The total synthesis of chlorotonil A

Article abstract of DOI:10.1002/anie.200703930

(Chemical Equation Presented) Stacking the deck: The natural product chloronitril A contains an unprecedented motif in which a CCl₂ unit in a 14-membered macrolide framework is flanked by two carbonyl groups. In the first total synthesis of chl

Full text of DOI:10.1002/anie.200703930

Angewandte
Chemie
DOI: 10.1002/anie.200703930
Total Synthesis
The Total Synthesis of Chlorotonil A**
Nicola Rahn and Markus Kalesse*
This work is dedicated to Professor Ekkehard Winterfeldt on the occasion of his 75th birthday
Myxobacteria are a particularly rich source of structurally
novel metabolites with a broad range of biological functions.
In 2004 researchers at the Helmholtz Centre for Infection
Research (formerly GBF) isolated chlorotonil from the
myxobacterium Sorangiumcellulosum .[1] Its structure, which
was determined by NMRstudies and X-ray crystallography, [2]
features an unprecedented motif in which a CCl2 unit in a 14-
membered macrolide framework is flanked by two carbonyl
groups. This highly unusual and intriguing molecular archi-
tecture together with promising first biological data[3] ren-
dered chlorotonil a highly attractive synthetic target. Herein,
we report the first total synthesis of chlorotonil A (1)
employing a biomimetic and highly stereocontrolled route
which also established the absolute configuration of chlor-
otonil unambiguously.
Our retrosynthetic analysis identified fragment 3 to be the
pivotal intermediate, which could be extended into hydroxy
ester 2 through a Z-selective olefination[4] reaction and
addition of ethyl-2-methyl acetoacetate (Scheme 1). We
envisioned that the configuration at C2 could be set by
taking advantage of the different acidities of axial and
equatorial hydrogens on the macrolactone. In agreement
with the X-ray structure of chlorotonil A (1) the methyl group
in the axial orientation at C2 is coplanar with the adjacent p
orbitals; consequently the proton at C2 not conjugated to the
carbonyl groups.
Scheme 1. Retrosynthetic analysis of chlorotonil A (1). TBS=tert-butyl-
dimethylsilyl, PMB=para-methoxybenzyl.
Unsaturated ester 4, in turn, should be generated through
an anti-selective Suzuki coupling between dibromoolefin 5
and vinyl boronate 6. To obtain the desired endo product in
the Diels–Alder reaction we took advantage of the discrim-
inating effect of the vinyl bromide substituent, which had
been successfully applied in total syntheses published by the
research groups led by Evans and Roush.[5] In our case the
bromine substituent would disfavor the endo-II transition
Scheme 2. Comparison of endo transition states.
ꢀ
state owing to electronic and steric repulsion of the C9 C10
double bond (Scheme 2). Dibromoolefin 5 was obtained in six
steps from the TBS-protected aldehyde 7 derived from the
Roche ester[6] (Scheme 3). A Still–Gennari reaction[7] fol-
lowed by DIBAL-H reduction afforded allyl alcohol 8,[8]
[*] Dr. N. Rahn, Prof. Dr. M. Kalesse
Institut für Organische Chemie
Leibniz Universität Hannover
Schneiderberg 1B, 30167 Hannover (Germany)
Fax: (+49)511-762-3011
E-mail: Markus.Kalesse@oci.uni-hannover.de
Scheme 3. Synthesis of 5: a) (CF3CH2O)2P(O)CH(CH3)CO2CH3,
[18]crown-6, KHMDS, THF, ꢀ40 to ꢀ788C, 85%; b) DIBAL-H, CH2Cl2,
ꢀ788C, 87%; c) LiCl, 2,6-lutidine, CF3SO2Cl, DMF, 79%; d) NaCN,
DMF, 08C, 96% based on recovered starting material; e) DIBAL-H,
CH2Cl2, ꢀ788C, EtOH, 81%; f) CBr4, PPh3, Zn, CH2Cl2, RT, 91%.
KHMDS=potassium bis(trimethylsilyl)amide, DIBAL-H=diisobutyl-
aluminum hydride, DMF=N,N-dimethylformamide.
[**] We thank R. Jansen, H. Steinmetz, and K. Gerth (HZI) for providing
authentic chlorotonil A.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 597 –599
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
597

Communications
which was homologated to aldehyde 9 in three steps.[9]
Notably, no isomerization to the a,b-unsaturated nitrile or
aldehyde was observed. Corey–Fuchs homologation then
gave the desired dibromoolefin 5 in high yields.[10]
The synthesis of compound 6 utilized a cross-metathesis
reaction between olefin 10[11] and vinyl boronate 11[12] as the
final transformation (Scheme 4) and delivered the desired
boronic ester 6 in good yield and E/Z selectivity (E/Z > 50:1).
Figure 1. X-ray crystal structure of compound 13 (ellipsoids at the
50% probability level).[14]
Scheme 4. Synthesis of 13: a) Grubbs 2nd generation catalyst, CH2Cl2,
reflux, 81%, E/Z>50:1; b) [Pd(PPh3)4], TlOEt, H2O/THF, RT, 76%;
c) HF·pyridine, THF/pyridine (1:1), RT, 96%; d) Dess–Martin period-
=
inane, CH2Cl2, RT, 82%; e) PPh3 CHCO2Et, CH2Cl2, RT, 89%;
f) BF3·Et2O, toluene, 858C, 58%.
Scheme 5. a) Na/Hg (5%), MeOH, RT, 2 h, 61%; b) BF3·Et2O, toluene,
858C, 49%, 15/16/17=3:1:1.
Elaboration of ester 4 commenced with the coupling of
boronic ester 6 and dibromoolefin 5 according to Suzukiꢀs
protocol (Scheme 4).[13] After the cross-coupling reaction the
TBS group of 12 was removed by treatment with HF·pyridine,
and the resulting free hydroxy group was subsequently
oxidized and the carbon chain was extended through a
Wittig olefination. Gratifyingly, the pivotal Diels–Alder
reaction could be effected by treatment with BF3·Et2O
(RT!858C, 3 h). To our delight, these reaction conditions
led to simultaneous removal of the PMB group, transesteri-
fication, and Diels–Alder reaction, yielding 13 in very good
diastereoselectivity (d.r. 13:1). The configuration of this
crucial intermediate was assigned unambiguously by X-ray
crystallography to be the desired endo product (Figure 1). The
bromine substituent that dictated the endo selectivity could
be removed later using sodium mercury alloy in methanol
(Scheme 7).
To further assess the directing effect of the bromine
substituent, we prepared the dehalogenated tetraene 14 from
compound 4 by treatment with sodium–mercury alloy and
subjected the resulting Diels–Alder substrate to the reaction
conditions used before (Scheme 5). Analysis of the complex
mixture of products revealed a 3:1:1 ratio of the desired endo
product 15 together with endo II product 17 and exo product
16 in 49% combined yield. This result clearly showed the
importance of the bromine substituent for obtaining useful
selectivity and the advantage of halogen-directed Diels–
Alder reaction over standard Diels–Alder routes.
lactone opening could be achieved by employing KOH/
MeOH, and the resulting acid was then immediately trans-
formed into its methyl ester by adding diazomethane. The
liberated primary C17 hydroxy group was subsequently
oxidized to give aldhehyde 21. For its extension to 22, the
novel allyl phosphonate reagent 20 was envisioned, which was
generated through a cross-metathesis reaction between allyl
phosphonate 18 and olefin 19 (Scheme 6). It should be
pointed out that the anion generated from phosphonate 20
did not eliminate the allylic OPMB group as one might
expect. Optimization of reaction conditions identified
KHMDS in Et2O to provide the best yields and selectivities.
Finally, addition of the dianion generated from ethyl-2-
methylacetoacetate led to intermediate 23 (Scheme 7).
Having recognized that BF3·Et2O could efficiently cleave
PMB ethers and would lead to concomitant lactonization, we
treated 23 with BF3·Et2O at room temperature. After 20 min
For extension of fragment 13 careful optimization of
reaction conditions was required (Scheme 7). First, the
Scheme 6. a) Bu4NI (cat.), 1808C, neat, 61%; b) Grubbs 2nd gener-
ation catalyst, CH2Cl2, 64%.
598
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 597 –599

Angewandte
Chemie
conformation-based selective acidification relying on the
different acidities of the axial and equatorial positiond in
the macrocyclic lactone was exploited for the stereoselective
construction of the chiral center in a position to the lactone.
Additionally, this example serves as an extension of the
halogen-directed Diels–Alder approach and adds to the
picture of general applicability of this strategy.
Received: August 27, 2007
Published online: November 28, 2007
Keywords: chlorotonil · cross-coupling · cycloaddition ·
.
olefination · total synthesis
[1] The structure was presented at the 16. Irseer Naturstofftage, R .
Jansen, H. Irschick, K. Gerth, H. Reichenbach, G. Höfle,
DECHEMA, handbook, 2004, p. 32.
[2] K. Gerth, H. Steinmetz, G. Höfle, R. Jansen, Angew. Chem. 2008,
120, 610; Angew. Chem. Int. Ed. 2008, 47, 600.
[3] The biological activity will be disclosed in due course.
[4] For examples of vinylogous Still–Gennari reactions see: a) S. J.
Miller, M. D. Ennis, D. A. Evans, J. Org. Chem. 1993, 58, 471;
b) S. R. Chemler, D. S. Coffey, W. R. Roush, Tetrahedron Lett.
1999, 40, 1269.
[5] a) R. Riva, M. Kageyama, B. B. Brown, J. S. Warmus, K. J.
Moriarty, W. R. Roush, J. Org. Chem. 1991, 56, 1192; b) B. B.
Brown, W. R. Roush, J. Am. Chem. Soc. 1993, 115, 2268; c) M. L.
Reilly, K. Koyama, B. B. Brown, W. R. Roush, J. Org. Chem.
1997, 62, 8708; d) J. T. Starr, D. A. Evans, Angew. Chem. 2002,
114, 1865; Angew. Chem. Int. Ed. 2002, 41, 1787.
Scheme 7. Synthesis of chlorotonil A (1): a) Na/Hg, MeOH, RT, 92%;
b) KOH, MeOH then diazomethane, RT, 97%; c) Dess–Martin period-
inane, CH2Cl2, RT, 79%; d) KHMDS, 20, Et2O, ꢀ808C, 55%; e) NaH,
nBuLi, ethyl-2-methylacetoacetate, THF, ꢀ788C to RT; f) BF3·EtO2,
toluene, RT, 47% for two steps; g) NCS, 2,6-lutidine, CH2Cl2, RT, 65%.
NCS=N-chlorosuccinimide.
the starting material had been consumed, and a new
compound was isolated which could be identified as the
dehalogenated analogue of chlorotonil. Gratifyingly, the new
compound was generated as a single isomer, and the config-
uration could be assigned by comparing the NMRspectra of
this synthetic intermediate to those of the compound
obtained by dehalogenation of the natural product. Finally,
the halogenation was performed employing NCS[15] to com-
plet the total synthesis of chlorotonil. Comparison of the
optical rotation of synthetic chlorotonil to that of an authentic
sample confirmed the absolute configuration of the natural
product as that described here.[16]
In summary, we have reported the first total synthesis of
chlorotonil A (1) by a highly stereoselective and modular
route which should be readily amenable to the preparation of
analogues. Notable features of our synthetic route include a
halogen-directed Diels–Alder reaction in conjunction with a
one-pot esterification and PMP removal, and a novel
olefination for the rapid construction of one side chain. A
[6] K. P. Chary, M. Quitschalle, A. Burzlaff, C. Kasper, T. Scheper,
M. Kalesse, Chem. Eur. J. 2003, 9, 1129.
[7] W. C. Still, C. Gennari, Tetrahedron Lett. 1983, 24, 4405.
[8] J. A. Marshall, B. E. Blough, J. Org. Chem. 1990, 55, 1540.
[9] A. Villalobos, S. J. Danishefsky, J. Org. Chem. 1990, 55, 2776.
[10] E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 3768.
[11] E. Kattnig, O. Lepage, A. Fꢁrstner, J. Am. Chem. Soc. 2006, 128,
9194.
[12] C. Morrill, R. H. Grubbs, J. Org. Chem. 2003, 68, 6031.
[13] a) A. Suzuki, J. Organomet. Chem. 1999, 576, 147; b) S. A. Frank,
W. R. Roush, J. Org. Chem. 2002, 67, 4316.
[14] We thank M. Wiebcke for performing X-ray crystallographic
measurements.
[15] W. S. Weiner, N. Maslouh, R. V. Hoffman, J. Org. Chem. 2001,
66, 5790. Halogenation was performed on synthetic material as
well as the material derived by dehalogenation of the natural
product.
[16] Synthetic chlorotonil: [a]2D5 = ꢀ180.9 (c = 1.05, CHCl3); authen-
tic chlorotonil: [a]2D2 = ꢀ159.5 (c = 1.05, CHCl3).
Angew. Chem. Int. Ed. 2008, 47, 597 –599
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
599