Enantioselective total synthesis of (+)-neosymbioimine

Article abstract of DOI:10.1021/ol070049a

Equation presented The enantioselective total synthesis of (+)-neosymbioimine was accomplished in 18 steps from (-)-(S)-citronellol utilizing an organocatalytic α-oxidation of aldehyde 6. The carbon core was constructed by a tandem Horner-Wadsworth-Emmons

Full text of DOI:10.1021/ol070049a

ORGANIC
LETTERS
2007
Vol. 9, No. 8
1461-1464
Enantioselective Total Synthesis of
)-Neosymbioimine
(+
Georgy N. Varseev and Martin E. Maier*
Institut fu¨r Organische Chemie, UniVersita¨t Tu¨bingen, Auf der Morgenstelle 18,
72076 Tu¨bingen, Germany
martin.e.maier@uni-tuebingen.de
Received January 8, 2007
ABSTRACT
The enantioselective total synthesis of (
+
)-neosymbioimine was accomplished in 18 steps from (
-oxidation of aldehyde 6. The carbon core was constructed by a tandem Horner Wadsworth
Alder cyclization. All double bonds of 12 were made in a stereoselective manner by Wittig-type reactions. Selective formation of the
−
)-(S)-citronellol utilizing an organocatalytic
r
−
−Emmons (HWE) reaction and an intramolecular
Diels
−
monosulfate monoester was accomplished by one-pot excessive sulfation followed by kinetic hydrolysis of bissulfate 18 in 79% yield.
Neosymbioimine (1) is a minor amphoteric metabolite of
the symbiotic marine dinoflagellate Symbiodinium sp.1 Like
symbioimine (2), this alkaloid is composed of a tricyclic
iminium core with an attached resorcinol monosulfate unit
that causes these molecules to occur as inner salts (Figure
1).2-4 Organic sulfate monoesters are widespread in biologi-
cal systems, occurring in proteins, carbohydrates, steroids,
and other small molecules.5,6 In contrast, zwitterionic com-
pounds containing sulfates are still very rare4 and could be
of interest for either organic or medicinal chemistry. For
example, the parent compound symbioimine (2) inhibits the
differentiation of precursor osteoclast cells (RAW264) into
mature osteoclasts. Also, it slightly inhibits cyclooxygenase-2
(COX-2) activity. Furthermore, it did not affect cell viability
even at 100 µg mL-1.1,2 Because of their low toxicity, these
alkaloids could be used in the development of efficient
antiosteoporosis drugs. A concise total synthesis of these
metabolites would be of considerable interest to further
delineate the biological activity.
(1) Kita, M.; Ohishi, N.; Washida, K.; Kondo, M.; Koyama, T.; Yamada,
K.; Uemura, D. Bioorg. Med. Chem. 2005, 13, 5253-5258.
(2) Kita, M.; Kondo, M.; Koyama, T.; Yamada, K.; Matsumoto, T.; Lee,
K.-H.; Woo, J.-T.; Uemura, D. J. Am. Chem. Soc. 2004, 126, 4794-4795.
(3) Kita, M.; Uemura, D. Chem. Lett. 2005, 34, 454-459.
(4) For some examples, see: (a) Herlt, A.; Mander, L.; Rombang, W.;
Rumampuk, R.; Soemitro, S.; Steglich, W.; Tarigan, P.; von Nussbaum, F.
Tetrahedron 2004, 60, 6101-6104. (b) Gray, C. A.; de Lira, S. P.; Silva,
M.; Pimenta, E. F.; Thiemann, O. H.; Oliva, G.; Hajdu, E.; Andersen, R.
J.; Berlinck, R. G. S. J. Org. Chem. 2006, 71, 8685-8690.
(5) (a) Simpson, L. S.; Widlanski, T. S. J. Am. Chem. Soc. 2006, 128,
1605-1610 and references therein. (b) Huibers, M.; Manuzi, AÄ .; Rutjes,
F. P. J. T.; van Delft, F. L. J. Org. Chem. 2006, 71, 7473-7476.
(6) Liu, Y.; Lien, I. F. F.; Ruttgaizer, S.; Dove, P.; Taylor, S. D. Org.
Lett. 2004, 6, 209-212.
Figure 1. Retrosynthetic analysis for neosymbioimine (1).
10.1021/ol070049a CCC: $37.00
© 2007 American Chemical Society
Published on Web 03/23/2007

Recently, we reported the first total synthesis of (()-
symbioimine (2) using an intramolecular endo-Diels-Alder
(endo-IMDA) reaction of an (E,E,E)-undeca-2,8,10-trien-1-
al as a key step.7 Meanwhile, a synthesis of the octalene
core of symbioimine, using a related endo-IMDA reaction
of an (E,E,E)-undeca-2,8,10-trien-1-amide, was reported by
Uemura.8 Another approach to the synthesis of 2 was realized
by Snider et al., who prepared (()-symbioimine through an
IMDA reaction of a 2,3-dihydropyridinium cation.9 Although
2 was synthesized by two different ways, the synthesis of
neosymbioimine (1) was not published until this time. Herein,
we report the first total synthesis of neosymbioimine (1) and
the assignment of its absolute stereochemistry.
Our retrosynthetic strategy is outlined in Figure 1. The
target compound should be available from nitrile 3, which
possesses all stereogenic centers of 1. The octalene unit of
3 might be accessed by an endo-IMDA of triene 4.10-12 All
double bonds of 4 could be made using Wittig chemistry,
and the secondary alcohol function could be introduced by
a proline-catalyzed R-hydroxylation of the corresponding
aldehyde.13 As the starting material, we identified (-)-(S)-
citronellol (5), which is readily available from rose oil (ee
ca. 92%).14,15
4 was required.7 Initially, the alcohol function of (-)-(S)-
citronellol (5) was protected as a TBS ether and the double
bond was cleaved by ozonolysis to obtain aldehyde 6
(Scheme 1). Aldehyde 6 was converted to 4-hydroxy-8-
silyloxy-enoate 7 by a one-pot MacMillan procedure, includ-
ing R-hydroxylation followed by Horner-Wadsworth-
Emmons (HWE) reaction and cleavage of the N-O bond
in 55% yield.17 The secondary alcohol of enoate 7 was
protected with a TBS group, followed by selective libera-
tion of the primary alcohol and its oxidation to provide
aldehyde 8. Wittig reaction of 8 with 2-(triphenylphos-
phoranylidene)propanal18 (9) gave only the E-isomer of
aldehyde 10 in 88% yield. In the subsequent condensation
of 10 with diethyl 3,5-dimethoxybenzylphosphonate19 (11),
we isolated approximately a 1:1 mixture of triene 12 and
cyclized product 13. Heating this mixture in chloroform for
2 h at 60 °C induced conversion of trienoate 12 to 13 to
about 95% (by NMR). The conversion did not increase even
after 12 h. We assume that only the desired diastereomer of
12 cyclizes under these conditions. The diastereomeric
impurity in substrate 12 most likely originates from the minor
enantiomer contained in the commercial (-)-(S)-citronellol
(92% ee).14 In fact, around 5% of impurity can be seen in
the 1H NMR spectra of compounds 7-13. The minor
diastereomer of 12 does not cyclize due to the steric
hindrance in the transition state (Scheme 2). After reduction
of ester 13, the target alcohol could be easily obtained in
pure form. This crucial effect of the methyl group on the
IMDA of 12 allowed us to produce enantio- and diastereo-
merically pure products from the enantioimpure starting
compound citronellol.
Even though C-5 of neosymbioimine is sp2-hybridized,
we planned to temporarily use a secondary alcohol function
at this position to guarantee a high diastereoselectivity in
the Diels-Alder step. As we found in the symbioimine
case, a keto function at C-5 gave a lower endo/exo-
diastereoselectivity in this crucial transformation.16 On the
basis of the results from the symbioimine study, the (S)-
configuration for the hydroxyl-carrying stereocenter of triene
Mesylation of the primary alcohol derived from cycload-
duct 13 followed by SN2 reaction with sodium cyanide gave
nitrile 14 in exellent yield. Methylation of acetonitrile deriv-
ative 14 proceeded very cleanly and efficiently, using 2 equiv
of LDA and 2 equiv of MeI in THF at -80 °C. Despite the
fact that a small amount (10%) of the undesired diastereomer
was also formed, pure 3 was isolated in 87% yield on a gram-
scale experiment. The structure of nitrile 3, featuring all
stereocenters of the natural product, was confirmed by single-
crystal X-ray analysis (Scheme 1). In this molecule, the
methyl and propanenitrile groups occupy equatorial positions,
whereas the bulky units (Ar, OTBS) are oriented axially. In
the 1H NMR spectra of nitrile 3, the signal of the R-methyl
group is shifted toward the high field (δH 0.6 ppm) due to
the shielding by the aromatic ring current. Concerning the
origin of the diastereoselective alkylation, one could argue
that the X-ray structure of nitrile 3 shows the available air
corridor of the electrophile. Thus, the methyl iodide could
approach the nitrile anion above the aryl ring and opposite
to the large tert-butyldimethylsilyloxy group.
(7) Varseev, G. N.; Maier, M. E. Angew. Chem. 2006, 118, 4885-4889;
Angew. Chem., Int. Ed. 2006, 45, 4767-4771.
(8) Sakai, E.; Araki, K.; Takamura, H.; Uemura, D. Tetrahedron Lett.
2006, 47, 6343-6345.
(9) (a) Snider, B. B.; Che, Q. Angew. Chem. 2006, 118, 946-949; Angew.
Chem., Int. Ed. 2006, 45, 932-935. (b) Zou, Y.; Che, Q.; Snider, B. B.
Org. Lett. 2006, 8, 5605-5608.
(10) For recent reviews about IMDA reactions, see: (a) Roush, W. R.
In ComprehensiVe Organic Synthesis; Paquette, L. A., Ed.; Pergamon
Press: Oxford, 1991; Vol. 5, pp 513-550. (b) Nicolaou, K. C.; Snyder, S.
A.; Montagnon, T.; Vassilikogiannakis, G. Angew. Chem. 2002, 114, 1742-
1773; Angew. Chem., Int. Ed. 2002, 41, 1668-1698. (c) Takao, K.-i.;
Munakata, R.; Tadano, K.-i. Chem. ReV. 2005, 105, 4779-4807.
(11) For some related examples (secondary alcohol next to dienophile),
see: (a) Funk, R. L.; Zeller, W. E. J. Org. Chem. 1982, 47, 180-182. (b)
Burke, S. D.; Magnin, D. R.; Oplinger, J. A.; Baker, J. P.; Abdelmagid,
A. Tetrahedron Lett. 1984, 25, 19-22. (c) Marshall, J. A.; Audia, J.
E.; Grote, J. J. Org. Chem. 1984, 49, 5277-5279. (d) Marshall, J. A.;
Audia, J. E.; Grote, J.; Shearer, B. G. Tetrahedron 1986, 42, 2893-
2902. (e) Trost, B. M.; Holcomb, R. C. Tetrahedron Lett. 1989, 30, 7157-
7160.
(12) See also: Mergott, D. J.; Frank, S. A.; Roush, W. R. Proc. Natl.
Acad. Sci. U.S.A. 2004, 101, 11955-11959.
(13) (a) Zhong, G. Angew. Chem. 2003, 115, 4379-4382; Angew. Chem.,
Int. Ed. 2003, 42, 4247-4250. (b) Brown, S. P.; Brochu, M. P.; Sinz, C.
J.; MacMillan, D. W. C. J. Am. Chem. Soc. 2003, 125, 10808-10809. (c)
Hayashi, Y.; Yamaguchi, J.; Hibino, K.; Shoji, M. Tetrahedron Lett. 2003,
44, 8293-8296. (d) Co´rdova, A.; Sunde´n, H.; Bøgevig, A.; Johansson, M.;
Himo, F. Chem.-Eur. J. 2004, 10, 3673-3684. (e) Hayashi, Y.; Yamaguchi,
J.; Sumiya, T.; Hibino, K.; Shoji, M. J. Org. Chem. 2004, 69, 5966-
5973.
Nitrile 3 was treated with TBAF to remove the TBS group.
The obtained alcohol was oxidized to the corresponding
(14) Ravid, U.; Putievsky, E.; Katzir, I.; Ikan, R.; Weinstein, V. FlaVour
(17) Mangion, I. K.; MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127,
3696-3697.
Fragrance J. 1992, 7, 235-238.
(15) For the synthesis of citronellol by hydrogenation, see: Takaya, H.;
Ohta, T.; Inoue, S.-i.; Tokunaga, M.; Kitamura, M.; Noyori, R. Org. Synth.
1995, 72, 74-85; Org. Synth., Coll. 1998, 9, 169.
(18) Schlessinger, R. H.; Poss, M. A.; Richardson, S.; Lin, P. Tetrahedron
Lett. 1985, 26, 2391-2394.
(19) Chen, G.; Shan, W.; Wu, Y.; Ren, L.; Dong, J.; Ji, Z. Chem. Pharm.
Bull. 2005, 53, 1587-1590.
(16) Varseev, G. N. unpublished results.
1462
Org. Lett., Vol. 9, No. 8, 2007

Scheme 1. Synthesis of (+)-Neosymbioimine (1) from (-)-(S)-Citronellol (5)a
a
Reagents and conditions: (a) TBSCl, imidazole, DMF, 23 °C, 12 h (95%); (b) O3, -78 °C, CH2Cl2/MeOH (1:1), then Me2S, -78 to
20 °C, 16 h (93%); (c) nitrosobenzene (1 equiv), D-proline (0.4 equiv), DMSO, 20 °C, 25 min, then triethylphosphono acetate, DBU, LiCl,
0 °C, 15 min, then MeOH, NH4Cl, Cu(OAc)2, 24 h; (d) TBSCl, imidazole, DMAP (0.1 equiv), CH2Cl2, 23 °C, 14 h (88%); (e) CSA (0.05
equiv), MeOH, 20 °C, 1 h (87%); (f) Dess-Martin periodinane, CH2Cl2, 20 °C, 2 h (96%); (g) 2-(triphenylphosphoranylidene)propanal (9),
toluene, 100 °C, 30 h (88%); (h) diethyl 3,5-dimethoxybenzylphosphonate (11), KOtBu, THF, -78 to 0 °C, then CHCl3, 60 °C, 2 h (86%);
(i) HAl(iBu)2, CH2Cl2, 23 °C, 24 h (88%); (j) MsCl, Et3N, CH2Cl2, 0 °C, 1 h (94%); (k) NaCN, DMSO, 50 °C, 3 days (98%); (l) LDA (2
equiv), THF, -80 °C, 1 h, then MeI (2 equiv), 0.5 h (87%); (m) TBAF, THF, 23 °C, 24 h (99%); (n) Dess-Martin periodinane, CH2Cl2,
23 °C, 12 h (96%); (o) (CH2OH)2, CSA (0.5 equiv), C6H6, 80 °C, 10 h (94%); (p) LiAlH4, Et2O, 23 °C, 24 h, then 3 M HCl, THF, 50 °C,
2 h (90%); (q) BBr3 (5 equiv), CH2Cl2, -80 to 0 °C, 1 h, then 0 °C, 1.5 h (84%); (r) SO3‚Py, pyridine, 60 °C, 24 h, then H2O/MeOH (1:2),
36 °C, 18 h (79%) brsm. TBS ) tert-butyldimethylsilyl. DMF ) dimethylformamide. DBU ) 1,8-diazabicycloundec-7-en. DMAP )
4-dimethylaminopyridine. CSA ) camphorsulfonic acid. DMP ) Dess-Martin periodinane. DIBAL ) diisobutylaluminum hydride. MsCl
) methanesulfonyl chloride. LDA ) lithium diisopropylamide. TBAF ) tetrabutylammonium fluoride. Py ) pyridine.
ketone which was protected as 1,3-dioxolane 15 in excellent
yield. Reduction of the nitrile to the amine with LiAlH4
followed by acid-catalyzed imine formation gave imine 16.
Finally, the aryl ether functions were cleaved with BBr3
providing resorcinol 17 in 84% yield.
routine flash chromatography, on silica gel using MeOH/
CHCl3 as eluent. Synthetic neosymbioimine was identical
by NMR to the natural compound. The optical rotation of
synthetic neosymbioimine ([R]23 +172, c 0.1, MeOH)
D
correlates well with the authentic one ([R]25 +149). Thus,
D
the absolute stereochemistry of (+)-neosymbioimine was
assigned.
Scheme 2. Stereochemistry of the IMDA Reaction of
Trienoate 12
Scheme 3. Selective Monohydrolysis of 18
As a final challenge, efficient monosulfation of resorcinol
17 remained. Excessive sulfation of 17 with SO3‚py (5 equiv)
in pyridine gave bisulfate 18. As expected, one sulfate group
in 18 is more sensitive to hydrolysis (Scheme 3). Accord-
ingly, keeping a solution of 18 in aqueous methanol at 36
°C for 18 h led to complete monohydrolysis of 18 (70% of
1 was isolated). At the same time, neosymbioimine (1) was
not hydrolyzed significantly (10% of 17 was recovered). It
seems that the formation of the inner salt contributes to the
stability of the sulfate group. The product was purified by
In summary, we developed an efficient route to the natural
alkaloid (+)-neosymbioimine, starting from readily available
(-)-(S)-citronellol. The synthetic strategy is based on the
extremely facile tandem HWE-IMDA reaction for construc-
tion of the neosymbioimine core, which allows for the
preparation of significant amounts of 1 in 10% total yield
over 18 steps.
Org. Lett., Vol. 9, No. 8, 2007
1463

Acknowledgment. We thank Graeme Nicholson (Institute
of Organic Chemistry) for measuring the high-resolution
mass spectra and Sonja Tragl (Institute of Inorganic Chem-
istry) for the X-ray analysis. Financial support from the
Deutsche Forschungsgemeinschaft (Ma 1012/21-1) and
the Fonds der Chemischen Industrie is gratefully acknowl-
edged.
Supporting Information Available: Experimental pro-
cedures and characterization for all new compounds reported
and copies of NMR spectra for important intermediates. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL070049A
1464
Org. Lett., Vol. 9, No. 8, 2007