Synthetic studies on mycolactones: Synthesis of the mycolactone core structure through ring-closing olefin metathesis

Article abstract of DOI:10.1055/s-2007-967943

The mycolactone core structure 2a has been prepared through ring-closing olefin metathesis from diene 3 with exquisite E selectivity. The preparation of diene 3 included a highly efficient stereoselective synthesis of carboxylic acid 5. The mycolactone co

Full text of DOI:10.1055/s-2007-967943

LETTER
415
Synthetic Studies on Mycolactones: Synthesis of the Mycolactone Core
Structure through Ring-Closing Olefin Metathesis
S
ynthetic Stud
a
ies onMyc
b
olactones ian Feyen,1 Andrea Jantsch, Karl-Heinz Altmann*
ETH Zürich, Department of Chemistry and Applied Biosciences, Institute of Pharmaceutical Sciences, HCI H 405, Wolfgang-Pauli-Str. 10,
8093 Zürich, Switzerland
Fax +41(44)6331360; E-mail: karl-heinz.altmann@pharma.ethz.ch
Received 24 November 2006
an appropriately protected seco acid comprising the C1-
Abstract: The mycolactone core structure 2a has been prepared
C20 fragment. No other syntheses of mycolactone have
yet been described in the literature, although several re-
ports have addressed the synthesis of the unsaturated fatty
through ring-closing olefin metathesis from diene 3 with exquisite
E selectivity. The preparation of diene 3 included a highly efficient
stereoselective synthesis of carboxylic acid 5. The mycolactone
core structure 2a may serve as a versatile intermediate for the acid portion attached to the macrolactone ring via an ester
bond (C1¢-C16¢).10–12
synthesis of mycolactone and analogues thereof.
Key words: Buruli, mycolactone, ring-closing olefin metathesis,
natural products, stereoselectivity, total synthesis
In spite of the biological importance of mycolactones, no
systematic studies on structure–activity relationships have
been reported so far for this class of natural products.9,13
Apart from elucidating the importance of individual struc-
Mycolactones A and B [1 (Figure 1); ‘mycolactone’2] are
tural elements for the pathogenic potential of the naturally
secondary metabolites of polyketide origin, and are
believed to be the toxic agents responsible for the patho-
genic effects of Mycobacterium ulcerans.3 Infection with
M. ulcerans leads to Buruli ulcer, a disfiguring disease
associated with tissue destruction and suppression of the
immune system that affects large parts of tropical and sub-
tropical regions of Africa, Asia, Latin America and the
Western Pacific. Mycolactone has been shown to possess
cytotoxic, analgesic and immunosuppressive activity4–6
and intradermal inoculation of the purified toxin in guinea
pigs leads to lesions similar to those of Buruli ulcer in
humans. Treatment of Buruli ulcer with antibiotics has
been unsuccessful so far and the only treatment option
currently available consists in surgery (to remove the
lesion) followed by a skin graft if necessary.
produced toxins, such studies could conceivably lead to
analogues with antibiotic activity against Buruli, e.g. by
blocking specific steps in the biosynthesis of mycolac-
tones. Based on these considerations, we have embarked
on the development of a new, flexible route to myco-
lactones that would enable the incorporation of various
structural modifications, in particular as part of the mac-
rolactone core or the C12-C20 side chain. In this paper
we would like to report on the successful construction of
the C13-functionalized macrolactone core structure of
mycolactones (C1-C13) based on ring-closing olefin
metathesis (RCM) and some preliminary work on the
elaboration of the entire C12-C20 side chain on this inter-
mediate. The disclosure of these results at an intermediate
stage of our work towards mycolactones and their ana-
logues is prompted by a recent report by Alexander et al.
on the synthesis of the mycolactone core with a pendant
4-ketopent-2-yl moiety on C11.14 As in our case, ring clo-
sure was achieved through RCM, but a distinctly different
route was followed for the construction of the requisite
RCM precursor (including a completely different synthe-
sis of carboxylic acid 5).
20
17
13
11
OH
OH
1
O
O
8
5
6
OH
16'
5'
1'
O
At the most advanced level, our approach to mycolactones
(as in Kishi’s synthesis) foresees ester formation between
an appropriately protected version of the C1-C20 core al-
cohol and the C1¢-C16¢ fatty acid side chain (or between
independently modified versions of these building blocks)
followed by protecting-group removal. The C1-C20 frag-
ment should be accessible from macrocyclic precursors
2b or 2a (Scheme 1) through cross-metathesis with olefin
6 or transition-metal-catalyzed cross-coupling with vinyl
iodide 7,7–9 respectively (or any other appropriate olefin or
vinyl iodide, in the case of analogue synthesis). While 2b
could be directly derived from 2a, a suitable precursor
for the latter appeared to be the diene 3, provided that
macrocyclization of 3 through RCM would proceed with
4'
12'
OH OH
O
1
Figure 1
The total synthesis of mycolactones A, B, and C has been
reported by Kishi and co-workers in a series of three
publications7–9 and is based on the construction of the
macrocyclic core structure through macrolactonization of
SYNLETT 2007, No. 3, pp 0415–0418
1
5.
0
2.
2
0
0
7
Advanced online publication: 07.02.2007
DOI: 10.1055/s-2007-967943; Art ID: G34406ST
© Georg Thieme Verlag Stuttgart · New York

416
F. Feyen et al.
LETTER
intermediacy of oxetane 11.17 In contrast, attempts to pre-
pare 12 through Cu-catalyzed coupling of the primary
alkyl iodide derived from TES-protected 10 via Finkel-
stein exchange and isopropenylmagnesium bromide18
proved to be unsuccessful. Olefin 12 was then elaborated
into the desired building block 5 via TES protection of the
secondary hydroxyl group, oxidative cleavage of the PMB
ether with DDQ and finally conversion of the resulting
primary alcohol into the carboxylic acid. The latter trans-
formation was based on a two-step protocol involving (i)
oxidation to the aldehyde with Dess-Martin periodinane
(DMP)19 followed by (ii) aldehyde oxidation with buff-
ered NaClO2 in the presence of 2-methyl-2-butene as
chlorine scavenger.20 Carboxylic acid 5 could thus be
obtained from 8 in ten steps and 20% overall yield.
OTs
OH
X
OTs
4
O
O
O
O
+
HO
O
OTES
2a: X = I
OTES
3
5
OTES
2b: X = C(CH3)=CH2
I
TBSO
OTBS
TBSO
OTBS
6
7
Scheme 1
O
O
i, ii
iii
PMBO
acceptable selectivity.15 Although this was not clear at the
outset of our work, we felt strongly attracted to this strat-
egy, as it would provide very efficient access to the de-
sired core structures 2a or 2b. Diene 3 would in turn be
obtained through esterification of carboxylic acid 5 with
alcohol 4.
HO
O
13
14
OH
OH
iv, v
TsO
PMBO
Scheme 2 summarizes the synthesis of carboxylic acid 5,
which starts from cheap 1,5-pentanediol 8. Monopro-
tection of 8 as a p-methoxybenzyl (PMB) ether followed
by Swern oxidation and subsequent aldol reaction of the
resulting aldehyde with N-propionyl-(2R)-bornane-
(10,2)-sultam16 provided the aldol product 9 in excellent
yield and with >95% diastereomeric excess. Reductive re-
moval of the auxiliary and subsequent selective tosylation
of the primary hydroxyl group gave tosylate 10, which
could be directly converted into olefin 12, employing
methodology that had been previously developed by
Carreira and co-workers for phenyllithium reagents.17
Thus, treatment of 10 with an excess of isopropenyl-
lithium gave 12 in 61% yield, presumably through the
15
4
Scheme 3 Reagents and conditions: (i) PMB-trichloroacetimidate,
TfOH, Et2O, 0 °C, 2 h, 58%; (ii) DIBAL-H, toluene, -90 °C, 1.5 h,
quant. (crude); (iii) Bu3SnCH2CH=CH2, SnCl4, CH2Cl2, -90 °C, 1 h,
82%, >95% de; (iv) (a) DDQ, CH2Cl2, buffer pH 7.2, r.t., 1.75 h; (b)
LiAlH4, Et2O, 0 °C → r.t., 1 h, 76%; (v) TsCl, Et3N, DMAP, CH2Cl2,
35 °C, 5 h, 67%.
The synthesis of alcohol 4 proceeded via PMB-protected
aldehyde 14, which was readily obtained in two steps
from (S)-Roche ester 13 (Scheme 3). Allylstannylation of
14 with allyltributyltin gave the desired anti product 15 in
82% yield and with excellent diastereoselectivity (>20:1).
O
OH
iv, v
i–iii
HO
OH
N
OPMB
8
9
SO2
OH
O
vi
OPMB
O
TsO
OPMB
10
11
OH
OTES
vii–x
OPMB
OH
12
5
Scheme 2 Reagents and conditions: (i) PMBCl, NaH, benzene, reflux, 6 h, 75%; (ii) (COCl)2, DMSO, Et3N, CH2Cl2, -78 °C → r.t., 1 h,
94%; (iii) (a) N-propionyl-(2R)-bornane-(10,2)-sultam, Et2BOTf, CH2Cl2, -5 °C, 20 min; (b) addition of aldehyde, -78 °C, 1.25 h, 83%, >95%
de; (iv) LiAlH4, THF, 0 °C → r.t., 2.5 h, 77%; (v) TsCl, Et3N, DMAP, CH2Cl2, 0 °C → r.t., 5 h, 91%; (vi) (a) LiC(CH3)=CH2 (1.5 equiv), Et2O,
-20 °C → r.t., 45 min, r.t., 30 min; (b) LiC(CH3)=CH2 (3 equiv), BF3·Et2O (3 equiv), -78 °C, 25 min, 61%; (vii) TESOTf, 2,6-lutidine, CH2Cl2,
-78 °C → r.t., 1 h, 98%; (viii) DDQ, CH2Cl2, buffer pH 7.2, r.t., 45 min, 92%; (ix) DMP, CH2Cl2, 0 °C → r.t., 2 h; (x) NaClO2, 2-methyl-2-
butene, t-BuOH, NaH2PO4, H2O, r.t., 30 min, 85% (two steps).
Synlett 2007, No. 3, 415–418 © Thieme Stuttgart · New York

LETTER
Synthetic Studies on Mycolactones
417
Subsequent treatment of 15 with DDQ produced a mixture Gratifyingly, treatment of this diene with second-genera-
of the free diol and the corresponding p-methoxybenzoate tion Grubbs catalyst15 under reflux conditions in CH2Cl2
ester as the major product. This mixture was not separat- for 4 hours resulted in the efficient formation of the 12-
ed, but directly treated with LiAlH4 to provide the desired membered ring, with E-isomer 16 being the only isolable
free alcohol in 76% yield. Selective tosylation of the pri- product.25 As indicated above, similar findings have re-
mary hydroxyl group then gave 4. The tosyl group simul- cently been reported by Alexander et al. for a somewhat
taneously serves as a protecting group (during ester different substrate,14 thus indicating that the trans prefer-
formation) and as an activating group for nucleophilic ence of the RCM is not unique for a specific (exocyclic)
substitution reactions at C13 subsequent to macrocycle C11-substituent. Conversion of tosylate 16 into iodide 2a
formation. The incorporation of this group prior to RCM could be achieved in good yield, but reaction time proved
obviates the need for a deprotection/activation sequence to be a highly critical variable for the successful execution
at a more advanced stage of the synthesis, which we felt of this transformation. Thus, treatment of 16 with NaI in
could be more problematic.
acetone at reflux for 24 hours (rather than ca. 6 hours)
produced none of the desired 2a, but only other un-
identified reaction products. Preliminary attempts at the
conversion of 2a into 2b {treatment of 2a with
LiCu[C(CH3)=CH2]3; Pd(0)-catalyzed Negishi coupling
with ClZnC(CH3)=CH226} have been unsuccessful, with
most of the starting 2a being recovered unchanged.27
However, other options for this transformation will be
explored as will be the Negishi coupling of 2a with vinyl
iodide 7. The results of these experiments will be reported
in due course.
The stereochemical outcome of the allylstannylation reac-
tion of 14 is in line with previous results reported by
White and co-workers for the allylstannylation of the
benzyl-protected variant of 14.21 Unfortunately, the direct
allylstannylation of tosylated aldehyde 14a (Figure 2; pre-
pared from 13 by tosylation and DIBAL-H reduction in
74% yield22) gave only a 4:6 mixture of anti and syn iso-
mers (79% chemical yield, including traces of residual
Bu3Sn-derivatives) and thus did not represent a viable
alternative to the route depicted in Scheme 3.
In summary, we have achieved a highly efficient stereo-
selective synthesis of functionalized macrolactones 16
and 2a, which should be versatile precursors for the syn-
thesis of natural mycolactones and structurally modified
analogues thereof.
O
TsO
14a
Figure 2
Acknowledgment
A ca. 10:1 mixture of 4 and its (2S,3S)-syn isomer could
be obtained after allylboration of 14a with (+)-
Ipc2B(CH2CH=CH2). However, this material was only
obtained in 19% yield and after extensive chromato-
graphic purification, thus significantly limiting the practi-
cability of the allylboration approach.23
We are indebted to Dr. Bernhard Pfeiffer for help with the NMR
experiments to determine the geometry of the C8–C9 double bond.
References and Notes
(1) Current Address: CARBOGEN AMCIS AG, 5502
Hunzenschwil, Switzerland.
As depicted in Scheme 4, esterification of 4 with 5 under
Höfle-Steglich conditions24 proceeded smoothly to pro-
vide the RCM substrate diene 3 in 82% yield.
(2) The term ‘mycolactone’ usually refers to the naturally
occurring mixture of mycolactones A and B, which are
geometric isomers at the C4¢=C5¢ double bond. Although
both isomers can be isolated separately, each of them rapidly
isomerizes to the mixture of mycolactones A and B.
(3) George, K. M.; Chatterjee, D.; Gunawardana, G.; Welty, D.;
Hayman, J.; Lee, R.; Small, P. L. Science 1999, 283, 854.
(4) Adusumilli, S.; Mve-Obiang, A.; Sparer, T.; Meyers, W.;
Hayman, J.; Small, P. L. Cell. Microbiol. 2005, 7, 1295.
(5) Coutanceau, E.; Marsollier, L.; Brosch, R.; Perret, E.;
Goossens, P.; Tanguy, M.; Cole, S. T.; Small, P. L.;
Demangel, C. Cell. Microbiol. 2005, 7, 1187.
OTs
X
i
ii
O
O
O
O
+
4
5
OTES
OTES
3
(6) George, K. M.; Pascopella, L.; Welty, D. M.; Small, P. L.
Infect. Immun. 2000, 68, 877.
(7) Benowitz, A. B.; Fidanze, S.; Small, P. L.; Kishi, Y. J. Am.
Chem. Soc. 2001, 123, 5128.
(8) Song, F.; Fidanze, S.; Benowitz, A. B.; Kishi, Y. Org. Lett.
2002, 4, 647.
(9) (a) The synthesis and structure confirmation of mycolactone
C, which is a ca. 1:1 mixture of 12¢-deoxymycolactones A
and B is described in: Judd, T. C.; Bischoff, A.; Kishi, Y.;
Adusumilli, S.; Small, P. L. C. Org. Lett. 2004, 6, 4901.
(b) Mycolactone C is the major metabolite of Australian
strains of M. ulcerans and it is 10,000-fold less cytopathic
16: X = OTs
2a: X = I
iii
Scheme 4 Reagents and conditions: (i) DCC, DMAP, CH2Cl2, 0 °C
→ r.t., 15 h, 82%; (ii) Grubbs’ II catalyst (8 mol%), CH2Cl2, reflux,
4 h, 72% (only E-isomer); (iii) NaI, acetone, r.t. → 65 °C, 6.5 h, 73%.
Synlett 2007, No. 3, 415–418 © Thieme Stuttgart · New York

418
F. Feyen et al.
LETTER
than mycolactones A and B: Mve-Obiang, A.; Lee, R. E.;
Portaels, F.; Small, P. L. C. Infect. Immun. 2003, 71, 774.
(c) West African strains of M. ulcerans produce
mycolactone C only as a minor metabolite.
catalyst being added after 2 h). After cooling to r.t., H2O (50
mL) was added to the reaction mixture and a part of the
solvent was removed under reduced pressure. The layers
were separated and the aqueous solution was extracted with
CH2Cl2 (2 × 30 mL). The combined organic extracts were
dried over MgSO4 and the solvent was evaporated in vacuo.
Purification of the residue by flash chromatography in
EtOAc-hexane (1:10) gave 16 (391 mg, 72%) as a faintly
yellow, viscous oil; [a]20D -44.0° (c = 0.61, CHCl3). 1H
NMR (400 MHz, CDCl3): d = 7.79 (d, J = 8.3 Hz, 2 H), 7.35
(d, J = 8.0 Hz, 2 H), 4.85-4.97 (m, 2 H), 4.02-4.08 (m, 1 H),
3.80-3.86 (m, 1 H), 3.37-3.42 (m, 1 H), 2.46 (s, 3 H), 2.35-
2.43 (m, 2 H), 2.00-2.11 (m, 2 H), 1.58-1.94 (m, 6 H), 1.65
(s, 3 H), 1.30-1.45 (m, 2 H), 0.90-0.99 (m, 15 H), 0.60 (q,
J = 8.0 Hz, 6 H). 13C NMR (100 MHz, CDCl3): d = 173.2,
144.8, 138.1, 132.9, 129.8, 128.0, 120.4, 77.6, 72.5, 71.6,
45.3, 37.6, 35.6, 33.5, 33.3, 31.4, 21.7, 21.6, 18.7, 15.7, 13.4,
7.0, 5.1. IR (film): 2954, 2912, 2876, 1731, 1366, 1244,
1176, 1022, 969, 815, 672 cm-1. HRMS (ESI, +ve): m/z
[M + Na]+ calcd for C29H48O6SSi: 575.2833; found:
575.2827.
(10) Gurjar, M. K.; Cherian, J. Heterocycles 2001, 55, 1095.
(11) Van Summeren, R. P.; Feringa, B. L.; Minnaard, A. J. Org.
Biomol. Chem. 2005, 3, 2524.
(12) Yin, N.; Wang, G.; Qian, M.; Negishi, E. Angew. Chem. Int.
Ed. 2006, 45, 2916; Angew. Chem. 2006, 118, 2982.
(13) Leadlay and co-workers have recently reported the isolation
and structure elucidation of new mycolactones from a
pathogenic strain of M. ulcerans from China (MU98912). In
comparison with mycolactones A and B from the African
strain MUAgy99, these new mycolactones incorporate an
additional methyl group attached to C2¢. MU98912 does not
produce mycolactones A/B: Hong, H.; Spencer, J. B.; Porter,
J. L.; Leadlay, P. F.; Stinear, T. ChemBioChem 2005, 6, 643.
(14) Alexander, M. D.; Fontaine, S. D.; La Clair, J. J.;
DiPasquale, A. G.; Rheingold, A. L.; Burkart, M. D. Chem.
Commun. 2006, 4602.
(15) For recent reviews on olefin metathesis cf., e.g.:
(a) Grubbs, R. H. Tetrahedron 2004, 60, 7117. (b) Trnka,
T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(16) (a) Oppolzer, W.; Blagg, J.; Rodriguez, I.; Walther, E. J. Am.
Chem. Soc. 1990, 112, 2767. (b) Nguyen, G.; Perlmutter, P.;
Rose, M. L.; Vounatsos, F. Org. Lett. 2004, 6, 893.
(17) (a) Carreira, E. M.; Du Bois, J. J. Am. Chem. Soc. 1995, 117,
8106. (b) See also: Eis, M. J.; Wrobel, J. E.; Ganem, B. J.
Am. Chem. Soc. 1984, 106, 3693.
The assignment of the newly formed double bond as E was
based on the absence of cross peaks between the C8-methyl
group and C9-H in both NOESY and ROSY experiments. As
expected for an E configured double bond between C8 and
C9 a strong NOE cross peak was observed between C9-H
and the 7-CH2 moiety.
(26) Zhou, J.; Fu, G. C. J. Am. Chem. Soc. 2003, 125, 12527.
(27) Encouraged by the efficiency and selectivity of ring closure
observed with RCM substrate 3, we have also investigated
triene 17 as a possible substrate for RCM-mediated
formation of the 12-membered ring (Figure 3). However,
treatment of this compound with Grubbs’ second-generation
catalyst did not produce any of the desired 12-membered
macrolactone. Instead, and perhaps not too surprisingly, the
major product formed in the reaction appears to be
cyclohexene 18 (based on MS analysis of the reaction
mixture).
(18) Wang, Y.; Dong, X.; Larock, R. C. J. Org. Chem. 2003, 68,
3090.
(19) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
(20) Kraus, G. A.; Roth, B. J. Org. Chem. 1980, 45, 4825.
(21) White, J. D.; Blakemore, P. R.; Green, N. J.; Hauser, E. B.;
Holoboski, M. A.; Keown, L. E.; Nylund Kolz, C. S.;
Phillips, B. W. J. Org. Chem. 2002, 67, 7750.
(22) Aïssa, C.; Riveiros, R.; Ragot, J.; Fürstner, A. J. Am. Chem.
Soc. 2003, 125, 15512.
(23) The reaction of 14a with (+)-Ipc2B(CH2CH=CH2) represents
the mismatched reactant combination. The matched case
[reaction of ent-14a with (+)-Ipc2B(CH2CH=CH2)] has been
reported by Fürstner and co-workers to produce the syn
product in good yield and with excellent selectivity (see ref.
22).
OTBS OTBS
O
O
(24) Höfle, G.; Steglich, W.; Vorbrüggen, H. Angew. Chem. Int.
Ed. Engl. 1978, 17, 569; Angew. Chem. 1978, 90, 602.
(25) Preparation of 16: To a solution of diene 3 (565 mg, 0.97
mmol) in CH2Cl2 (320 mL, 0.003 M) was added [1,3-
bis(2,4,6-trimethylphenyl)-2-imidazolidinylidene]
dichloro(phenylmethylene)(tricyclohexylphosphine)Ru
(Grubbs’ II catalyst; 42 mg, 0.075 mmol) and the mixture
was heated to reflux for 4 h (with additional 21 mg of
O
O
OTES
OTES
17
18
Figure 3
Synlett 2007, No. 3, 415–418 © Thieme Stuttgart · New York