A convergent strategy for the pamamycin macrodiolides: Total synthesis of pamamycin-607, pamamycin-593, and pamamycin-621D precursors

Article abstract of DOI:10.1002/anie.200701749

A convergent total synthesis of pamamycin-607 (1), isolated from Streptomyces alboniger, was achieved by an E-Z isomerization of a tetrahydrofuran alkylidene and a regio- and diastereoselective solvent-dependent cyclo-C₆H₁₁BCl/Et

Full text of DOI:10.1002/anie.200701749

Communications
DOI: 10.1002/anie.200701749
Natural Product Synthesis
A Convergent Strategy for the Pamamycin Macrodiolides: Total
Synthesis of Pamamycin-607, Pamamycin-593, and Pamamycin-621D
Precursors**
Steve Lanners, Hassan Norouzi-Arasi, Xavier J. Salom-Roig, and Gilles Hanquet*
Dedicated to Professor Guy SolladiØ
The pamamycins are a family of naturally occurring homol-
ogous macrodiolides and are isolated from various Strepto-
myces species.[1,2] Besides displaying autoregulatory, aniono-
phoric, and antifungal activities, several members of this
family have been shown to be highly active against Gram-
positive bacteria including multiple antibiotic-resistant strains
of Mycobacterium tuberculosis.[3] Owing to their complex
structure, the pamamycins are attractive synthetic targets.
Several total syntheses of the most prominent member,
pamamycin-607 (1a, Scheme 1), have been reported,[4] and
synthetic studies towards the pamamycins have been
reviewed recently.[5] On the other hand, the promising
biological properties of the pamamycins call for the develop-
ment of general routes to these molecules to establish
structure–activity relationships, improve the pharmacological
properties, and probe their mode of action. Metz and co-
workers described the synthesis of pamamycin-621A and
pamamycin-635B by means of their sultone-based strategy.[6]
Herein we describe the asymmetric total synthesis of pama-
mycin-607 (1a) from enantiomerically pure (S)-p-tolyl methyl
sulfoxide as the only chiral starting material and illustrate the
potential of our strategy for an analogous synthesis of
pamamycin-621D (1b) and pamamycin-593( 1c) by the
preparation of advanced precursors.
Scheme 1. Retrosynthetic analysis of Pamamycin-607. TBS=tert-butyl-
dimethylsilyl.
In a retrosynthetic sense, our approach is based on the
disconnection of the two lactone linkages to afford the C1’–
C11’ fragment 2 and the C1–C18 fragment 3 (Scheme 1). In
the latter, the tetrahydrofuran ring and dimethylamino group
can be further simplified to obtain 4. An aldol disconnection
of the C7-C8 bond leads to the precursors 5 and 6. Although
the convergence of our strategy is based on the assembly of
three fragments (2, 5, and 6) of similar molecular weight, the
observation that 2 and 5 only differ by the configuration a to
the tetrahydrofuran ring allows a further simplification to a
common intermediate 7.
We have previously reported the diastereodivergent syn-
thesis of fragments 2 and 5 from (S)-p-tolyl methyl sulfoxide[7]
by a diastereoselective reduction[8] of a suitably protected b,d-
diketosulfoxide.[9] We have also developed efficient condi-
tions for the E–Z isomerization of the double bond of
intermediate 7[10] that enables us to access both 2 and 5 by
diastereoselective hydrogenation of either of the double bond
isomers.
Although one possible approach to ketones 6 relied on the
ring opening of g-caprolactone,[11] the most reliable and
scalable route proved to be a five-step procedure starting
from commercially available 4-pentenoic acid that afforded
ketones 6a and 6b (Scheme 2) in 45 to 50% overall yield.[12]
Ketone 6c was obtained by Wacker oxidation of an inter-
mediate of this sequence.[13,14]
[*] Dr. S. Lanners, Dr. H. Norouzi-Arasi, Dr. G. Hanquet
Laboratoire de StØrØochimie associØ au CNRS
ECPM, UniversitØ Louis Pasteur
25, rue Becquerel, 67087 Strasbourg (France)
Fax: (+33)3-9022-42742
E-mail: ghanquet@chimie.u-strasbg.fr
Dr. X. J. Salom-Roig
Institut de BiomolØcules Max Mousseron (IBMM)
UMR 5247, Synth›ses StØrØosØlectives
UniversitØ Montpellier II
Place Eug›ne Bataillon, Montpellier (France)
[**] The authors thank the Minist›re de l’Enseignement SupØrieur et de
la Recherche for research fellowships (S.L. and X.J.S.R.) and
Prof. Dr. Peter Metz for useful discussions and for kindly providing
NMR data of 1b.
7086
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7086 –7089

Angewandte
Chemie
Scheme 2. Reagents and conditions: a) Chx2BCl, Et3N, Et2O, 08C, then
5, ꢀ78!ꢀ238C, (4a, 52%; 4b, 61%; 4c, 48%); b) LiHMDS, THF,
ꢀ788C, 63%. Chx=cyclohexyl, LiHMDS=lithium hexamethyldisilaza-
nide, Pr=propyl.
With precursors 2, 5, and 6 in hands, we examined the
planned aldol coupling between aldehyde 5 and ketone 6a. To
obtain the desired 2,3-anti-3,4-syn relationship between the
newly formed stereogenic centres in aldol adduct 4a, we
decided to use the Chx2BCl/Et3N system[15] to prepare the
E enolate of 6a selectively. We anticipated that enolization of
6a might give rise to regioselectivity issues through compet-
itive deprotonation at C5 and C7. Indeed, when the reaction is
conducted in pentane as the solvent, 4a is isolated as the
minor product in a 1:4 ratio with respect to the aldol products
derived from internal enolization. Fortunately, switching the
solvent to diethyl ether inverted the regioselectivity. Aldol
adduct 4a could be isolated in good yields along with
regioisomeric products (3:1 ratio with respect to all undesired
products; Scheme 2). A detailed examination of the factors
governing this unusual solvent effect will be published
elsewhere.[12]
Scheme 3. Reagents and conditions: a) SmI2 cat., MeCHO, THF,
ꢀ108C, 59%; b) oxalic acid, CH2Cl2, H2O; c) TBAF, THF, RT, 78% over
two steps; d) H2, 4 bar, Rh/Al2O3, MeOH, RT, 90%; e) HN3, PPh3,
DIAD, toluene, RT, 72%; f) H2, Pd/C, HCHO, AcOH, MeOH, RT, 85%.
TBAF=tetrabutylammonium fluoride, DIAD=diisopropyl azodicarbox-
ylate.
Similarly, the aldol coupling with ketone 6b delivered the
desired adduct 4b in good yield. On the other hand, for
methylketone 6c, the same reaction conditions led to
complete regioselectivity but with an important decrease in
the diastereoselectivity (syn/anti = 6:4). Fortunately, diaste-
reoselectivity was increased without affecting regioselectivity
by using the lithium enolate,[16] and 4c was isolated in 63%
yield. Although 4a was subsequently used to complete the
total synthesis of pamamycin-607 (1a), the preparation of
aldol adducts 4b and 4c—precursors of pamamycin-621D
(1b) and pamamycin-593( 1c), respectively—illustrates the
flexibility of our synthetic approach as far as the incorpo-
ration of different alkyl substituents (R1, R2) is concerned.
The subsequent elaboration of the C1–C18 fragment 3
required the 1,3-anti selective reduction of the ketone group
of 4a with concomitant differentiation of the two secondary
hydroxy groups, a transformation that was achieved by using
the samarium(II) iodide catalyzed Evans–Tishchenko con-
ditions.[17] The use of acetaldehyde as a reducing agent
converted 4a into hydroxyacetate 8 (Scheme 3). Acid-cata-
lyzed dioxolane deprotection and cyclization followed by TBS
deprotection afforded the unsaturated intermediate 9. A
diastereoselective hydrogenation under Bartlett conditions[18]
afforded intermediate 10. The requisite configuration of the
amine function was obtained by Mitsunobu inversion of the
secondary alcohol by using hydrazoic acid, and the resulting
azide 11 was reduced to the primary amine and methylated by
Pd/C catalyzed hydrogenation in the presence of formalde-
hyde, thus completing the synthesis of the C1–C18 fragment
3.
The final sequence of steps followed the order of
esterifications established Metz and co-workers to avoid
epimerization at C2.[4a] Deprotection of tert-butyl ester 3 by
using trifluoroacetic acid (TFA) in dichloromethane and a
first esterification between the resulting acid 12 and the small
fragment 2 under Yamaguchi conditions[19] led to the
seco ester 13 in good yield (Scheme 4). Unfortunately,
attempted deprotection of the acetate group under a wide
variety of conditions (KCN, MeOH; 1,8-diazabicyclo-
[5.4.0]undec-7-ene (DBU), toluene; LiOH, THF/water
Angew. Chem. Int. Ed. 2007, 46, 7086 –7089
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7087

Communications
by the described replacement of 2 by 5 as the small fragment
precursor.[24] The high convergence and flexibility of our
strategy makes it suitable for the production of analogues.
Herein we have shown how the incorporation of an alter-
native substituent at C2 and C7 (Scheme 1, R1 and R2) can be
achieved. Similarly, a late-stage modification in the synthesis
of 2 and 5[7] should be a viable route for the introduction of
alternative substituents at C9 and C2’. These modifications
are complementary to those recently reported by Metz and
co-workers,[6] and thus our efforts complete the available tools
required for the preparation of the structurally and biolog-
ically intriguing pamamycin macrodiolides. The total syn-
thesis of pamamycin-621D (1b) and -593( 1c) starting from
precursors 4b and 4c, respectively, are in progress in our
group.
Received: April 19, 2007
Revised: June 4, 2007
Published online: August 7, 2007
Keywords: aldol reaction · natural products · pamamycins ·
.
regioselectivity · total synthesis
[1] For reviews on isolation, structural elucidation, and biological
activities of the pamamycins, see: a) M. Natsume, Recent Res.
Dev. Agric. Biol. Chem. 1999, 3, 11 – 22; b) M. Natsume,
Actinomycetologica 1999, 13, 11 – 20; c) B. M. Pogell, Cell. Mol.
Biol. 1998, 44, 461 – 463.
[2] For isolation of De-N-methylpamamycins, see I. Kozone, H.
Abe, M. Natsume, J. Pestic. Sci. 2001, 26, 149 – 153.
[3] P. Lef›vre, P. Peirs, M. Braibant, M. Fauville-Dufaux, R.
Vanhoof, K. Huygen, X. M. Wang, B. M. Pogell, Y. Wang, P.
Fisher, P. Metz, J. Content, J. Antimicrob. Chemother. 2004, 54,
824 – 827.
Scheme 4. Reagents and conditions: a) TFA, CH2Cl2, RT, 88%;
b) 2,4,6-Cl3PhCOCl, Et3N, THF, then 3, DMAP, toluene, 74%; c) lipase
type VII, DMF, H2O; d) MgBr2, CH2Cl2, 84% over two steps; e) 2,4,6-
Cl3PhCOCl, DMAP, CH2Cl2, 62%. DMAP=4-(N,N-dimethylamino)-
pyridine, DMF=N,N-dimethylformamide.
[4] a) Y. Wang, H. Bernsmann, M. Gruner, P. Metz, Tetrahedron
Lett. 2001, 42, 7801 – 7804; b) O. Germay, N. Kumar, E. J.
Thomas, Tetrahedron Lett. 2001, 42, 4969 – 4974; c) E. Lee,
E. J. Jeong, E. .J. Kang, L. T. Sung, S. K. Song, J. Am. Chem. Soc.
2001, 123, 10131 – 10132; d) S. H. Kang, J. W. Jeong, Y. S.
Hwang, S. B. Lee, Angew. Chem. 2002, 114, 1450 – 1453;
Angew. Chem. Int. Ed. 2002, 41, 1392 – 1395; e) E. J. Jeong,
E. J. Kang, L. T. Sung, S. K. Hong, E. Lee, J. Am. Chem. Soc.
2002, 124, 14655 – 14662.
(2:1); 0.5m NH2NH2, pyridine/AcOH (4:1); guanidine)[20]
resulted either in the decomposition of the substrate, C1
ester cleavage, or intramolecular trans esterification between
the resulting free alcohol at C8 and the C1 ester. The
conditions of NH3/MeOH/water afforded the desired product
in only 27% yield. Reductive deacetylation by using diisobu-
tylaluminum hydride (DIBAL-H) in toluene at ꢀ788C
improved the yield to 60%. Finally, the use of the enzymatic
method described by Holmes and co-workers[21] followed by
tert-butyl ester removal with MgBr2 in dichloromethane
delivered the isolated seco acid in 84% yield. The synthesis
was completed in 62% yield by using a modified Yamaguchi
macrolactonization under the conditions reported by Fleming
and Ghosh.[22,23]
In summary, a short and highly stereoselective access to
the larger fragment precursors 4a, 4b, and 4c of pamamycins-
607 (1a), -621D (1b), and-593( 1c) has been developed by
means of an unprecedented solvent-modulated regio- and
diastereoselective aldol reaction and application of b-keto-
sulfoxide methodology (all 13stereogenic centers of 1a are
derived from (S)-p-tolyl methyl sulfoxide). Starting from
intermediate 4a, a total synthesis of pamamycin-607 (1a) has
been accomplished in only 11 steps. Thus, our route for the
formation of 1a compares favorably with the recently
disclosed alternatives. The synthesis can be further shortened
[5] P. Metz, Top. Curr. Chem. 2005, 244, 215 – 249.
[6] P. Fischer, A. B. Garcia Segovia, M. Gruner, P. Metz, Angew.
Chem. 2005, 117, 6387 – 6390; Angew. Chem. Int. Ed. 2005, 44,
6231 – 6234.
[7] a) G. Solladiꢀ, X. J. Salom-Roig, G. Hanquet, Tetrahedron Lett.
2000, 41, 551 – 554; b) G. Solladiꢀ, X. J. Salom-Roig, G. Hanquet,
Tetrahedron Lett. 2000, 41, 2737 – 2740.
[8] For reviews on the use of chiral sulfoxides in natural product
synthesis, see: a) M. C. Carreꢁo, Chem. Rev. 1995, 95, 1717 –
1760; b) G. Hanquet, F. Colobert, S. Lanners, G. Solladiꢀ,
ARKIVOC 2003, (vii), 328 – 401.
[9] G. Hanquet, X. J. Salom-Roig, L. Gressot-Kempf, S. Lanners, G.
Solladiꢀ, Tetrahedron: Asymmetry 2003, 14, 1291 – 1301.
[10] G. Hanquet, X. J. Salom-Roig, S. Lemeitour, G. Solladiꢀ, Eur. J.
Org. Chem. 2002, 2112 – 2119.
[11] S. Lanners, N. Khiri, G. Solladiꢀ, G. Hanquet, Tetrahedron Lett.
2005, 46, 619 – 622.
[12] S. Lanners, H. Norouzi-Arasi, N. Khiri, G. Hanquet, Eur. J. Org.
Chem. 2007, 4065 – 4075. The aldol reaction of ketones 6a and
6b has been performed on a model aldehyde and the relative
7088 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7086 –7089

Angewandte
Chemie
configurations were ascertained by 1H NMR studies, see: C.
Gennari, S. Vieth, A. Comotti, A. Vulpetti, J. M. Goodman, I.
Paterson, Tetrahedron 1992, 48, 4439 – 4446.
[17] D. A. Evans, A. H. Hoveyda, J. Am. Chem. Soc. 1990, 112, 6447 –
6449.
[18] P. A. Bartlett, J. D. Meadows, E. Ottow, J. Am. Chem. Soc. 1984,
106, 5304 – 5311.
[13]
[19] J. Inanaga, K. Hirata, H. Saeki, T. Katsuki, M. Yamaguchi, Bull.
Chem. Soc. Jpn. 1979, 52, 1989 – 1993.
[20] T. W. Greene, P. G. M. Wuts, Protecting Groups in Organic
Synthesis 3rd ed., Wiley, New York, 1999.
[21] P. T. OꢂSullivan, W. Buhr, M. A. M. Fuhry, J. R. Harrison, J. E.
Davies, N. Feeder, D. R. Marshall, J. W. Burton, A. B. Holmes, J.
Am. Chem. Soc. 2004, 126, 2194 – 2207.
[22] I. Fleming, S. K. Ghosh, J. Chem. Soc. Perkin Trans. 1 1998,
2733 – 2747.
[23] Analytical data of synthetic 1b match exactly those described
previously in the literature. See reference [4].
[24] H. Bernsmann, M. Gruner, R. Frꢃhlich, P. Metz, Tetrahedron
Lett. 2001, 42, 5377 – 5380.
bmim = 1-n-butyl-3-methylimidazolium.
[14] I. A. Ansari, S. Joyasawal, M. K. Grupka, Y. S. Yadav, R. Gree,
Tetrahedron Lett. 2005, 46, 7507 – 7510.
[15] a) H. C. Brown, K. Ganesan, R. K. Dhar, J. Org. Chem. 1993, 58,
147 – 153; b) H. C. Brown, K. Ganesan, J. Org. Chem. 1993, 58,
7162 – 7169.
[16] D. R. Gustin, M. S. VanNeuwenhze, W. Roush, Tetrahedron Lett.
1995, 36, 3443 – 3446.
Angew. Chem. Int. Ed. 2007, 46, 7086 –7089
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7089