Total synthesis of brasoside and littoralisone

Article abstract of DOI:10.1021/ja050064f

The first total syntheses of littoralisone (1) and brasoside (2) have been achieved in 13 overall steps. Both natural products are forged from a common intermediate which is rapidly assembled using organocatalytic technology, including a proline-catalyzed

Full text of DOI:10.1021/ja050064f

Published on Web 02/26/2005
Total Synthesis of Brasoside and Littoralisone
Ian K. Mangion and David W. C. MacMillan*
DiVision of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125
Received January 5, 2005; E-mail: Dmacmill@caltech.edu
Extracts of Verbena littoralis, a plant used widely in traditional
folk remedies, possess intriguing activity as enhancers for the neuro-
trophic properties of nerve growth factor (NGF).1 Littoralisone (1),
isolated in 2001, was demonstrated to be the active agent for in-
creased NGF-induced neurite outgrowth in PC12D cells.2,3 While
littoralisone is presumed2 to be biochemically derived from bra-
soside (2),4 no relevant intermediates to support this pathway have
been found to date. A uniquely complex member of the iridoid
class,5 littoralisone presents a variety of challenges for chemical
synthesis including the assembly of a stereochemically elaborate
cyclobutane ring, an adjacent nine-membered lactone, and 14
stereocenters within a 24-carbon framework. In this communication,
we report the first total syntheses of littoralisone and brasoside in
13 overall steps. These syntheses confirm the absolute and relative
stereochemistries of isolates 1 and 2 and provide chemical support
for a biosynthetic pathway between the two. More importantly, these
syntheses demonstrate the capacity of proline catalysis to (i) override
the inherent stereochemical bias of iridoid formation and (ii)
selectively control aldol additions and C-O bond formation in a
complex target setting.
Figure 1. An organocatalytic approach toward (-)-littoralisone.
Scheme 1. Enantioselective Synthesis of the Iridolactone Corea
Given the presumed biochemical origins of littoralisone, we
envisioned the late-stage construction of both the nine- and four-
membered rings via an intramolecular photolytic [2+2] cycload-
dition. This strategy would require an unsaturated iridoid and a
a Reagents and conditions: (a) MesCl, DMAP, pyridine, CH2Cl2. (b)
cinnamoyl olefin held in proximity by an appropriately function-
alized glucoside tether 3 that, in turn, would arise from the
glycosidic coupling of iridolactone 4 and the 2-cinnamoyl saccharide
5 (Figure 1). As the pivotal step in our synthesis, we hoped the
iridoid stereochemical core might be accessed via a catalyst-
controlled intramolecular Michael addition with the formyl-enal 6.
While the elegant work of Schreiber6 has demonstrated that iridoid
systems can be generated under thermodynamic control via stoi-
chiometric enamine Michael additions, we recognized that the
resident (3S,5S) stereogenicity of enal 6 would lead to the undesired
trans-(1S,2S) cyclopentyl product 10 (see Table 1). As such, a
central goal of this study was to determine if proline might be
employed as an external source of stereoinduction to override
substrate-directed selectivities in iridoid synthesis. With respect to
the glycosidic coupling partner, we identified the glucose system
5 as an ideal target for our recently developed two-step approach
to differentially protected carbohydrates.7
O3, MeOH, CH2Cl2, -78 °C. (c) PhNO, D-proline (40 mol %), DMSO;
(EtO)2P(O)CH2CO2Me, LiCl, DBU; NH4Cl, MeOH. (d) TBDPSCl, imi-
dazole, DMF. (e) DIBAL, Et2O, -78 °C. (f) DMP, CH2Cl2. (g) Table 1,
entry 5; Ac2O, DMAP, pyridine, 0 °C. (h) POCl3, DMF, 40 °C. (i) NaClO2,
NaH2PO4, t-BuOH. (j) HF‚pyridine, THF. (k) DCC, CH2Cl2.
Synthesis of the iridolactone core 4 began with protection of
(-)-citronellol as its mesitoate ester followed by ozonolysis of the
resident olefin (Scheme 1). Exposure of the resulting aldehyde 7
to the proline-catalyzed R-formyl oxidation protocol using ni-
trosobenzene furnished the corresponding R-oxyamino aldehyde
with high levels of catalyst-enforced induction.8,9 Conveniently,
olefination of the resultant aldehyde using Horner-Wadsworth-
Emmons conditions and cleavage of the aminoxy bond (upon
standing in MeOH) could be accomplished in the same protocol to
provide γ-chiral R,â-unsaturated ester 8 in a single operation from
7.10 Protection of the resulting alcohol as its TBDPS-ether followed
by treatment with DIBAL and then the Dess-Martin oxidant,
provided the requisite formyl-enal Michael substrate 6 in six steps.
9
3696
J. AM. CHEM. SOC. 2005, 127, 3696-3697
10.1021/ja050064f CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Examination of the Key Intramolecular Michael Addition
Scheme 3. Completion of the Synthesis of Brasoside and
Littoralisonea
entry
catalyst
solvent
yielda (%)
9:10
1b
2
3
L-proline
L-proline
PhNH(Me)
L-proline
L-proline
(()-proline
D-proline
CHCl3
CHCl3
CHCl3
MeOH
DMSO
DMSO
DMSO
61
54
87
26
91
86
83
3:1
1:19
1:19
7:1
10:1
2:1
4
5c
6c
7c
1:2
a Combined yield of isomers 9, 10. b After 12 h. c Performed at 40 °C.
Scheme 2. Two Step Assembly of Selectively Substituted
Glucosea
a Reagents and conditions: (a) 1-O-(TMS)-â-D-glucose tetraacetate,
TMSOTf (0.4 equiv), CH3CN, -30 °C. (b) Et3N, H2O, MeOH, CH2Cl2,
-15 °C.
respects to the natural isolate. It should be noted that the [2+2]
cycloaddition was also observed to occur slowly in ambient light,
a result that lends support to the proposed biochemical formation
of littoralisone from brasoside.2
In summary, the first total synthesis of littoralisone has been
achieved in 13 steps and in 13% overall yield. Prominent features
of this synthesis include the use of proline to (i) overcome the
inherent stereoinduction of enamine-Michael reactions and (ii)
enable the two-step asymmetric construction of a polyol differenti-
ated glucose coupling partner.
a Reagents and conditions: (a) Ag2O, BnBr. (b) Pd/Al2O3, HCO2NH4.
(c) TMSCl, Et3N, 80 °C.
At this point, we sought to test the feasibility of the proposed
contra-thermodynamic intramolecular Michael addition. To our
delight treatment of 6 with L-proline in CHCl3 provided the desired
lactol 9 in good yield, albeit as a 3:1 mixture with its monocyclic
isomer 10 (Table 1, entry 1). Unfortunately, prolonged exposure
of 9 to catalytic conditions (entry 2) resulted in its complete
conversion to 10, revealing that the trans isomer is thermodynami-
cally favored. Exposure of enal 6 to Schreiber’s protocol6a also
results in trans-cyclopentyl formation (entry 3). However, proline
catalysis can be employed in high dielectric media to provide the
desired kinetic outcome,11 in accord with our retrosynthetic
hypothesis. Indeed, exposure of 6 to L-proline in DMSO provides
the lactol 9 in 91% yield and with 10:1 cis-selectivity (entry 5).
Interestingly, catalyst turnover appears to be rate limiting, a process
that can be accelerated by heat or by the addition of water.12 The
former was found to be amenable to in situ acetylation of 9, leading
directly to 11 in 83% yield in a single operation. Conversion of
the iridoid 11 to the lactone moiety 4 was then accomplished in
four steps and 56% yield using standard methods (Scheme 1).
Synthesis of the requisite glycosidic coupling partner began with
proline-catalyzed dimerization of benzyloxyacetaldehyde to provide
12,13 which upon treatment with enolsilane 1313 led to trisbenzyl-
2-cinnamoyl glucose 14 (Scheme 2). Benzylation of this carbohy-
drate was then followed by regio- and diastereoselective function-
alization of the anomeric alcohol to provide exclusively the TMS-
ether â-anomer 15.
Acknowledgment. Financial support was provided by the
NIHGMS (R01 GM66142-01) and kind gifts from Amgen and
Merck. I.K.M. is grateful for a NSF predoctoral fellowship. Amanda
Reider is thanked for experimental contributions.
Supporting Information Available: Experimental procedures and
spectral data. This material is available free of charge via the Internet
at http://pubs.acs.org.
References
(1) Umana, E.; Castro, O. Int. J. Crude Drug Res. 1990, 28, 175.
(2) Li, Y.-S.; Matsunaga, K.; Ishibashi, M.; Ohizumi, Y. J. Org. Chem. 2001,
66, 2165.
(3) For reviews on biological activities of iridoid natural products see: (a)
Buzogany, K.; Cucu, V. Farmacia 1983 31, 129. (b) Tietze, L.-F. Angew.
Chem., Int. Ed. Engl. 1983, 22, 828.
(4) (a) Schafer, B.; Rimpler, H. Z. Naturforsch. 1979, 34, 311. (b) Franke,
A.; Rimpler, H. Phytochemistry 1987, 26, 3015.
(5) For iridoid syntheses see: (a) Bu¨chi, G.; Carlson, J. A.; Powell, J. E.;
Tietze, L.-F. J. Am. Chem. Soc. 1970, 92, 2165. (b) Callant, P.; Storme,
P.; Van der Eycken, E.; Vandewalle, M. Tetrahedron Lett. 1983, 24, 5797.
(c) Trost, B. M.; Balkovec, J. M.; Mao, M. K.-T. J. Am. Chem. Soc. 1986,
108, 4974. (d) Piccinini, P.; Vidari, G.; Zanoni, G. J. Am. Chem. Soc.
2004, 126, 5088 and references therein.
(6) (a) Schreiber, S. L.; Meyers, H. V.; Wiberg, K. B. J. Am. Chem. Soc.
1986, 108, 8274. For an intramolecular Michael addition using our
imidazolidinone catalyst that provides the trans-cyclopentyl isomer see:
(b) Fonseca, M. T. H.; List, B. Angew. Chem., Int. Ed. 2004, 43, 3958.
(7) Northrup, A. B.; MacMillan, D. W. C. Science 2004, 305, 1752.
(8) Brown, S. P.; Brochu, M. P.; Sinz, C. J.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2003, 125, 10808.
The glycosidic union of 11 and 15 was accomplished in a facile
process using TMSOTf 14 to provide the desired glucose-tethered
diene 16 in 74% yield (Scheme 3). In a similar fashion, iridoid 11
was coupled with 1-O-(TMS)-â-D-glucose tetracetate, which upon
deacetylation, furnished (-)-brasoside in 13 overall steps. To
complete the synthesis of littoralisone, we next turned our attention
to the proposed intramolecular [2+2] photocyloaddition. To our
delight, exposure of 16 to UV light (350 nm) for 2 h followed by
in situ hydrogenolysis furnished synthetic (-)-littoralisone as a
single isomer in 84% yield, a substance that was identical in all
(9) For other reports on proline-catalyzed oxidation of aldehydes, see: (a)
Zhong, G. Angew. Chem., Int. Ed. 2003, 42, 4247. (b) Hayashi, Y.;
Yamaguchi, J.; Hibino, K.; Shoji, M. Tetrahedron Lett. 2003, 44, 8293.
(10) For a two-step approach, see: Zhong, G.; Yu, Y. Org. Lett. 2004, 6, 1637.
(11) Exposure of 9 to D-proline in DMSO for 48 h does not lead to formation
of 10. This reveals that the outcomes of entries 5, 6, and 7 are kinetic.
(12) H2O (2% v/v) enables full conversion to 9 in 48 h at 23 °C in DMSO.
(13) Northrup, A. B.; Mangion, I. K.; Hettche, F.; MacMillan, D. W. C. Angew.
Chem., Int. Ed. 2004, 43, 2152.
(14) See Supporting Information for details.
JA050064F
9
J. AM. CHEM. SOC. VOL. 127, NO. 11, 2005 3697