Total synthesis of (-)-himandrine

Article abstract of DOI:10.1021/ja903790y

(Chemical Equation Presented) We describe the first total synthesis of (-)-himandrine, a member of the class II galbulimima alkaloids. Noteworthy features of this chemistry include a diastereoselective Diels-Alder reaction in the rapid synthesis of the tr

Full text of DOI:10.1021/ja903790y

Published on Web 06/25/2009
Total Synthesis of (-)-Himandrine
Mohammad Movassaghi,* Meiliana Tjandra, and Jun Qi
Massachusetts Institute of Technology, Department of Chemistry, Cambridge, Massachusetts 02139
Received May 11, 2009; E-mail: movassag@mit.edu
Scheme 1. Retrosynthetic Analysis of (-)-Himandrine (1)
The galbulimima alkaloid (-)-himandrine (1) is a topologically
fascinating compound isolated from the bark of Galbulimima
belgraVeana, a tree indigenous to Papua New Guinea and northern
Australia.1 The promise that natural and synthetic derivatives of
galbulimima alkaloids have shown for treatment of human ailments2
has resulted in substantial attention in both academia and industry.
We previously reported the first enantioselective total synthesis of
(-)-galbulimima alkaloid 13 (class III)3 and revised the absolute
stereochemical assignment of the class II and III derivatives.4 While
there have been several outstanding syntheses of class I5 and III6
galbulimima alkaloids, no total synthesis of the class II7 galbu-
limima alkaloids possessing the unique N-C9 spirofused polycyclic
framework has been reported. Herein we describe our total synthesis
of class II alkaloid 1 guided by our previously disclosed hypothesis
for its biogenesis,3b featuring a final stage oxidative spirocyclization
to secure the BCF ring juncture.
Scheme 2. Enantioselective Synthesis of Tricyclic Enone (-)-15a
Inspired by Mander, Ritchie, and Taylor’s 1967 proposal relating
all galbulimima alkaloids to a single polyacetate precursor,1 and
d
consistent with our specific hypothesis for the biogenesis of class
II and III galbulimima alkaloids,3b we identified aminoketoester 3
(Scheme 1) as a plausible point of divergence en route to more
complex alkaloids. Our retrosynthetic analysis of (-)-1 follows this
hypothesis,3b in which the N-C9 bond is introduced by a late stage
oxidative spirocyclization of the pentacyclic aminoketoester 3. We
envisioned that oxidation of 3, potentially facilitated by dienol
formation, would afford allylic alcohol 2, which is poised for the
critical condensative spirocyclization. We expected that application
of our annulation methodology8 to enone 4 and iminium chloride
5 would convergently assemble a pentacycle primed for our
proposed biomimetic oxidative spirocyclization.
Our enantioselective synthesis of a tricyclic enone 4 mimic is
illustrated in Scheme 2. The C14-stereochemistry, introduced
through the use of MacMillan’s D-proline-catalyzed R-oxidation9,10
of hept-6-enal afforded the enantiomerically enriched diol (-)-6
(>98.5% ee).11 The C14-methyl ether was then secured by
etherification12 of alcohol (+)-7 to provide the methoxyalcohol (+)-
8, setting the stage for the substrate-directed synthesis of the trans-
decalin AB-ring system. Oxidation13 of alcohol (+)-8 followed by
conversion of the corresponding aldehyde to dibromoolefin provided
(-)-9. A highly efficient Suzuki cross-coupling reaction3,14 involv-
ing boronic acid 1010 and dibromide (-)-9 afforded the cis-vinyl
bromide 11 in 97% yield. A copper-promoted coupling15 of vinyl
bromide 11 with 2-azetidinone followed by conversion of the C20-
silyl ether to the corresponding C20-silyl enol ether gave the desired
tetraene (-)-12. The 2-azetidinone grouping at C16 was strategically
introduced to serve a dual role in facilitating the planned Diels-Alder
reaction by providing a 2-N-acylaminodiene with a greater prefer-
ence for the s-cis C16-C17 conformation and masking the C16-
carbonyl for subsequent transformations. A Ru-catalyzed olefin
cross-metathesis reaction16 with acrolein enabled selective func-
tionalization of the acid sensitive tetraene (-)-12 to the corre-
sponding unsaturated aldehyde (-)-13 in 85% yield. Heating a
a Conditions: (a) TBSCl, imidazole, DMAP, DMF, 0 °C, 94%. (b) 4
Å-MS, Proton-sponge, Me3O·BF4, CH2Cl2, 23 °C, 93%. (c) HCl, MeOH,
23 °C, 98%. (d) DMSO, iPr2NEt, SO3 ·pyr, CH2Cl2, 23 °C. (e) CBr4, PPh3,
CH2Cl2, 0 °C, 65% (two steps). (f) Pd(PPh3)4, Tl2CO3, THF, H2O, 23 °C,
97%. (g) 2-Azetidinone, CuI, K2CO3, (MeNHCH2)2, PhMe, 120 °C, 85%.
i
(h) TBAF, THF, 0f23 °C. (i) DMSO, Pr2NEt, SO3 ·pyr, CH2Cl2, 23 °C,
80% (two steps). (j) TBSOTf, Et3N, CH2Cl2, -78 °C, 82%. (k) Acrolein,
4,5-dihydroIMesCl2RudCH(2-iPrO)Ph (10 mol %), CH2Cl2, 23 °C, 85%.
(l) BHT, N,N-diethylaniline, MeCN, 95 °C, 75%, 5:1 dr. (m) TiCl4, CH2Cl2,
-78 °C. (n) Martin sulfurane, PhH, 23 °C, 57% (two steps).
solution of tetraenal (-)-13 in acetonitrile at 95 °C afforded the
desired trans-decalin aldehyde (-)-14 as the major endo Diels-Alder
product (75%, dr ) 5:1) as supported by NOE studies.10 Treatment
of aldehyde (-)-14 with titanium tetrachloride provided the
corresponding Mukaiyama aldol17 product, which upon exposure
to Martin sulfurane18 afforded the oxygen and acid sensitive enone
(-)-15 in 57% yield over two steps (Scheme 2).
Lithiation of the readily available iminium chloride (-)-53
followed by copper-promoted conjugate addition8 to the enantio-
merically enriched enone (-)-15 afforded the highly air sensitive
pentacyclic iminoalcohol 18 (Scheme 3). Importantly, blocking of
the si-face of the enone by the trans-decalin AB-ring system
imposed exquisite stereochemical control during the C7-C8 bond
9
9648 J. AM. CHEM. SOC. 2009, 131, 9648–9650
10.1021/ja903790y CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
Scheme 3. Enantioselective Total Synthesis of (-)-Himandrine (1)a
a
n
i
Conditions: (a) BuLi, THF, (-)-5, -78f 0 °C; CuBr ·SMe2; (-)-15, -78 f -10 °C. (b) NaBH4, EtOH, 0 °C. (c) ClCO2Bn, Pr2NEt, K2CO3, THF,
H2O, 50% (three steps). (d) TsOH ·H2O, PhH, 23 °C, 81%. (e) POCl3, DMF, CH2Cl2, 0 f 23 °C, 71%. (f) DDQ, SiO2, MeCN, H2O, 23 °C. (g) NaClO2,
NaH2PO4 ·H2O, 2-methyl-2-butene, tBuOH, H2O, 23 °C. (h) CH2N2, THF, 0 °C, 61% (three steps). (i) TMS-I, 2,6-di-tBu-4-Me-Pyr, CH2Cl2, 0 °C, 66%. (j)
Et3N·(HF)3, THF, 23 °C, 90%. (k) NCS, MeCN, 23 °C, 45 min, 89%. (l) NaBH4, EtOH, 0 °C, 90%. (m) BzCl, pyridine, 23 °C, 7 d, 87%.
formation. Rapid tautomerization of the transiently formed imi-
noketone 16 enabled nucleophilic addition of C5 to the C20-ketone
of 17, consistent with our biosynthetic hypothesis for the D-ring
formation.3 Addition of sodium borohydride to crude imino alcohol
18 resulted in a completely diastereoselective C6-imine reduction,
at which point benzyloxycarbonylation of the resulting aminoalcohol
19 provided the desired product (-)-20 (Scheme 3, Z ) benzy-
loxycarbonyl) in 50% yield over three steps. Importantly, the formal
cycloaddition between iminium chloride (-)-5 and enone (-)-15
and subsequent C6-reduction of imine 18 secured four stereogenic
centers and expediently assembled the pentacyclic substructure of
(-)-himandrine (1) as a single diastereomer.
Mild hydrolysis of N-vinyl-carbamate (-)-20 with p-toluene
sulfonic acid monohydrate in benzene afforded the key intermediate
ketone (-)-21 in 75% yield. The mild reaction conditions used in
hydrolysis of the C16 N-vinyl lactam circumvent the competing
acid catalyzed ꢀ-elimination of the C14-methyl ether in ketone (-)-
21.19 At this juncture, we required a mild method for introduction
of the C17-methoxycarbonyl group to access the aminoketoester
3. After exploring a variety of strategies, we found that treatment
of ketone (-)-21 with Vilsmeier’s reagent provided the vinyl ether
(+)-23 in 71% yield, likely occurring through nucleophilic addition
of enol 22 (Scheme 3) followed by trapping with the C20-alcohol.
Exposure of vinyl ether (+)-23 to 2,3-dichloro-5,6-dicyano-1,4-
benzoquinone (DDQ)20 furnished the corresponding unsaturated
ꢀ-ketoaldehyde 24. Immediate oxidation of the acid sensitive C18-
aldehyde 24 to the corresponding carboxylic acid 25 followed by
treatment with diazomethane provided the desired ꢀ-ketoester (-)-
26 in 61% yield over three steps (Scheme 3). Sequential treatment
of carbamate (-)-26 with trimethylsilyliodide (TMS-I)3,21 and
triethylamine trihydrofluoride provided pentacyclic aminoketoester
(-)-3, our proposed common biosynthetic intermediate to the class
II and class III galbulimima alkaloids.
oxidation as a prelude to intramolecular allylic displacement by
the amine to give the N-C9 spirocycle. Deuterium labeling studies
were employed in monitoring H/D exchange at C9 (Scheme 4).
Compellingly, simple dilution of aminoketoester (-)-3 in methanol-
d4 resulted in immediate and quantitative deuterium incorporation
at C9 in the form of aminoketoester 3-d3 (Scheme 4).22 Dissolution
of aminoketoester 3-d3 in methanol returned aminoketoester (-)-
3, indicating the exchange of C9-methine occurs with retention of
C9-stereochemistry.23 Cumulatively, our results are consistent with
the amino group being intimately involved in facilitating C9-depro-
tonation.24,25 The preservation of the C9-stereochemistry is con-
sistent with intramolecular protium/deuterium delivery by the
corresponding ammonium ion from the more sterically hindered
face of the C19-C9 tetrasubstituted alkene of 27.
Guided by our biosynthetic hypothesis and with evidence for
rapid dienol formation from aminoketoester (-)-3, we wished to
capitalize on the inherent chemistry of this plausible biosynthetic
intermediate. In the event, treatment of aminoketoester (-)-3 with
N-chlorosuccinimide (NCS) in acetonitrile at 23 °C over 45 min
afforded the desired spirofused hexacyclic enone (+)-28 in 89%
yield. The structure of (+)-28 was supported through detailed 2D
NMR analysis.26 Treatment of ketone (+)-28 with sodium boro-
hydride in ethanol effected a completely diastereoselective C16-
reduction to the desired diol (+)-29 in 90% yield. Benzoylation of
the C16-hydroxyl group of diol (+)-29 proceeded to give the first
synthetic sample of (-)-himandrine (1) in 87% yield ([R]22D ) -21
(c, 0.12 CHCl3); Lit.1a ([R] ) -38 (c 1.22, CHCl3))).27 All
spectroscopic data for synthetic (-)-1 matched those reported for
the natural compound.10 The structure of our synthetic (-)-1 was
unequivocally confirmed by X-ray crystallographic analysis.
Intrigued by the high efficiency in the conversion of aminoke-
toester (-)-3 to hexacyclic enone (+)-28, we sought to gain further
mechanistic insight into this transformation. Close monitoring of
the reaction mixture indicated that exposure of aminoketoester (-)-3
to NCS in acetonitrile resulted in concomitant formation of
hexacycle (+)-28 and the light-sensitive N-chloro pentacycle 30
(Scheme 4) that would disappear by the end of the reaction. Use
With the key intermediate in hand, we were able to evaluate the
feasibility of our postulated biomimetic late stage N-C9 bond
formation. We expected that ꢀ-ketoester 3 might undergo rapid
tautomerization to the electron rich dienol 27, enabling facile C17
9
J. AM. CHEM. SOC. VOL. 131, NO. 28, 2009 9649

C O M M U N I C A T I O N S
of benzene as solvent for this transformation reduced the overall
rate of hexacycle (+)-28 formation and allowed for the isolation
of N-chloro pentacycle 30.28 Importantly, dissolution of N-chloro
pentacycle 30 in acetonitrile for 12 h did not result in formation of
any hexacycle (+)-28 and completely returned 30.29,30 Additionally,
when the deuterium incorporation studies described above were
conducted with 30, there was no evidence for formation of the
corresponding dienol, suggesting the nitrogen of 30 is not basic
enough to enable deprotonation at C9. Interestingly, exposure of
aminoketoester (-)-3 to samples of N-chloro pentacycle 30 in
acetonitrile resulted in formation of hexacycle (+)-28 and (-)-
3.31 This result is consistent with an intermolecular N to C halogen
transfer from N-chloro pentacycle 30 to aminoketoester (-)-3 (likely
via 27). A plausible mechanism for conversion of aminoketoester
(-)-3 to spirofused hexacycle (+)-28 is halogenation of the dienol
27 to give R-chloroester 31,32 followed by intramolecular allylic
displacement by the amine. Significantly, this mechanism is
consistent with our proposed biomimetic hypothesis for the
advanced stage oxidative spirocyclization of aminoketoester (-)-
3.3b
References
(1) (a) Binns, S. V.; Dustan, P. J.; Guise, G. B.; Holder, G. M.; Hollis, A. F.;
McCredie, R. S.; Pinhey, J. T.; Prager, R. H.; Rasmussen, M.; Ritchie, E.;
Taylor, W. C. Aust. J. Chem. 1965, 18, 569. (b) Guise, G. B.; Mander,
L. N.; Prager, R. H.; Rasmussen, M.; Ritchie, E.; Taylor, W. C. Aust.
J. Chem. 1967, 20, 1029. (c) Mander, L. N.; Prager, R. H.; Rasmussen,
M.; Ritchie, E.; Taylor, W. C. Aust. J. Chem. 1967, 20, 1473. (d) Mander,
L. N.; Prager, R. H.; Rasmussen, M.; Ritchie, E.; Taylor, W. C. Aust.
J. Chem. 1967, 20, 1705.
(2) For SCH-530348, a galbulimima alkaloid derived antiplatelet agent, in
phase-III clinical trials for acute coronary syndrome, see: (a) Chackala-
mannil, S.; Wang, Y.; Greenlee, W. J.; Hu, Z.; Xia, Y.; Ahn, H.-S.; Boykow,
G.; Hsieh, Y.; Palamanda, J.; Agans-Fantuzzi, J.; Kurowski, S.; Graziano,
M.; Chintala, M. J. Med. Chem. 2008, 51, 3061. (b) Malaska, M. J.; Fauq,
A. H.; Kozikowski, A. P.; Aagaard, P. J.; McKinney, M. Bioorg. Med.
Chem. Lett. 1995, 5, 61.
(3) (a) Movassaghi, M.; Hunt, D. K.; Tjandra, M. J. Am. Chem. Soc. 2006,
128, 8126. (b) Please see Supporting Information of ref 3a (page S4, Scheme
S2) for a detailed unified biosynthetic hypothesis for class II and III
galbulimima alkaloids.
(4) (a) The direct stereochemical and structural relationship between (-)-
galbulimima alkaloid 13 (class III) and (-)-himandrine (1, class II) was
established through instructive chemical degradation studies in the context
of the isolation work (ref 1c). For revision of the stereochemical assignment
based on new X-ray analysis, see: (b) Willis, A. C.; O’Connor, P. D.; Taylor,
W. C.; Mander, L. N. Aust. J. Chem. 2006, 59, 629.
(5) For prior syntheses of class I galbulimima alkaloids, see: (a) Hart, D. J.;
Wu, W.-L.; Kozikowski, A. P. J. Am. Chem. Soc. 1995, 117, 9369. (b)
Chackalamannil, S.; Davies, R. J.; Aserom, T.; Doller, D.; Leone, D. J. Am.
Chem. Soc. 1996, 118, 9812. (c) Tchabanenko, K.; Adlington, R. M.;
Cowley, A. R.; Baldwin, J. E. Org. Lett. 2005, 7, 585.
(6) For prior syntheses of class III galbulimima alkaloids, see: (a) Mander,
L. N.; McLachlan, M. M. J. Am. Chem. Soc. 2003, 125, 2400. (b) Shah,
U.; Chackalamannil, S.; Ganguly, A. K.; Chelliah, M.; Kolotuchin, S.;
Bulevich, A.; McPhail, A. J. Am. Chem. Soc. 2006, 128, 12654. (c) Evans,
D. A.; Adams, D. J. J. Am. Chem. Soc. 2007, 129, 1048.
Scheme 4. Key Observations Relevant to N-C9 Bond Formation
(7) For an intriguing synthesis of the skeleton of 1, see: O’Connor, P. D.;
Mander, L. N.; McLachlan, M. M. W. Org. Lett. 2004, 6, 703.
(8) Movassaghi, M.; Chen, B. Angew. Chem., Int. Ed. 2007, 46, 565.
(9) Brown, S. P.; Brochu, M. P.; Sinz, C. J.; MacMillan, D. W. C. J. Am.
Chem. Soc. 2003, 125, 10808.
(10) See Supporting Information for details.
(11) Diol (-)-6 is prepared in three steps from commercially available 7-octene-
1,2-diol on greater than 8-g scale.
(12) Meerwein, H. Methoden Org. Chem. (Houben-Weyl) 1965, 6, 325.
(13) Parikh, J. P.; Doering, W. E. J. Am. Chem. Soc. 1967, 89, 5505.
(14) (a) Miyaura, N.; Yamada, K.; Suginome, H.; Suzuki, A. J. Am. Chem. Soc.
1985, 107, 972. (b) For rate enhancement with TlOH, see: Uenishi, J.-i.;
Beau, J.-M.; Armstrong, R. W.; Kishi, Y. J. Am. Chem. Soc. 1987, 109,
4756. (c) Roush, W. R.; Moriarty, K. J.; Brown, B. B. Tetrahedron Lett.
1990, 31, 6509. (d) Evans, D. A.; Starr, J. T. J. Am. Chem. Soc. 2003, 125,
13531.
(15) Jiang, L.; Job, G. E.; Klapars, A.; Buchwald, S. L. Org. Lett. 2003, 5,
3667.
(16) (a) Chatterjee, A. K.; Morgan, J. P.; Scholl, M.; Grubbs, R. H. J. Am. Chem.
Soc. 2000, 122, 3783. (b) Garber, S. B.; Kingsbury, J. S.; Gray, B. L.;
Hoveyda, A. H. J. Am. Chem. Soc. 2000, 122, 8168.
(17) Mukaiyama, T.; Marasaka, K.; Banno, K. Chem. Lett. 1973, 1011.
(18) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327.
(19) The more forcing reaction conditions needed for hydrolysis of an N-vinyl
oxazolidinone (ref 3) derivative of 20 led to competing ꢀ-elimination of
the C14-methyl ether.
(20) (a) Liu, H. J.; Tran, D. D. P. Tetrahedron Lett. 1999, 40, 3827. (b) Walker,
D.; Hiebert, J. D. Chem. ReV. 1967, 67, 153.
We have described the first total synthesis of (-)-himandrine
(1), a member of the class II galbulimima alkaloids. Noteworthy
features of this chemistry include the diastereoselective Diels-Alder
reaction for rapid synthesis of the trans-decalin containing tricycle
(-)-15 in an enantiomerically enriched form, the formal [3+3]
annulation strategy to secure the CDE-ring system with complete
diastereoselection, and successful implementation of our bioge-
netically inspired oxidative spirocyclization in converting 3 to (+)-
28. The successful and direct conversion of (-)-3 to (+)-28 drew
on the power of biogenetic considerations and fully utilized the
inherent chemistry of this plausible biosynthetic intermediate.
(21) Jung, M. E.; Lyster, M. A. J. Chem. Soc., Chem Comm. 1978, 315.
(22) The 1H NMR spectrum of 3-d3 was the same as that for the starting
aminoketoester (-)-3 with the exception of the C9-methine spin systems.
(23) Similar results were obtained using a C20 O-trimethylsilylated derivative
of (-)-3.
(24) Attempts at intermolecular C9 deprotonation were unsuccessful; the
proximity of the amine to C9-methine seems to facilitate tautomerization.
(25) Use of derivatives of aminoketoester 3 not possessing a basic amine (i.e.,
N-Cbz, N-Cl) resulted in no C9-deuterium incorporation over 24 h.
(26) Key HMBC correlations between C9/C2-H and C9/C6-H confirmed the
N-C9 bond connectivity.
Acknowledgment. M.M. is an Alfred P. Sloan Research Fellow,
a Beckman Young Investigator, and a Camille Dreyfus Teacher-
Scholar. M.T. acknowledges BMS and Novartis graduate fellow-
ships. We thank Justin Kim and Dr. Peter Mu¨ller for X-ray
crystallographic analysis of (-)-1. We thank Professor Robert G.
Griffin and Dr. Tony Bielecki for use of a high field instrument at
the MIT-Harvard Center for Magnetic Resonance (EB002026). We
acknowledge financial support by NIH-NIGMS (GM074825).
(27) The C14-methyl ether and the C17-methoxycarbonyl substituents greatly
shield the C16 alcohol leading to slow benzoylation.
(28) The reaction had to be stopped within 10 min; otherwise significantly more
hexacycle (+)-28 would be generated.
(29) Addition of succinimide does not lead to conversion of 30 to (+)-28.
(30) While the sensitivity of 30 precluded its derivatization, use of its more
stable C20-O-trimethylsilyl derivative under basic, acidic, or photochemical
conditions predominantly led to elimination and decomposition.
(31) Complete mass balance was observed, and the amount of hexacycle (+)-
28 formed was exactly proportional to the amount of 30 used.
(32) While C9 halogenation cannot be ruled out, C17 halogenation is consistent
with the lack of product formation using the C20-O-trimethylsilyl derivative
of 30 with significantly blocked access to C17.
Supporting Information Available: Experimental procedures,
spectroscopic data, copies of 1H and 13C NMR spectra, and X-ray
structure of (-)-1. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA903790Y
9
9650 J. AM. CHEM. SOC. VOL. 131, NO. 28, 2009