Total synthesis of the microtubule stabilizing antitumor agent laulimalide and some nonnatural analogues: The power of sharpless' asymmetric epoxidation

Article abstract of DOI:10.1021/jo026743f

Three different routes are described for the synthesis of deoxylaulimalide (3), which is the immediate precursor of the marine sponge metabolite laulimalide (1). These routes mainly differ with respect to their ring closing step. Thus, route 1 uses a Still-Gennari olefination, route 2 a Yamaguchi lactonization, and route 3 an intramolecular allylsilane-aldehyde addition for establishing the macrocyclic structure. The unprotected deoxy derivative 3 was subjected to Sharpless' asymmetric epoxidation (SAE). With (R,R)-tartrate the 16,17-epoxide laulimalide (1) is formed selectively, whereas (S,S)-tartrate generates the 21,22-epoxide 142. This demonstrates the high reagent control involved in the SAE process, which in this case is used to achieve high stereo- and regioselectivity. Laulimalide and some derivatives thereof have been tested with respect to antitumor activity and compared to standard compounds paclitaxel and epothilone B.

Full text of DOI:10.1021/jo026743f

Tota l Syn th esis of th e Micr otu bu le Sta bilizin g An titu m or Agen t
La u lim a lid e a n d Som e Non n a tu r a l An a logu es: Th e P ow er of
Sh a r p less’ Asym m etr ic Ep oxid a tion
Anjum Ahmed, E. Kate Hoegenauer, Valentin S. Enev, Martin Hanbauer, Hanspeter Kaehlig,
Elisabeth O¨ hler, and J ohann Mulzer*
Institut fu¨r Organische Chemie, Universita¨t Wien, Wa¨hringer Strasse 38, A-1090 Wien, Austria
johann.mulzer@univie.ac.at
Received November 20, 2002
Three different routes are described for the synthesis of deoxylaulimalide (3), which is the immediate
precursor of the marine sponge metabolite laulimalide (1). These routes mainly differ with respect
to their ring closing step. Thus, route 1 uses a Still-Gennari olefination, route 2 a Yamaguchi
lactonization, and route 3 an intramolecular allylsilane-aldehyde addition for establishing the
macrocyclic structure. The unprotected deoxy derivative 3 was subjected to Sharpless’ asymmetric
epoxidation (SAE). With (R,R)-tartrate the 16,17-epoxide laulimalide (1) is formed selectively,
whereas (S,S)-tartrate generates the 21,22-epoxide 142. This demonstrates the high reagent control
involved in the SAE process, which in this case is used to achieve high stereo- and regioselectivity.
Laulimalide and some derivatives thereof have been tested with respect to antitumor activity and
compared to standard compounds paclitaxel and epothilone B.
In tr od u ction
of the MSAA (microtubule stabilizing antitumor agents)
family of compounds, which share the same mechanism
of action as the frontline anticancer drugs Taxol (pacli-
taxel)4 and Taxotere (docetaxel). Thus, the family of
compounds with “taxol-like” activity5 currently includes
the following members: taxanes (isolated from yew
trees), marine metabolites (sarcodictyins/eleutherobin,
discodermolide, laulimalide, and peloruside A6), microbial
metabolites (the epothilones,7 which are already under
clinical investigation, and the polycyclic compound
FR182877, formerly known as WS9885B8), other natural
products (taccalonolide,9a tryprostatin,9b and xantho-
chymol9c), and nonnatural compounds (for instance, an
Laulimalide (1),1a also known as fijianolide B,1b is a
20-membered macrolide that was isolated together with
its tetrahydrofuran containing isomer isolaulimalide (2)
(fijianolide A) from various marine sponges such as
Hyattella sp.,1a Cacospongia mycofijiensis,1b Fasciospon-
gia rimosa,1c,d and very recently a sponge in the genus
Dactylospongia.1e Interestingly, both compounds were
also isolated from a nudibranch, Chromodoris lochi, that
was found grazing on the sponge.1a,b The structure of 1
was initially established by NMR analysis,1a,b and later
corroborated through X-ray diffraction studies.1c Lauli-
malide is a potent inhibitor of cellular proliferation, with
IC50 values against numerous drug sensitive cell lines
in the low nanomolar range [KB cell line, 15 ng/mL;1a
P388, A549, HT29, MEL28, 10-50 ng/mL;1d MDA-MB-
435, 6 ng/mL;2 SK-OV-3, 12 ng/mL;2 MCF-7, 7 ng/mL3],
whereas isolaulimalide is much less potent with IC50
values in the low micromolar range. The high cytotoxicity
of 1 originates from its ability to promote and stabilize
tubulin polymerization, leading to the formation of
abnormal mitotic spindles, mitotic arrest, and apoptosis.2
Thus, laulimalide has to be considered as a new member
(4) For reviews, see: (a) Nicolaou, K. C.; Dai, W.-M.; Guy, R. K.
Angew. Chem., Int. Ed. Engl. 1994, 33, 15-44. (b) Kingston, D. G. I.
Chem. Commun. 2001, 867-880.
(5) (a) Stachel, S. J .; Biswas, K.; Danishefsky, S. J . Curr. Pharm.
Des. 2001, 7, 1277-1290. (b) He, L.; Orr, G. A.; Horwitz, S. B. Drug
Discovery Today 2001, 6, 1153-1164. (c) Altmann, K.-H. Curr. Opin.
Chem. Biol. 2001, 5, 424-431. (d) J ordan, M. A. Curr. Med. Chem.:
Anti-Cancer Agents 2002, 2, 1-17. (e) J imenez-Barbero, J .; Amat-
Guerri, F.; Snyder, J . P. Curr. Med. Chem.: Anti-Cancer Agents 2002,
2, 91-122.
(6) Sarcodictyins: D’Ambrosio, M.; Guerriero, A.; Pietra, F. Helv.
Chim. Acta 1987, 70, 2019-2027. D’Ambrosio, M.; Guerriero, A.;
Pietra, F. Helv. Chim. Acta 1988, 71, 964-976. Eleutherobin: Lindel,
T.; J ensen, P. R.; Fenical, W.; Long, B. H.; Casazza, A. M.; Carboni J .;
Fairchild, C. R. J . Am. Chem. Soc. 1997, 119, 8744-8745. Discoder-
molide: Gunasekera, S. P.; Gunasekera, M.; Longley, R. E.; Schulte,
G. K. J . Org. Chem. 1990, 55, 4912-4915. The original structure of
discodermolide was revised in: J . Org. Chem. 1991, 56, 1346. Peloruside
A: (a) West, L. M.; Northcote, P. T.; Battershill, C. N. J . Org. Chem.
2000, 65, 445-449. (b) Hood, K. A.; West, L. M.; Rouwe, B.; Northcote,
P. T.; Berridge, M. V.; Wakefield, St. J .; Miller, J . H. Cancer Res. 2002,
62,(73)3F5o6r-r3e3v6ie0w. s, see: (a) Nicolaou, K. C.; Roschangar, F.; Vourloumis,
D. Angew. Chem., Int. Ed. 1998, 37, 2015-2045. (b) Mulzer, J .
Monatsh. Chem. 2000, 131, 205-238. (c) Nicolaou, K. C.; Ritzen, A.;
Namoto, K. Chem. Commun. 2001, 1523-1535. (d) Wartmann, M.;
Altmann, K.-H. Curr. Med. Chem.: Anti-Cancer Agents 2002, 2, 123-
148.
(1) (a) Corley, D. G.; Herb, R.; Moore, R. E.; Scheuer, P. J .; Paul, V.
J . J . Org. Chem. 1988, 53, 3644-3646. (b) Quinoa`, E.; Kakou, Y.;
Crews, P. J . Org. Chem. 1988, 53, 3642-3644. (c) J efford, C. W.;
Bernardinelli, G.; Tanaka, J .-i.; Higa, T. Tetrahedron Lett. 1996, 37,
159-162. (d) Tanaka, J .-i.; Higa, T.; Bernardinelli, G.; J efford, C. W.
Chem. Lett. 1996, 255-256. (e) Cutignano, A.; Bruno, I.; Bifulco, G.;
Casapullo, A.; Debitus, C.; Gomez-Paloma, L.; Riccio, R. Eur. J . Org.
Chem. 2001, 775-778.
(2) Mooberry, S. L.; Tien, G.; Hernandez, A. H.; Plubrukarn, A.;
Davidson, B. S. Cancer Res. 1999, 59, 653-660.
(3) Pryor, D. E.; O’Brate, A.; Bilcer, G.; Diaz, J . F.; Wang, Yu.; Wang,
Yo.; Kabaki, M.; J ung, M. K.; Andreu, J . M.; Ghosh, A. K.; Gianna-
kakou, P.; Hamel, E. Biochemistry 2002, 41, 9109-9115.
10.1021/jo026743f CCC: $25.00 © 2003 American Chemical Society
Published on Web 03/25/2003
3026
J . Org. Chem. 2003, 68, 3026-3042

Total Synthesis of Laulimalide
no more than three reports from the Ghosh11a and
Nishiyama11d,e groups were available on the synthesis of
major fragments of 1. To date, in addition to an impres-
sive number of 15 approaches to key fragments by several
groups,11 as many as 10 total syntheses of 1 have been
completed.12-16 The first synthesis was accomplished by
Ghosh and Wang,12a,c who later refined their first ap-
proach by a stereoselective introduction of the 2,3-cis-
enoate.12b,c These two reports were followed in close
succession by three syntheses from our group13 and one
from Paterson,14 which all avoid low-yielding protective
group manipulations during the endgame by using a
regio- and stereoselective epoxidation of the unprotected
macrocycle as the last step. This strategy has also been
adopted in the total synthesis by Wender and co-work-
ers,15 which features a diastereoselective Sakurai reaction
for fragment coupling and a regioselective macrolacton-
ization of a 2,3-alkynoic acid with an unprotected vicinal
19,20-diol. Lately, the series of total syntheses of 1 has
been complemented by two closely related approaches
from the Crimmins16a and the Williams16b groups which
both focus on a diastereoselective allylic transfer of a C1-
C14 allylstannane16a (or silane16b) to a C15-C27 R,â-
epoxyaldehyde. A very recent synthesis by Nelson16c is
characterized by extensive use of asymmetric acyl halide-
aldehyde cyclocondensation methodology for the con-
struction of main fragments.
F IGURE 1. Laulimalide (1) and isolaulimalide (2).
analogue of estradiol,10a a combretastatin D analogue,10b
and GS-16410c). However, while the epothilones, disco-
dermolide, and eleutherobin inhibit the binding of [3H]-
paclitaxel to tubulin polymer in a competitive manner,
and thus apparently bind to the same or an overlapping
site at the protein, it was recently discovered3 that
laulimalide binds at a distinct site.
Our interest in laulimalide was kindled in February
1999 by the work of Mooberry and co-workers,2 which
underlined that 1 is as much as 100-fold more potent
than paclitaxel against SKVLB-1 cells, a P-glycoprotein
overexpressing ovarian cancer cell line, which exhibits
multidrug resistance. Very recently, the high therapeutic
potential of 1 was further underlined by Hamel,3 who
found that laulimalide also kills human ovarian carci-
noma cells which, due to taxoid site mutations in the M
40 human â-tubulin gene, are resistant to paclitaxel
(PTX10, PTX22), epothilone A (A8), and epothilone B
(B1).
Apart from the significant clinical potential of 1 and
its restricted natural supply, the attraction of laulimalide
as a synthetic target originates from its unique and
complex molecular architecture. Specifically, its 16,17-
epoxide is susceptible to nucleophilic attack from the 20-
hydroxy group to form the more stable and biologically
less active tetrahydrofuran isomer 2 (Figure 1),1a and the
2,3-cis-enoate moiety readily undergoes Z/E-isomeriza-
tion. When we started our synthetic efforts in late 1999,
Herein, we report the full details of our synthetic
studies, which so far culminated in three total syntheses.
All of them use the same endgame, namely the regio- and
stereoselective epoxidation of 16,17-deoxylaulimalide (3).
In view of the easy isomerization of 1 to 2, we decided to
avoid any protecting group manipulation at the C15- and
C20-OH functions after the introduction of the C16-C17
epoxide17 and to rely on a chirally induced epoxidation18
(11) (a) Ghosh, A. K.; Mathivanan, P.; Cappiello, J . Tetrahedron Lett.
1997, 38, 2427-2430. (b) Ghosh, A. K.; Wang, Y. Tetrahedron Lett.
2000, 41, 2319-2322. (c) Ghosh, A. K.; Wang, Y. Tetrahedron Lett.
2000, 41, 4705-4708. (d) Shimizu, A.; Nishiyama, S. Tetrahedron Lett.
1997, 38, 6011-6014. (e) Shimizu, A.; Nishiyama, S. Synlett 1998,
1209-1210. (f) Mulzer, J .; Hanbauer, M. Tetrahedron Lett. 2000, 41,
33-36. (g) Dorling, E. K.; O¨ hler, E.; Mulzer, J . Tetrahedron Lett. 2000,
41, 6323-6326. (h) Dorling, E. K.; O¨ hler, E.; Mantoulidis, M.; Mulzer,
J . Synlett 2001, 1105-1108. (i) Ahmed, A.; O¨ hler, E.; Mulzer, J .
Synthesis 2001, 2007-2010. (j) Nadolski, G. T.; Davidson B. S.
Tetrahedron Lett. 2001, 42, 797-800. (k) Messenger, B. T.; Davidson
B. S. Tetrahedron Lett. 2001, 42, 801-804. (l) Sivaramakrishnan, A.;
Nadolski, G. T.; McAlexander, I. A.; Davidson B. S. Tetrahedron Lett.
2002, 43, 213-216. (m) Paterson, I.; De Savi, C.; Tudge, M. Org. Lett.
2001, 3, 213-216. (n) Lee, H. W.; J eong, C.-S.; Yoon, S. H.; Lee, I.-Y.
C. Bull. Korean Chem. Soc. 2001, 22, 791-792. (o) Lee, H. W.; Yoon,
S. H.; Lee, I.-Y. C.; Chung, B. Y. Bull. Korean Chem. Soc. 2001, 22,
1179-1180.
(12) (a) Ghosh, A. K.; Wang, Y. J . Am. Chem. Soc. 2000, 122, 11027-
11028. (b) Ghosh, A. K.; Wang, Y. Tetrahedron Lett. 2001, 42, 3399-
3401. (c) Ghosh, A. K.; Wang, Y.; Kim, J . T. J . Org. Chem. 2001, 66,
8973-8982.
(8) (a) Sato, B.; Muramatsu, H.; Miyauchi, M.; Hori, Y.; Takase, S.;
Hino, M.; Hashimoto, S.; Terano, H. J . Antibiot. 2000, 53, 123-130.
(b) Sato, B.; Nakajima, H.; Hori, Y.; Hino, M.; Hashimoto, S.; Terano,
H. J . Antibiot. 2000, 53, 204-206. (c) Yoshimura, S.; Sato, B.;
Kinoshita, T.; Takase, S.; Terano, H. J . Antibiot. 2000, 53, 615-622.
The structure of FR182877 was revised in: J . Antibiot. 2002, 55, C1.
(9) (a) Chen, Z.; Wang, B.; Chen, M. Huaxue Xuebao 1988, 46, 1201-
1206. (b) Kondoh, M.; Usui, T.; Mayumi, T.; Osada, H. J . Antibiot. 1998,
51, 801-804. (c) Roux, D.; Hadi, H. A.; Thoret, S.; Guenard, D.;
Thoison, O.; Paies, M.; Sevenet, T. J . Nat. Prod. 2000, 63, 1070-1076.
(10) (a) Wang, Z.; Yang, D.; Mohanakrishnan, A. K.; Fanwick, P.
E.; Nampoothiri, P.; Hamel, E.; Cushman, M. J . Med. Chem. 2000,
43, 2419-2429. (b) Couladouros, E. A.; Li, T.; Moutsos, V. I.; Pitsinos,
E. N.; Soufli, I. C. Bioorg. Med. Chem. Lett. 1999, 9, 2927-2928. (c)
Shintani, Y.; Tanaka, T.; Nozaki, Y. Cancer Chemother. Pharmacol.
1997, 40, 513-520.
(13) (a) Mulzer, J .; O¨ hler, E. Angew. Chem., Int. Ed. 2001, 40, 3842-
3846. (b) Enev, V. E.; Ka¨hlig, H.; Mulzer, J . J . Am. Chem. Soc. 2001,
123, 10764-10765. (c) Mulzer, J .; Hanbauer, M. Tetrahedron Lett.
2002, 43, 3381-3383.
(14) Paterson, I.; De Savi, C.; Tudge, M. Org. Lett. 2001, 3, 3149-
3152.
(15) Wender, P. A.; Hegde, S. G.; Hubbard, R. D.; Zhang, L. J . Am.
Chem. Soc. 2002, 124, 4956-4957.
(16) (a) Crimmins, M. T.; Stanton, M. G.; Allwein, S. P. J . Am. Chem.
Soc. 2002, 124, 5958-5959. (b) Williams, D. R.; Mi, L.; Mullins, R. J .;
Stites, R. E. Tetrahedron Lett. 2002, 43, 4841-4844. (c) Nelson, S. G.;
Cheung, W. S.; Kassick, A. J .; Hilfiker, M. A. J . Am. Chem. Soc. 2002,
124, 13654-13655.
(17) The recent work of Crimmins16a revealed that removal of two
TBS protecting groups from the C15 and C20 hydroxyls is possible
without rearrangement to isolaulimalide (2).
J . Org. Chem, Vol. 68, No. 8, 2003 3027

Ahmed et al.
F IGURE 2. Retrosynthetic routes to deoxylaulimalide (3) and the corresponding ring closure reactions.
of the unprotected macrocycle (vide infra). In conse-
quence, 3 was envisaged as the common intermediate of
our three routes which, otherwise, mainly differ with
respect to the macrocyclization step (Figure 2). The first
route13a employs an intramolecular Still-Gennari olefi-
nation of phosphonate aldehyde 4. Our second route13c
involves a final Yamaguchi macrolactonization of seco-
acid 5, and in the third and so far only fully stereocon-
trolled route,13b the ring closure has been achieved by
adding the C14-allylsilane to the C15-acetal in seco-
compound 6.
Resu lts a n d Discu ssion
Rou te 1. F r a gm en t Un ion by J u lia -Kocien sk i
Olefin a tion , Ma cr ocycliza tion by Still-Gen n a r i Re-
a ction . The underlying synthetic strategy is illustrated
in Figure 3. Inspired by precedence from the syntheses
of phorboxazole A,19 we envisaged an intramolecular
Still-Gennari olefination20 of seco-precursor 4 for mac-
rocyclization and introduction of the C2-C3 (Z)-enoate.
Identical protective groups (i.e. MOM) at the C15 and C20
allylic alcohols of 4 would significantly simplify the
protective group strategy.
An E-selective one-step J ulia-Kocienski olefination21,22
FIGURE 3. Retrosynthetic disconnection of key intermediate 4.
was envisioned to connect the two advanced synthetic
intermediates 7 and 8, which were to be assembled from
the smaller fragments 9, 11, and 10, 12, respectively. It
was decided to prepare the dihydropyrans 9 and 10 not
only by the relative obvious ring-closing olefin metathesis
(RCM),23 but also by other competing methodology. The
C17-C27 sulfone 7 should be assembled via Horner-
Wadsworth-Emmons olefination from aldehyde 9 and
ketophosphonate 11, whereas C3-C16 aldehyde 8 should
be prepared by addition of the anion derived from sulfone
10 to glycidyl ether 12, and subsequent introduction of
the C13 methylene group. To keep costs low, intermedi-
ates 9, 10, and 11 should be derived from inexpensive
(18) (a) Rossiter, B. E. Chiral Catalysis. In Asymmetric Synthesis;
Morrison, J . D., Ed.; Academic Press: Orlando, FL, 1985; Vol. 5, pp
193-246. (b) J ohnson, R. A.; Sharpless, K. B. In Catalytic Asymmetric
Synthesis; Ojima, I., Ed.; VCH Publishers: New York, 1993; p 103.
(19) (a) Forsyth, C. J .; Ahmed, F.; Cink, R. D.; Lee, C. S. J . Am.
Chem. Soc. 1998, 120, 5597-5598. (b) Williams, D. R.; Clark, M. B.
Tetrahedron Lett. 1999, 40, 2291-2294. (c) Smith, A. B., III; Minbiole,
K. B.; Verhoest, P. R.; Schelhaas, M. J . Am. Chem. Soc. 2001, 123,
4834-4836 and 10942-10953.
(20) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405-4408.
3028 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
disulfide.26 Epoxide 18 was obtained without isolation of
intermediates in 89% overall yield via mesylation of the
secondary hydroxy group, deprotection of the primary
silyl ether with tetrabutylammonium fluoride, and ring
closure under inversion with sodium hydroxide. Thus, the
desired configurations at C11 and C9 had been introduced
without any additional chiral auxiliary.
SCHEME 1. Syn th esis of Ep oxid e 18
Elaboration of epoxide 18 into the dihydropyran 21
(Scheme 2) was accomplished with Ghosez’s one-pot
lactonization procedure.27 Boron trifluoride etherate me-
diated addition of of lithium methyl phenylsulfonyl
orthopropionate28 (19) to epoxide 18, followed by acid-
catalyzed lactonization with trifluoroacetic acid in dichlo-
romethane and subsequent elimination of phenyl sulfinic
acid with DBU, afforded the R,â-unsaturated lactone 20
in 88% overall yield. Lactone 20 was converted into
anomerically pure ethyl glycoside 2129 in 93% overall
yield by sequential oxidation of the C13 thioether with
magnesium monoperoxy phthalate (MMPP) in ethanol
to the corresponding sulfone, reduction of the lactone with
diisobutyl aluminum hydride, and subsequent treatment
of the resulting lactol with PPTS in ethanol.
A second synthesis of 21 was performed via a RCM
strategy (Scheme 2), by analogy to our previous work.11f
This included oxidation of the phenyl sulfide 18 to the
corresponding sulfone 22 (MMPP, 96% yield), followed
by copper(I) iodide catalyzed addition of vinylmagnesium
bromide (4 equiv, THF, -40 °C), to provide the desired
homoallylic alcohol 23 in 96% yield. Next, PPTS-
catalyzed transacetalization of alcohol 23 with acrolein
diethylacetal was carried out by a modification of Crim-
mins’ procedure30 in toluene under reduced pressure with
azeotropic removal of ethanol11f to give mixed acetal 24
as an epimeric mixture in 86% yield. RCM of diene 24 in
boiling dichloromethane with 5 mol % of Grubbs’ ruthe-
nium catalyst I23a provided, after treatment with PPTS
in ethanol, ethyl glycoside 21 in 85% yield.
compounds of the chiral carbon pool, such as D-mannitol,
D-glucose, and (S)-malic acid.
Syn th esis of th e C3-C16 F r a gm en t. For the con-
struction of the C5-C9 dihydropyran moiety, epoxide 18
was envisaged as a suitable intermediate (Scheme 1). We
started from the known R,â-unsaturated lactone 13,
which is readily prepared from commercially available
triacetyl D-glucal in four steps.24 Lactone 13 was silylated
to give 14, which underwent axial conjugate addition
with lithium dimethyl cuprate to provide the desired 1,4-
adduct 15 with laulimalide stereochemistry at C11 as a
single diastereomer (by 1H NMR) in 84% yield.25 The
expected trans-disubstitution in 15 was confirmed by a
strong NOE between the signals of the C11 methyl group
and H-9. To enable the introduction of the sulfonyl group
and inversion at C9, lactone 15 was reduced with lithium
borohydride to provide the 1,5-diol 16 almost quantita-
tively. The primary alcohol in 16 was selectively trans-
formed to the corresponding phenyl sulfide 17 in 83%
yield by treatment with tributyl phosphine/diphenyl
The stereoselective introduction of the C3-C4 side
chain was attempted by treatment of acetal 21 with tert-
butyldimethylsilyl vinyl ether31 in dry dichloromethane
at 0 °C, using montmorillonite K-10 clay32 as a Lewis acid
(Scheme 3). Under these conditions the desired trans-
disubstituted C3-aldehyde 25a was obtained stereoselec-
tively, but depending on the reaction time, substantial
amounts of the corresponding C3-diethyl acetal 25b were
also formed. After some experimentation, it was found
that 25b was easily reconverted to aldehyde 25a by
(21) Blakemore, P. R.; Cole, W. J .; Kocienski, P. J .; Morley, A. Synlett
1998, 26-28. For recent applications of the J ulia-Kocienski protocol
in natural product syntheses, see: (a) Smith, A. B., III; Wan, Z. J .
Org. Chem. 2000, 65, 3738-3753 and references therein. (b) Liu, P.;
J acobsen, E. N. J . Am. Chem. Soc. 2001, 123, 10772-10773. (c) Smith,
A. B., III; Safonov, I. G.; Corbett, M. C. J . Am. Chem. Soc. 2001, 123,
12426-12427. (d) Takano, D.; Nagamitsu, T.; Ui, H.; Shiomi, K.;
Yamaguchi, Y.; Masuma, R.; Kuwajima, I.; Omura, S. Org. Lett. 2001,
3, 2289-2291. (e) Lee, E.; Song, H. Y.; Kang, J . W.; Kim, D.-S.; J ung,
C.-K.; J oo, J . M. J . Am. Chem. Soc. 2002, 124, 384-385. (f) Lautens,
M.; Colucci, J . T.; Hiebert, S.; Smith, N. D.; Bouchain, G. Org. Lett.
2002, 4, 1879-1882.
(26) Nakagawa, I.; Abi, K.; Hata, T. J . Chem. Soc., Perkin Trans. 1
1983, 1315-1318.
(27) Carretero, J . C.; Ghosez, L. Tetrahedron Lett. 1988, 29, 2059-
2062. For recent applications of Ghosez’s methodology, see: (a)
Nicolaou, K. C.; Patron, A. P.; Ajito, K.; Richter, P. K.; Khatuya, H.;
Bertinato, P.; Miller, R. A.; Tomaszewski, M. J . Chem. Eur. J . 1996,
2, 847-868. (b) Murakami, N.; Wang, W.; Aoki, M.; Tsutsui, Y.;
Sugimoto, M.; Kobayashi, M. Tetrahedron Lett. 1998, 39, 2349-2352.
(c) Surivet, J .-P.; Vate`le, J .-M. Tetrahedron 1999, 55, 13011-13028.
(d) Williams, D. R.; Ihle, D. C.; Plummer, S. V. Org. Lett. 2001, 3, 1383-
1386.
(22) A standard but low-yielding J ulia olefination to connect a C27
-
C17 phenyl sulfone with a C12-C16 aldehyde fragment has been used
initially by Nishiyama and Shimizu,11d and was later also adopted by
Ghosh and Wang12a,c to generate the C16-C17 double bond.
(23) For reviews, see: (a) Grubbs, R. H.; Chang, S. Tetrahedron
1998, 54, 4413 - 4450. (b) Armstrong, S. K. J . Chem. Soc., Perkin
Trans. 1 1998, 371-388. (c) Fu¨rstner, A. Angew. Chem., Int. Ed. 2000,
39, 3012-3043. (d) Trinka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001,
34, 18-29.
(28) De Lombaert, S.; Nemery, I.; Roekens, B.; Carretero, J . C.;
Kimmel, T.; Ghosez, L. Tetrahedron Lett. 1986, 27, 5099-5102.
(29) The stereochemistry of 21 was assigned by NOESY experi-
ments. See also ref 12c.
(24) (a) Corey, E. J .; Pyne, S. G.; Su, W.-g. Tetrahedron Lett. 1983,
24, 4883-4886. (b) Roth, B. D.; Roark, W. H. Tetrahedron Lett. 1988,
29, 1255-1258. See also: (c) Rollin, P.; Sinay, P. Carbohydr. Res. 1981,
98, 139-142. (d) Lichtenthaler, F. W.; Ro¨nninger, S.; J arglis, P. Liebigs
Ann. Chem. 1989, 1153-1161.
(25) For related protocols, see: (a) Takano, S.; Shimazaki, Y.;
Iwabuchi, Y.; Ogasawara, K. Tetrahedron Lett. 1990, 31, 3619-3622.
(b) Kim, D.; Lee, J .; Shim, P. J .; Lim, J . I.; Doi, T.; Kim, S. J . Org.
Chem. 2002, 67, 772-781.
(30) Crimmins, M. T.; King, B. W. J . Am. Chem. Soc. 1998, 120,
9084-9085.
(31) Srisiri, W.; Padias, A. B.; Hall, H. K., J r. J . Org. Chem. 1994,
59, 5424-5435.
(32) Toshima, K.; Miyamoto, N.; Matsuo, G.; Nakata, M.; Mat-
sumura, S. Chem. Commun. 1996, 1379-1380.
J . Org. Chem, Vol. 68, No. 8, 2003 3029

Ahmed et al.
SCHEME 3. Syn th esis of C3-C16 F r a gm en t 30
SCHEME 2. Tw o Syn th eses of Eth yl Glycosid e 21
subsequent treatment of the crude reaction mixture with
K-10 clay in wet dichloromethane.33 The crude aldehyde
was reduced with sodium borohydride to deliver alcohol
26 in 86% overall yield from 21.34 After TBS protection
of the hydroxyl function, sulfone 10 was deprotonated
with n-butyllithium and treated with epoxide 12 at -78
°C in the presence of boron trifluoride etherate to afford
a
See text and Supporting Information.
steps,38 was tosylated and transformed to cyanide 32
(NaCN, DMSO, 80 °C) in 86% yield for the two steps.
Reduction of 32 with DIBALH in THF provided aldehyde
33 (80% yield), which was allylated with (-)-allyl
diisopinocampheylborane39a according to Brown’s im-
proved protocol39b to provide homoallylic alcohol 34 in
93% yield with high diastereoselectivity (de ) 90%).
Allylation of aldehyde 33 according to Duthaler’s proce-
dure40 with (R,R)-TADDOLTiCpCl and allylmagnesium
bromide in diethyl ether at -78 °C afforded alcohol 34
with better diastereoselection (de ) 96%), but in a
disappointing yield of 51%. Homoallylic alcohol 34 was
then transformed to C3-C12 dihydropyran 38 via transac-
etalization to 35 (84% yield), RCM to 36 (94% yield),
stereoselective C-glycosidation at C5 with vinyl-OTBS,
followed by reduction of the resulting aldehyde with
sodium borohydride to alcohol 37 (85% for the two steps),
and silyl protection at C3-O. Treatment of 38 with DDQ
in wet dichloromethane cleaved the PMB group, and the
resulting primary alcohol 39 was transformed to the
corresponding iodide 40 in 91% yield, under standard
conditions.41,42 Treatment of 40 with the nucleophile
derived from methyl phenyl sulfone and n-BuLi in THF/
HMPA (20:1) led to the desired sulfone 10 in 85% yield.
1
a mixture of epimeric sulfones 27 (1:1, by H NMR) in
86% yield. When sulfones 27 were subjected to the J ulia
methylenation procedure,35,36 the yield of the desired C13-
methylene derivative 2937 was unvariably less than 30%.
Therefore, the C15-OH group in 27 was protected as the
MOM-ether sulfones 28, which underwent methylenation
smoothly (deprotonation in THF/HMPA at -78 °C, then
addition of 5 equiv of the carbenoid prepared from
diiodomethane and isopropylmagnesium chloride), to
furnish C13-methylene compound 30 in 75% yield.
An alternative route to C3-C13-sulfone 1011f also em-
ploying Grubbs RCM is shown in Scheme 4. Known
alcohol 31, which is readily prepared from commercially
available methyl (S)-3-hydroxy-2-methylpropionate in 2
(33) Gautier, E. C. L.; Graham, A. E.; McKillop, A.; Standen, S. P.;
Taylor, R. J . K. Tetrahedron Lett. 1997, 38, 1881-1884.
(34) The stereochemistry at C5 was determined by NOESY experi-
ments and was further confirmed by a strong NOE between the protons
at C5 and C9 in the NMR spectrum of the cis-isomer 25c, which was
isolated in traces (e2%) by chromatography.
(35) De Lima, C.; J ulia, M.; Verpeaux, J .-N. Synlett 1992, 133-134.
(36) For recent applications of J ulia’s methodology, see: (a) Romo,
D.; Meyer, S. D.; J ohnson, D. D.; Schreiber, S. L. J . Am. Chem. Soc.
1993, 115, 7906-7907. (b) Smith, A. B., III; Doughty, V. A.; Lin, Q.;
Zhuang, L.; McBriar, M. D.; Boldi, A. M.; Moser, W. H.; Murase, N.;
Nakayama, K.; Sobukama, M. Angew. Chem., Int. Ed. 2001, 40, 191-
195. (c) Smith, A. B., III; Lin, Q.; Doughty, V. A.; Zhuang, L.; McBriar,
M. D.; Kerns, Y. K.; Brook, C. S.; Murase, N.; Nakayama, K. Angew.
Chem., Int. Ed. 2001, 40, 196-199.
(38) Walkup, R. D.; Boatman, P. D., J r.; Kane, R. R.; Cunningham,
R. T. Tetrahedron Lett. 1991, 32, 3937-3940.
(39) (a) Brown, H. C.; Bhat, K. S.; Randad, R. S. J . Org. Chem. 1989,
54, 1570-1576. (b) Racherla, U. S.; Brown, H. C. J . Org. Chem. 1991,
56, 401-404.
(40) Hafner, A.; Duthaler, R. O.; Marti, R.; Rihs, G.; Rothe-Streit,
P.; Schwarzenbach, F. J . J . Am. Chem. Soc. 1992, 114, 2321-2336.
(41) Garegg, P. J .; Samuelsson, B. Chem. Commun. 1979, 978-980.
(37) Compound 29 is a key fragment in Ghosh’s synthesis of 1.12a
3030 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
SCHEME 4. Alter n a tive Ap p r oa ch to C3-C13
Su lfon e 10
SCHEME 5. F ir st Ap p r oa ch to Ald eh yd e 9
SCHEME 6. Secon d Ap p r oa ch to Ald eh yd e 9
converted into the volatile aldehyde 9 by Parikh-Doering
oxidation.44
In the second and more convenient approach to alde-
hyde 9,11g the trisubstituted double bond was formed by
RCM (Scheme 6). Thus, glycidyl trityl ether 45 was
regioselectively opened with isopropenylmagnesium bro-
mide (2 equiv) under copper(I) catalysis in THF at -30
°C to furnish the homoallylic alcohol 46 quantitatively,
which was alkylated with allyl bromide to give diene 47
in 92% yield. When 47 (0.015 M in dichloromethane) was
exposed to 2.5-3% of Grubbs’ ruthenium catalyst I for
only 1-2 h at room temperature,45 the corresponding
dihydropyran 48 was obtained in quantitative yield. This
result was particularly remarkable, as literature prece-
dence indicated that RCM of dienes bearing one gem-
disubstituted double bond does not work with Grubbs’
catalyst I.46 Removal of the trityl group with hydrogen
chloride in dichloromethane without aqueous workup
provided alcohol 44 in 84% yield.47
Syn th esis of Su lfon e 7. As shown in Figure 3, the
C17-C27 sulfone 7 was to be prepared by HWE-reaction
of aldehyde 9 with C17-C21 â-oxophosphonate 11. Alto-
gether three approaches to 9 were elaborated, two of
which were based on RCM. In the first one (Scheme 5),11g
we started from glycidyl ether 12, which was regioselec-
tively opened with the lithium salt of ethyl propiolate in
the presence of BF3‚Et2O to give alcohol 41 in near
quantitative yield. Stereoselective addition of lithium
dimethyl cuprate to the triple bond in 41 and in situ
lactonization25a of the resulting (Z)-enoate with acetic acid
in toluene at 80 °C furnished lactone 42 in 91% yield.
Reduction of 42 to the lactol with diisobutyl aluminum-
hydride and in situ ionic hydrogenation43 with triethyl-
silane/BF3‚Et2O at -78 °C gave the dihydropyran 43 in
77% yield. Removal of the PMB group with DDQ pro-
duced the highly water soluble alcohol 44, which was
In a third approach to aldehyde 911i (Scheme 7), two-
(44) Parikh, J . R.; von Doering, W. E. J . Am. Chem. Soc. 1967, 89,
5505-5507.
(45) Kirkland, T. A.; Grubbs, R. H. J . Org. Chem. 1997, 62, 7310-
7318.
(46) (a) Kinoshita, A.; Mori, M. Synlett 1994, 1020-1022. (b) Callam,
C. S.; Lowary, T. Org. Lett. 2000, 2, 167-169. (c) Clark, J . S.; Kettle,
J . G. Tetrahedron Lett. 1997, 38, 123-126. (d) Miller, J . F.; Termin,
A.; Koch, K.; Piscpio, A. D. J . Org. Chem. 1998, 63, 3158-3159.
(47) A closely related approach to the C22-C27 fragment of 1 was
independently developed by Ghosh and Wang11c and was later on also
used by other groups.11k,l,n,o,14
(42) Iodide 40 was also an intermediate in Ghosh’s synthesis of 1,12a,c
who independently started from the benzyl protected analogue of
alcohol 31 via a RCM route.
(43) Kursanov, D. N.; Parnes, Z. N.; Loim, N. M. Synthesis 1974,
633-651.
J . Org. Chem, Vol. 68, No. 8, 2003 3031

Ahmed et al.
SCHEME 8. Com p letion of Key F r a gm en t 7
SCHEME 7. Th ir d Ap p r oa ch to Ald eh yd e 9
a
See text and Supporting Information.
directional synthesis48 was applied by starting from the
known diepoxide 50, which is available from the D-
mannitol-acetonide 49 in three high-yielding steps.49
Diepoxide 50 was transformed quantitatively in two steps
into tetraene 51, which was subjected to RCM under high
dilution in CH2Cl2 at room temperature. No medium ring
sized cycloolefins were formed across the central ac-
etonide ring which, under these conditions, served as a
barrier to crossover metathesis.50 In this way the sym-
metric bis-dihydropyran 52 was obtained in 83% yield.
To complete the synthesis of aldehyde 9, the acetonide
ring in 52 was cleaved with trifluoroacetic acid in
dichloromethane at 0 °C (93% yield), and the resulting
diol 53 was oxidized with sodium periodate/silica gel in
CH2Cl2,51 or with lead tetraacetate/sodium carbonate in
CH2Cl2,52 or with periodic acid in Et2O. Simple filtration
and careful evaporation of the low-boiling solvents gave
aldehyde 9 in excellent purity and yield. This approach
had the advantage of generating the highly volatile
aldehyde in higher purity under neutral conditions,
compared to preparation from alcohol 44.
started from PMB-protected R-hydroxy butyrolactone
54,53 which is readily available from natural (S)-malic
acid.54 Lactone 54 was treated with an equimolar amount
of the lithium salt of diethyl methanephosphonate. The
resulting anionic adduct was further deprotonated with
1 equiv of LDA leading to dianion 55a , which was
silylated with 2 equiv of TESCl. The silylenol ether in
bis-silyl ether 55b was selectively hydrolyzed during
workup with aqueous NH4Cl solution, providing the
triethylsilyl-protected â-oxophosphonate 11 without iso-
lation of intermediates in 86% yield.55 HWE olefination
of phosphonate 11 with aldehyde 9 under Masamune-
Roush conditions56 with 2.5 equiv of triethylamine and
lithium chloride in THF at 0 °C afforded enone 56
E-stereoselectively (E:Z > 40:1) in high yield. The Fel-
Com p letion of Key F r a gm en t 7. As illustrated in
Scheme 8, the synthesis of the chiral phosphonate 11
(48) Poss, C. S.; Schreiber, S. L. Acc. Chem. Res. 1994, 27, 9. For a
review, see: Magnuson, S. R. Tetrahedron 1995, 51, 2167-2213. For
some recent examples of two-directional RCM, see: (a) Burke, S. D.;
Quinn, K. J .; Chen, V. J . J . Org. Chem. 1998, 63, 8626-8627. (b)
Lautens, M.; Hughes, G. Angew. Chem., Int. Ed. 1999, 38, 129-162.
(c) Baylon, C.; Heck, M.-P.; Mioskowski, C. J . Org. Chem. 1999, 64,
3354-3360. (d) Heck, M.-P.; Baylon, C.; Noian, S. P.; Mioskowski, C.
Org. Lett. 2001, 3, 1989-1991. (e) Clark, J . S.; Hamelin, O. Angew.
Chem., Int. Ed. 2000, 39, 372-374.
(49) Le Merrer, Y.; Dure´ault, A.; Greck, C.; Micas-Languin, D.;
Gravier, C.; Depezay, J .-C. Heterocycles 1987, 25, 541-548.
(50) For a review, see: Maier, M. E. Angew. Chem., Int. Ed. 2000,
39, 2073-2077. For the clean conversion of a bis-homoallylic alcohol
derived from diepoxide 50 to the corresponding cyclooctene derivative
by RCM, see: Gravier-Pelletier, C.; Andriuzzi, O.; Le Merrer, Y.
Tetrahedron Lett. 2002, 43, 245-248.
(53) Mulzer, J .; Mantoulidis, A.; O¨ hler, E. J . Org. Chem. 2000, 65,
7456-7467.
(54) Collum, D. B.; McDonald, J . H.; Still, W. C. J . Am. Chem. Soc.
1980, 102, 2118-2120.
(55) For related protocols, see: (a) Ditrich, K.; Hoffmann, R. W.
Tetrahedron Lett. 1985, 26, 6325-6328. (b) Hanessian S.; Roy, P. J .;
Petrini, M.; Hodgesi, P. J .; Di Fabrio, R.; Carganico, G. J . Org. Chem.
1990, 55, 5766-5777.
(56) Blanchette, M. A.; Choy, W.; Davis J . T.; Essenfeld, A. P.;
Masamune, S.; Roush W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183-2186.
(51) Zhong, Y. L.; Shing, T. K. M. J . Org. Chem. 1997, 62, 2622-
2624.
(52) Banwell, M. G.; Forman, G. S. J . Chem. Soc., Perkin Trans. 1
1996, 2565-2566.
3032 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
SCHEME 9. F r a gm en t Un ion a n d
Ma cr ocycliza tion
kin-Anh selective reduction of the C20-carbonyl group
in 56 in the presence of the chelating vicinal PMB-ether
was a challenging task. Preliminary experiments with
L-Selectride were hampered by tedious workup, and
standard Luche reduction57 at -78 °C furnished an
unsatisfactory 5.6:1 syn:anti ratio of epimers 57. Finally,
optimal conditions were found by performing the Luche
reduction at -95 °C under slow addition of sodium
borohydride to a vigorously stirred mixture of enone 56
and CeCl3‚7H2O in methanol. This procedure resulted in
1
an 8:1 mixture (monitored by H NMR) in favor of the
desired epimer syn-57 in 99% combined yields. After
separation by HPLC, the C20-(S)-configuration of syn-57
was confirmed by Mosher-ester analysis,58 and the un-
desired epimer anti-57 was recycled by oxidation with
sulfur trioxide-pyridine complex.44 Alcohol syn-57 was
converted to alcohol 58 in 93% yield by sequential MOM
protection of the secondary hydroxyl group and depro-
tection of the primary triethylsilyl ether with tetrabuty-
lammonium fluoride. Alcohol 58 was then treated with
1-phenyl-1H-tetrazol-5-thiol (PT-SH) under Mitsunobu
conditions59 to give sulfide 59 in 90% yield. Careful
oxidation of thioether 59 with hydrogen peroxide in the
presence of ammonium heptamolybdate tetrahydrate in
ethanol60 produced after 4.5 h a mixture of sulfone 7,
unreacted thioether 59, and the intermediate sulfoxides,
from which most of the crystalline sulfone 7 could be
separated by filtration. The residue was resubmitted to
oxidation to produce a second crop of sulfone in 74%
combined yield.61 Thus, the C17-C27 fragment 7 was
available from butyrolactone 54 in six steps and 42%
overall yield, and from glycidyl ether 45, respectively, in
10 steps and 38% total yield.
F r a gm en t Assem bly a n d Com p letion of th e Syn -
th esis. For the completion of the synthesis, fragments 7
and 8 were connected by using an E-selective one-step
J ulia-Kocienski olefination21,22 (Scheme 9). Thus, the
C16-OPMB ether in 30 was removed with DDQ to give
alcohol 60 in 89% yield. Swern62 or Dess-Martin63
oxidation of 60 provided crude aldehyde 8, which was
treated with the anion derived from sulfone 7 and
KHMDS. When the olefination was performed in 1,2-
dimethoxyethane (DME) at -60 °C, we obtained E/Z-
mixtures of olefins 61 varying from 11.4:1 to 6.3:1 in 60-
67% combined yields, depending on slight variations of
the reaction conditions (deprotonation time for sulfone
7, rate of rise in temperature after union of components).
In THF as solvent, the yield rose to 71%, but the E/Z
ratio dropped to 2.8:1. Isomer E-61 was readily separated
a
For reaction conditions, yield, and ratio of isomers, see text
and Supporting Information.
by chromatography, and the PMB ether was cleaved with
DDQ to give alcohol 62 in 89% yield, which was acylated
with bis(2,2,2-trifluoroethyl) chlorocarbonylmethyl-phos-
phonate64 to afford phosphonate 63 in 91% yield. Treat-
ment of 63 with aqueous acetic acid in THF at room
temperature removed the silyl ether smoothly in 3.5 h
to generate alcohol 64 in 95% yield.65 Oxidation with
Dess-Martin periodinane63 provided cyclization precur-
sor 4 in 96% yield, which was subjected to an intramo-
lecular Horner-Emmons olefination using Still’s opti-
mized base system20 (0.95 equiv of KHMDS, 18-crown-6,
(57) Luche, J .-L. J . Am. Chem. Soc. 1978, 100, 2226-2227.
(58) (a) Dale, J . A.; Mosher, H. S. J . Am. Chem. Soc. 1973, 95, 512-
519. (b) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J . Am.
Chem. Soc. 1991, 113, 4092-4096. (c) Rieser, M. J .; Hui, Y.-h.;
Rupprecht, J . K.; Kozlowski, J . F.; Wood, K. V.; McLaughlin, J . L.;
Hanson, P. R.; Zhuang, Z.; Hoye, T. R. J . Am. Chem. Soc. 1992, 114,
10203-10213, and references therein.
(59) Mitsunobu, O. Synthesis 1981, 1-28.
(60) Schultz, H. S.; Freyermuth, H. B.; Buc, S. R. J . Org. Chem.
1963, 28, 1140-1142.
(61) Prolonged reaction times resulted in olefin epoxidation products.
In a recent synthesis of the C15-C27 fragment of 1, Davidson’s group11l
also observed competing formation of a C25-C26 epoxy derivative during
oxidation of a PT-thioether derived from alcohol 44. However, the
epoxide could be reconverted to the desired alkene, by treatment with
Ph3P/I2.
(62) Mancuso, A. J .; Swern, D. Synthesis 1981, 165-185.
(63) Dess, D. B.; Martin, J . C. J . Am. Chem. Soc. 1991, 113, 7277-
7287.
(64) The acylating agent was prepared from commercially available
methyl ester by means of enzymatic hydrolysis with PLE (Fluka 46058)
and subsequent reaction of the acid with oxalyl chloride.
(65) Attempted desilylation of 63 with TBAF resulted in decomposi-
tion of the phosphonate moiety.
J . Org. Chem, Vol. 68, No. 8, 2003 3033

Ahmed et al.
SCHEME 10. Syn th esis of th e Deoxyla u lim a lid es 3
a n d 67
THF, 50 min, -78 °C). This procedure led to a disap-
pointing 1.8:1 E/Z-mixture of macrolactones 65 and 66
in 80% yield. Alternatively, treatment of phosphonate-
aldehyde 4 with K2CO3 (6 equiv)/18-crown-6 (12 equiv)
for 1 h in toluene at room temperature66 provided the
olefination products 65 and 66 quantitatively, but did not
improve the isomeric ratio (E:Z ) 2.1:1).67 Separation of
65 and 66 by chromatography on silica gel, followed by
removal of both MOM groups with dimethylboron bro-
mide (3 equiv) in dichloromethane at -78 °C,68 generated
the deoxylaulimalides 67 and 3 in 96 and 85% yield,
respectively (Scheme 10).
Secon d -Gen er a tion Syn th eses (Rou tes 2 a n d 3):
F r a gm en t Un ion by Allyl Tr a n sfer , Ma cr ocycliza -
tion by Ya m a gu ch i La cton iza tion (Rou te 2). To
obtain the Z-enoate moiety stereoselectively, an alterna-
tive strategy was developed by envisaging an asymmetric
allyl transfer for the connection of appropriately func-
tionalized C15-C27 and C3-C14 fragments11g,h,13c (Figure
4). The simplest version of the C15-C27 part would have
been aldehyde 74, where the allylsilane or allylstannane
moiety would add in the presence of a chiral Lewis
acid69,70 to achieve notable stereocontrol at the stereogenic
center C15. This option, however, did not appeal to us,
considering the known toxicity of organostannanes and
the uncertain stereodirecting efficiency of the chiral
catalyst. Rather, derivatives 6911g and 7313c were taken
into consideration, both of which could provide the
desired stereochemistry at C15, either in the form of
Felkin-Anh type 1,2-induction (69)71 or via a diaste-
reotopos-selective SN2-type opening of the chiral acetal
F IGURE 4. Second retrosynthetic analysis.
in 73.72-75 For the C3-C14 allylic part two variations were
obvious, namely bromide 70, which would call for a
Nozaki-Hiyama-type addition76 to epoxy aldehyde 69,
or the silanes 71/72, which could be added to acetal 73
under Lewis acid catalysis.72-75
Syn th esis of C15-C27 F r a gm en ts. First, we aimed
for the synthesis of epoxy aldehyde 69 (Scheme 11).11g
Thus, our previous intermediate 41 was protected as the
TBDPS ether 75 and the ester group in 75 was reduced
(DIBALH, CH2Cl2, -78 °C) to provide the primary alcohol
76 (91% yield), which was then converted to the THP
ether 77. Next, the PMB ether in 77 was cleaved with
DDQ to give alcohol 78. Oxidation of 78 with pyridine-
SO3 complex,44 followed by Pinnick oxidation to the acid77
and esterification with diazomethane gave methyl ester
(71) Similarly to our retrosynthetic design,11g Williams16b utilized
an analogue of epoxyaldehyde 69, which was treated with an allylsilane
to provide the (15S)-coupling product selectively in 53% yield. The
groups of Davidson11l and Crimmins16a synthesized C15-C27 epoxyal-
dehyde fragments analogous to 69 but with (19R)-stereochemistry, as
the coupling partners. To date, the critical allylation reaction was
performed only by Crimmins and gave a rather low 3:1 stereoselec-
tivity.
(72) (a) Bartlett, P. A.; J ohnson, W. S.; Elliott, J . D. J . Am. Chem.
Soc. 1983, 105, 2088-2090. (b) Choi, V. M. F.; Elliott, J . D.; J ohnson,
W. S. Tetrahedron Lett. 1984, 25, 591-594. (c) J ohnson, W. S.;
Crackett, P. H.; Elliott, J . D.; J agodzinski, J . J .; Lindell, S. D.;
Natajaran, S. Tetrahedron Lett. 1984, 25, 3951-3954.
(66) Improved Z-selectivity at higher temperatures was observed
during phorboxazole A synthesis.19c
(67) For similar results obtained with differently protected ana-
logues of aldehyde phosphonate 4, see refs 12a,c.
(68) Guindon, Y.; Yoakim, C.; Morton, H. E. J . Org. Chem. 1984,
49, 3912-3920.
(69) (a) Ishihara, K.; Mouri, M.; Gao, Q.; Maruyama, T.; Furuta,
K.; Yamamoto, H. J . Am. Chem. Soc. 1993, 115, 11490-11495. (b)
Marshall, J . A.; Tang, Y. Synlett 1992, 653-654. (c) Keck, G.; Tarbet,
K. H.; Geraci, L. S. J . Am. Chem. Soc. 1993, 115, 8467-8468. (d) Costa,
A. L.; Piazza, M. G.; Tagliavini, E.; Trombini, C.; Umani-Ronchi, A. J .
Am. Chem. Soc. 1993, 115, 7001-7002. (e) Yanagisawa, A.; Na-
kashima, H.; Ishaba, A.; Yamamoto, H. J . Am. Chem. Soc. 1996, 118,
4723-4724 and references therein.
(73) For a review on chiral acetals derived from optically active
alcohols, see: Alexakis, A.; Mangeney, P. Tetrahedron: Asymmetry
1990, 1, 477-511.
(74) For extensive studies on the mechanism and origin of stereo-
selective opening of chiral dioxane acetals, and for leading references,
see: (a) Denmark, S. E.; Willson, T. M.; Almstead, N. G. J . Am. Chem.
Soc. 1989, 111, 9258-9260. (b) Denmark, S. E.; Almstead, N. G. J .
Org. Chem. 1991, 56, 6485-6487. (c) Denmark, S. E.; Almstead, N. G.
J . Am. Chem. Soc. 1991, 113, 8089-8110.
(75) For recent examples, see: (a) Yamamoto, Y.; Abe, H.; Nishii,
S.; Yamada, J .-i. J . Chem. Soc., Perkin Trans. 1 1991, 3253-3257. (b)
Chen, S.-H.; Sun, X.; Boyer, R.; Paschal, J .; Zeckner, D.; Current, W.;
Zweifel, M.; Rodriguez, M. Bioorg. Med. Chem. Lett. 2000, 10, 2107-
2110.
(70) A TBS-protected analogue of aldehyde 74 has been
a key
fragment in Wender’s recent total synthesis of laulimalide,15 and was
connected stereoselectively with a 5-vinyl-substituted analogue of allyl
silanes 71 and 72 by means of Yamamoto’s chiral acyloxyborane.69a
In the cyclization step, Wender and co-workers referred to the work of
Ghosh12b,c by lactonization of a hydroxy alkynoic acid.
(76) For reviews, see: (a) Cintas, P. Synthesis 1992, 248-257. (b)
Wessjohann, L. A.; Scheid, G. Synthesis 1999, 1-36.
3034 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
SCHEME 11. Syn th esis of C15-C27 F r a gm en ts 86
a n d 73
syn-alcohol 82 as a single diastereomer. However, exten-
sive silyl migration from C19-O to C20-O was observed.
This problem was overcome by applying Luche condi-
tions57 under which silyl migration was reduced to less
than 5%. The cerium(III) counterion significantly reduced
the nucleophilicity of the alkoxide, and alcohol 82 was
isolated in 80% yield. MOM-protection of the newly
formed hydroxy group led to 83 in 90% yield. Selective
removal of the THP group in 83 with 2.5% aqueous HCl
in methanol provided propargylic alcohol 84 in 89% yield.
Reduction of the triple bond in 84 with Red-Al in diethyl
ether furnished stereoselectively the (E)-allylic alcohol
85 in 70% yield. Finally, Sharpless epoxidation18 with (+)-
diethyl tartrate (DET) gave epoxy alcohol 86 in 46% yield
and a diastereomeric ratio of 94:6 (determined by 1H
NMR). At that point we thought it advisable to test
conditions for removing the MOM group at C20-O in the
presence of the C16-C17 epoxide, using the TBS-protected
derivative 87 as a model compound. However, to our
dismay, even under the mildest conditions available (Me2-
BBr, CH2Cl2, EtNiPr2, -78 °C) formation of the tetrahy-
drofuran 88 was observed, so that this approach was
abandoned. Instead, allylic alcohol 85 was oxidized to
aldehyde 74 with the Dess-Martin periodinane63 and
transformed to acetal 73 in 98% yield by reaction with
commercially available (R,R)-(+)-pentane-2,4-diol in the
presence of montmorillonite K-10 clay32 in toluene under
azeotropic removal of water.
A more efficient approach to C15-C27 aldehyde 7413c is
illustrated in Scheme 12. Commercially available (S)-R-
hydroxybutyrolactone 8954 was protected as the TBDPS
ether 90 (93% yield), which was transformed to â-oxo-
phosphonate 91 by the one-pot procedure described in
detail for analogue 11 in Scheme 8. HWE-olefination of
91 with aldehyde 9 (prepared by oxidation of alcohol 44)
under Masamune-Roush conditions56 led stereoselec-
tively to (E)-enone 92 in 81% yield. In contrast to the
PMB-protected analogue 56, 1,2-reduction of 92 under
Luche conditions57 delivered the desired syn-alcohol 93
1
with a diastereoselection of 20:1 (by H NMR) in 94%
yield. No problems with silyl migration were encountered
when the reaction was performed in methanol at -78 °C
and quenched after 10 min with a solution of ammonium
chloride in methanol. Next, the secondary OH-function
in 93 was protected as the MOM-ether by using a large
excess of MOMCl and Hu¨nig’s base in the presence of
tetrabutylammonium iodide in DMF. Without purifica-
tion, the sensitive primary triethylsilyl-ether was cleaved
with a catalytic amount of TsOH in methanol at room
temperature to provide alcohol 94 in 85% yield after two
steps. Parikh-Doering oxidation44 of 94 yielded aldehyde
95 in 98% yield. Chain elongation to (E)-enal 74 was
achieved by a stereoselective HWE-olefination of 95 with
commercial available phosphonate 96 (NaH, THF, 92%
yield) to give Weinreb amide 97, which on reduction with
DIBALH in THF at -78 °C furnished aldehyde 74 in 89%
yield. Compared with other syntheses11k,l,14-16 this ap-
proach to a C15-C27 aldehyde is by far the most effective
with respect to the number of steps (8 steps from
butyrolactone 89, 13 steps from trityl glycidol 45) and
overall yield (41% from 89, 32% from 45).
79 in 75% overall yield. Reaction of ester 79 with the
lithium salt of dimethyl methanephosphonate afforded
the â-oxophosphonate 80 in 91% yield. HWE-olefination
of 80 with aldehyde 9 was first performed with lithium
hydroxide, prepared in situ from n-butyllithium and
water, and led selectively to (E)-olefin 81 in 80% yield.11g
The reaction between phosphonate 80 with aldehyde 9
in THF in the presence of 2 equiv of lithium chloride and
triethylamine56 afforded enone 81 in 96% yield.11i The
high yield in this case presumably results mainly from
the fact that 9 had been prepared by glycol cleavage of
5311i and not by Parikh Doering oxidation of 44. Reduc-
tion of the C20 carbonyl group with L-Selectride yielded
Syn th esis of th e Allylic C3-C14 F r a gm en ts 70-72.
Although the prospect of performing the envisaged
Hiyama-Nozaki coupling of allyl bromide 70 to epoxy-
(77) Bal, B. S.; Childers, W. E.; Pinnick, H. W. Tetrahedron 1981,
37, 2091-2096.
J . Org. Chem, Vol. 68, No. 8, 2003 3035

Ahmed et al.
SCHEME 12. Im p r oved Ap p r oa ch to th e C15-C27
yield for the two steps). In this way, aldehyde 98 was
converted into epoxide 99 with an overall yield of 76%.
For the conversion of 99 into lactone 103, we first used
Ghosez’s lactonization method,27 following the protocol
that had been previously applied to epoxide 18 (cf.
Scheme 2). However, due to concomitant attack on the
trisubstituted double bond, the acid-promoted lactoniza-
tion step could not be performed with TFA. When we used
TsOH instead (toluene, rt), lactone 103 was obtained in
yields up to 85%. Alternatively, the trusted method81
previously used for the synthesis of the “upper” dihydro-
pyran fragment (cf. Scheme 5) was also applied with
success. Regioselective ring opening of epoxide 99 with
lithium ethyl propiolate (THF, -95 °C) mediated by BF3‚
Et2O led to alkynoate 102 in 91% yield. Lindlar hydro-
genation in ethanol in the presence of quinoline afforded
the (Z)-enoate, which was cyclized in situ to form lactone
103 in 96% yield. The electron-rich double bond in 103
was selectively epoxidized with m-chloroperbenzoic acid,
the resulting epoxide was opened with perchloric acid in
THF/H2O, and the diol thus obtained was cleaved with
NaIO4 to provide aldehyde 104 in 77% overall yield. To
introduce the C13 methylene group, 104 was subjected
to Eschenmoser methylenation,82 which led to aldehyde
105 in 66% yield. Reduction of 105 with DIBALH
converted the enal into the allylic alcohol and the lactone
into the corresponding lactol, which was transformed
without purification into ethyl glycoside 106 with ethanol
in the presence of TsOH (78% overall yield). After
acetylation, compound 107 was ready for the introduction
of the C2-C3 appendage. Commercially available vinyl-
OTMS was added in the presence of lithium perchlorate83
to provide the corresponding C3 aldehyde as a single
F r a gm en t; Syn th esis of Ald eh yd e 74
a
For reaction conditions, see text.
aldehyde 69 appeared rather bleak for the reasons
outlined above, the synthesis of 70 was completed
nonetheless. Our approach (Scheme 13a,b)11h was based
on the similarity of the C9-C14 section of laulimalide with
naturally occurring (-)-citronellal 98.78
1
epimer (by H NMR), which was reduced to alcohol 108
with NaBH4 (85% yield over the two steps). TBS-
protection delivered intermediate 109 (73% yield), which
was selectively deprotected at the C14-OAc position with
potassium carbonate in methanol (85% yield) to give
alcohol 110. Standard bromination (CBr4, PPh3, MeCN)
led to bromide 70 in 56% yield.
Among several options, conversion of aldehyde 98 into
epoxide 99 appeared attractive for introduction of the
laulimalide stereochemistry at C9 and subsequent elabo-
ration of the dihydropyran moiety. The epoxide, in turn,
should be generated in diastereomerically pure form by
J acobsen’s hydrolytic kinetic resolution (HKR).79 In this
event, aldehyde 98 was converted to an 1:1 mixture of
epoxide diastereomers 99 and 9-epi-99 via Corey’s sul-
fonium ylide addition.80 J acobsen-HKR with catalyst 100
in tert-butyl methyl ether (TBME) for 36 h led to the
formation of diol 101 along with the desired epoxide 99,
both diastereomerically pure according to standard cri-
The synthesis of allylsilane 7113b started from com-
mercially available ethyl hydrogen (R)-3-methylglutarate
111 (Scheme 14). The carboxyl group in 111 was reduced
regioselectively with BH3‚Me2S complex in THF, and the
resulting alcohol was oxidized with the Dess-Martin
periodinane63 to give aldehyde 11284 in 97% yield after 2
steps. Brown’s asymmetric allylboration39 under the
conditions described in Scheme 4 for the conversion of
33 to 34 furnished homoallylic alcohol 113 diastereose-
1
1
teria (HPLC, H and 13C NMR spectra). Diol 101 was
lectively (de ) 91%, by H NMR of the crude product) in
effectively transformed to epoxide 99 by a dehydrative
cyclization under inversion of configuration at C9, via
regioselective protection of the primary hydroxy group
(TBDPSCl, imidazole, DMF, 94% yield), followed by
mesylation of the secondary alcohol, silyl deprotection
with TBAF, and ring closure with sodium hydroxide (89%
87% yield. Under the conditions necessary for transac-
etalization with acrolein diethylacetal, ester 113 under-
went lactonization. Therefore, 113 was transformed to
the more stable dimethylamide 114, which was converted
by transacetalization and subsequent RCM of the result-
ing dienes 115 into an anomeric mixture (1:1, by 1H
(78) The synthetic potential of (-)-citronellal (98) was also used by
Davidson11j and Crimmins16a to prepare C1-C14 allyl stannane frag-
ments of 1.
(79) (a) J acobsen, E. N.; Tokunaga, M.; Larrow, J . F.; Kakiuchi, F.
Science 1997, 277, 936-938. (b) J acobsen, E. N.; Furrow, M. E.; Schaus,
S. E. J . Org. Chem. 1998, 63, 6776-6777. (c) J acobsen, E. N.;
Hinterding, K. J . Org. Chem. 1998, 63, 2164-2165.
(80) Borredon, E.; Delmas, M.; Gaset, A. Tetrahedron Lett. 1982,
23, 5283-5286.
(81) For related protocols, see: Oizumi, M.; Takahashi, M.;
Ogasawara, K. Synlett 1997, 1111-1113. Also see ref 25a.
(82) For a related protocol, see: Paquette, L.; Dyck, B. P. J . Am.
Chem. Soc. 1998, 120, 5953-5960.
(83) Grieco, P. A.; Speake, J . D. Tetrahedron Lett. 1998, 39, 1275-
1278.
(84) The methyl ester analogue of aldehyde 112, available from (R)-
citronellic acid, has been used in Wender’s recent total synthesis of
115 to prepare a close analogue of allyl silanes 71 and 72.
3036 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
SCHEME 13. (a ) J a cobsen HKR of Ep oxid e 99 a n d (b) Syn th esis of Allyl Br om id e 70
NMR) of the ethyl glycosides 116. Introduction of the side
chain at C5 was achieved as before by treatment with
vinyl-OTBS, using a solution of lithium perchlorate in
ethyl acetate as the Lewis acid.83 Aldehyde 117, obtained
subjected to the reaction with trimethylsilylmethylmag-
nesium chloride (6 equiv) in the presence of Pd(PPh3)4
(30 mol %) to give, after 1 h, an inseparable mixture of
compound 71 and its ∆12,13 isomers. Quite obviously, the
large amount of the catalyst as well as the long reaction
time led to isomerization. The similarity of the described
protocol to the Stille coupling87 prompted us to perform
the reaction in the presence of lithium chloride. We were
pleased to observe that in the presence of 5 equiv of LiCl
and only 5 mol % of Pd(PPh3)4, triflate 120 reacted with
1
as a single epimer (according to H NMR of the crude
product) in 88% yield, was reduced with sodium borohy-
dride and the resulting alcohol protected as triethylsilyl
ether 118 (96% yield for the 2 steps). Treatment of amide
118 with methyllithium in Et2O at -78 °C provided
methyl ketone 119 in excellent yield.
For the introduction of the allylsilane moiety, methyl
ketone 119 was converted to the kinetic enolate (KH-
MDS, 1.5 equiv) and treated with PhNTf2 (1.6 equiv)85
to afford enol triflate 120 regioselectively in 79% yield.
Next, following Kuwajima’s protocol,86 compound 120 was
(86) (a) Morihira, K.; Hara, R.; Kawahara, S.; Nishimori, T.;
Nakamura, N.; Kusama, H.; Kuwajima, I. J . Am. Chem. Soc. 1998,
120, 12980-12981. (b) Kusama, H.; Hara, R.; Kawahara, S.; Nishimori,
T.; Kashima, H.; Nakamura, N.; Morihira, K.; Kuwajima, I. J . Am.
Chem. Soc. 2000, 122, 3811-3820.
(87) Stille, J . K.; Scott, J . W. J . Am. Chem. Soc. 1986, 108, 3033-
3040.
(85) Scott, W.; McMurry, E. Acc. Chem. Res. 1988, 21, 47-59.
J . Org. Chem, Vol. 68, No. 8, 2003 3037

Ahmed et al.
SCHEME 14. Syn th esis of Allylsila n e 71
transformed into allylsilane 72 via enoltriflate 123 as
described before.
Com p letion of Rou te 2. Ya m a gu ch i Ma cr ola cton -
iza tion . Extensive experimentation was necessary to
obtain satisfactory results in the crucial allylation of
acetal 73 with allylsilanes 71 or 72. Traces of water and/
or HCl arising from handling the moisture-sensitive
Lewis acids (SnCl4/DMF, EtAlCl2, TiCl4) led to concomi-
tant loss of the protective groups (C3-OTES, C3-OTBS,
C20-OMOM). Moreover, the chiral auxiliary was partially
lost, so that the coupling proceeded with low stereocon-
trol. Reproducible results were achieved by treating a
solution of TiCl4 (from a freshly opened bottle) in CH2-
Cl2 with a few drops of Et3N, and adding an aliquot (1.2
equiv) of the resulting mixture via syringe filter to a
precooled (-60 °C) solution of acetal 73 (1 equiv) and the
more stable allylsilane 72 (1.2 equiv). This procedure
reproducibly furnished the fully protected coupling prod-
1
uct 124 (de ) 92%, monitored by H NMR) in 65% yield.
After Dess-Martin oxidation of alcohol 124 to methyl
ketone 125, base-induced retro-hetero-Michael reaction
with potassium carbonate in methanol at room temper-
ature88 provided alcohol 126 in 89% yield. Alcohol 126
was protected as the MOM ether 127 in high yield with
20 equiv of MOMCl in DMF in the presence of tetrabu-
tylammonium iodide. Next, both silyl groups in 127 were
removed with TBAF and replaced by triethylsilyl ethers
to give intermediate 129 (85% yield for the 2 steps).
Selective Swern oxidation of the primary TES ether89 in
129 afforded aldehyde 130 in excellent yield.
To complete the carbon skeleton, aldehyde 130 was
subjected to an Ando-Horner-Emmons olefination90
with 2-trimethylsilylethyl (TSE) (diphenoxyphosphoryl)-
acetate.91 Deprotonation of the phosphonate with KH-
MDS and reaction with aldehyde 130 in the presence of
18-crown-6 in THF at -78 °C led to the desired (Z)-enoate
131 in 80% yield. Treatment of the fully protected
intermediate 131 with TBAF in THF removed both the
C19-OTBDPS ether and the TSE ester, and provided seco-
acid 5 in 89% yield. However, Yamaguchi macrocycliza-
tion92 was accompanied by Z/E-isomerization of the 2,3-
enoate, and we isolated the MOM-protected macrolides
65 and 66 as an E/Z-mixture (E/Z ) 2.7:1) in 60%
combined yields.
SCHEME 15. Syn th esis of Allylsila n e 72
Base-induced C2,3-Z/ E isomerization during Yamagu-
chi-type macrolactonization has been previously observed
by Roush during a synthesis of verrucarin B,93 and was
assumed to occur through a reversible Michael addition
of nucleophilic reagents to the active ester intermediate.
The same Z/ E isomerization was also painfully experi-
enced by Ghosh, who submitted a close analogue of seco-
acid 5 to cyclization,12c and by Paterson during a syn-
(88) Sugimura, T.; Yoshikawa, M.; Futagawa, T.; Tai, A. Tetrahedron
1990, 46, 5955-5966.
(89) Rodriguez, A.; Nomen, M.; Spur, B. W.; Godfroid, J . J . Tetra-
hedron Lett. 1999, 40, 5161-5164 and references therein.
(90) Ando, K. J . Org. Chem. 1999, 64, 8406-8408 and references
therein.
(91) The olefination reagent was prepared in two steps by acylation
of 2-TMS-ethanol with bromoacetyl bromide and treatment of the
resulting bromoacetate with diphenyl phosphite in the presence of
triethylamine.90
(92) Inanaga, J .; Hirata, K.; Saeki, H.; Katsuki, T.; Yamaguchi, M.
Bull. Chem. Soc. J pn. 1979, 52, 1989-1993.
(93) Roush, W. R.; Blizzard, T. A. J . Org. Chem. 1984, 49, 4332-
4339 and references therein.
TMSCH2MgCl (2 equiv) to furnish, after only 10 min,
pure allylsilane 71 in 97% yield.
For the synthesis of allylsilane 7213c (Figure 4 and
Scheme 15), intermediate 39 was transformed to the
cyanide 121 in 87% yield by treatment of the tosylate
with sodium cyanide in DMSO at 80 °C. Treatment of
nitrile 121 with methyllithium in diethyl ether at 0 °C
provided methyl ketone 122 (96% yield), which was
3038 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
SCHEME 16. Dia ster eoselective Cou p lin g of
F r a gm en ts 72 a n d 73 a n d Ya m a gu ch i
Ma cr ola cton iza tion of Seco Acid 5
F IGURE 5. Third retrosynthetic analysis.
ether 71 was cleaved with K2CO3 in methanol at 0 °C to
give alcohol 134 in 96% yield, which was subjected to
Dess-Martin oxidation63 to give aldehyde 133 in 87%
yield. For the synthesis of key fragment 132, the TBDPS
ether in acetal 73 was removed with TBAF and the
resulting alcohol 135 acylated with bis(2,2,2-trifluoroet-
hyl) chlorocarbonylmethyl-phosphonate64 to form phos-
phonate 132 in 96% yield.
F r a gm en t Con n ection a n d Com p letion of th e
Syn th esis. To connect fragments 132 and 133 (Scheme
18), phosphonate 132 was deprotonated with KHMDS in
THF at -78 °C, in the presence of 18-crown-6 (5 equiv),20
carefully avoiding an excess of base. Then, freshly
prepared aldehyde 133 was added slowly at the same
temperature. Under these conditions, (Z)-enoate 6 was
1
obtained isomerically pure (monitored by H NMR) in
85% yield. A 4:1 (E/Z)-mixture was obtained if more than
1 equiv of KHMDS was used, and the reaction time was
extended over 1 h at -78 °C.
The cyclization was performed by adding seco-com-
pound 6 slowly at -50 °C to a 4 × 10-4 M solution of
ethylaluminum dichloride94 in dichloromethane, provid-
ing macrocycle 137 as a single isomer in 82% yield. On
monitoring the reaction by TLC, a transient intermediate
was observed that was isolated and identified as the
desilylated seco-compound 136. We thus reasoned that
the conversion of 6 to 137 had proceeded via two parallel
pathways. One would be the direct cyclization of allylsi-
lane 6. The second way would have involved first a
moisture-induced protodesilylation of 6 to 136, which
then underwent cyclization via an ene reaction.95 Indeed,
when we subjected compound 136 directly to the cycliza-
tion conditions described above, macrocycle 137 was
obtained in 56% yield, along with 30% of the starting
material. When repeated under rigorously anhydrous
conditions, the cyclization of 6 proceeded without the
appearance of 136 in 86% yield.
a
See text and Supporting Information.
thesis of the C1-C20 core of 1.11m Ghosh overcame the
problem by applying the Yamaguchi protocol to a 2,3-
alkynoic acid,12b and Paterson used a Mitsunobu macro-
lactonization of a (2Z,19R)-seco acid.11m,14
Rou te 3. F r a gm en t Assem bly by Still-Gen n a r i
Olefin a tion , Ma cr ocycliza tion via Allyl Tr a n sfer . To
avoid the base-induced isomerization of the 2,3-(Z)-
enoate, we decided to reverse the order of the final steps.
As shown in Figure 5, our synthetic design focused now
on compound 6 as the macrocyclization precursor,13b
which should be assembled from phosphonate 132 and
aldehyde 133 by an intermolecular Still-Gennari olefi-
nation.20
The removal of the protective groups from macrocycle
137 required some care. To initiate the envisaged retro-
(94) Snider, B. B.; Rodini, D. J .; Karras, M.; Kirk, T. C.; Deutsch,
E. A.; Cordova, R.; Price, R. T. Tetrahedron 1981, 37, 3927-3934.
(95) For allyl transfer to chiral dioxane acetals via ene reaction,
see: Cambie, R. C.; Higgs, K. C.; Rustenhoven, J . J .; Rutledge, P. S.
Aust. J . Chem. 1996, 49, 677-688.
Key fragments 132 and 133 were readily obtained from
previous intermediates (Scheme 17). Thus, triethylsilyl
J . Org. Chem, Vol. 68, No. 8, 2003 3039

Ahmed et al.
SCHEME 17. Syn th esis of Cou p lin g P a r tn er s 132
a n d 133
SCHEME 19. SAE Con sid er a tion s for
Deoxyla u lim a lid e 3
SCHEME 18. F r a gm en t Con n ection a n d
Com p letion of th e Syn th esis
maining MOM ether in intermediate 140 was then
cleaved as before by treatment with dimethylboron
bromide (5 equiv) in dichloromethane at -78 °C to
provide deoxylaulimalide 3 in high yield. This compound
was identical in all respects with the compound obtained
in our first approach and thus confirmed that the correct
stereochemistry at C15 had be generated in the cyclization
step. Additionally, the conversion of the fully protected
macrocycle 138 to the monoprotected derivatives 139 and
140 by treatment with Me2BBr and TsOH, respectively,
demonstrates the orthogonality of MOM ether and the
4-oxopent-2-yl group, even under acidic conditions.
Ep oxid a tion of 3 a n d 67. As mentioned in the
Introduction, there is a strong tendency of 1 to undergo
isomerization to isolaulimalide (2) even under mildly
acidic conditions. Therefore, it was thought unwise to try
epoxidations of the 16,17-double bond in the presence of
the MOM-protective group at C20-O. Instead we decided
to apply a SAE reaction to “naked” deoxylaulimalide 3
in the hope that the reagent control of this procedure
would suffice to bring about the desired regioselectivity
of the matched C15-C17-allylic alcohol moiety over the
mismatched C20-C22 section (Scheme 19).
Indeed, exposure of 3 to Sharpless epoxidation18 with
natural (+)-(R,R)-diisopropyl tartrate (DIPT) at -20 °C
for 2 h (Scheme 20, eq 1) did provide a 2:1 mixture of 1
and unreacted compound 3, from which 1 was separated
by HPLC in 86% yield, based on recovered starting
material (BORSM). Still, this selectivity might have been
the result of a substrate-mediated epoxidation due to the
exposed position of the C16,17-olefin on the outside of the
macrocyclic ring. Therefore, the epoxidation of 3 was
repeated without the DIPT additive (Scheme 20, eq 2).
Again, 1 was formed selectively; however, the reaction
proceeded much more slowly and took about 18 h. This
underscores the intrinsic preference for the 16,17-epoxi-
dation and the catalytic effect of the tartrate, and leads
to the conclusion that eq 1 represents the case in which
substrate and reagent control are matched to each other.
To test the power of the SAE process, epoxidation of 3
with (-)-(S,S)-tartrate was also performed, which indeed
gave epoxide 142 (eq 3).
Michael reaction under acidic conditions, intermediate
137 was oxidized with Dess-Martin periodinane to
methyl ketone 138 in 90% yield. Treatment of methyl
ketone 138 with Me2BBr68 in dichloromethane at -78 °C
removed the MOM ether at C20-O selectively, leading to
compound 139. Treatment of intermediate 139 with
TsOH in chloroform, however, led to 2,3-Z-E isomeriza-
tion.
Finally, we were successful by reversing the order of
deprotection steps. Thus, treatment of methyl ketone 138
with TsOH (4 equiv) in chloroform at 0 °C for 24 h,
smoothly furnished alcohol 140 in 77% yield. The re-
Figure 6 shows how these selectivities are reflected in
the 1H NMR spectra of compounds 1, 3, and 142. The
spectrum of 3 exhibits both a pseudo-AB pattern centered
3040 J . Org. Chem., Vol. 68, No. 8, 2003

Total Synthesis of Laulimalide
F IGURE 6. 1H NMR spectra (600 MHz) of 1, 3, and 142.
TABLE 1. An tip r olifer a tive Effects of La u lim a lid esa
SCHEME 20. Regio- a n d Ster eocon tr olled
Ep oxid a tion of 3 a n d 67
IC50 [nM]
compd
MCF-7
NCI/ADR
MaTu
MaTu/ADR
3
67
1
89
ni
3.8
54
ni
ni
36
ni
43
ni
3.8
38
170
ni
6.0
250
141
epothilone B
paclitaxel
0.59
3.2
3.5
>1000
0.46
3.3
1.2
600
a
ni: no inhibition measured up to 100 nM. MCF-7: human
breast tumor cell line. NCI/ADR: human multi-drug-resistant
breast tumor cell line. MaTu: human breast tumor cell line. MaTu/
ADR: human multi-drug-resistant breast tumor cell line.
Biologica l Activities. Compounds 1, 3, 67, and 141
were tested for their effects on the proliferation of four
tumor cell lines (Table 1), along with the standard
MSAAs paclitaxel and epothilone B.96,97It turned out that
1 is about as active as paclitaxel, but significantly less
active than epothilone B. The laulimalide derivatives 3
and 141 show much lower activity, and 67 has no activity
at all. Against multidrug resistant tumor cell lines, 1,
by contrast to paclitaxel, still reveals considerable activ-
ity, albeit less than epothilone B.
at 5.63 ppm for the 16,17-vinylic part and the signals of
the 21,22-moiety (H-22, ddd at 5.87 ppm; H-21, ddd at
5.73 ppm; J 21,22 ) 15.6 Hz). In the spectrum of 1, the
signals of the 21,22-unit are still present (H-22, ddd at
5.87 ppm; H-21, ddd at 5.75 ppm), whereas the signals
Con clu sion
In conclusion, we have described three different routes
to deoxylaulimalide 3. Route 1 is not stereocontrolled
with respect to the 2,3-enoate moiety; however, the
overall yield and the reliability and simplicity of the
reactions involved still render this route a viable one.
Route 3 is entirely regio- and stereocontrolled, but it
requires considerable experimental skill as some tricky
reactions are involved. The same is true of route 2, which
for H-17 (ddd at 3.06 ppm) and H-16 (br t with J 16,17
)
2.5 Hz) now appear in the epoxide range. In contrast,
the spectrum of epoxide 142 still shows the AB pattern
of the 16,17-unit, whereas the signals of H-21 (dd with J
) 2.4 and 3.1 Hz at 3.06 ppm) and H-22 (dd with J ) 2.4
and 5.7 Hz at 3.18 ppm) now are shifted to the epoxide
range. As an additional experiment (eq 4), compound 67
was subjected to the “matched” SAE conditions. Not
surprisingly, the 16,17-epoxide 141 was obtained as the
main product, however, along with another unidentified
epoxide.
(96) Stein, U.; Walther, W.; Lemm, M.; Naundorf, H.; Fichtner, I.
Int. J . Cancer 1997, 72, 885-891.
(97) Kueng, W.; Silber, E.; Eppenberger, U. Anal. Biochem. 1989,
182, 6-19.
J . Org. Chem, Vol. 68, No. 8, 2003 3041

Ahmed et al.
additionally has the disadvantage of lacking 2,3-(Z)-
selectivity. Nevertheless, we have presented a broad
spectrum of options for the preparation of 3 and various
major fragments thereof, which should also be applicable
to the synthesis of interesting analogues. For the ultimate
conversion of 3 into the desired natural product (-)-
laulimalide (1), we have shown that SAE reaction may
be used for achieving regioselectivity in a molecule
containing two allylic alcohol moieties with opposite
topicity. Thus, 3 could be alternatively converted into
laulimalide (1) and its regioisomer 142 simply by switch-
ing the chiral additive from (+)- to (-)-tartrate. The
biological tests have shown that among our laulimalide
derivatives 1, 3, 67, and 141 the natural compound 1 is
by far the most active one, and that, compared to
paclitaxel and epothilone B, 1 is superior to paclitaxel,
but inferior to epothilone B.
Ack n ow led gm en t. This paper is dedicated to Prof.
Werner Skuballa on the occasion of his 60th birthday.
We thank Prof. Dr. R. Metternich, Dr. G. Siemeister,
and Dr. U. Klar, Schering AG, Berlin, for providing us
with the biological data and Ing S. Schneider for HPLC
separations. Financial support from the FWF (project
P-13941-CHE) is gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
1
cedures and characterization data for all new compounds, H
and 13C NMR spectra of compounds 1, 3, 5, 6, 7, 8, 10, 18, 30,
syn-57, 62, 64-67, 72, 73, 124, 137, 138, 140, and 141, and
figures showing the effects of 1, 3, 67, and 141 on tumor cell
proliferation. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O026743F
3042 J . Org. Chem., Vol. 68, No. 8, 2003