Design, synthesis and biological evaluation of novel, simplified analogues of laulimalide: Modification of the side chain

Article abstract of DOI:10.1016/j.bmcl.2005.03.018

Novel, simplified analogues of the microtubule-stabilizing anticancer agent laulimalide, including the first derivatives with unnatural side chains, were designed by molecular modelling, synthesized by a late-stage diversification strategy, and evaluated in vitro for growth inhibition of human ovarian carcinoma cell lines (A2780, A2780/AD10).

Full text of DOI:10.1016/j.bmcl.2005.03.018

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2243–2247
Design, synthesis and biological evaluation of novel,
simplified analogues of laulimalide: modification of the side chain
a
a
a
a
Ian Paterson,a, Dirk Menche, Anders E. Hakansson, Adrian Longstaff, David Wong,
*
˚
Isabel Barasoain, Ruben M. Buey and J. Fernando Dıaz
b
b
b
´
´
aUniversity Chemical Laboratory, Lensfield Road, Cambridge CB2 1EW, UK, EU
bCentro de Investigaciones Biologicas, Consejo Superior de Investigaciones Cientificas, Ramiro de Maeztu, 9. 28040 Madrid, Spain, EU
Received 25 January 2005; revised 2 March 2005; accepted 4 March 2005
Abstract—Novel, simplified analogues of the microtubule-stabilizing anticancer agent laulimalide, including the first derivatives with
unnatural side chains, were designed by molecular modelling, synthesized by a late-stage diversification strategy, and evaluated in
vitro for growth inhibition of human ovarian carcinoma cell lines (A2780, A2780/AD10).
Ó 2005 Elsevier Ltd. All rights reserved.
Pacific marine sponges (Fasciospongia rimosa, Hyatella
sp. and Spongia mycofijiensis) are the natural source of
laulimalide (1, Fig. 1), a 20-membered macrolide of
polyketide origin.1,2 It is a highly potent antimitotic
agent, which inhibits proliferation of a range of human
cancer cell lines, including multidrug resistant cells, at
nanomolar concentrations.3 Like TaxolÒ (paclitaxel), it
acts by microtubule stabilization and disruption of mito-
tic spindle formation.3a However, it appears to have a
different, as yet undefined, binding site on tubulin,4,5 is
superior in its ability to circumvent P-glycoprotein-medi-
ated drug resistance, and retains activity against paclit-
axel resistant cells.3a This promising biological profile
renders laulimalide a highly attractive lead for the devel-
opment of new anticancer agents, however, this is se-
verely hampered by its low natural abundance. Despite
considerable synthetic efforts, which have culminated
in a multitude of total syntheses,2,6,7 this supply problem
still remains, such that the development of simplified
and more readily available analogues retaining biologi-
cal activity is an important goal.
active.7 Recently, we have developed a novel syn-
thetic approach towards laulimalide using a versatile
So far, only a limited range of analogues, relying on
modifications at C2–C3, C8, C15–C17 and C20 have
been reported, all of which are however significantly less
Keywords: Laulimalide; Analogues; Anticancer agent; Nozaki–Kishi
coupling; Modelling.
*
Corresponding author. Tel.: +44 1223 336407; fax: +44 1223
336362; e-mail: ip100@cam.ac.uk
Figure 1. Laulimalide and simplified analogues thereof.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.03.018

2244
I. Paterson et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2243–2247
Nozaki–Kishi coupling8 for late-stage introduction of
the dihydropyran side chain and successfully applied
this procedure to access 11-desmethyl-laulimalide (2)
and related structures.7d Herein, we describe an im-
proved protocol for the preparation of 11-desmethyl-
laulimalide, report on the design and synthesis of the
first laulimalide analogues with unnatural (or truncated)
side chains (3–5), and disclose the cancer cell growth
inhibitory activity of all of these compounds relative
to other microtubule-stabilizing agents (laulimalide,
paclitaxel and discodermolide).
The key step in our recent synthesis of 11-desmethyl-lau-
limalide (2, Scheme 2) was a catalytic asymmetric Noza-
ki–Kishi coupling of macrocyclic aldehyde 10 with vinyl
iodide 11 in the presence of DIANANE-type ligand
(R,R)-12.7d,8,13 This overturned the unfavourable sub-
strate-based selectivity in this transformation and gave
the desired allylic alcohol 13 as the major diastereomer.
As an improvement to this protocol, we found that
increasing the nickel content (from 2 to 10 mol % NiCl2)
significantly speeds up this reaction.14 Together with a
modified procedure for cleaving the intermediate TMS
ether by use of HF/pyridine, instead of the previously
used two-step procedure,7d,8 led to a 50% enhancement
of the chemical yield (65% vs 43%) together with slightly
improving the diastereomeric ratio (83:17 vs 78:22). The
minor C20 epimer is not lost as any unfavourable diaste-
reomeric mixture can be used by employing an oxida-
tion–reduction sequence, as shown for 20-epi-13. Here,
it was found that K-Selectride not only leads to higher
chemoselectivity but also better diastereoselectivity, as
compared to previously used L-Selectride and allows
access to diastereomerically pure 13 (dr > 20:1 vs 9:1).15
Completion of the synthesis of 11-desmethyl-laulimalide
Our starting point for rational analogue design was con-
formational analysis (MM2, MacroModel 8.0) of lauli-
malide and related simplified structures, which was
carried out both in vacuo and in solution (chloroform,
water).9 These calculations revealed that deletion of
the 11-methyl group only has a minor impact on the
respective 3D structures. Likewise, changes in the side
chain, for example, by replacing the dihydropyran with
a cyclohexyl group (Fig. 1), or its complete removal,
only had a minor influence on the macrocyclic conform-
ation, as expected. These findings prompted us to evalu-
ate the general importance of the side chain and the
11-methyl group on biological activity, together with
preparing and evaluating side chain modified analogues
3–5.10
As a first target, we decided to delete the side chain alto-
gether and access macrocyclic epoxide 3. Its preparation
started from known lactone 8 (Scheme 1), which was
synthesized in a convergent and scalable manner using
our previously established route from D-malic acid de-
rived C15–C19 building block 6 and the C1–C14 subunit
7.6c Stepwise deprotection of the primary and the
11
secondary hydroxyls of 8 using TBAF and H2SiF6
revealed the secondary allylic alcohol 9 in a straightfor-
ward sequence. A highly selective Sharpless epoxida-
tion6c,e completed the synthesis of the desired
epoxide 3.12
Scheme 2. Synthesis of 11-desmethyl-laulimalide (2) by asymmetric
Nozaki–Kishi coupling. Reagents and conditions: (a) (i) (R,R)-12
(10 mol%), CrCl2 (10 mol%), NiCl2 (10 mol%), Et3N (20 mol%), Mn,
TMSCl, THF, rt; (ii) HF/pyridine, THF, 0 °C; (b) (COCl)2, DMSO,
Et3N, CH2Cl2, ꢀ78 to ꢀ10 °C; (c) K-Selectride, THF, ꢀ78 °C.
Scheme 1. Synthesis of macrocyclic epoxide 3. Reagents and condi-
tions: (a) TBAF/AcOH (pH 7), THF, 0 °C to rt; (b) H2SiF6, CH3CN,
rt; (c) (+)-DIPT, Ti(OiPr)4, t-BuOOH, CH2Cl2, ꢀ20 °C.

I. Paterson et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2243–2247
2245
via 14 by regio- and diastereoselective Sharpless epoxi-
dation proceeded under our previously reported
conditions.7d,16
With these improved protocols in hand, we then pro-
ceeded to synthesize the target laulimalide derivatives
with simplified, unnatural side chains using aldehyde
10 as a common intermediate (Scheme 3). The cyclo-
hexyl analogue 4 was chosen as a sterically similar, but
electronically different mimic of the authentic dihydro-
pyran ring, while the thiazole analogue 5 was selected
for generation of novel structural hybrids with the epo-
thilones.17 Vinyl iodide 15 was readily available from
aldehyde 16 by homologation with the Ohira–Bestmann
reagent,18 hydroalumination of the derived alkyne and
trapping of the intermediate organometallic species with
iodine.19 Generation of 17 likewise involved use of the
Ohira–Bestmann reagent on a thiazole aldehyde ob-
tained by cyclocondensation20 and subsequent reduc-
tion21 from bromoketone 18 and thioacetamide 19.
This time, hydroiodination was effected by hydro-
zirconation and subsequent treatment with iodine.22
Catalytic, asymmetric Nozaki–Kishi coupling of
macrocyclic aldehyde 10 with both 15 and 17 proceeded
smoothly and with preparatively useful yields, in the
presence of ligand (R,R)-12, to give 20 and 21, respec-
tively.23 Notably, the diastereoselectivities were im-
proved as compared to the coupling with vinyl iodide
11, which might be attributed to an influence from the
additional C23 stereocentre or, in the case of 17, the pos-
sibility for further coordination of the thiazole ring to
the catalytically active intermediate chromium(II)-
species. Subsequent TBS deprotection gave 22 and
23, which was followed again by Sharpless epoxidation to
produce 4 and 5, respectively, in a highly selective
manner.12
Scheme 3. Synthesis of the side chain analogues 4 and 5 via common
late-stage intermediate 10. Reagents and conditions: (a) (MeO)2-
POC(N2)C(O)Me, K2CO3, MeOH, rt; (b) (i) DIBALH, hexane,
ꢀ40 °C to 50 °C; (ii) I2, hexane/THF, ꢀ40 °C to rt; (c) (i) KHCO3,
DME, rt; (ii) (CF3CO)2O, pyridine, DME, 0 °C to rt; (d) DIBALH,
CH2Cl2, ꢀ78 °C; (e) (MeO)2POC(N2)C(O)Me, K2CO3, MeOH, rt; (f)
(i) Cp2ZrHCl, CH2Cl2, rt; (ii) I2, CH2Cl2, rt; (g) (i) (R,R)-12 (10 mol%),
CrCl2 (10 mol%), NiCl2 (10 mol%), Et3N (20 mol%), Mn, TMSCl,
THF, rt; (ii) HF/pyridine, THF, 0 °C; (h) HF/pyridine, THF; 0 °C to
rt; (i) (+)-DIPT, Ti(OiPr)4, t-BuOOH, CH2Cl2, ꢀ20 °C.
For biological evaluation of the foregoing analogues, we
first checked if they could promote the polymerization
of tubulin for microtubule assembly, that is, modify
the critical concentration (Table 1).24 For this purpose,
pure GTP-tubulin was incubated, in
a glycerol
Table 1. Effect on in vitro tubulin assembly and inhibitory effects of ligands on growth of human ovarian carcinoma cells
Compds
Critical concentration
GABa,b,c (lM)
Cytotoxicity A2780
d
IC50 (nM)
Cytotoxicity A2780/AD10
d
IC50 (nM)
Laulimalide (1)
2
0.72 0.1
1.1 0.1
3.0 1.2
1.4 0.2
3.6 0.4
3.4 0.3
1.8 0.3
3.8 0.6
3.4 0.4
3.3 0.1
0.46 0.24
0.51 0.09
3.4 1.0
50 13
40,000 1000
9000 310
na
7.5 0.7
74 14
44,000 1500
8300 2300
na
3
4
5
9
14
na
na
430 70
na
na
1100 230
na
na
22
23
DMSO
Paclitaxel
Discodermolide
na
1.1 0.3
10 0.1
na
1500 60
100 19
a GAB: 3.4 M glycerol, 10 mM phosphate, 1 mM ethylene glycol-bis(b-aminoethylether)-N,N,N0,N0-tetraacetic acid (EGTA), 1 mM GTP, 6 mM
MgCl2 and pH 6.7 buffer.
b Average of four measurements (with standard errors).
c All the data measured at 37 °C.
d IC50 were determined after three days exposure to drugs using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide cell proliferation
assay24 and values ( standard error) are means of at least four independent experiments. Na = not active below 50 lM.

2246
I. Paterson et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2243–2247
containing buffer (GAB-1mM GTP) at 37 °C, in the
presence of 10% excess of each of the ligands. Under
these conditions, tubulin can assemble in the absence
of exogenous ligands with a critical concentration of
ꢁ3.3 lM. Out of the analogues examined, only 11-desm-
ethyl-laulimalide (2), its desepoxy-precursor 14 and the
cyclohexyl analogue 4, with laulimalide (1) as a control,
significantly increased tubulin assembly, having critical
concentrations of 1.1 0.1 lM, 1.8 0.3 lM, 1.4
0.2 lM and 0.72 0.1 lM, respectively, (analogue
3 only modified the critical concentration weakly, 3.0
1.2 lM). As in the previous studies, paclitaxel
(critical concentration = 0.46 0.2 lM) modified assem-
bly more powerfully than laulimalide and its ana-
logues.3a All the polymers observed were microtubules.
fully acknowledged. We thank Prof. Dr. A. Berkessel
(University of Cologne) for a gift of DIANANE-type
´
ligand (R,R)-12, Dr. Marıa A. Silva-Martınez (Cam-
´
bridge) for helpful discussions over the modelling
´
´
studies and Matadero Madrid Norte S.A. and Jose Luis
Gancedo S. L. for providing the calf brains from which
the tubulin used was purified.
References and notes
1. (a) Quinoa, E.; Kakou, Y.; Crews, P. J. Org. Chem. 1988,
53, 3642; (b) Corley, D. G.; Herb, R.; Moore, R. E.;
Scheuer, P. J.; Paul, V. J. J. Org. Chem. 1988, 53, 3644; (c)
Jefford, C. W.; Bernardinelli, G.; Tanaka, J.; Higa, T.
Tetrahedron Lett. 1996, 37, 159.
2. For reviews, see: (a) Crimmins, M. T. Curr. Opin. Drug
Table 1 also summarizes the inhibitory effect of laulima-
lide and its analogues, as compared with paclitaxel and
discodermolide, on the growth of the ovarian carcinoma
cell line A2780 and the multidrug-resistant (MDR, P-
glycoprotein overexpressing) cell line A2780/AD10.25
The parental A2780 cells were 3-fold more sensitive to
paclitaxel than to laulimalide. However, the MDRcell
line exhibited a strong resistance to paclitaxel (1360-
fold), while laulimalide had only a small reduction in
potency of 2.2-fold. Among the laulimalide analogues
tested, the most potent is its desmethyl-derivative 2,
followed by its desepoxy precursor 14, the cyclohexyl
analogue 4 and the side chain truncated analogue 3.
Significantly, analogue 2 was only 1.5-fold less potent
in the MDRcell as compared to the parental line. More-
over, 11-desmethyl-laulimalide (2) is of similar potency
to discodermolide26 and represents one of the most ac-
tive laulimalide analogues tested so far.7 The observa-
tion that removal of the epoxide leads to loss in
activity is in agreement with previous data.3a,7b The
other compounds were essentially inactive having an
IC50 above 50 lM.
¨
Discovery Dev. 2002, 5, 944; (b) Mulzer, J.; Ohler, E.
Chem. Rev. 2003, 103, 3753.
3. (a) Mooberry, S. L.; Tien, G.; Hernandez, A. H.;
Plubrukarn, A.; Davidson, B. S. Cancer Res. 1999, 59,
653; (b) Cragg, G. M.; Newman, D. J. J. Nat. Prod. 2004,
67, 232.
4. Pryor, D. E.; OÕBrate, A.; Bilcer, G.; Diaz, J. F.; Wang,
Yu.; Wang, Yo.; Kabaki, M.; Jung, M. K.; Andreu, J. M.;
Ghosh, A. K.; Giannakakou, P.; Hamel, E. Biochemistry
2002, 41, 9109.
5. Peloruside has recently been found to bind to the same
site: Gaitanos, T. N.; Buey, R. M.; Diaz, J. F.; Northcote,
P. T.; Teesdale-Spittle, P.; Andreu, J. M.; Miller, J. H.
Cancer Res. 2004, 64, 5063.
6. (a) Ghosh, A. K.; Wang, Y. J. Am. Chem. Soc. 2000, 122,
11027; (b) Ghosh, A. K.; Wang, Y.; Kim, J. T. J. Org.
Chem. 2001, 66, 8973; (c) Paterson, I.; De Savi, C.; Tudge,
M. Org. Lett. 2001, 3, 213; (d) Paterson, I.; De Savi, C.;
¨
Tudge, M. Org. Lett. 2001, 3, 3149; (e) Mulzer, J.; Ohler,
E. Angew. Chem., Int. Ed. 2001, 40, 3843; (f) Evev, V. S.;
Kaehlig, H.; Mulzer, J. J. Am. Chem. Soc. 2001, 123,
10764; (g) Crimmins, M. T.; Stanton, M. G.; Allwein, S. P.
J. Am. Chem. Soc. 2002, 124, 5958; (h) Williams, D. R.;
Mi, L.; Mullins, R. J.; Stites, R. E. Tetrahedron Lett. 2002,
43, 4841; (i) Nelson, S. G.; Chueng, W. S.; Kassick, A. J.;
Hilfiker, M. A. J. Am. Chem. Soc. 2002, 124, 13654; (j)
Wender, P. A.; Hegde, S. G.; Hubbard, R. D.; Zhang, L.
J. Am. Chem. Soc. 2002, 124, 4956.
In conclusion, we have developed a series of analogues
of laulimalide with simplifications in the macrocycle
(deletion of the 11-methyl group) and also the side chain
(truncation or substitution). The synthesis of these com-
pounds was enabled by a late-stage diversification strat-
egy relying on asymmetric Nozaki–Kishi methodology.
The most potent analogue with low nanomolar IC50 val-
ues was 11-desmethyl-laulimalide. Truncation or substi-
tution of the side chain, however, leads to significant
loss of activity, which suggests that this must be part
of the pharmacophore region. These results, combined
with other analogue studies,7 suggest that it may be pos-
sible to simplify the laulimalide structure, yet still retain
potency.
7. (a) Ahmed, A.; Hoegenauer, K.; Enev, V. S.; Hanbauer,
¨
M.; Kaehlig, H.; Ohler, E.; Mulzer, J. J. Org. Chem. 2003,
68, 3026; (b) Wender, P. A.; Hegde, S. G.; Hubbard, R.
D.; Zhang, L.; Mooberry, S. L. Org. Lett. 2003, 5, 3507;
(c) Gallagher, B. M., Jr.; Fang, F. G.; Johannes, C. W.;
Pesant, M.; Tremblay, M. R.; Zhao, H.; Akasaka, K.; Li,
X.-y.; Liu, J.; Littlefield, B. A. Bioorg. Med. Chem. Lett.
2004, 14, 575; (d) Paterson, I.; Bergmann, H.; Menche, D.;
Berkessel, A. Org. Lett. 2004, 6, 1293; (e) Mooberry, S. L.;
Randall-Hlubek, D. A.; Leal, R. M.; Hegde, S. G.;
Hubbard, R. D.; Zhang, L.; Wender, P. A. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101, 8803; (f) Gallagher, B. M.,
Jr.; Zhao, H.; Pesant, M.; Fang, F. G. Tetrahedron Lett.
2005, 46, 923.
8. (a) Berkessel, A.; Menche, D.; Sklorz, C.; Schro¨der, M.;
Paterson, I. Angew. Chem., Int. Ed. 2003, 42, 1032; (b)
Berkessel, A.; Schro¨der, M.; Sklorz, C. A.; Tabanella, S.;
Vogl, N.; Lex, J.; Neudo¨rfl, J. M. J. Org. Chem. 2004, 69,
3050.
9. 10,000 step conformational searches were carried out with
the generalized Born/surface area (CB/SA) solvent model
and the crystal structure data of Laulimalide (1) as input
geometries; (a) Mohamadi, F.; Richards, N. G. J.; Guida,
Acknowledgements
Financial support from the European Commission (IHP
Networks HPRN-CT-2000-18 and HPRN-CT-2000-
00014; postdoctoral fellowship to D.M.), the EPSRC
(GR/N08520), Merck and the Ministerio Educacion y
Ciencia of Spain (grants BFU2004-0358/BMC to
J.F.D.; FPU predoctoral fellowship to R.M.B.) is grate-
´

I. Paterson et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2243–2247
2247
W. C.; Kiskamp, R.; Lipton, M.; Caufield, C.; Chang, G.;
Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11,
440; (b) Still, W. C.; Tempczyk, A.; Hawley, R. C.;
Hendrickson, J. Am. Chem. Soc. 1990, 112, 6127; (c) The
solution structures of 1 and its analogues 2–5 have recently
been determined: Paterson, I.; Menche, D.; Britton, R.;
2.5, 1.3 Hz, 1H), 5.89 (dddd, J = 10.3, 4.3, 1.9, 1.9 Hz,
1H), 5.78 (dddd, J = 10.3, 2.8, 2.8, 1.1 Hz, 1H), 5.26 (ddd,
J = 11.4, 4.7, 1.9 Hz, 1H), 4.61 (br s, 1H), 4.87 (s, 1H),
4.79 (s, 1H), 4.36 (m, 2H), 4.07 (ddd, J = 8.5, 3.9, 2.2 Hz,
1H), 3.82 (dddd, J = 16.9, 11.2, 11.2, 1.1 Hz, 1H), 3.73
(dddd, J = 9.0, 9.0, 3.2, 3.1 Hz, 1H), 3.04 (ddd, J = 7.9,
4.4, 2.3 Hz, 1H), 2.92 (dd, J = 2.2, 2.2 Hz, 1H), 2.65 (s,
3H), 2.26 (ddd, J = 14.2, 4.4, 1.73 Hz, 1H), 2.18–1.82 (m,
7H), 1.64 (ddd, J = 14.2, 11.4, 1.9 Hz, 1H), 1.55–1.20 (m,
5H); 13C NMR d (100 MHz, CD3OD): 168.5, 167.2, 154.2,
150.5, 147.7, 131.6, 130.1, 126.1, 125.4, 121.6, 116.6, 111.4,
74.5, 73.7, 73.6, 68.5, 67.8, 62.5, 53.3, 39.1, 38.5, 36.5, 34.2,
33.8, 32.4, 25.7, 18.7; HRMS-ES m/z: measured 502.2268
(MH+, calcd 502.2263 for C27H36NO6S); MS-ES m/z
´
˚
Hakansson, A. E.; Silva-Martınez, M. A. Tetrahedron
Lett., submitted for publication.
´
10. For an in silico approach to predicting the binding site for
`
laulimalide, see: Pineda, O.; Farras, J.; Maccari, L.;
Manetti, F.; Botta, M.; Vilarrasa, J. Bioorg. Med. Chem.
Lett. 2004, 14, 4825.
11. Pilcher, A. S.; Hill, D. K.; Shimshock, S. J.; Waltermire,
R. E.; DeShong, P. J. Org. Chem. 1992, 57, 2492.
12. All new compounds had spectroscopic data in support of
the assigned structures.
(relative intensity): 502 (MH+, 100), 343 (52); FTIR mmax
3413, 2922, 1717, 1178, 1079, 810 cmꢀ1
13. The catalytic Nozaki–Kishi coupling has been developed
by the group of Furstner: (a) Furstner, A.; Shi, N. J. Am.
:
.
20
3: ½aꢂD ꢀ96 (c 0.047, CH2Cl2); 1H NMR d (700 MHz,
CD3OD): 6.42 (ddd, J = 11.5, 10.4, 3.4 Hz, 1H), 5.94 (ddd,
J = 11.5, 2.6, 1.1 Hz, 1H), 5.86 (ddt, J = 10.3, 5.6, 2.2 Hz,
1H), 5.76 (ddt, 10.4, 2.9, 1.2 Hz, 1H), 5.15–5.12 (m, 1H),
4.83 (br s, 1H), 4.80 (d, J = 1.5 Hz, 1H), 4.34 (dt, J = 11.3,
2.4 Hz, 1H), 4.04 (ddd, J = 8.5, 3.9, 2.1 Hz, 1H), 3.82
(dddd, 17.0, 10.3, 10.2, 1.0 Hz, 1H), 3.69–3.65 (m, 1H),
3.59 (dd, J = 11.8, 4.5 Hz, 1H), 3.56 (dd, J = 11.8, 5.3 Hz,
1H), 3.00 (ddd, J = 7.3, 4.6, 2.2 Hz, 1H), 2.89 (t,
J = 2.2 Hz, 1H), 2.22 (dq, J = 17.0, 3.1 Hz, 1H), 2.15
(dd, J = 15.6, 3.8 Hz, 1H), 2.12 (ddd, J = 14.4, 4.7, 2.2 Hz,
1H), 2.07 (dt, J = 14.3, 7.5 Hz, 1H), 2.01 (dt, J = 14.9,
7.5 Hz, 1H), 1.98–1.86 (m, 3H), 1.57 (ddd, J = 14.4, 11.1,
7.9 Hz, 1H), 1.49–1.42 (m, 3H), 1.42–1.36 (m, 1H); 13C
NMR d (126 MHz, CD3OD): 167.3, 150.8, 147.7, 130.1,
126.1, 121.7, 111.5, 74.5, 72.4, 68.5, 67.8, 64.7, 62.5, 53.1,
39.1, 38.6, 36.6, 34.7, 34.2, 32.4, 25.8; HRMS-EI m/z:
measured 401.1954 ([M+Na]+, calcd 401.1940 for
C21H30O6Na); MS-EI m/z (relative intensity): 401
([M+Na]+, 100), 219 (47); FTIR mmax: 3351, 2925, 1716,
¨
¨
Chem. Soc. 1996, 118, 12349; (b) Furstner, A. Chem. Rev.
1999, 99, 991.
14. A similar observation has been reported: Choi, H.-w.;
Nakajima, K.; Demeke, D.; Kang, F.-A.; Jun, H.-S.; Wan,
Z.-K.; Kishi, Y. Org. Lett. 2002, 4, 4435.
15. No conjugate reduction of the C2–C3-cis-enoate was
observed, which was difficult to avoid, in particular upon
scale-up, when using L-Selectride (lithium tri-s-butylboro-
hydride), and then requires extensive chromatography for
purification of 13. The yield of 13 was further enhanced to
87% by a modified work-up using sodium perborate:
Urban, F. J.; Jasys, V. J. Org. Process Res. Dev. 2004, 8,
169.
¨
16. The data for 2 have previously been reported.7d
17. For leading reviews on epothilones, see: (a) Nicolaou, K.
C.; Roschangar, F.; Vourloumis, D. Angew. Chem., Int.
Ed. 1998, 37, 2014; (b) Stachel, S. J.; Biswas, K.;
Danishefsky, S. J. Curr. Pharm. Design. 2001, 7, 1277;
(c) Altmann, K.-H. Curr. Opin. Chem. Biol. 2001, 5, 424.
18. (a) Muller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J.
¨
Synlett 1996, 521; (b) Ohira, S. Synth. Commun. 1989, 19,
561.
19. Stille, J. K.; Simpson, J. H. J. Am. Chem. Soc. 1987, 109,
2138.
20. Bredenkamp, M. W.; Holzapfel, C. W.; van Zyl, W. J.
Synth. Commun. 1990, 20, 2235.
21. Nicolaou, K. C.; Ninkovic, S.; Sarabia, F.; Vourloumis,
D.; He, Y.; Vallberg, H.; Finlay, M. R. V.; Yang, Z.
J. Am. Chem. Soc. 1997, 119, 7974.
22. Schwartz, J.; Carr, D. B. J. Am. Chem. Soc. 1979, 101,
2927.
23. As before, the diastereomeric ratio may readily be
upgraded by the oxidation-reduction sequence.
24. (a) Oosawa, F.; Asakura, S. Thermodynamics of the
Polymerization of Protein; Academic: London, 1975; (b)
24
1647, 1420, 1081, 814 cmꢀ1. 4: ½aꢂD ꢀ182 (c 0.050,
CH2Cl2); 1H NMR d (500 MHz, CD3OD): 6.39 (ddd,
J = 11.4, 10.4, 3.5 Hz, 1H), 5.94 (dd, J = 11.5, 1.4 Hz, 1H),
5.90–5.80 (m, 1H), 5.72 (J = ddd, 10.1, 2.4, 1.1 Hz, 1H),
5.62 (dd, J = 15.6, 6.9 Hz, 1H), 5.34 (ddd, J = 15.6, 6.6,
1.1 Hz, 1H), 5.10 (ddd, J = 11.9, 11.4, 4.7 Hz, 1H), 4.79 (s,
1H), 4.72 (s, 1H), 4.58 (br s, 1H), 4.31 (d, J = 11.4 Hz,
1H), 4.07 (dd, J = 5.8, 5.4 Hz, 1H), 3.91 (ddd, J = 8.7, 3.9,
2.1 Hz, 1H), 3.81 (ddd, J = 16.4, 10.1, 10.1 Hz, 1H), 3.70–
3.60 (m, 1H), 2.95 (ddd, J = 7.3, 4.4, 2.2 Hz, 1H), 2.81 (dd,
J = 2.2, 2.0 Hz, 1H), 2.20–1.85 (m, 9H), 1.75–1.60 (m, 5H),
1.52 (ddd, J = 14.3, 11.4, 7.9 Hz, 1H), 1.45–1.0 (m, 10H);
13C NMR d (100 MHz, CD3OD): 167.2, 150.7, 147.8,
140.3, 130.1, 127.2, 126.1, 121.7, 111.5, 74.6, 74.5, 73.9,
68.5, 67.9, 62.7, 53.4, 41.8, 39.0, 38.4, 36.6,34.3, 34.3, 34.0,
33.9, 32.4, 27,3, 27.0, 25.7. HRMS-ES m/z: measured
487.3036 (MH+, calcd 487.3060 for C29H43O6). MS-ES
´ ´
Dıaz, J. F.; Menendez, M.; Andreu, J. M. Biochemistry
1993, 32, 10067.
m/z (relative intensity): 509 (100), 487 (46). FTIR mmax
:
24
3426, 2922, 1719, 1168, 810 cmꢀ1. 5: ½aꢂD ꢀ198 (c 0.033,
1
CH2Cl2); H NMR d (500 MHz, CD3OD): 7.18 (s, 1H),
6.67 (dd, J = 15.7, 1.3 Hz, 1H), 6.50 (dd, J = 15.7, 5.8 Hz),
6.44 (ddd, J = 11.5, 10.4, 3.5 Hz, 1H), 5.99 (ddd, J = 11.5,
25. Mossman, T. J. Immunol. Methods 1983, 65, 55.
26. For a leading reference, see: Paterson, I.; Florence, G. Eur.
J. Org. Chem. 2003, 12, 2193.