Enantioselective construction of vicinal stereogenic quaternary centers by dialkylation: Practical total syntheses of (+)- and meso-chimonanthine

Article abstract of DOI:10.1002/(SICI)1521-3773(20000103)39:1<213::AID-ANIE213>3.0.CO;2-Z

All three stereoisomers of the hexacyclic 3a,3a'-bispyrrolidino[2,3- b]indoline moiety found in complex indole alkaloids can be prepared, as illustrated by total syntheses of meso-chimonanthine (1) and (+)- chimonanthine (2). A rare example of high diaste

Full text of DOI:10.1002/(SICI)1521-3773(20000103)39:1<213::AID-ANIE213>3.0.CO;2-Z

COMMUNICATIONS
1871 ± 1873; b) N. Yoshikawa, Y. M. A. Yamada, J. Das, H. Sasai, M.
Shibasaki, J. Am. Chem. Soc. 1999, 121, 4168 ± 4178.
[7] The absolute configuration was determined by Mosherꢁs method; see
J. A. Dale, H. S. Mosher, J. Am. Chem. Soc. 1973, 95, 512 ± 519. The
enantiomeric excess was determined by HPLC analysis using DAI-
CEL CHIRALPAK AD (2-propanol/hexane 1/100).
[8] a) T. Mandai, T. Matsumoto, M. Kawada, J. Tsuji, Tetrahedron Lett.
1993, 34, 2161 ± 2164; b) J. Tsuji, T. Sugiura, I. Minami, Synthesis 1987,
603 ± 606.
Enantioselective Construction of Vicinal
Stereogenic Quaternary Centers by
Dialkylation: Practical Total Syntheses
of (‡)- and meso-Chimonanthine**
Larry E. Overman,* Jay F. Larrow, Brian A. Stearns,
and Jennifer M. Vance
[9] a) K. Harada, H. Urabe, F. Sato, Tetrahedron Lett. 1995, 36, 3203 ±
3206; b) F. Sato, H. Ishikawa, M. Sato, Tetrahedron Lett. 1981, 22, 85 ±
88. This method could lead to the fragment for the synthesis of
epothilone B (2).
[10] For related work with oximes, see E. J. Corey, L. S. Melvin, Jr., M. F.
Haslanger, Tetrahedron Lett. 1975, 3117 ± 3120.
[11] We can also prepare syn-aldol from cis-enone by this method.[13]
Through coupling with the stereocontrolled catalytic asymmetric
epoxidation of the cis- and trans-enones recently developed in our
group,[5] it is possible to prepare syn- and anti-aldol from cis- and trans-
enones, respectively, with excellent ee values (Table 1).
Enantioselective formation of vicinal stereogenic quater-
nary carbon centers is one of the more formidable challenges
in contemporary organic synthesis.[1, 2] Here we report a
powerful method for creating such arrays by reaction of
achiral dienolates with a chiral, enantiopure dielectrophile.
The method was developed to address the vicinal stereogenic
quaternary benzylic carbon centers that are the signature
structural feature of indole alkaloids containing the hexacyclic
3a,3a'-bispyrrolidino[2,3-b]indoline unit.
The chimonanthines are the simplest members of this
family and are found in nature in all three stereochemical
motifs: meso-chimonanthine (1)[3] and ( )-chimonanthine[4]
from plant sources; (‡)-chimonanthine (2) from the skin of a
Colombian frog[5] and from plant sources.[6] In addition, many
Table 1. Epoxide opening of cis-epoxy-oximes by cuprate reagents.
R3
O
R1
R2
OH
R1
R3Li, CuCN
Et2O
R2
N
N
MeO
MeO
Entry
R1
R2
R3
Yield [%]
1
2
3
4
5
6
7
methyl
propyl
pentyl
pentyl
pentyl
pentyl
pentyl
pentyl
phenethyl
methyl
methyl
methyl
butyl
phenyl
vinyl
77
72
85
80
80
50
81
phenethyl
phenethyl
phenethyl
phenethyl
pentyl
methyl
[12] The relative configuration of the alcohol resulting from reduction of
21 with Me4NBH(OAc)3 was determined by analyzing the NOE of 22,
and the relative configuration of the methyl group of 22 resulting from
allylation of 20 was further deter-
mined by transforming 22 into 31 and
analyzing the coupling constant be-
Ha
OBn
tween Ha and Hb (10.2 Hz).
Hb
[13] The absolute configurations were de-
termined by Mosherꢁs method,[8] and
the enantiomeric excesses were deter-
mined by HPLC analysis using DAI-
CEL CHIRALPAK AD (2-propanol/
O
OH
O
31
hexane 1/100). The relative configurations were determined by
transforming 24 into 32 and 25 into 33 (cyclization of 27 and 34,
respectively), and analyzing their NOEs.
[*] Prof. L. E. Overman, Dr. J. F. Larrow, B. A. Stearns, J. M. Vance
Department of Chemistry
OTBS
OTBS
University of California, Irvine
TBSO
OH
O
O
516 Rowland Hall, Irvine, CA 92697-2025 (USA)
Fax: (‡1)949-824-3866
O
O
E-mail: leoverma@uci.edu
33
32
[**] This research was supported by the NIH (grant GM-12389), Bristol-
Myers Squibb Pharmaceuticals (graduate fellowship to B.A.S.), the
American Cancer Society (postdoctoral fellowship to J.F.L., PF-98-
003), and Pfizer (undergraduate research fellowship to J.M.V). NMR
and mass spectra were determined at University of California, Irvine
with instruments purchased with the assistance of the NSF and NIH
shared instrumentation programs. We are grateful to Dr. Joseph Ziller
and Dr. John Greaves for their assistance with X-ray and mass
spectrometric analyses.
OPh
OH
OH OTBS O
34
[14] R. Göttlich, K. Yamakoshi, H. Sasai, M. Shibasaki, Synlett 1997, 971 ±
973.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
Angew. Chem. Int. Ed. 2000, 39, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3901-0213 $ 17.50+.50/0
213

COMMUNICATIONS
Pb(OAc)4, and the resulting labile dialdehyde was reduced
with NaBH4 to provide diol 10. This intermediate could be
converted into meso-chimonanthine (1) in six steps and 65%
overall yield.[12, 15]
The critical dialkylation step of the approach to 1 summar-
ized in Scheme 1 is more amenable to scale-up than the
related step of our earlier route.[12] Use of the tartrate-derived
dielectrophile 7 avoids problems with competing SN2' cycliza-
tion that are seen in the reaction of the disodium dienolate of
6 with cis-1,4-dichloro-2-butene.[12] Stereoselection in the
second alkylation step is consistent with preferential reaction
of chelated intermediate 8.[12, 16]
more complex indole alkaloids contain the 3a,3a'-bispyrroli-
dino[2,3-b]indoline moiety. For example, ( )-idiospermuline
(3) consists of a ( )-chimonanthine fragment functionalized
with one additional pyrrolidinoindoline unit,[7] while the
dodecacyclic alkaloid quadrigemine C (4) is proposed to have
a meso-chimonanthine core functionalized with two pyrroli-
dinoindoline units of the same absolute stereochemistry.[8]
Early synthetic efforts directed towards this class of indole
alkaloids produced largely the racemic C2-symmetric isomers
by oxidative dimerization of indole[9] or oxindole deriva-
tives,[10] or by dialkylation of dihydroisoindigo.[11] We have
reported a stereoselective approach to meso-bispyrrolidinoin-
doline alkaloids by samarium-mediated reductive dialkylation
of an isoindigo,[12] and more recently a flexible approach to
either the meso or C2 isomers using an intramolecular Heck
reaction cascade was presented.[13] Herein we disclose a highly
efficient synthesis of 3a,3a'-bispyrrolidino[2,3-b]indolines
having either meso or C2 symmetry. The approach utilizes
the nature of a metal dienolate (chelated or nonchelated) to
control the relative configuration, while exploiting the
chirality of a tartrate-derived dielectrophile to control abso-
lute stereochemistry.
The outcome of the dialkylation reaction was dramatically
altered by changing the dienolate counter ion and reaction
solvent (Scheme 2). When the reaction of 6 with ditriflate 7
was carried out with LiHMDS (2.1 equiv) in THF containing
Isoindigo 5,[12] which is available in two steps and 85% yield
from commercially available oxindole and isatin, was con-
verted into the corresponding dihydroisoindigo 6 by reduction
with zinc in acetic acid at room temperature (Scheme 1).
Alkylation of the disodium dienolate of 6 with 1.1 equiv of
ditriflate 7[14] in THF at 788C provided in 92% yield a single
pentacyclic product 9, having the cis relationship of the two
spirooxindole groups. The acetonide was removed by expo-
sure of 9 to camphorsulfonic acid (CSA) and MeOH to
liberate the cyclohexanediol, which was cleaved with
Scheme 2. a) LiHMDS (2.1 equiv), THF/DMPU, 788C, 55%; b) CSA,
MeOH/CH2Cl2, 100%; c) Pb(OAc)4, PhH; NaBH4, MeOH, 90%; d) tBu-
Li, Et2O, 788C then O2; Boc2O, DMAP, CH2Cl2, 61%. Boc ˆ tert-
butoxycarbonyl, DMAP ˆ 4-dimethylaminopyridine, DMPU ˆ 1,3-dime-
thylhexahydro-2-pyrimidinone.
10% DMPU at 788C, the major product formed was the C2-
symmetric bisoxindole 12, which could be isolated in 55%
yield. Under these conditions, the other possible C2 product
was generated in trace amounts only (1 ± 3%), while isomer 9
having the spirooxindole groups cis was formed in 20% yield.
Both of these minor isomers were easily removed from 12 by
chromatography.
The relative stereochemistry of 12 was established by
removing the benzyl groups with tert-butyllithium/oxygen[17]
followed by N-acylation with di-tert-butyl dicarbonate to
provide BOC derivative 13. Single-crystal X-ray analysis of
this derivative established the relative stereochemistry of
12.[18] Using the same sequence employed to elaborate 9, C2-
symmetric bisoxindole 12 was converted in two steps and 90%
overall yield into 14 ([a]2D3 ˆ ‡235, c ˆ 0.9 in CHCl3). This
intermediate was processed in six additional steps to complete
the first enantioselective total synthesis of (‡)-chimonanthine
(2; 68% overall yield; [a]2D3 ˆ ‡274, c ˆ 0.5 in EtOH).[13, 19]
In summary, the dialkylation approach described herein
constitutes a practical way to prepare either the meso or
enantiopure C2 stereoisomers of 3a,3a'-bispyrrolidino[2,3-b]-
indolines on large scales. Starting from commercially avail-
Scheme 1. a) Zn, AcOH, room temperature, 83%; b) NaHMDS
(2.1 equiv), THF,
788C, 92%; c) CSA, MeOH/CH2Cl2, 100%;
d) Pb(OAc)4, PhH; NaBH4, MeOH, 92%. Bn ˆ benzyl, HMDS ˆ
1,1,1,3,3,3-hexamethyldisilazane, OTf ˆ trifluoromethansulfonate (tri-
flate),
214
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3901-0214 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 1

COMMUNICATIONS
able oxindole and isatin, the total syntheses of meso- (1) and
(‡)-chimonanthine (2) were accomplished in 39 and 21%
overall yield respectively. The alkylation of the dilithium
dienolate of 6 in 10% DMPU/THF is notable in producing
one of the two possible C2 products with high selectivity
(20:1). This alkylation reaction is a rare example of high
diastereoselectivity arising from the combination a proster-
eogenic enolate with a chiral electrophile containing an sp3
carbon center.[20]
How Strong Is the Coordination Bond between
a Histidine Tag and Ni ± Nitrilotriacetate?
An Experiment of Mechanochemistry on
Single Molecules**
Matteo Conti, Giuseppe Falini, and Bruno Samorì*
A number of extensively used methods of increasing
importance in molecular biology are based on the ability of
the nickel(ii) ion, when chelated by nitrilotriacetate (nta), to
selectively bind proteins containing stretches of consecutive
histidine residues (His). Hochuli et al. established this field of
applications when they discovered that proteins containing
isolated histidines lead to complexes that are less stable than
those arising from proteins having two consecutive His tags at
one terminus (2 Â His-tag).[1] Indeed, a stretch of six consec-
utive His tags (6 Â His-tag) is now commonly appended to the
primary sequence of recombinant proteins, making it possible
to isolate them selectively from a flow of crude cell lysate
Received: September 2, 1999 [Z13956]
[1] Recent reviews of asymmetric synthesis of quaternary centers: a) E. J.
Corey, A. Guzman-Perez, Angew. Chem. 1998, 110, 402 ± 415; Angew.
Chem. Int. Ed. 1998, 37, 388 ± 401; b) K. Fuji, Chem. Rev. 1993, 93,
2037 ± 2066.
[2] For an excellent recent strategy, see R. M. Lemieux, A. I. Meyers, J.
Am. Chem. Soc. 1998, 120, 5453 ± 5457.
Â
[3] Y. Adjibade, B. Weniger, J. C. Quirion, B. Kuballa, P. Cabalion, R.
Anton, Phytochemistry 1992, 31, 317 ± 319.
[4] H. F. Hodson, B. Robinson, G. F. Smith, Proc. Chem. Soc. London
1961, 465 ± 466.
by means of
a Ni ± nta-functionalized chromatographic
matrix[2a,b] and to immobilize them on biosensors for biomo-
lecular interaction analysis.[3] The stability of the anchoring of
the His-tagged proteins is challenged in both cases by the
frictional force exerted on them by the flow. Such interplay
between external forces and chemical processes can now be
transferred to the single-molecule level thanks to recent
developments in nanoscale manipulation techniques.[4, 5] Sin-
gle-molecule experiments can give crucial insights into bind-
ing processes that cannot be accessed by solution methods. In
the single-molecule world, bond making and breaking be-
comes a digital ªyes-or-noº process. Determination of binding
constants by traditional means provides only information that
is averaged on the entire ensemble of the molecules and on
the overall time scale of the experiment.
[5] T. Tokuyama, J. W. Daly, Tetrahedron 1983, 39, 41 ± 47.
Á
[6] L. Verotta, T. Pilati, M. Tato, E. Elisabetsky, T. A. Amador, D. S.
Nunes, J. Nat. Prod. 1998, 61, 392 ± 396.
[7] R. K. Duke, R. D. Allan, G. A. R. Johnston, K. N. Mewett, A. D.
Mitrovic, C. C. Duke, T. W. Hambley, J. Nat. Prod. 1995, 58, 1200 ±
1208.
Â
Â
[8] F. Gueritte-Voegelein, T. Sevenet, J. Pusset, M.-T. Adeline, B. Gillet,
Â
J.-C. Beloeil, D. Guenard, P. Potier, R. Rasolonjanahary, C. Kordon, J.
Nat. Prod. 1992, 55, 923 ± 930.
[9] E. S. Hall, F. McCapra, A. I. Scott, Tetrahedron 1967, 23, 4131 ± 4141.
[10] a) R. Göschke, J. B. Hendrickson, R. Rees, Tetrahedron 1964, 20,
565 ± 579; b) C. L. Fang, S. Horne, N. Taylor, R. Rodrigo, J. Am. Chem.
Soc. 1994, 116, 9480 ± 9486.
[11] T. Hino, S. Yamada, Tetrahedron Lett. 1963, 1757 ± 1760.
[12] J. T. Link, L. E. Overman, J. Am. Chem. Soc. 1996, 118, 8166 ± 8167.
[13] L. E. Overman, D. V. Paone, B. A. Stearns, J. Am. Chem. Soc. 1999,
121, 7702 ± 7703.
We simulated the process involved in the Ni ± nta protein
separation techniques by reproducing the approach, binding,
stretching, and unbinding under external forces between
individual His tags and individual Ni ± nta chelating groups in
a single-molecule experiment with the scanning force micro-
scope (SFM, see Figure 1).[6] This experiment revealed that:
1) the probability that the encounters between a 6 Â His-tag
and a Ni ± nta group develop in a stable bond is much higher
than that for a 2 Â His-tag; 2) both tags can make either of two
types of complex with markedly different stabilities and with
different ªenergy landscapesº along their force-driven disso-
[14] Ditriflate 7 is available on a large scale by reaction of (‡)-2,3-O-
isopropylidene-l-threitol with triflic anhydride and iPr2NEt in Et2O at
08C. Both enantiomers of this intermediate are available from (‡)-
and ( )-tartaric acid: E. A. Mash, K. A. Nelson, S. Van Deusen, S. B.
Hemperly in Organic Syntheses, Vol. 68 (Ed.: J. D. White), Wiley, New
York, 1990, pp. 92 ± 103.
[15] Diol 10 had previously been converted into meso-chimonanthine.[12]
The efficiency of this conversion is increased to 65% by utilizing the
conditions we recently developed to convert a related chiral diol into
(
)-chimonanthine.[13]
[16] Consistent with this interpretation, stereoselection was eroded in the
presence of a sodium-selective cryptand (Kryptofix 221).
[17] R. M. Williams, E. Kwast, Tetrahedron Lett. 1989, 30, 451 ± 454.
[18] Crystallographic data (excluding structure factors) for the structure
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication no.
CCDC-134670. Copies of the data can be obtained free of charge on
application to CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax:
(‡44)1223-336-033; e-mail: deposit@ccdc.cam.ac.uk).
[*] Prof. B. Samorì, Dr. M. Conti
Á
Dip. di Biochimica, Universita di Bologna
via Irnerio, 48, 40126 Bologna (Italy)
Fax: (‡39)051-2094387
E-mail: samori@alma.unibo.it
[19] Literature value:[6] [a]2D3 ˆ ‡264.5 (c ˆ 1.0 in EtOH).
Dr. G. Falini
Á
Dip. di Chimica G. Ciamician, Universita di Bologna
[20] a) A recent survey of stereoselective C C bond formation cites no
examples in sections dealing with alkylation reactions with chiral
electrophiles; see D. S. Matteson, Methoden Org. Chem. (Houben-
Weyl) 4th ed. 1996, Vol. E21/2, pp. 1077 ± 1118; b) Interconversion of
atropisomers of the dienolates derived from 6 is likely to be slow
under the alkylation conditions. Whether the 20:1 selectivity arises
solely in the initial alkylation step or is enhanced by selective
partitioning of the minor diastereomer in the cyclization step has not
yet been determined.
[**] We thank G. Zuccheri (Bologna) for crucial and helpful advice, one
referee for his insightful comments and enlightening suggestions, V.
Albano and A. Ripamonti (Bologna) for their advice and support, and
Biacore AB (Uppsala, Sweden) for providing us with samples of their
NTA chip and product information. This work was supported by
Programmi Biotecnologie legge 95/95 (MURST 5%), Progetto
Biologia Strutturale MURST (ex quota 40%), Progetto CNR MSTA
Á
II Biomateriali, and Universita di Bologna (Fund for Selected
Research Topics).
Angew. Chem. Int. Ed. 2000, 39, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3901-0215 $ 17.50+.50/0
215