Toward the total synthesis of disorazole A(1) and C(1): asymmetric synthesis of a masked southern segment.

Article abstract of DOI:10.1021/ol026468j

[reaction: see text] A highly convergent asymmetric synthesis of the masked southern segment of the antimitotic agent disorazole A(1) involves a Sonogashira coupling between a C1'-C10' enyne and a suitably protected C11'-C19' vinyl iodide. The central E,Z,Z-triene moiety is masked as a more stable ynediene.

Full text of DOI:10.1021/ol026468j

ORGANIC
LETTERS
2002
Vol. 4, No. 19
3239-3242
Toward the Total Synthesis of
Disorazole A1 and C : Asymmetric
1
Synthesis of a Masked Southern
Segment
Ingo V. Hartung, Barbara Niess, Lars Ole Haustedt, and H. Martin R. Hoffmann*
Department of Organic Chemistry, UniVersity of HannoVer, Schneiderberg 1B,
D-30167 HannoVer, Germany
hoffmann@mbox.oci.uni-hannoVer.de
Received July 4, 2002
ABSTRACT
A highly convergent asymmetric synthesis of the masked southern segment of the antimitotic agent disorazole A1 involves a Sonogashira
coupling between a C1′−C10′ enyne and a suitably protected C11′−C19′ vinyl iodide. The central E,Z,Z-triene moiety is masked as a more
stable ynediene.
The disorazoles are a family of 29 unique macrocyclic
polyketides, which were isolated in 1994 from the fermenta-
tion broth of the gliding bacteria Sorangium cellulosum
(strain So ce12) by Ho¨fle et al.1 Disorazole A1 is by far the
major component, being isolated in 17.2% from the crude
extraction residue. While the structure of disorazole A1 was
reported in 1994, the correct absolute configuration of all
seven stereocenters had not been assigned unambiguously
until 2000.2 Disorazole A1 comprises a highly functionalized
30-membered macrodiolide, which is built up from two
different hydroxy acids, whereas the C2-symmetric disorazole
C1 is a homodimer of the southern half of disorazole A1.
Disorazole A1 initiates decay of microtubules in subnano-
molar concentrations. It causes cell cycle arrest in the G2/M
phase and competes in vitro with vinblastin for the tubulin
binding site.3 With an IC50 value of 3 pg/mL (cell line L
929, mouse fibroblasts)4 disorazole A1 is too cytotoxic for a
direct application in anticancer therapy, but structural deriva-
tives and their mode of action are of great scientific and
pharmaceutical interest.
In 2000 Meyers et al. reported the asymmetric synthesis
of a southern half of disorazole A1 and C1 showing a syn
relationship between the C14′ and C16′ hydroxy groups and
the nonnatural configuration at C6′.5
An asymmetric synthesis of a nonnatural C12′-C19′
polyketide fragment has been published from our laboratories
very recently.6
Herein, we report the first asymmetric synthesis of the
masked southern segment of disorazole A1 and C1 with the
natural absolute configuration of all stereocenters along with
a modeling study that has guided our synthetic strategy.
(4) Irschik, H.; Jansen, R.; Gerth, K.; Ho¨fle, G.; Reichenbach, H. J.
Antibiot. 1995, 48, 31-35.
(1) Jansen, R.; Irschik, H.; Reichenbach, H.; Wray, V.; Ho¨fle, G. Liebigs
Ann. Chem. 1994, 759-773.
(2) Ho¨fle, G. GBF Annual Report 1999/2000, 102-103.
(3) Elnakady, Y. A. Ph.D. Thesis, TU Braunschweig, 2001.
(5) Hillier, M. C.; Park, D. H.; Price, A. T.; Ng, R.; Meyers, A. I.
Tetrahedron Lett. 2000, 41, 2821-2824.
(6) Vakalopoulos, A.; Smits, R.; Hoffmann, H. M. R. Eur. J. Org. Chem.
2002, 5, 1538-1545.
10.1021/ol026468j CCC: $22.00 © 2002 American Chemical Society
Published on Web 08/24/2002

Scheme 1
Our retrosynthetic disconnections of disorazole A1 and
conformations were used as input for a Monte Carlo search
to find the global minimum.10 Our calculations imply a more
strained 15-membered macrolactone in the case of the C9′-
C10′ alkyne, while the relative energy of the macrodiolide
disorazole C1 are outlined in Scheme 1. Cleavage of the
dilactone provides the southern segment 2 and the northern
segment 3. Because of the expected instability of the triene
moiety toward isomerization, we decided to mask one of
the two Z-olefins of segment 2 as an alkyne (see below).7
The southern segment was envisaged to be assembled in
convergent fashion by a Sonogashira coupling of the suitably
protected vinyl iodide 4 and terminal enyne 5, which can be
traced back to oxazole aldehyde 6.
The position of the triple bond relative to the ring-closing
functionalities at C1′ and C14′ was thought to be critical for
the cyclization steps leading to disorazole C1. A direct
dimerization of the hydroxy acid to the macrodiolide would
significantly shorten our synthesis. Therefore an intra-
molecular lactonization yielding the 15-membered macro-
lactone has to be suppressed. A suitably placed triple bond
should preclude this unwanted intramolecular cyclization due
to ring strain.
To assess the best fitting location we started a modeling
study to compare the relative energies of the 30-membered
macrodiolides I and II and the corresponding 15-membered
macrolactones III and IV (Figure 1).
All starting geometries were minimized using the MMFF
ForceField8 within MacroModel 7.2.9 These minimized
(7) Late-stage transformation of alkynes to Z-olefins has been used in a
variety of natural product synthesis. For recent examples, see: (a) Wender,
P. A.; Hegde, S. G.; Hubbard, R. D.; Zhang, L. J. Am. Chem. Soc. 2002,
124, 4956-4957. (b) Nazare´, M.; Waldmann, H. Chem. Eur. J. 2001, 7,
3363-3376. (c) Babudri, F.; Fiandanese, V.; Marchese, G.; Punzi, A.
Tetrahedron 2000, 56, 327-331. See also [18]annulene: Sto¨ckel, K.;
Sondheimer, F. Organic Syntheses; Wiley: New York, 1988; Collect. Vol.
VI, pp 68-75.
Figure 1.
(8) Halgren, T. A. J. Comput. Chem. 1996, 17, 490-512.
3240
Org. Lett., Vol. 4, No. 19, 2002

seems to be unperturbed by the presence of the C9′-C10′
triple bond (465.76 kJ/mol for II vs 463.30 kJ/mol for
disorazole C1, 1). Therefore, the C9′-C10′ Z-olefin was
chosen to be masked as an alkyne.11
The synthesis of the C1′-C10′ fragment 5 starts from
commercially available 2-benzyloxy acetaldehyde 7 (Scheme
2). Keck allylation12 using (R)-BINOL leads in 84% yield
the benzyl protecting group by hydrogenation, oxidation of
the primary alcohol leads to R-methoxy aldehyde 6. The
synthesis of enyne 5 is completed by a three-carbon chain
elongation involving Wittig olefination of aldehyde 6 with
TMS-protected propargyl triphenylphosphonium bromide
(2.5:1 E/Z selectivity) and removal of the terminal TMS
group.
The synthesis of vinyl iodide 4 is summarized in Scheme
3. Starting from readily available PMB-protected 3-hydroxy-
Scheme 2a
Scheme 3a
a Reaction conditions: (a) (R)-BINOL, Ti(OiPr)4, MS 4 Å,
allyltributylstannane, CH2Cl2, -20 °C, 84%; (b) NaH, MeI, THF,
rt, 94%; (c) O3, CH2Cl2, -78 °C; PPh3, rt, 86%; (d) NaClO2,
KH2PO4, H2O2, CH3CN/MeOH/H2O 1:1:2, 10 °C, 98%; (e)
L-SerOMe‚HCl, IBCF, NMM, THF, -25 °C f rt, 71%; (f) DAST,
K2CO3, CH2Cl2, -78°C; (g) DBU, BrCCl3, CH2Cl2, 0 °C f rt,
79% from 10; (h) H2, Pd/C, EtOH, rt, 97%; (i) SO3‚pyr, Et3N,
CH2Cl2/DMSO 4:1, 0 °C, 75%; (j) Br-Ph3P+CH2CdCTMS,
n-BuLi, THF, -78 to 0 °C, 49% (E:Z ) 2.5:1); (k) K2CO3, MeOH,
rt, 77%. IBCF ) isobutyl chloroformate; NMM ) N-methylmor-
pholine; DAST ) diethylaminosulfuryl trifluoride.
a Reaction conditions: (a) N-Tos-D-valine, BH3‚THF, CH2Cl2,
-78 °C; K2CO3, MeOH, 96%; (b) TBSOTf, 2,6-lutidine, DMAP,
CH2Cl2, rt, 99%; (c) DIBAH, toluene, -78 °C, 94%; (d) Dess-
Martin periodinane, CH2Cl2, 0 °C to rt, 83%; (e) trans-1-
bromopropene, t-BuLi, Et2O/THF 1:1, -95 °C, 99% (16:17 ) 1.1:
1); (f) separation of diastereomers; (g) SEMCl, Hu¨nig’s base, Bu4NI,
CH2Cl2; (h) DDQ, CH2Cl2/H2O 10:1, 0 °C, 98% from 17; (i)
SO3‚pyr, Et3N, CH2Cl2/DMSO 6:1, 0 °C, 80%; (j) I-Ph3P+CH2I,
NaHMDS, THF/HMPA 10:1, -78 °C, 82%.
to homoallylic alcohol 8 with the desired (R)-configuration
at C6′.13 O-Methylation followed by ozonolysis and subse-
quent oxidation14 of the resulting aldehyde provides car-
boxylic acid 9 (79% from 8).
The elaboration of carboxylic acid 9 into oxazole ester
11 is achieved employing a three-step sequence consisting
of amide formation with L-serine methyl ester hydrochloride,
cyclodehydration using DAST,15 and oxidation of the ox-
azoline to oxazole 11 by DBU/BrCCl3.16 After removal of
propanal 13,17 asymmetric Mukaiyama-aldol addition of silyl
ketene acetal 12 under Kiyooka’s conditions (in situ forma-
tion of the chiral oxazaborolidine promotor from N-Tos-D-
valine and BH3‚THF)18 gives â-hydroxy ester 14 in 96%
yield and 88% ee.19 After protection of the secondary
(9) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem.
1990, 11, 440-467.
(10) All minimizations are calculated without a solvent model to complete
convergence. Monte Carlo searches are run until the global minimum is
found more than once. Goodmann, J. M.; Still, W. C. J. Comput. Chem.
1991, 12, 1110-1117.
(11) In the meantime Meyers et al. have reported that the direct
dimerization of the southern half of disorazole A1 containing a C11′-C12′
alkyne leads preferentially to the 15-membered macrolactone: Hillier, M.
C.; Price, A. T.; Meyers, A. I. J. Org. Chem. 2001, 66, 6037-6045.
(12) Keck, G. E.; Krishnamurthy, D. Org. Synth. 1998, 75, 12-18.
(13) The enantiomeric excess is estimated to be higher than 94% by two
independent methods (chiral GC, Mosher ester analysis).
(14) Dalcanale, E. J. Org. Chem. 1986, 51, 567-569.
(16) Williams, D. R.; Lowder, P. D.; Gu, Y.-G.; Brooks, D. A.
Tetrahedron Lett. 1997, 38, 331-334.
(17) PMB-protected 3-hydroxypropanal 13 is prepared from 1,3-pro-
pandiol by mono PMB protection and Parikh-Doering oxidation in 77%
overall yield.
(18) (a) Kiyooka, S.-i.; Kaneko, Y.; Komura, M.; Matsuo, H.; Nakano,
M. J. Org. Chem. 1991, 56, 2276-2278. (b) Kiyooka, S.-i.; Kaneko, Y.;
Kume, K.-i. Tetrahedron Lett. 1992, 33, 4927-4930. (c) Kiyooka, S.-i.;
Hena, M. A. J. Org. Chem. 1999, 64, 5511-5523.
(15) Lafargue, P.; Guenot, P.; Lellouche, J.-P. Heterocycles 1995, 41,
947-958.
Org. Lett., Vol. 4, No. 19, 2002
3241

hydroxy group as the TBS ether, aldehyde 15 is formed via
reduction with DIBAH and Dess-Martin oxidation.20 Ad-
dition of lithiated trans-bromopropene (t-BuLi, -95°C)21
leads to a mixture of syn and anti alcohols 16 and 17 with
slight preference for the syn diastereomer (16:17 ) 1.1:1).22
The epimeric alcohols are easily separated by column
chromatography; the undesired syn diastereomer 16 can be
recycled to anti epimer 17 by oxidation to the enone and
diastereoselective reduction with Corey’s Me-(R)-oxazaboro-
lidine reagent.23 The C16′ hydroxy group is protected as
2-trimethylsilylethoxymethyl (SEM) ether. PMB cleavage
with DDQ in wet dichloromethane24 is followed by Parikh-
Doering oxidation25 of the primary alcohol to aldehyde 19.
Finally, the vinyl iodide 4 is generated with freshly prepared
I-Ph3P+CH2I applying the Stork-Zhao procedure.26 With
both fragments in hand, the crucial Sonogashira coupling27
is approached. Using the combination of PdCl2(PPh3)2/CuI/
Et3N, coupling of vinyl iodide 4 and enyne 5 (1:1.2 ratio)
furnishes the masked southern segment of disorazole A1 and
C1 2 (58% yield; 97% yield based on recovered vinyl iodide
4) (Scheme 4). The C7′-C8′ Z/E isomers are easily separated
by column chromatography. At -20°C both isomers are
stable for months. In addition, syn alcohol 16 is indepen-
dently converted into a masked nonnatural southern hemi-
Scheme 4a
a Reaction conditions: (a) PdCl2(PPh3)2 (10 mol %), CuI (30
mol %), Et3N, CH3CN, -20 °C to rt, 58% (97% based on recovered
4).
sphere of disorazole A1. By using a slightly modified
procedure (degassed DMF instead of CH3CN as solvent;
addition of the more labile enyne 5 after premixing catalyst,
vinyl iodide, and base) a satisfying yield of 75% for the final
Sonogashira coupling is achieved.
In conclusion, a masked and fully resolved southern
segment of disorazole A1 and C1 has been synthesized in a
highly convergent manner in 12 linear steps. The synthesis
of the northern segment 3 and a cyclization sequence are
currently under investigation in our laboratories.
(19) The enantiomeric excess is quantified by 1H NMR shift measurement
with europium tris[heptafluoropropyl-hydroxymethylene]-(+) camphorate
Eu(hfc)3. The measured enantiomeric excess of 88% is in agreement with
comparable literature results: Mulzer, J.; Mantoulidis, A.; O¨ hler, E. J. Org.
Chem. 2000, 65, 7456-7467.
(20) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155-4156.
(b) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-7287.
(c) Frigerio, M.; Santagostino, M.; Sputore, S. J. Org. Chem. 1999, 64,
4537-4538.
(21) Lebsack, A. D.; Overman, L. E.; Valentekovich, R. J. J. Am. Chem.
Soc. 2001, 123, 4851-4852.
(22) The syn and trans relationship is assigned by Rychnovsky’s
acetonide method: (a) Rychnovsky, S. D.; Skalitzky, D. J. Tetrahedron
Lett. 1990, 31, 945-948. (b) Rychnovsky, S. D.; Rogers, B.; Yang, G. J.
Org. Chem. 1993, 58, 3511-3515.
(23) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1986-
2012. As in similar cases with sterically very demanding ketones, the Me-
CBS reagent has to be used in excess: Trost, B. M.; Gunzner, J. L. J. Am.
Chem. Soc. 2001, 123, 9449-9450.
(24) Horita, K.; Yoshioka, T.; Tanaka, T.; Oikawa, Y.; Yonemitsu, O.
Tetrahedron 1986, 42, 3021-3028.
(25) Tidwell, T. T. Org. React. 1990, 39, 297-572.
Acknowledgment. We thank Dr. Jonathan Goodman
(University of Cambridge, U.K.) for an introduction to
computational chemistry (L.O.H.) and Dr. Christian B. W.
Stark and Ulrike Eggert for helpful discussions. We are
grateful to the Innovation Campaign of the University of
Hannover and the Fonds der Chemischen Industrie for
support.
Supporting Information Available: Physical and spec-
troscopic data for all compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(26) Stork, G.; Zhao, K. Tetrahedron Lett. 1989, 30, 2173-2174.
(27) (a) Sonogashira, K.; Tohda, Y.; Hagihara, N. Tetrahedron Lett. 1975,
16, 4467-4470. (b) Sonogashira, K. In Metal-Catalyzed Cross-Coupling
Reactions; Diederich, F., Stang, P. J., Eds.; Wiley VCH: Weinheim,
Germany, 1998; pp 203-229.
OL026468J
3242
Org. Lett., Vol. 4, No. 19, 2002