Total synthesis of (-)-aurantioclavine

Article abstract of DOI:10.1021/ol100470g

Figure presented The concise total synthesis of (-)-aurantioclavine has been achieved by taking advantage of strategies for the asymmetric alkenylation of N-tert-butanesulfinyl imines. The enantiomerically pure natural product was prepared in 6 steps and 27% overall yield by using Rh-catalyzed addition of a N-methyliminodiacetic acid (MIDA) boronate and in 5 steps and 29% yield by employing a Grignard reagent addition sequence.

Full text of DOI:10.1021/ol100470g

ORGANIC
LETTERS
2010
Vol. 12, No. 9
2004-2007
Total Synthesis of (-)-Aurantioclavine
Katrien Brak and Jonathan A. Ellman*
UniVersity of California-Berkeley, Department of Chemistry, Berkeley, California 94720
jellman@berkeley.edu
Received February 24, 2010
ABSTRACT
The concise total synthesis of (-)-aurantioclavine has been achieved by taking advantage of strategies for the asymmetric alkenylation of
N-tert-butanesulfinyl imines. The enantiomerically pure natural product was prepared in 6 steps and 27% overall yield by using Rh-catalyzed
addition of a N-methyliminodiacetic acid (MIDA) boronate and in 5 steps and 29% yield by employing a Grignard reagent addition sequence.
(-)-Aurantioclavine ((-)-1) was isolated from Penicillium
aurantioVirens in 1981,1 and soon after its discovery, concise
syntheses of racemic aurantioclavine were reported in the
literature.2 This ergot alkaloid continues to attract consider-
able interest from the synthetic community due to its
proposed role as an intermediate in the biosynthesis of the
complex polycyclic alkaloids of the communesin family
(Figure 1).3,4 However, despite this attention, the asymmetric
synthesis of aurantioclavine has only recently been ac-
complished in 13 steps and <1% overall yield from com-
mercially available material.5 Herein, we report a new, highly
efficient synthesis of (-)-aurantioclavine that utilizes asym-
metric alkenylation of a densely functionalized N-tert-
butanesulfinyl imine.
Figure 1. (-)-Aurantioclavine as an intermediate en route to the
(1) Kozlovskii, A. G.; Soloveva, T. F.; Sakharobskii, V. G.; Adanin,
V. M. Dokl. Akad. Nauk SSSR 1981, 260, 230.
communesins.
(2) (a) Yamada, F.; Makita, Y.; Suzuki, T.; Somei, M. Chem. Pharm.
Bull. 1985, 33, 2162. (b) Hegedus, L. S.; Toro, J. L.; Miles, W. H.;
Harrington, P. J. J. Org. Chem. 1987, 52, 3319. (c) Yamada, K.;
Namerikawa, Y.; Haruyama, T.; Miwa, Y.; Yanada, R.; Ishikura, M. Eur.
J. Org. Chem. 2009, 5752.
Our approach to (-)-aurantioclavine is depicted in Scheme
1. We envisioned that the natural product could be accessed
from sulfinamide 2 via cyclization to form the azepine ring
followed by cleavage of the sulfinyl group. The asymmetric
synthesis of sulfinamide 2 would be accomplished by the
addition of the appropriate organometallic reagent to N-
sulfinyl imine 3.6 This imine could in turn be generated by
the condensation of N-tert-butanesulfinamide with the alde-
hyde resulting from formylation of 4-bromotryptophol.7
Installation of the formyl group was initially attempted
via a traditional approach involving KH-mediated deproto-
(3) (a) May, J. A.; Stoltz, B. M. Tetrahedron 2006, 62, 5262. (b) May,
J. A.; Zeidan, R. K.; Stoltz, B. M. Tetrahedron Lett. 2003, 44, 1203
.
(4) The only syntheses to date of a communesin ((()-communesin F):
(a) Yang, J.; Wu, H.; Shen, L.; Qin, Y. J. Am. Chem. Soc. 2007, 129, 13794.
(b) Liu, P.; Seo, J. H.; Weinreb, S. M. Angew. Chem., Int. Ed. 2010, 49,
2000. For a review on synthetic efforts toward the communesins, see: (c)
Siengalewicz, P.; Gaich, T.; Mulzer, J. Angew. Chem., Int. Ed. 2008, 47,
8170
.
(5) Krishnan, S.; Bagdanoff, J. T.; Ebner, D. C.; Ramtohul, Y. K.;
Tambar, U. K.; Stoltz, B. M. J. Am. Chem. Soc. 2008, 130, 13745. In
accomplishing the total synthesis of (-)-aurantioclavine the authors assigned
the (R) absolute configuration.
10.1021/ol100470g  2010 American Chemical Society
Published on Web 03/31/2010

Scheme 1
.
Retrosynthetic Analysis of (-)-Aurantioclavine
Scheme 3. Initial Route Toward the Synthesis of (-)-1
nation of the acidic sites, followed by a lithiation-formylation
sequence with tBuLi and DMF.8 While the desired aldehyde
could be obtained by using this method, the reaction yield
was highly variable due to the poor solubility of the initially
formed dianion. We next turned to the highly efficient Pd-
catalyzed formylation of aryl and heteroaryl bromides
reported by Beller and co-workers.9 While application of
their general procedure resulted in complete conversion to
undesired lactone 4 (Scheme 2), in situ protection of the
alcohol with TMSCl prevented this cyclization and afforded
the desired aldehyde 5 (Scheme 3). Isolation of aldehyde 5
was complicated by polymerization via intermolecular hemi-
acetal formation, and therefore the unpurified material was
directly converted to N-tert-butanesulfinyl imine 3 in 53%
yield over the two steps.
boronates11 to N-tert-butanesulfinyl imines. Consistent with
the higher diastereoselectivity previously observed in the
addition of alkenylboron reagents, the additions of both
MIDA boronate 9 (entry 3) and trifluoroborate 10 (entry 4)
proceeded in moderate yields and with high diastereoselec-
tivities. Taking advantage of the hydrolytic stability of the
N-sulfinyl imine, higher yields could be attained by adding
the trifluoroborate in three portions (entry 5). Significantly,
the synthesis of sulfinamide 2 was achieved in high yield
and selectivity without requiring protection of the alcohol
or indole moieties.
Scheme 2
The next step en route to the synthesis of (-)-aurantio-
clavine was cyclization to provide the azepine ring. We
decided to use the Mitsunobu reaction for this cyclization
because this approach has been used successfully for the
formation of 5-membered rings with N-tert-butanesulfina-
mides and primary alcohols.12 To our surprise, the Mitsunobu
reaction of alcohol 2 resulted in formation of spiro[cyclo-
propyl]indolenine 613 (Scheme 3). A variety of acidic
conditions were explored for ring expansion of indolenine 6
to azepine 7. Unfortunately, decomposition was observed
with both Brønsted acids (HCl, TFA) and Lewis acids
(BF3OEt2, Yb(OTf)2, AuCl3), and upon treatment with
Schreiner’s thiourea catalyst,14 the indole-thiourea adduct
was instead isolated. Furthermore, indolenine 6 was unre-
active under basic conditions. When the sulfinamide moiety
was deprotonated with KH, no reaction was observed.
With the substrate for the key reaction in hand, we
explored the alkenylation of N-sulfinyl imine 3 (Table 1).
Addition of Grignard reagent 8 as a solution in ether resulted
in precipitation of the dianion and no desired product (entry
1). When a solution of the Grignard reagent in THF was
employed, the reaction proceeded in good yield but with poor
diastereoselectivity (entry 2). We recently developed a
method for the rhodium-catalyzed addition of alkenyltrif-
luoroborates10 and N-methyliminodiacetic acid (MIDA)
(6) For reviews on the asymmetric synthesis of amines via N-tert-
butanesulfinyl imines, see: (a) Ellman, J. A.; Owens, T. D.; Tang, T. P.
Acc. Chem. Res. 2002, 35, 984. (b) Morton, D.; Stockman, R. A.
Tetrahedron 2006, 62, 8869. (c) Ferreira, F.; Botuha, C.; Chemla, F.; Pe´rez-
Luna, A. Chem. Soc. ReV. 2009, 38, 1162.
(11) Brak, K.; Ellman, J. A. J. Org. Chem. 2010, 75. Submitted for
publication. See the Supporting Information for full reference.
(12) (a) Tang, T. P.; Ellman, J. A. J. Org. Chem. 2002, 67, 7819. (b)
Jakobsche, C. E.; Peris, G.; Miller, S. J. Angew. Chem., Int. Ed. 2008, 47,
6707.
(7) 4-Bromotryptophol is commercially available or can be synthesized
from 4-bromoindole according to the following: Yang, J.; Wu, H.; Shen,
L.; Qin, Y. J. Am. Chem. Soc. 2007, 129, 13794.
(13) For previous characterization of spiro[cyclopropane-1,3′-indolenine]
compounds, see: (a) Johansen, J. E.; Christie, B. D.; Rapoport, H. J. Org.
Chem. 1981, 46, 4914. (b) Clossen, W. D.; Roman, S. A.; Kwiatkowski,
G. T.; Corwin, D. A. Tetrahedron Lett. 1966, 21, 2271.
(8) For the successful formylation of indoles containing one acidic site
via this methodology, see: Moyer, M. P.; Shiurba, J. F.; Rapoport, H. J.
Org. Chem. 1986, 51, 5106.
(9) Klaus, S.; Neumann, H.; Zapf, A.; Strubing, D.; Hubner, S.; Almena,
J.; Riermeier, T.; Grob, P.; Sarich, M.; Krahnert, W.-R.; Rossen, K.; Beller,
M. Angew. Chem., Int. Ed. 2006, 45, 154.
(14) N,N′-Bis[3,5-bis(trifluoromethyl)phenyl]thiourea, see: (a) Schreiner,
P. R.; Wittkopp, A. Org. Lett. 2002, 4, 217. (b) Schreiner, P. R.; Wittkopp,
A. Chem.sEur. J. 2003, 9, 407.
(10) Brak, K.; Ellman, J. A. J. Am. Chem. Soc. 2009, 131, 3850.
(15) Cogan, D. A.; Liu, G.; Ellman, J. A. Tetrahedron 1999, 55, 8883.
Org. Lett., Vol. 12, No. 9, 2010
2005

Table 1. Alkenylation of N-Sulfinyl Imine 3 with Various
Organometallic Reagents
Scheme 4. Total Synthesis of (-)-Aurantioclavine via the
Rh-Catalyzed Addition of MIDA Boronate 9
yield
(%)a
conversion
(%)a
entry
M (equiv)
drb
1c
2c
3d
4e
5e
MgBr in Et2O (3.5 equiv)
MgBr in THF (3.5 equiv) 69
BMIDA (2 equiv)
BF3K (2 equiv)
BF3K (3 × 1.5 equiv)
0
38
89
50
64
94
33:67
97:3
97:3
97:3
23
42
84 (81)f
a Yield and conversion were determined by 1H NMR relative to an
external standard. b Diastereoselectivity was determined by HPLC com-
parison to authentic diastereomers. c Reaction was run in CH2Cl2.15
d Reaction was performed with [Rh(OH)(cod)]2, dppbenz, and K3PO4 in
H2O/dioxane.11 e Reaction was performed with [Rh(OH)(cod)]2, dppbenz,
and NEt3 in H2O/DMF.10 f Isolated yield of diastereomerically pure material.
The inability to carry forward spiro-indolenine 6
required us to revise our strategy. We hypothesized that
cyclization without formation of the undesired indolenine
would require deactivation of the indole. Selective protec-
tion of the indole moiety of 2 would be inefficient because
it would require protecting group manipulations of the
more nucleophilic alcohol. A much more appealing
approach was tosylation of both the indole and alcohol,
which would deactivate the indole while simultaneously
activating the alcohol for a subsequent SN2-mediated
cyclization. Bis-tosylation of N-sulfinyl allylic amine 2
was met with limited success. Careful evaluation of the
reaction products suggested that the sulfinamide moiety
was unstable under these conditions. Fortunately, tosyla-
tion of the N-sulfinyl imine precursor (3) proceeded in
good yield as long as the reaction solution was maintained
at low temperature (Scheme 4).
of the sulfinyl group (Scheme 5). These straightforward
deprotections could be carried out separately, in either order,
or in one-pot. The ability to selectively deprotect either
functional group provides great flexibility for subsequent
synthetic approaches to the communesins, and the one-pot
double-deprotection resulted in the formation of (-)-auran-
tioclavine in quantitative yield (Scheme 4).
Scheme 5. Selective Deprotection of Tosyl and Sulfinyl Groups
While the Rh(I)-catalyzed addition of trifluoroborate 10
to bis-tosylated N-sulfinyl imine 11 was not very efficient,
the newly developed MIDA boronate slow release condi-
tions11 provided sulfinamide 12 in high yield and diastereo-
selectivity (Scheme 4). After chromatography, diastereo-
merically pure 12 was isolated in 78% yield. Furthermore,
cyclization of sulfinamide 12 proceeded smoothly upon
deprotonation with NaH.
To complete the synthesis, only the removal of the
sulfinamide and tosyl groups remained. Initial attempts to
remove both of the protecting groups with a single reagent
were unsuccessful. Treatment with MeLi or TBAF resulted
in deprotection of the tosyl group as well as multiple other
side products. Magnesium in methanol proved to be a mild
and highly effective method for deprotection of the tosyl
group (Scheme 5). Additionally, acidic alcoholysis with HCl
in methanol, the general conditions for removing tert-
butanesulfinyl groups,15 resulted in quantitative deprotection
Despite the precedence for lower selectivity, we realized
that Grignard reagent addition could prove more efficient
than the addition of boron reagents if spontaneous
cyclization occurred upon formation of the nucleophilic
sulfinamide anion (Scheme 6).16 The solvent of the
Grignard solution was again found to be critical to reaction
success, with no cyclization observed in ether. However,
addition of Grignard 8 as a solution in the more highly
2006
Org. Lett., Vol. 12, No. 9, 2010

coordinating solvent THF cleanly afforded cyclized prod-
uct 15 with moderate diastereoselectivity. After chroma-
tography, diastereomerically pure product was isolated in
72% yield. It is noteworthy that because Grignard addition
provides the opposite diastereoselectivity to that for the
Rh-catalyzed addition of the MIDA boronate, both N-tert-
butanesulfinyl azepine diastereomers 7 and 15 are acces-
sible. The one-pot double-deprotection conditions were
equally successful for azepine diastereomer 15, providing
(-)-aurantioclavine in high yield.
Scheme 6
.
Total Synthesis of (-)-Aurantioclavine via the
Addition of Grignard Reagent 8
In summary, the asymmetric total synthesis of (-)-
aurantioclavine has been accomplished in 6 steps and 27%
overall yield by using a MIDA boronate and in 5 steps
and 29% overall yield by using a Grignard reagent. The
syntheses are considerably shorter and higher yielding than
the previously reported synthesis5 and provide rapid access
to significant quantities of (-)-aurantioclavine. This work
also highlights the synthetic utility of the various meth-
odologies for the alkenylation of N-tert-butanesulfinyl
imines.
Acknowledgment. This work was supported by a grant
from the National Science Foundation (CHE-0742565).
Supporting Information Available: Experimental pro-
cedures, characterization data, copies of H and 13C NMR
spectra, and copies of HPLC traces. This material is available
free of charge via the Internet at http://pubs.acs.org.
1
(16) For the one-pot preparation of aziridines and pyrrolidines by
Grignard reagent additions to halo-substituted N-tert-butanesulfinyl imines,
see: (a) Denolf, B.; Mangelinckx, S.; To¨rnroos, K. W.; De Kimpe, N. Org.
Lett. 2006, 8, 3129. (b) Reddy, L. R.; Prashad, M. Chem. Commun. 2010,
46, 222.
OL100470G
Org. Lett., Vol. 12, No. 9, 2010
2007