2E,4S,5R,6E,8S)-5-[(tert-butylsilyl)oxy]-10-
phenyl-2,4,6,8-tetramethyldodeca-2,6-dienoate (9) (2.11 g, 4.6 mmol)
in THF (0.2 M, 25 mL) at -40 °C was added DIBAL (1.5 M, 9.2 mL,
13.8 mmol) rapidly as a THF solution. The mixture was allowed to
stir for 1 h and was quenched by the slow addition of methanol (1
mL) followed by saturated aqueous Rochelle salts (5 mL). The solution
was allowed to stir until aqueous and organic layers separated into
distinct phases (approximately 3 h). The THF and water layers were
separated, and the aqueous layer was extracted with methlyene chloride
(2 × 15 mL). The organic phases were combined, dried over MgSO4,
and distilled down to a crude oil which was passed through a plug of
silica gel (8 × 4 cm) and distilled down to give 1.56 g of alcohol
(81%).
1
diastereomers: Rf ) 0.43 in 20% EtOAc/Hex; H NMR (200 MHz,
CDCl3) δ 7.33-7.12 (m, 5H), 5.87-5.66 (m, 1H), 5.26-5.09 (m, 3H),
3.65 (d, J ) 8 Hz, 1H), 2.66-2.22 (m, 4H), 1.73-1.51 (m, 6H), 0.96
(d, J ) 7.2 Hz, 3H), 0.89 (d, J ) 7.2 Hz, 3H); 13C NMR (50 MHz,
CDCl3) δ 143.2, 141.8, 135.4, 134.3, 128.9, 128.7, 126.1, 116.9, 81.8,
42.7, 39.7, 34.4, 32.3, 21.5, 17.4, 11.8; MS (CI) m/z 241 (1.0, M + H
(-H2O)). Anal. Calcd for C18H26O: C, 83.67; H, 10.14. Found: C,
83.35; H, 10.08.
Preparation of (3S,4R,5E,7S)-4-[(Dimethyl-tert-butylsilyly)oxy]-
9-phenyl-3,5,7-trimethylnona-1,5-diene. To a stirring solution of
(3S,5E,4R,7S)-4-hydroxy-9-phenyl-3,5,7-trimethylnona-1,5-diene (7)
(3.37 g, 13 mmol) in DMF (0.6 M, 27 mL) were added imidazole (2.6
g, 39.1 mmol) and tert-butyldimethylsilyl chloride (3.92 g, 26.1 mmol)

Synthesis of Stipiamide and Designed Polyenes
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12167
This was divided into halves and radial chromatographed on a 4
mm plate using gradient elutions (5%, 7%, and 10% EtOAc/hexanes).
The overlapping fractions from the two separations were combined and
rechromatographed. In this manner, the minor isomers from the
Wittig-crotylboration-Wittig reactions were eliminated to give 1.16
g of pure alcohol (61%, 1H NMR). To a stirring solution of the alcohol
(338 mg, 0.94 mmol) in 2 mL of methylene chloride were added oven-
dried, crushed 3 Å molecular sieves (470 mg), NMO (166 mg, 1.41
mmol) and TPAP (33 mg, 0.094 mmol). The solution was stirred for
10 min and passed through a 4 cm diameter course fritted funnel
containing sand (0.5 cm) on top, filter paper in between, and silica gel
(2.5 cm) on the bottom. The material was rinsed through with ether.
The filtered crude was concentrated, giving 338 mg of crude aldehyde
(87%, one spot by TLC).
IR (film) 2926, 1718, 1154. Anal. Calcd for C19H36O2Sn; C, 54.96;
H, 8.74. Found: C, 54.24; H, 9.00.
Preparation of Ethyl (2E,4E,6Z)-2-Methyl-7-(tri-n-butylstannyl)-
hepta-2,4,6-trienoate (15). To a stirring solution of ethyl (2E,4Z)-5-
(tri-n-butylstannyl)penta-2,4-dienoate (13) (1.09 g, 2.6 mmol, (ZE:ZZ
≈ 4:1) in THF (0.1 M, 25 mL) at -40 °C was added DIBAL (1.5 M,
5.3 mL, 7.9 mmol) rapidly as a THF solution. The mixture was allowed
to stir for 1 h and was quenched by the slow addition of methanol (1
mL) followed by saturated aqueous Rochelle salts (3 mL). The solution
was allowed to stir until aqueous and organic layers separated into
distinct phases (approximately 3 h). The THF and water layers were
separated, and the aqueous layer was extracted with methlyene chloride
(2 × 15 mL). The organic phases were combined, dried over MgSO4,
and distilled down to a crude oil which was passed through a plug of
silica gel (8 × 4 cm) and distilled down to give the alcohol. This
alcohol was radial chromatographed (4 mm plate) using 5% EtOAc/
hexanes, allowing the separation of the minor diastereomer to give 580
mg of pure (ZE)-product (60% or 79% if only the (ZE)-isomer of the
starting material is counted). The overlapping fractions were discarded.
To a stirring solution of alcohol (792 mg, 2.12 mmol) in 5 mL of
methylene chloride were added oven-dried, crushed 3 Å molecular
sieves (1 g), NMO (375 mg, 3.19 mmol), and TPAP (30 mg, 0.08
mmol). The solution was stirred for 15 min and passed through a 5
cm diameter course fritted funnel containing sand (0.5 cm) on top, filter
paper in between, and silica gel (3 cm) on the bottom. The material
was rinsed through with ether. The filtered crude was concentrated,
giving 720 mg of crude aldehyde (one spot by TLC). Butyllithium
(0.85 mL, 2.13 mmol) was added to a solution of ethyl 2-(diethylphos-
phoryl)propionate (0.48 mL, 2.23 mmol) in THF (11 mL, 0.2 M) at 0
°C. To this was added the crude aldehyde, and the mixture was allowed
to warm to rt. After 5 min, the reaction mixture was filtered through
a plug of silica gel (4 × 3 cm) using ether. The filtrate was then
concentrated and radial chromatographed (4 mm plate) using 5%
EtOAc/hexanes to give 867 mg (71% from ester) of product (a clear
oil) as one diastereomer (S/N > 20:1 in 13C NMR): Rf ) 0.4 in 10%
EtOAc/Hex; 1H NMR (200 MHz, CDCl3) δ 7.28-7.12 (m, 2H), 6.60-
6.38 (m, 3H), 4.24 (q, J ) 8.4 Hz, 2H), 1.97 (s, 3H), 1.61-1.23 (m,
18H), 0.97 (t, J ) 8.4 Hz, 3H), 0.89 (t, J ) 7.8 Hz, 9H); 13C NMR (50
MHz, CDCl3) δ 11.1, 13.2, 14.2, 14.8, 27.7, 29.6, 61.1, 76.9, 77.5,
78.1, 128.0, 129.6, 138.4, 141.3, 142.1, 146.5, 168.9; IR (film) 2958,
1706, 1102. Anal. Calcd for C12H40O2Sn; C, 58.04; H, 8.86. Found:
C, 58.15; H, 9.08.
CrCl2 (220 mg, 1.79 mmol) was added to a 25 mL round bottom
flask and gently flame dried under high vacuum. Upon cooling, the
flask was released under argon and charged with 2 mL of THF. The
slurry was cooled to 0 °C and allowed to stir for 10 min. To another
flask containing aldehyde (119 mg, 0.29 mmol) was added 1 mL of
THF followed by iodoform (235 mg, 0.60 mmol). These were then
added dropwise by syringe to the CrCl2 slurry which turned a brown-
red color. The reaction was allowed to proceed for 3 h at 0 °C and
then filtered through a plug of silica gel (4 × 3 cm) using ether. The
filtrate was then concentrated, dissolved in 1 mL of ether, and treated
with 1 mL of TBAF (1 M in THF) with stirring to remove the excess
iodoform. This solution was allowed to stir for 10 min and filtered
through a plug of silica gel (4 × 3 cm) using ether. The filtered solution
was concentrated and radial chromatographed on a 1 mm plate using
2% EtOAc/hexanes to give 109 mg (70%) of product (a clear oil) as a
mixture of diastereomers (E/Z, 20:1, by 500 MHz 1H NMR). The
overall yield for the three steps was 37% (50% including the minor
isomers): [R]D + 7.68° (c 0.24, CH2Cl2); Rf ) 0.5 in 0.5% EtOAc/
1
Hex; H NMR (500 MHz, CDCl3) δ 7.32-7.26 (m, 2H), 7.21-7.17
(m, 3H), 7.06 (d, J ) 15.1 Hz, 1H), 6.09 (d, J ) 15.1 Hz, 1H), 5.36
(d, J ) 10.1 Hz, 1H), 5.14 (d, J ) 10.1 Hz, 1H), 3.72 (d, J ) 7.8 Hz,
1H), 2.68-2.56 (m, 2H), 2.54-2.46 (m, 1H), 2.46-2.36 (m, 1H), 1.71
(s, 3H), 1.69-1.59 (m, 1H), 1.59-1.48 (m, 1H), 1.56 (s, 3H), 0.95 (d,
J ) 7.2 Hz, 3H), 0.86 (s, 9H), 0.84 (d, J ) 0.72 Hz, 3H), -0.01 (s,
6H); 13C NMR (50 MHz, CDCl3) δ 150.5, 143.4, 139.0, 135.7, 134.4,
133.7, 128.84, 128.78, 126.1, 83.0, 73.2, 40.0, 37.9, 34.4, 32.3, 26.3,
21.7, 18.6, 18.2, 12.8, 12.4, -3.9, -4.4; HRMS (CI) calcd for C27H43
IOSi 539.2206, found 539.2222.
Preparation of N-[(2E,4E,6Z)-2-Methyl-7-(tri-n-butylstannyl)-
hepta-2,4,6-trienoyl]-(S)-alaninol (16). To a stirring solution of ethyl
(2E,4E,6Z)-2-methyl-7-(tri-n-butylstannyl)hepta-2,4,6-trienoate (15) in
tert-butyl alcohol (11 mL, 0.03 M) was added aqueous LiOH (2.3 mL,
2 M). After 2 d, the mixture was filtered through a plug of silica gel
(4 × 3 cm) with EtOAc and concentrated to give 137 mg of the crude
acid. This acid was then dissolved in methylene chloride (3 mL, 0.1
M) and cooled to 0 °C. To this solution were added diisopropylethy-
lamine (0.17 mL, 0.97 mmol), (S)-alaninol (72 mg, 0.97 mmol), and
PyBrOP (180 mg, 0.39 mmol). After 12 h, the reaction was filtered
through a plug of silica gel (4 × 3 cm) using EtOAc and radial
chromatographed on a 1 mm plate to give 94 mg (59% from the ester)
of a yellow oil. This material was quite unstable, forming a yellow
solid even after -20 °C storage for one week. Hence, the material
had to be synthesized just prior to subsequent use: Rf ) 0.25 in 50%
Preparation of Ethyl (2E,4Z)-5-(Tri-n-butylstannyl)penta-2,4-
dienoate (13). To a solution of hexabutylditin (7.6 mL, 15 mmol) in
50 mL of THF at -20 °C was added n-butyllithium (6 mL, 15 mmol,
2.5 M) via syringe. After 15 min, the reaction was further cooled to
-78 °C, and methyllithium was added (10.7 mL, 15 mmol, 1.4 M)
and allowed to stir for an additional 10 min. Copper(I) cyanide was
then added (1.34 g, 15 mmol), giving an orange-red solution. After
the solution was stirred for an additional 30 min, a balloon of acetylene
(∼1 L) was placed upon the flask. The flask was placed under a mild
vacuum (by means of a three-way valve in the balloon apparatus), and
the contents of the flask were exposed to the acetylene gas. Over the
course of 30 min, the reaction mixture turned to a dark green color.
After 30 min at -78 °C, the balloon was removed and ethyl propiolate
(0.25 mL, 2.5 mmol) and methanol (0.2 mL, 5 mmol) were added as
a THF solution (1 mL) dropwise to the flask. After 10 min, 0.5 mL of
NaHCO3 was added. The reaction mixture was warmed to rt and diluted
with 10 mL of ether. The reaction mixture was then passed through
an 8 cm diameter course fritted funnel containing sand (0.5 cm) on
top, filter paper in between, and silica gel (4 cm) on the bottom. The
material was rinsed through with ether. The filtered crude was
concentrated and further purified by column chromatography (4 × 17
cm) using 250 mL of 0%, 1%, 3%, and 5% EtOAc/hexanes and
concentrated to give 854 mg (82%) of product oil as one diastereomer
by proton NMR (S/N > 25:1). Data for the major diastereomer: Rf )
1
EtOAc/Hex; H NMR (500 MHz, CDCl3) δ 7.16 (dd, J ) 11.7, 10.8
Hz, 1H), 7.03 (d, J ) 10.8 Hz, 1H), 6.49 (dd, J ) 11.7, 10.8 Hz, 1H),
6.43-6.36 (m, 2H), 5.89 (s, 1H), 4.22-4.14 (m, 1H), 3.78-3.70 (m,
1H), 3.64-3.55 (m, 1H), 2.93 (s, 1H), 1.98 (s, 3H), 1.57-1.18 (m,
18H), 0.96 (d, J ) 8.4 Hz, 3H), 0.87 (t, J ) 7.8 Hz, 9H); 13C NMR
(50 MHz, CDCl3) δ 11.0, 13.5, 14.2, 17.6, 27.7, 29.6, 48.6, 67.7, 129.4,
129.9, 134.9, 140.6, 141.3, 146.5, 169.9; IR (film) 3336, 1636, 1458,
990; HRMS (CI) calcd for C23H43NO2Sn 482.2389, found 482.2379.
Preparation of (1E,3E,4S,5R,7E,8S)-5-Hydroxy-1-iodo-10-phenyl-
3,5,7,9-tetramethylundeca-1,3,7-triene (17). To a flask containing
neat (1E,3E,4S,5R,7E,8S)-5-[(tert-butylsilyl)oxy]-1-iodo-10-phenyl-
3,5,7,9-tetramethylundeca-1,3,7-triene (11) (109 mg, 0.202) was added
a THF solution of tetrabutylammonium fluoride (3 mL, 3.04 mmol, 1
M) at 0 °C. The reaction was allowed to warm to rt and to stir for 30
h. Filtration of the reaction mixture through a plug of silica gel (4 ×
1
0.44 in 2% EtOAc/Hex; H NMR (500 MHz, CDCl3) δ 7.23-7.10
(m, 2H), 6.71 (d, J ) 12.2 Hz, 1H), 5.88 (d, J ) 14.8 Hz, 1H), 4.21
(q, J ) 7.4 Hz, 2H), 1.56-1.47 (m, 9H), 1.35-1.27 (m, 9H), 1.01 (t,
J ) 7.4 Hz, 3H), 0.88 (t, J ) 6.6 Hz, 9H); 13C NMR (50 MHz, CDCl3)
δ 10.6, 13.8, 14.1, 27.8, 29.8, 60.7, 122.2, 143.5, 146.1, 147.3, 165.9,

12168 J. Am. Chem. Soc., Vol. 119, No. 50, 1997
Andrus et al.
3 cm) with ether followed by concentration and radial chromatography
on a 1 mm plate using 5% EtOAc/hexanes gave 74 mg (86%) of
product: [R]D +29.4° (c ) 0.0125, CH2Cl2); Rf ) 0.20 in 10% EtOAc/
δ 7.16 (d, J ) 12 Hz, 1H), 7.99-6.84 (m, 1H), 5.93 (d, J ) 15.2 Hz,
1H), 4.23 (q, J ) 8 Hz, 2H), 1.97 (s, 3H), 1.32 (t, J ) 7.2 Hz, 3H),
0.22 (s, 9H); 13C NMR (50 MHz, CDCl3) δ 168.3, 137.9, 136.9, 130.4,
117.8, 104.4, 101.3, 61.2, 14.7, 13.4, 0.2; HRMS (EI) calcd for
C13H20O2Si 236.1233, found 236.1223.
1
Hex; H NMR (500 MHz, CDCl3) δ 7.31-7.25 (m, 2H), 7.21-7.15
(m, 3H), 7.10 (d, J ) 15.0 Hz, 1H), 6.21 (d, J ) 15.0 Hz, 1H), 5.39
(d, J ) 9.6 Hz, 1H), 5.21 (d, J ) 9.6 Hz, 1H), 3.69 (d, J ) 9.0 Hz,
1H), 2.72-2.52 (m, 3H), 2.48-2.39 (m, 1H), 1.79 (s, 3H) 1.70-1.57
(m, 2H), 1.60 (s, 3H), 1.62-1.55 (m, 1H), 0.97 (d, J ) 7.2 Hz, 3H),
0.86 (d, J ) 7.2 Hz, 3H); 13C NMR (50 MHz, CDCl3) δ 150.0, 143.2,
137.3, 136.4, 135.6, 134.4, 128.9, 128.7, 126.0, 82.7, 74.7, 39.6, 37.2,
34.4, 32.4, 21.5, 17.8, 12.9, 11.9; HRMS (CI) calcd for C21H29 IO
425.1341, found 425.1349.
Preparation of (1′S,2E,4E,6Z,8E,10E,12R,13R,14E,16S)-(-)-13-
Hydroxy-N-(2′-hydroxy-1′-methylethyl)-2,10,12,14,16-pentamethyl-
18-phenyl-2,4,6,8,10,14-octadecahexaenamide ((-)-Stipiamide, 1).
To a flask containing neat (1E,3E,4S,5R,7E,8S)-5-hydroxy-1-iodo-10-
phenyl-3,5,7,9-tetramethylundeca-1,3,7-triene (17) (5.8 mg, 0.0136
mmol) and neat N-[(2E,4E,6Z)-2-methyl-7-(tri-n-butylstannyl)hepta-
2,4,6-trienoyl]-(S)-alaninol (16) (6.6 mg, 0.0136 mmol) at 0 °C were
added 1-methyl-2-pyrrolidinone (0.2 mL) and bis(acetonitrile)palladium
dichloride (0.4 mg, 0.0014 mmol). After being stirred for 15 min, the
reaction mixture was filtered through a plug of alumina (3 × 3 cm),
rinsing with 5% MeOH/CH2Cl2. The crude was then concentrated and
placed under a 200 µm Hg vacuum for 12 h to remove the pyrrolidi-
none.
Preparation of (1′S,2E,4E)-N-(2′-Hydroxy-1′-methylethyl)-2-
methyl-2,4-hept-6-ynedienamide (22). To a stirring solution of ethyl
(2E,4E]-5-bromo-2-methyl-7-(trimethylsilyl)-2,4-hept-6-ynedienoate (20)
(226 mg, 0.96 mmol) in ethanol (10 mL, 0.1 M) at rt was added all at
once potassium carbonate (159 mg, 1.15 mmol). After 2 h the reaction
mixture was filtered through a plug of silica gel (3.5 × 4 cm) and
concentrated to give 140 mg of the desilylated ester. The crude ester
was then dissolved in THF/MeOH/H2O (3, 1, and 1 mL, respectively).
To this was added LiOH (61 mg, 2.56 mmol) at rt, and the reaction
was stirred vigorously. After several hours, the reaction mixture was
filtered through a plug of silica gel (3.5 × 4 cm) with methanol and
concentrated to give the acid as an orange solid. To this solid was
added methylene chloride (9 mL, ∼0.1 M) followed by diisopropyl-
ethylamine (440 µL, 2.55 mmol) and (S)-alaninol (128 mg, 1.7 mmol).
The acid was only slightly soluble in this mixture. Finally, the PyBrOP
reagent was added (475 mg, 1.02 mmol), and the reaction was allowed
to stir for 24 h in the absence of light. The reaction was worked up by
filtration through a plug of silica gel (3.5 × 4 cm) with methanol
followed by concentration. This orange material was further purified
by radial chromatography on a 1 mm plate using ethyl acetate as the
eluent. Concentration of the product-containing fractions gave 101 mg
(54%, three steps) of the title compound as a clear oil that readily turned
orange in less than 15 min: Rf ) 0.29 in 100% EtOAc; 1H NMR (200
MHz, CDCl3) δ 7.01-6.83 (m, 2H), 6.13 (d, 1H), 5.91-4.73 (m, 1H),
4.18-4.02 (m, 1H), 3.68 (dd, J ) 11.4, 4.6 Hz, 1H), 3.54 (dd, J )
8.4, 6.1 Hz, 1H), 3.37 (s, 1H), 3.18 (s, 1H), 2.03 (s, 1H), 1.97 (s, 3H),
1.28-1.16 (m, 1H), 1.21 (d, J ) 7.2 Hz, 3H); 13C NMR (50 MHz,
CDCl3) δ 169.5, 138.5, 133.3, 132.8, 115.9, 83.1, 82.7, 67.3, 48.5, 17.5,
13.8; HRMS (EI) calcd for C11H15NO2 194.1181, found 194.1178.
The crude was then further purified on a minicolumn (0.5 × 6 cm)
of alumina using 5 mL of 2%, 5%, and 7% MeOH/CH2Cl2. Product-
containing fractions were concentrated on a rotary evaporator, keeping
the water bath at rt to give 5.3 mg (80%) of a bright yellow oil. 1H
NMR (500 MHz) shows the product to be a ∼2:1 mixture of stipiamide/
(1′S,2E,4Z,6E,8E,10E,12R,13R,14E,16S)-13-hydroxy-N-(2′-hydroxy-1′-
methylethyl)-2,10,12,14,16-pentamethyl-18-phenyl-2,4,6,8,10,14-octa-
decahexaenamide by comparison with the authentic spectra: [R]D
-100° (c ) 0.0015, MeOH). The literature value for pure stipiamide
is [R]D -189° (c ) 1, MeOH), and that for the (E,Z,E,E,E)-isomer is
[R]D -40.3° (c ) 0.3, MeOH). For a 2:1 mixture the expected rotation
should be [R]D 0.66° (-189°) + 0.33° (-40.3°) ) -138°: Rf ) 0.32
Preparation of Ethyl (E)-2-Methyl-2-hepten-6-ynoate (24). Py-
ridinium dichromate (0.971 g, 4.50 mmol) was added at 0 °C to a stirred
solution of 4-pentyn-1-ol (0.252, 3.0 mmol), sodium acetate (0.737 g,
9.0 mmol), and 4 Å sieves (0.367 g). The mixture was warmed to
room temperature and allowed to stir for 5 h. The solution was then
filtered through silica gel with ether and concentrated to give crude
1
in EtOAc; H NMR (500 MHz, CDCl3) δ 7.29-7.23 (m, 2H), 7.19-
7.13 (m, 3H), 7.05 (d, J ) 11.5 Hz, 1H), 6.98 (d, J ) 11.5 Hz, 1H),
6.67 (dd, J ) 15, 11.5 Hz, 1H), 6.47 (dd, J ) 15, 11.5 Hz, 1H), 6.37
(d, J ) 15 Hz, 1H), 6.17 (t, J ) 11.5 Hz, 1H), 6.09 (t, J ) 11.5 Hz,
1H), 5.88 (d, J ) 7 Hz, 1H), 5.49 (d, J ) 9.5 Hz, 1H), 5.22 (d, J ) 9.5
Hz, 1H), 4.22-4.13 (m, 1H), 3.75-3.63 (m, 3H), 3.62-3.56 (m, 1H),
3.07-3.01 (m, 1H), 2.78-2.70 (m, 1H), 2.64-2.51 (m, 2H), 2.47-
2.39 (m, 1H), 2.00 (s, 3H), 1.90 (s, 3H), 1.68-1.49 (m, 2H), 1.60 (s,
3H), 1.24 (d, J ) 6.5 Hz, 3H), 0.96 (d, J ) 6.5 Hz, 3H), 0.87 (d, J )
6.5 Hz, 3H); HRMS (CI) calcd for C32H45NO3 492.3478, found
492.3492.
1
4-pentyn-1-al: Rf ) 0.58 (EtOAc/hexane, 1:4); H NMR (200 MHz,
CDCl3) δ 9.8 (s, 1H, CHO). The crude aldehyde was dissolved in
toluene (15 mL), (carbethoxyethylidene)triphenylphosphorane (1.50 g,
4.14 mmol) was added, and the reaction mixture was heated to 90 °C
for 22 h. The solution was filtered through silica gel (Et2O/pentane,
1:4) and concentrated to give the crude vinyl ester. 1H NMR showed
E/Z selectivity >20:1 (200 MHz, CDCl3): δ 6.75 (t, J ) 7.1 Hz, (E)-
vinyl-H), 6.0 (t, J ) 8.0 Hz, (Z)-vinyl-H). Purification of the crude
mixture via radial chromatography (Et2O/pentane, 1:19) gave the (E)-
vinyl ester as a clear oil (0.226 g, 54% yield): Rf ) 0.60 (EtOAc/
Preparation of Ethyl (2E,4E)-5-Bromo-2-methyl-2,4-pentadi-
enoate. To a stirring solution of ethyl (2E,4E)-5-(tributylstannyl)-2-
methyl-2,4-pentadienoate (19) (367 mg, 0.86 mmol) in CCl4 (5 mL) at
-20 °C was added bromine (138 mg, 0.86 mmol) as a CCl4 solution
(2 mL) dropwise until a faint yellow color persisted. The mixture was
concentrated, filtered though a plug of silica gel (3.5 × 3 cm) with
ether, and concentrated. The oil was then radial chromatographed,
giving 172 mg (92%) of the title compound: Rf ) 0.39 in 10% EtOAc/
1
hexane, 1:4); IR (neat) 3300, 1712, 1652 cm-1; H NMR (200 MHz,
CDCl3) δ 6.75 (t, J ) 7.1 Hz, 1H), 4.15 (q, J ) 7.0 Hz, 2H), 2.45-
2.20 (m, 4H), 1.96 (t, J ) 2.5 Hz, 1H), 1.83 (s, 3H), 1.24 (t, J ) 7.0
Hz, 3H); 13C NMR (50 MHz, CDCl3) δ 168.3, 139.8, 129.7, 83.6, 69.5,
60.9, 28.2, 18.2, 14.7, 13.0; MS (CI) 167 (M + H, 100%). Anal. Calcd
for C10H14O2: C, 72.26; H, 8.49. Found: C, 71.88; H, 8.83.
1
Hex; H NMR (200 MHz, CDCl3) δ 7.13-6.94 (m, 2H), 6.79-6.63
Preparation of (S)-(-)-N-(2′-Hydroxy-1′-methylethyl)-(E)-2-
methyl-2-hepten-6-ynamide (26). Ethyl (E)-2-methyl-2-hepten-6-
ynoate (0.173 g, 1.04 mmol) was added to a 5 mL THF/MeOH/H2O
(3:1:1) solution. The mixture was cooled to 0 °C and LiOH (0.050 g,
2.08 mmol) was added followed by stirring at room temperature for
20 h. An additional amount of LiOH (0.050 g, 2.08 mmol) was added,
and stirring of the reaction was continued for an additional 4 h, upon
which TLC showed disappearance of the ester. The solution was
filtered through silica gel (EtOAc/MeOH, 1:4) to give the yellow acid
residue. The crude acid was then dissolved in CH2Cl2 (15 mL). PyBrop
(0.826 g, 1.77 mmol) and (S)-(+)-2-amino-1-propanol (0.222 g, 2.95
mmol) were sequentually added followed by addition of DIEA (0.77
mL, 4.4 mmol) at 0 °C. The reaction mixture was stirred for 12 h at
room temperature. The solution was filtered through silica gel (EtOAc)
and purified via radial chromatography (gradient elution, EtOAc/hexane,
2:1, to 100% EtOAc), giving a yellow-white solid (0.228 g, 92%): Rf
(m, 1H), 4.21 (q, J ) 8.0 Hz, 2H), 1.92 (s, 3H), 1.29 (t, J ) 8.0, 3H);
13C NMR (50 MHz, CDCl3) δ 168.3, 135.1, 133.8, 128.8, 116.6, 61.3,
14.7, 13.3; HRMS (EI) calcd for C8H11BrO2 217.9942, found 217.9938.
Preparation of Ethyl (2E,4E)-5-Bromo-2-methyl-7-(trimethylsi-
lyl)-2,4-hept-6-ynedienoate (20). To a stirring solution of ethyl
(2E,4E)-5-bromo-2-methyl-2,4-pentadienoate (159 mg, 0.73 mmol) in
benzene (0.15 M, 6 mL) at room temperature were added n-butylamine
(100 µL, 1.02 mmol) and Pd(Ph3P)4 (8 mg, 0.01 mmol). The resulting
solution was protected from light by wrapping the flask with foil. The
solution was allowed to stir for 45 min, after which (trimethylsilyl)-
acetylene (216 µL, 1.52 mmol) was added followed by copper(I) iodide
(22 mg, 0.12 mmol). After 5 h the mixture was filtered through a
plug of silica gel (3.5 × 4 cm) and concentrated. The resulting oil
was radial chromatographed to give 157 mg (91%) of the title
compound: Rf ) 0.40 in 10% EtOAc/Hex; 1H NMR (200 MHz, CDCl3)

Synthesis of Stipiamide and Designed Polyenes
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12169
) 0.263 (EtOAc); [R]D23 ) -4.08° (c 3.56 × 10-2, EtOAc); IR (neat)
Preparation of (1′S,2E,8E,10R,11R,12E)-11-Hydroxy-N-(2′-hy-
droxy-1′-methylethyl)-15-phenyl-2,11,13-trimethylpentadeca-2,8,12-
trien-6-yneamide (3). To a flask containing (1E,3S,4R,5E)-4-hydroxy-
1-iodo-8-phenyl-3,5-dimethylocta-1,5-diene (35) (11.1 mg, 0.031 mmol)
at -20 °C was added alkyne 26 (9.1 mg, 0.047 mmol) as an ethyl
acetate solution (1.0 mL). (PPh3)PdCl2 (1.7 mg, 2.4 × 10-3 mmol),
CuI (0.6 mg, 3.1 × 10-3 mmol), and i-Pr2NH (0.16 mL, 0.2M) were
subsequently added. The reaction mixture was protected from light,
warmed to room temperature, and allowed to stir for 1 h, whereupon
TLC showed disappearance of the starting alkyne. The solution was
quenched with 0.25 mL of saturated NH4Cl, filtered through a silica
plug (ca. 7 g) using 1:20 MeOH/EtOAc, and concentrated to give the
crude material. Purification via radial chromatography (100%) gave
11.5 mg (87%) of the title compound as an oil: [R]D ) -10.0° (c )
0.010, CHCl3); Rf ) 0.32 in EtOAc; IR (neat) 3355, 3025, 2969, 2925,
1
3297, 2877, 2117, 1661, 1617 cm-1; H NMR (200 MHz, CDCL3) δ
) 6.35 (t, J ) 6.4 Hz, 1H), 6.0 (br d, 1H), 4.15-4.0 (m, 1H), 3.71-
3.64 (dd, J ) 11.0, 3.75 Hz, 1H), 3.59-3.50 (dd, J ) 11.0, 6.0 Hz,
1H), 3.18 (br s, 1H), 2.41-2.26 (m, 4H), 1.97 (t, J ) 2.4 Hz, 1H),
1.85 (s, 3H), 1.2 (d, J ) 6.8 Hz, 3H); 13C NMR (50 MHz, CDCl3) δ
170.4, 134.4, 132.6, 83.8, 69.6, 67.5, 48.4, 27.9, 18.4, 17.6, 13.4; MS
(CI) m/z 196 (M + H, 100%). Anal. Calcd for C11H17O2N: C, 67.66;
H, 8.78. Found: C, 67.31; H, 8.92.
Preparation of (1′S,2E,4E,8E,10E,12R,13R,14E,16S)-(-)-13-Hy-
droxy-N-(2′-hydroxy-1′-methylethyl)-2,10,12,14,16-pentamethyl-18-
phenyl-2,4,8,10,14-octadecapentaen-6-ynamide (DHS, 2). To a flask
containing (1E,3E,4S,5R,7E,8S)-5-hydroxy-1-iodo-10-phenyl-3,5,7,9-
tetramethylundeca-1,3,7-triene (17) (5.0 mg, 0.0117 mmol) at -20 °C
was added alkyne 22 (3.9 mg, 0.020 mmol) as an ethyl acetate solution
(1.0 mL). (PPh3)PdCl2 (0.5 mg, 0.7 × 10-3 mmol), CuI (0.8 mg, 4.1
× 10-3 mmol), and i-Pr2NH (0.06 mL, 0.2M) were subsequently added.
The reaction mixture was protected from light, warmed to room
temperature, and allowed to stir for 30 min, whereupon TLC showed
disappearance of the starting alkyne. The reaction mixture was placed
directly by pipet onto a 20 × 20 cm analytical TLC plate and eluted
with 1:5 hexane/EtOAc in a TLC chamber. After eight elution/drying
sequences, the appropriate band was scraped off the plate, and the
scrapings were rinsed with 10 mL of ethyl acetate through a paper
filter. Concentration of the ethyl acetate gave 3.4 mg (59%) of the
title product as a yellow solid: [R]D ) 22.9° (c ) 0.0017, CHCl3); Rf
) 0.32 in 100% EtOAc; IR (neat) 3377, 3029, 2955, 2924, 1638, 1605,
1524, 1451, 1382 cm-1; 1H NMR (200 MHz, CDCl3) δ 7.30-7.23 (m,
2H), 7.21-7.11 (m, 3H), 6.97 (d, J ) 11.5 Hz, 1H), 6.82 (dd, J )
15.6, 15.1 Hz, 1H), 6.70 (d, J ) 17.0 Hz, 1H), 6.04 (d, J ) 16.0 Hz,
1H), 5.92 (d, 1H), 5.71 (d, J ) 16.0 Hz, 1H), 5.52 (d, J ) 10.1 Hz,
1H), 5.23 (d, J ) 9.6 Hz, 1H), 4.16 (m, 1H), 3.76-3.69 (m, 3H), 3.59
(dd, J ) 11.1, 6.0 Hz, 1H), 2.78-2.70 (m, 1H), 2.63-2.41 (m, 3H),
2.00 (s, 3H), 1.82 (s, 3H), 1.71-1.48 (m, 4H), 1.62 (s, 3H), 1.23 (d, J
1
1659, 1617, 1530, 1450 cm-1; H NMR (500 MHz, CDCl3) δ 7.29-
7.22 (m, 2H), 7.20-7.18 (m, 3H), 6.35 (t, J ) 6.9 Hz, 1H), 6.00 (dd,
J ) 15.5, 8.5 Hz, 1H), 5.94 (d, 1H), 5.54 (d, J ) 16.4 Hz, 1H), 5.42
(t, J ) 7.1 Hz, 1H), 4.16-4.08 (m, 1H), 3.73-3.67 (m, 1H), 3.64 (d,
J ) 8.6 Hz, 1H), 3.59-3.53 (m, 1H), 3.10 (s, 1H), 2.73-2.61 (m,
2H), 2.44-2.28 (m, 8H), 1.86 (s, 3H), 1.54 (s, 3H), 1.21 (d, J ) 7.3
Hz, 3H), 0.81 (d, J ) 6.9 Hz, 3H); 13C NMR (50 MHz, CDCl3) δ
170.5, 146.3, 142.4, 136.1, 134.6, 132.6, 128.9, 128.8, 126.3, 111.5,
88.7, 82.2, 80.2, 67.7, 48.4, 41.7, 36.1, 30.0, 28.1, 19.4, 17.6, 17.2,
13.4, 11.4; HRMS (CI) calcd for C27H37NO3 424.2852, found 424.2834.
Acknowledgment. We are grateful to the Purdue Research
Foundation, The National Science Foundation (Career Award),
and Procter and Gamble for funding. We are grateful to Ms.
Vicki Croy of the Purdue Cancer Center Cell Cytology Lab for
performing the MDR assays. We also thank Ms. Arlene
Rothwell for mass spectroscopy and D. Lee for elemental
analysis. We are again grateful to Professor Gerhard Ho¨fle for
providing a sample of natural stipiamide.
) 6.9 Hz, 3H), 0.99 (d, J ) 6.0 Hz, 3H), 0.91 (d, J ) 6.9 Hz, 3H); 13
C
Supporting Information Available: Experimental details,
NMR spectra, and MDR protocols for selected compounds (26
pages). See any current masthead page for ordering and Internet
access instructions.
NMR (50 MHz, CDCl3) δ 169.7, 147.4, 143.2, 139.5, 136.0, 135.5,
134.4, 133.7, 131.7, 128.9, 128.7, 126.1, 117.8, 106.2, 95.9, 91.2, 82.8,
77.7, 67.9, 48.6, 39.6, 37.6, 34.4, 32.3, 21.5, 17.9, 17.6, 13.7, 12.9,
11.9; FAB HRMS (DTT) calcd for C32H43NO3 490.3321, found
490.3331.
JA972603P