Cobalt-catalysed hydrovinylation as the key step in a short synthesis of moenocinol

Article abstract of DOI:10.1039/b822023a

A regio- and chemoselective cobalt(i)-catalysed 1,4-hydrovinylation reaction is the key step in the straightforward and convergent synthesis of moenocinol, the aglycone of moenomycin A.

Full text of DOI:10.1039/b822023a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Cobalt-catalysed hydrovinylation as the key step in a short synthesis
of moenocinolwz
Gerhard Hilt* and Jonas Treutwein
Received (in Cambridge, UK) 9th December 2008, Accepted 9th January 2009
First published as an Advance Article on the web 26th January 2009
DOI: 10.1039/b822023a
A regio- and chemoselective cobalt(I)-catalysed 1,4-hydrovinyla-
tion reaction is the key step in the straightforward and con-
vergent synthesis of moenocinol, the aglycone of moenomycin A.
Atom-economic reactions for the fast and efficient synthesis of
complex molecules are of great interest.1 Among widespread
structural subunits which are found in nature, 1,4-dienes are
relatively rare. Because of the ease of isomerisation to the con-
jugated 1,3-diene systems, mild methods are needed to generate
such subunits. While the ruthenium-catalysed Alder-ene reaction
proved to be very effective in this respect,2 the recently discovered
cobalt-catalysed Alder-ene reaction is still of academic interest.3
An alternative approach toward the synthesis of 1,4-dienes is the
cobalt-catalysed 1,4-hydrovinylation of 1,3-dienes with terminal
alkenes.4 A good example for a subunit that can be addressed by
the cobalt-catalysed 1,4-hydrovinylation reaction is the 1,4-diene
bearing an exo-methylene double bond within the aglycon of
moenomycin A. We envisaged the cobalt-catalysed 1,4-hydro-
Scheme 1 Retrosynthesis of the aglycone 1.
vinylation of a terminal alkene with commercially available
myrcene to be a suitable reaction to build the 1,4-diene moiety.
In addition, the cobalt-catalysed hydrovinylation reaction tolerates
various functional groups, permitting the fast atom-economic
assembly of 1,4-diene intermediates in the synthesis of moenocinol
1.5 Among several previously described syntheses of the isoprenoid
side chain moenocinol,6 the recent one by Yang and Huang is the
most efficient thus far. In their approach the target molecule 1 was
synthesised in 10 linear steps with an overall yield of 12%.6i
Our retrosynthetic analysis of 1 (Scheme 1) identified the
assembly of the 1,4-diene in 2 as the key step in the synthesis of
1 from myrcene and alkene 4.
The disconnection of the right-hand-side part of 1 could be
accomplished by an olefination reaction utilising the sulfone
3 via a Julia–Kocienski reaction with the corresponding
aldehyde of 2.7 Firstly, we addressed the synthesis of the
right-hand-side part 3, which can be easily accomplished by
a three-step synthesis starting from commercially available
5-chloropentan-2-one (5) and 1-phenyl-1H-tetrazol-5-thiole
(PT-SH). The substitution and the subsequent oxidation of
the sulfide to the sulfone (85% yield) are followed by a
Peterson olefination of the ketone functional group utilising
ethyl trimethylsilyl acetate (ETSA) producing 3 as a 2 : 1 Z–E
mixture of both isomers with 71% yield.8 The E–Z-isomers
were separated by column chromatography and the desired
Z-configured product 3 was isolated in 40% overall yield over
three steps (Scheme 2).
The alternative approach via a Mitsunobu reaction starting
from 5-hydroxypentan-2-one and PT-SH gave the substitution
product in quantitative yield.9 However, the desired product
could not be completely separated from the accompanying
Fachbereich Chemie, Philipps-Universitat Marburg, Hans-Meerwein-Straße,
¨
D-35043 Marburg, Germany. E-mail: Hilt@chemie.uni-marburg.de;
Fax: +49 (0)6421 2825677; Tel: +49 (0)6421 2825601
w In memoriam of Prof. Dr Peter Welzel.
z Electronic supplementary information (ESI) available: Experimental
procedures and analytical data. See DOI: 10.1039/b822023a
Scheme 2 Synthesis of 3.
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 1395–1397 | 1395

View Article Online
ligand sphere of the cobalt complex replacing the solvent
molecules to form intermediate B. After an insertion of the
alkene, the intermediate C is formed from which a b-hydride
elimination can occur to generate intermediate D. The pro-
posed reaction mechanism is concluded by a reductive elim-
ination to form the desired product 10 and regenerate the low
valent cobalt catalyst. The regiochemistry can be rationalised
in a similar way as we recently described for the cobalt-
catalysed Diels–Alder reaction. In this case the cycloaddition
product is formed with the substituents predominantly in a
1,4-arrangement when diphosphine ligands are used.11
The chain length of the diphosphine tether seems to be of
crucial importance for the hydrovinylation, as for diphosphine
ligands with a longer chain (43 carbons) between the two
donor centres, the reaction failed completely. A rational could
be that longer chained diphosphine ligands do not form stable
chelates and are more flexible.
Scheme 3 Synthesis of 4.
hydrazine by-product. Therefore, the reaction sequence
outlined in Scheme 2 was preferred.
For the synthesis of the left-hand-side part, 2, of the target
molecule the alkene 4 was assembled in a one-step synthesis
following a procedure described by Cohen et al. utililising
preformed di-tert-butyl-biphenyl (DTBB) lithium radical
anion as electron transfer reagent.10 Starting from commer-
cially available 3,3-dimethyloxetane 6 the intermediately
formed dianion was quenched with allyl bromide and the
alkene 4 was isolated in 75% yield (Scheme 3). It is worth
mentioning that the purity of the lithium source was crucial for
the success of the reaction, and the use of fresh lithium pieces
is advised to reproduce the literature yield.
Nevertheless, the very good level of regiocontrol in the
carbon–carbon bond formation process (B10 : 1 for dppp
as ligand) permitted us to proceed with the synthesis of 1
utilising the Co(dppp)Br2 system for the transformation. The
In order to achieve a high level of regioselectivity in the
carbon–carbon bond formation process during the cobalt-
catalysed 1,4-hydrovinylation—a key step of the total syn-
thesis of 1—we performed a ligand screening of different
simple commercially available diphosphine ligands. In this
respect we investigated the cobalt-catalysed conversion of
isoprene and 1-hexene to the corresponding hydrovinylation
products 10 and 11 as a test system (Scheme 4).
cobalt-catalysed 1,4-hydrovinylation reaction of
4 with
myrcene (Scheme 6) generates the triene 2 in 90% yield. Of
the two possible regioisomers, the major isomer is formed
where the new carbon–carbon bond is generated at the less
hindered side of the myrcene 1,3-diene subunit, in agreement
with the results obtained in the test system. To our great
delight, the regioselectivity of the cobalt-catalysed transforma-
tion was so high (98 : 2) that it was rather difficult to deter-
mine the amount of the undesired regioisomer. In addition,
the alcohol group could be applied unprotected and no
double bond isomerisation was detected under the reaction
conditions.
The best results were obtained for the dppp ligand system,
which gave the product 10 where the carbon–carbon bond
formation took place predominantly at the less hindered side
of the 1,3-diene. This regioselectivity can be rationalised by the
proposed reaction mechanism that is outlined in Scheme 5.
The reaction mechanism is initialised by a low valent cobalt
complex A, which is stabilised by dichloromethane solvent
molecules. The starting materials then coordinate within the
The purity of 2 was sufficiently high to continue the syn-
thesis of 1 without further purification. Intermediate 2 was
then oxidised prior to use by PCC (95% yield) adopting
known protocols described by Stumpp and Schmidt6g or
Coates and Johnson6c to the corresponding aldehyde 7. At
this point it is worth mentioning that the order of steps
concerning the cobalt-catalysed 1,4-hydrovinylation and the
PCC oxidation should not be inverted. While the cobalt-
catalysed transformation of 4 with myrcene (as outlined in
Scheme 6) proceeds without difficulties, the hydrovinylation of
the corresponding aldehyde derived from the alcohol 4
led even under the rather mild Lewis-acidic hydrovinylation
Scheme 4 Regioselectivity of the hydrovinylation process.
Scheme 5 Proposed reaction mechanism of the 1,4-hydrovinylation.
1396 | Chem. Commun., 2009, 1395–1397
Scheme 6 Synthesis of 2.
ꢀc
This journal is The Royal Society of Chemistry 2009

View Article Online
Chem., 2000, 37, 647; (c) L. Weber, K. Illgen and M. Almstetter,
Synlett, 1999, 366. For a review on boron functionalised building
blocks in cycloaddition reactions see: (d) G. Hilt and P. Bolze,
Synthesis, 2005, 2091.
2 Leading references: (a) B. M. Trost, J. A. Martinez, R. J.
Kulawiec and A. F. Indolese, J. Am. Chem. Soc., 1993, 115,
10402; (b) B. M. Trost, M. Machacek and M. J. Schnaderbeck,
Org. Lett., 2000, 2, 1761; (c) B. M. Trost, Acc. Chem. Res.,
2002, 35, 695; (d) B. M. Trost, H. C. Shen and A. B.
Pinkerton, Chem.–Eur. J., 2002, 8, 2341; (e) B. M. Trost,
M. U. Frederiksen and M. D. Rudd, Angew. Chem., Int. Ed.,
2005, 44, 6630.
3 J. Treutwein and G. Hilt, Angew. Chem., Int. Ed., 2007, 46,
8500.
4 (a) G. Hilt and S. Luers, Synthesis, 2002, 609; (b) G. Hilt, F.-X. du
¨
also: M. M. P. Grutters, C. Muller and D. Vogt, J. Am. Chem.
¨
Mesnil and S. Luers, Angew. Chem., Int. Ed., 2001, 40, 387; (c) See
¨
Scheme 7 Synthesis of 1.
Soc., 2006, 128, 7414; (d) C.-C. Wang, P.-S. Lin and C.-H.
Cheng, J. Am. Chem. Soc., 2002, 124, 9696; (e) H.-T.
Chang, T. T. Jayanth, C.-C. Wang and C.-H. Cheng, J. Am.
Chem. Soc., 2007, 129, 12032; (f) M. Petit, C. Aubert and
M. Malacria, Org. Lett., 2004, 6, 3937; (g) O. Buisine, C. Aubert
and M. Malacria, Chem.–Eur. J., 2001, 7, 3517; (h) D. Llerena,
C. Aubert and M. Malacria, Tetrahedron Lett., 1996, 37,
7027.
reaction conditions to an intramolecular Prins reaction.
Therefore, the oxidation of 2 to 7 should be performed under
Lewis-acid-free conditions, and once the aldehyde is generated
it should be used right away.
The sulfone 3 was deprotonated by LDA to generate the
lithiated sulfone precursor 8 in situ (Scheme 7). The Julia–
Kocienski olefination gave 9 in 80% yield with virtually exclusive
E-selectivity (97 : 3) of the newly formed double bond. The final
reduction of 9 to the alcohol 1 was accomplished by DIBAL in
quantitative yield (499%). The NMR data of 1 generated by this
route are identical with that appearing in the literature.6
The presented convergent synthesis presents the shortest
route to date to 1 over five steps in the longest linear sequence
starting from 5 and in seven total steps. The use of protecting
groups is not required and most reactions in the sequence are
carbon–carbon bond forming processes. The bottleneck of the
synthesis is the generation of the Z-configured double bond in
3 by the Peterson olefination. Nevertheless, the first applica-
tion of a cobalt-catalysed 1,4-hydrovinylation could be estab-
lished as a procedure to build sensitive 1,4-diene subunits in
linear target molecules in an atom-economic fashion in a high
yield and with excellent regioselectivity.
5 For biological activity of antibiotics such as moenomycin A, see:
(a) P. Welzel, Chem. Rev., 2005, 105, 4610; (b) M. Adachi,
Y. Zhang, C. Leimkuhler, B. Sun, J. V. LaTour and
D. E. Kahne, J. Am. Chem. Soc., 2006, 128, 14012.
6 For earlier syntheses of the aglycone 1, see: (a) R. Tschesche and
J. Reden, Liebigs Ann. Chem., 1974, 6, 853; (b) P. A. Grieco,
Y. Masaki and D. Boxler, J. Am. Chem. Soc., 1975, 97, 1597;
(c) R. Coates and M. W. Johnson, J. Org. Chem., 1980, 45, 2685;
(d) P. J. Kocienski, J. Org. Chem., 1980, 45, 2037;
(e) P. J. Kocienski and M. Todd, J. Chem. Soc., Perkin Trans. 1,
1983, 1777; (f) D. Bottger and P. Welzel, Tetrahedron Lett., 1983,
¨
24, 5201; (g) M. C. Stumpp and R. R. Schmidt, Tetrahedron, 1986,
42, 5941; (h) U. Schuricht, L. Hennig, M. Findeisen, P. Welzel and
D. Arigoni, Tetrahedron Lett., 2001, 42, 3835; (i) H. J. Huang and
W. B. Yang, Tetrahedron Lett., 2007, 48, 1429.
7 (a) K. Plesniak, A. Zarecki and J. Wicha, Top. Curr. Chem., 2007,
275, 163; (b) P. R. Blakemore, W. J. Cole, P. J. Kocienski and
A. Morley, Synlett, 1998, 26; (c) A. B. Charette, C. Berthelette and
D. St-Martin, Tetrahedron Lett., 2001, 42, 5149.
8 L. F. van Staden, D. Gravestock and D. J. Ager, Chem. Soc. Rev.,
2002, 31, 195.
9 C. Jasper, R. Wittenberg, M. Quitschalle, J. Jakupovic and
A. Kirschning, Org. Lett., 2005, 7, 479.
10 (a) B. Mudryk, C. A. Shook and T. Cohen, J. Am. Chem. Soc.,
1990, 112, 6389; (b) B. Mudryk and T. Cohen, J. Org. Chem., 1989,
54, 5657.
Notes and references
1 (a) G. Balme, E. Bossharth and N. Monteiro, Eur. J. Org. Chem.,
2003, 4101; (b) I. Ugi, A. Domling and B. Werner, J. Heterocycl.
11 P. Morschel, J. Janikowski, G. Hilt and G. Frenking, J. Am. Chem.
¨
Soc., 2008, 130, 8952.
¨
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 1395–1397 | 1397