Synthesis of (±)-Secosyrin 1 and a Formal Synthesis of (-) -Secosyrin 1

Article abstract of DOI:10.1021/ol0362313

(Equation presented) A short synthesis of (±)-secosyrin 1 is presented that starts from an electron-deficient furan; reductive alkylation under Birch conditions gives rapid access to the natural product skeleton. Two aspects of stereoselectivity are explo

Full text of DOI:10.1021/ol0362313

ORGANIC
LETTERS
2004
Vol. 6, No. 4
465-467
Synthesis of (±)-Secosyrin 1 and a
Formal Synthesis of (−)-Secosyrin 1
,†
Timothy J. Donohoe,* John W. Fisher,† and Paul J. Edwards‡
Dyson Perrins Laboratory, UniVersity of Oxford, South Parks Road,
Oxford, OX1 3QY, UK, and Sandwich Laboratories, Pfizer Global R&D,
Sandwich, Kent, CT13 9NJ, UK
timothy.donohoe@chem.ox.ac.uk
Received November 13, 2003
ABSTRACT
A short synthesis of (±)-secosyrin 1 is presented that starts from an electron-deficient furan; reductive alkylation under Birch conditions gives
rapid access to the natural product skeleton. Two aspects of stereoselectivity are explored, the first being directed dihydroxylation of a
homoallylic alcohol. Second, the facial selectivity obtained during reduction of a highly substituted cyclic ketone was examined. Finally, our
synthesis was rendered enantioselective by the reduction of a furan bearing a chiral auxiliary.
Secosyrins 1 and 2 and syributins 1 and 2 were isolated by
Sims and co-workers from pseudomonas syringae pv. tomato
as the major coproducts with the syringolide elicitors (Figure
1).1 The syringolides are of interest both for their unusual
response in resistant soybean plants. Though the secosyrins
and syributins do not display the same biological activity as
the syringolides, they are nonetheless of biosynthetic im-
portance and have been the subject of a number of syntheses.2
The secosyrins are formally related to the syringolides via
an intramolecular Claisen-type reaction and to the syributins
by an elimination, followed by 1,3 acyl migration.
In this paper, we wish to present a total synthesis of (()-
secosyrin, and a means to render this synthesis enantio-
selective, based on the methodologies we have developed
for the Birch reduction of electron-deficient furans3 and the
directed dihydroxylation of alkenes.4 Our approach to the
secosyrins involved disconnection to dihydrofuran 1 onto
which we would incorporate a trans-diol unit at C-3,4.
Adjustment of the oxidation state of the exocyclic carbon
(1) Midland, S.; Keen, N. T.; Sims, J. J. J. Org. Chem. 1995, 60, 1118.
(2) Previous syntheses of secosyrin 1: (a) Mukai, C.; Moharram, S. M.;
Hanaoka, M. Tetrahedron Lett. 1997, 38, 2511. (b) Yu, P.; Yang, Y.; Zhang,
Z. Y.; Mak, T. C. W.; Wong, H. N. C. J. Org. Chem. 1997, 62, 6359. (c)
Carda, M.; Castillo, E.; Rodriguez, S.; Falomir, E.; Marco, J. A. Tetrahedron
Lett. 1998, 39, 8895.
Figure 1.
(3) For examples of the Birch reduction of furans in synthesis see:
Semple, J. E.; Wang, P. C.; Lysenko, Z.; Joullie´, M. M. J. Am. Chem. Soc.
1980, 102, 7505. Donohoe, T. J.; Guillermin J.-B.; Walter, D. S. J. Chem.
Soc., Perkin Trans. 1 2002, 1369.
(4) Donohoe, T. J.; Moore, P. R.; Waring, M. J.; Newcombe, N. J.
Tetrahedron Lett. 1997, 38, 5027.
oxygen-rich tricyclic structure and their biological activity,
which centers around their ability to elicit a hypersensitive
† University of Oxford.
‡ Pfizer Global R&D.
10.1021/ol0362313 CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/28/2004

then revealed 2 as a likely precursor to 1 that could be
prepared itself by reductive alkylation of furan 3.
Our synthesis began with amide 4 (prepared in 98% yield
from furoyl chloride and pyrrolidine) and subjected it to
reductive alkylation in ammonia, quenching with p-meth-
oxybenzyl bromide (Scheme 1). In this reaction, we obtained
The syn and anti isomers of 7 were separable by chro-
matography, and the stereochemical identity of syn-7 was
proven by X-ray crystallography (Figure 2).
Scheme 1a
Figure 2.
We then proceeded to doubly protect the triol syn-7
intending to then invert the hydroxyl at C-3 (Scheme 2).8
Formation of 9 proceeded smoothly with 2 equiv of TBSOTf.
Initial attempts at forming and displacing a triflate derived
from 9 were unsuccessful (displacement at a neopentyl
center) and only yielded the ketone 10 (Scheme 2). Instead,
we oxidized the C-3 hydroxyl group to form ketone 11 with
Dess-Martin periodinane (DMP) and then attempted dia-
stereoselective reduction. Our first attempt at reduction of
11 used sodium borohydride (in THF),9 which gave a 1:1
mixture of 9 and 12. Further screening showed that two other
reducing agents (LiAlH4 and LiBH4) both gave more of the
desired configuration during the reduction process. Finally,
reduction with LiAlH4 at -78 °C gave a useful level of anti
selectivity (93:7). Interestingly, the bulky hydride source
lithium superhydride (LiEt3BH) yielded the undesired dia-
stereoisomer selectively.
a Reagents and conditions: (a) Na, NH3 then ArCH2Br; (b) (aq)
HCl; (c) LiAlH4.
a mixture of amides derived from both pyrrolidine and
ammonia and therefore modified the conditions (by warming)
to allow efficient conversion into 5. After hydrolysis and
reduction to yield alcohol 6, we investigated oxidation of
the alkene in order to prepare the trans-1,2-diol unit. The
first approach that we adopted involved epoxidation of 6 with
both mcpba and VO(acac)2/t-BuOOH. Unfortunately, forma-
tion of the epoxide was not observed under either of these
reaction conditions.
If we attempt to rationalize the reduction reactions, then
arguments based on steric interaction between the substrate
and hydride source would appear to favor the formation of
Therefore, we turned to tactics that involved dihydroxyl-
ation of the alkene, which would clearly have to be followed
by an inversion sequence. The dihydroxylation of 6 was
examined under a variety of different conditions. The results
in Scheme 1 show that Upjohn-type conditions5 (i) were only
moderately selective for formation of syn-7. However, this
bias could be increased by utilizing hydrogen bonding control
during the oxidation sequence. Conditions (ii) and (iii) were
capable of exhibiting a higher degree of hydrogen bonding
during oxidation with catalytic OsO4.6 Interestingly, we found
that these catalytic hydrogen bonding conditions were cleaner
if they were run in the presence of a polymer-bound DABCO
(8).7
Scheme 2a
Finally, we also showed that complete selectivity for syn-7
could be obtained using stoichiometric OsO4 with TMEDA
at low temperatures.4
(5) VanRheenen, V.; Kelly, R. C.; Cha, D. Y. Tetrahedron Lett. 1976,
1973. Quinuclidine was added because it increases the rate of reaction.
(6) For example, see: Blades, K.; Donohoe, T J.; Winter, J. J. G.; Stemp,
G. Tetrahedron Lett. 2000, 41, 4701.
(7) Compound 8 is 1,4-diazabicyclo[2.2.2]octane hydrochloride, polymer
bound, 1% DVB, 100-200 mesh. Prior to the reaction, the polymer was
added to a solution of OsO4 in cyclohexane; the solvent was then evaporated,
and the solid so obtained was added to the dihydroxylation mixture. See:
Cainelli, G.; Contento, M.; Manescalchi, Plessi, L. Synthesis 1989, 45.
a Reagents and conditions: (a) TBSOTf, -78 °C; (b) DMP,
CH2Cl2; (c) Tf2O then KOAc, 18-C-6.
466
Org. Lett., Vol. 6, No. 4, 2004

9. This is consistent with the result using bulky LiEt3BH.
As far as the contrasteric selectivity observed with the less
bulky reducing agents (to give 12) is concerned, we suggest
two possible reasons why sterics can be overturned. First,
the p-methoxy aryl group may be involved in an electronic
(π-stacking) interaction with the carbonyl, thus effectively
shielding the lower face of the carbonyl compound.10 Second,
the role of stereoelectronic effects in this system (such as
approach by a nucleophile so as to minimize torsional
strain11) may also be an important factor in determining the
observed stereoselectivity.
Scheme 4a
To complete the synthesis, the hexanoic ester side-chain
was attached at C-3 (12f13) and then the electron-rich aryl
ring was oxidized to carboxylic acid 14 using catalytic
ruthenium tetroxide12 (Scheme 3). Lactonization ensued after
a Reagents and conditions: (a) BuLi then TMSCl; (b) Na, NH3
then pMeOC6H4CH2Br; (c) (aq) HCl, ∆; (d) LiAH4.
Scheme 3a
setting the correct enolate geometry during the reduction
reaction. So, compound 16 was prepared by ortholithiation
of 15, followed by trapping with TMSCl. Compound 16 was
then reductively alkylated with p-methoxybenzyl bromide
to an inseparable mixture of diastereoisomers 17 (g20:1).
The sense of diastereoselectivity for this reaction is assumed
to be consistent with our previously described model for the
reductive alkylation of 16.14
Finally, acid hydrolysis of the amide auxiliary was
performed with 10 M HCl so as to promote concomitant
protodesilylation of the vinyl silane. After reduction of acid
18, the alcohol 6 was shown to be g90% ee by HPLC on a
chiral column (measured against a racemic standard).
Clearly, if we so desired, this material could be converted
into (-)-secosyrin 1 (g90% ee) by the sequence shown
above,15 and therefore this represents a formal synthesis of
(-)-secosyrin 1.
To conclude, we have demonstrated the effectiveness of
partial reduction reactions in total synthesis by completing
an 11-step synthesis of secosyrin 1 from readily available
starting materials. We have also further showcased the utility
of our directed dihydroxylation method for setting the alcohol
stereochemistry in this system. Finally, we have also been
able to apply our recently developed chiral auxiliary meth-
odology to develop a formal total synthesis of (-)-secosyrin
1.
a Reagents and conditions: (a) hexanoic anhydride; (b) NalO4,
(cat.) RuCl3‚2H2O, H2O, CCl4, MeCN; (c) (CF3CO)2O/CF3CO2H,
TBAF.
reaction of 14 with trifluoroacetic anhydride/trifluoroacetic
acid, and finally deprotection of the more robust secondary
OTBS group was achieved, in one-pot, with the addition of
TBAF. Secosyrin 1 produced in this way exhibited spectro-
scopic data that matched that of the natural product.13
Finally, we completed a formal synthesis of (-)-secosyrin
by performing a Birch reduction on the auxiliary laden furan
16 (Scheme 4). Previous work had shown that the presence
of a substituent at C-3 was essential to the success of the
asymmetric methodology.14 We believe that the requirement
for a substituent ortho to the acyl group has its origins in
Acknowledgment. We would like to thank the EPSRC
and Pfizer for funding this project. AstraZeneca, Pfizer, and
Novartis are also thanked for generous unrestricted support.
A. A. Calabrese is thanked for some initial experimental
studies.
Supporting Information Available: Copies of 1H NMR
spectra and detailed spectroscopic data for all new com-
pounds and representative experimental procedures. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(8) See: Donohoe, T. J.; Headley, C. E.; Cousins, R. P. C.; Cowley, A.
Org. Lett. 2003, 5, 999.
(9) See: Reductions by the Alumino- and Borohyrides in Organic
Synthesis; Seyden-Penne, J.; Wiley VCH: New York, 1997.
(10) Teixeira, L. H. P.; Barreiro, E. J.; Fraga, C. A. M. Synth. Commun.
1997, 27, 3241; for a review see: Jones, G. B. Tetrahedron 2001, 57, 7999.
(11) See: Larsen, C. H.; Ridagway, B. H.; Shaw, J. T.; Woerpel, K. A.
J. Am. Chem. Soc. 1999, 121, 12208.
(12) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. J.
Org. Chem. 1981, 46, 3936.
(13) See Supporting Information. Secosyrin 1: δC (CDCl3) 174.1, 173.7,
86.53, 81.99, 75.88, 75.62, 72.59, 35.37, 34.00, 31.16, 24.46, 22.21, 13.84;
lit.2b, δC (CDCl3) 174.4, 173.6, 86.61, 81.67, 76.40, 75.85, 72.83, 35.47,
34.01, 31.15, 24.46, 22.19, 13.75 ppm.
OL0362313
(14) Donohoe, T. J.; Calabrese, A. A.; Guillermin, J.-B.; Walter, D. S.
Tetrahedron Lett. 2001, 42, 5841. Donohoe, T. J.; Calabrese, A. A.;
Guillermin, J.-B.; Frampton C. S.; Walter, D. J. Chem. Soc., Perkin Trans.
1 2002, 1748.
(15) This argument assumes that compound (-)-6 does not racemize
under the dihydroxylation conditions, which seems unlikely.
Org. Lett., Vol. 6, No. 4, 2004
467