Six-step synthesis of (S)-brevicolline from (S)-nicotine

Article abstract of DOI:10.1021/ol061334h

A six-step synthesis of (S)-brevicolline from (S)-nicotine is reported. Regioselective trisubstitution of the pyridine ring of nicotine, followed by successive Suzuki cross-coupling and Buchwald animation reactions, afforded the enantiopure β-carboline al

Full text of DOI:10.1021/ol061334h

ORGANIC
LETTERS
2006
Vol. 8, No. 16
3549-3552
Six-Step Synthesis of (S)-Brevicolline
from (S)-Nicotine
Florence F. Wagner and Daniel L. Comins*
Department of Chemistry, North Carolina State UniVersity,
Raleigh, North Carolina 27695-8204
daniel_comins@ncsu.edu
Received May 31, 2006
ABSTRACT
A six-step synthesis of (S)-brevicolline from (S)-nicotine is reported. Regioselective trisubstitution of the pyridine ring of nicotine, followed by
successive Suzuki cross-coupling and Buchwald amination reactions, afforded the enantiopure -carboline alkaloid, brevicolline.
â
The â-carboline alkaloids (S)-brevicolline (1) and brevicarine
(2) are the major alkaloids isolated1 in the late 60’s from
the plant Carex breVicollis D.C. (Cyperacee), native to the
southwestern part of the former U.S.S.R. (Figure 1).
antineoplastic (tubuline binding),3 anticonvulsive, hypnotic
and anxiolytic (benzodiazepine receptor inhibitoric),4 anti-
viral,5 antimicrobial,4b topoisomerase II inhibition,6 and
inhibition of cGMP-dependent processes.7 (S)-Brevicolline’s
biological activities range from a phototoxic effect on
bacteria and fungi to an oxytocic effect when used against
uterine internia of pregnant women.8 Two racemic syntheses
of 1 have been reported,9 and a ten-step enantioselective
synthesis starting from (S)-proline was published by Mah-
boobi and co-workers in 1999.10
(3) (a) Molina, P.; Fresdena, P. M. J. Chem. Soc., Perkin Trans. 1 1988,
1819. (b) Molina, P.; Fresdena, P. M.; Garcia-Zafra, S.; Almendros, P.
Tetrahedron Lett. 1994, 35, 8851. (c) Behforouz, M.; Merriman, R. L. (Ball
State University), U.S. Pat. Appl. 5 646 150, 1997; Chem. Abstr. 1997,
127, 149053q.
(4) (a) Braestrup, C.; Nielsen, M.; Olsen, C. E. Proc. Natl. Acad. Sci.
U.S.A. 1980, 77, 228. (b) Schlecker, W.; Huth, A.; Ottow, E.; Mulzer, J.
Synthesis 1995, 1225. (c) Batch, A.; Dodd, R. H. J. Org. Chem. 1988, 63,
872. (d) Seidelmann, D.; Huth, A.; Otow, E.; Olesen, p. H.; Turner, J.;
Hillman, M.; Cole, B. (Schering A. G.) Ger. Offen. DE 195 14 524 A1,
1996: Chem. Abstr. 1997, 126, P 8106e.
(5) Molina, P.; Fresdena, P. M.; Garcia-Zafra, S. Tetrahedron Lett. 1995,
36, 3581.
(6) Pommier, Y.; MacDonald, T. L.; Madalengoitia, J. S. (Department
of Health and Human Services) PCT Int. Aplll. WO94/101751 A1, 1997;
Chem. Abstr. 1997, 126, 343431.
Figure 1. â-Carboline alkaloids (S)-brevicolline (1) and brevicarine
(2).
The structure and absolute configuration of 1 was deter-
mined by Lazurevski1 and Bla´ha2, respectively. The â-car-
bolines are a group of pharmacologically interesting and bio-
logically active compounds, whose reported effects include
(7) Daughan, A. C. M.; Labaudiniere, R. F. (Glaxo Welcome Labora-
tories), PCT Int. Appl. WO96/32003, 1996; Chem. Abstr. 1997, 126, P
18895a.
(8) Marcu, G. A. Tr. Tret’ei Nauchn. Konf. Molodykh Uch. Mold., Biol.
S’kh. Nauki 1965, 2, 243; Chem Abstr. 1965, 63, 2297a.
(9) (a) Mu¨ller, W. H.; Preuss, R.; Winterfeldt, E. Chem. Ber. 1977, 110,
2424. (b) Leete, E. J. Chem. Soc., Chem. Commun. 1979, 821.
(1) (a) Vember, P. A.; Terenteva, I. V.; Lazurevskij, G. Khim. Prir.
Soedin. 1967, 3, 249. (b) Terenteva, I. V.; Lazurevskij, G. V.; Shirshova,
T. I. Khim. Prir. Soedin. 1969, 5, 397. (c) Vember, P. A.; Terenteva, I. V.;
Uljanova, A. V. Khim. Prir. Soedin. 1968, 4, 98.
(2) Bla´ha, K.; Koblicova´, Z.; Pospˆı’’ek, J.; Troja´nek, J. Collect. Czech.
Commun. 1971, 36, 3448.
10.1021/ol061334h CCC: $33.50
© 2006 American Chemical Society
Published on Web 07/13/2006

(S)-Brevicolline possesses the core (S)-nicotine (5) struc-
ture. Recently, we have been investigating the synthesis of
enantiopure nicotine derivatives using natural (S)-nicotine
itself as an inexpensive starting material.11 We envisioned
that 1 could be obtained via a short synthesis starting from
(S)-nicotine. It was anticipated that enantiopure 1 could arise
from a trisubstituted nicotine derivative such as 3 (Scheme
1). Following this plan, methylation of 3 by substitution at
Scheme 2. Synthesis of (S)-5,6-Dichloro-4-iodonicotine (7)
and Formation of (S)-4-Phenylnicotine (9) to Unequivocally
Prove the Regioselectivity in the Iodination Step
Scheme 1. Retrosynthetic Analysis of (S)-Brevicolline
C-6 followed by an intramolecular catalytic Buchwald
amination at C-5, creating the carboline ring, would complete
the synthesis. Compound 3 would be derived from dichlo-
ronicotine 4 in two steps involving C-4 halogenation and a
cross-coupling reaction. Finally, 4 is easily accessible in two
steps from 5.11
With the 4-iodonicotine derivative 7 in hand, various
conditions reported in the literature for the Suzuki coupling
of amino boronic ester 10 with aryl iodides13 were tested
without success (Scheme 3). The recovery of deiodonated
Our initial approach was to use a Suzuki cross-coupling
reaction between (S)-5,6-dichloro-4-iodonicotine (7) and an
amino boronate ester to afford compound 3, which, in turn,
would be suitable for an intramolecular Buchwald amination.
The synthesis of 7 was achieved in good yield in three steps
from (S)-nicotine (Scheme 2). We previously reported the
formation of (S)-6-chloronicotine (6) from natural nicotine
via an ortho-directed lithiation process.11c,f,12 (S)-5,6-Dichlo-
ronicotine (4) resulted from a regioselective lithiation-
chlorination reaction on 6, using LiTMP as the base and
hexachloroethane as the electrophile.11h Finally, 7 was
obtained from 4 via a third regioselective lithiation-
substitution process, using n-BuLi as the base and iodine as
the electrophile. The regioselectivity of this reaction could
not be confirmed by NMR experiments. To verify the
structure assignment, a Suzuki coupling with phenylboronic
acid followed by reductive removal of both chlorines using
Raney nickel afforded known (S)-4-phenylnicotine (9), the
structure of which could be unequivocally confirmed by
NMR spectroscopy.11a,d
Scheme 3. Attempts at Suzuki Couplings on 7
compound 4 showed that the oxidative addition step was
occurring rapidly (less than 2 h by TLC). The use of
Pd(OAc)2(PPh3)2 as a catalyst in dimethylacetamide gave,
in a very low yield, undesired compound 11 where the Suzuki
coupling occurs with the activated chlorine at C-6 after rapid
reduction of the iodine at C-4.
(10) Mahboobi, S.; Wiegrebe, W.; Popp, A. J. Nat. Prod. 1999, 62, 577.
(11) (a) Comins, D. L.; Despagnet, E. U.S. Patent No. 6,995,265, 2006.
(b) King, L. S.; Despagnet, E.; Comins, D. L. U.S. Patent Application No.
10/715, 147. (c) Comins, D. L.; Fe´vrier, F. C.; Despagnet, E. D. U.S. Patent
Application No. 10/926, 821. (d) Comins, D. L.; King, L. S.; Smith, E. D.;
Fe´vrier, F. C. Org. Lett. 2005, 7, 5059. (e) Fe´vrier, F. C.; Smith, E. D.
Comins, D. L. Org. Lett. 2005, 7, 5457. (f) Smith, E. D.; Fe´vrier, F. C.;
Comins, D. L. Org. Lett. 2006, 8, 179. (g) Comins, D. L.; Smith, E. D.
Tetrahedron Lett. 2006, 47, 1449. (h) Wagner, F. F.; Comins, D. L. Eur. J.
Org. Chem. 2006, in press.
(12) LiDMAE ) Me2N(CH2)2OLi. For pyridine lithiations using n-BuLi-
LiDMAE, see: Gros, P.; Choppin, S.; Mathieu, J.; Fort, Y. J. Org. Chem.
2002, 67, 234 and references therein.
(13) (a) Herrbach, A.; Marinetti, A.; Baudoin, O.; Gue´nard, D. F. J. Org.
Chem. 2003, 68, 4897. (b) Broutin, P.-E.; EÅ ero`a, I.; Campaniello, M.;
Leroux, F.; Colobert, F. Org. Lett. 2004, 6, 4419.
3550
Org. Lett., Vol. 8, No. 16, 2006

Another route to brevicolline where a Buchwald amination
would be carried out first to provide 13 and followed by a
ring closure at C-4 to give 12 was investigated (Scheme 4).
Table 1. Amination at C-5 of 15a-c
Scheme 4. Second Retrosynthetic Route
entrya
1
conditions
results
15a + 1.2 equiv of 16 + 5% Pd2dba3,
1.5 equiv of Cs2CO3, 10% Xantphos,
dioxane, 110 °C
48% of 13c
The trihalo precursors to 13 were prepared as shown in
Scheme 5.11h
2
3
4
5
15b + 1.2 equiv of 16 + 5% Pd2dba3,
1.5 equiv of Cs2CO3, 10% Xantphos,
1,4-dioxane, 110 °C, 18 h
15b + 1.2 equiv of 16 + 5% Pd2dba3,
1.5 equiv of Cs2CO3, 10% Xantphos,
1,4-dioxane, 110 °C, 3 h
15c + 1.2 equiv of 16 + 5% Pd2dba3,
1.5 equiv of Cs2CO3, 10% Xantphos,
1,4-dioxane, 110 °C, 3 h
15c + 1.0 equiv of 10 + 5% Pd2dba3,
1.5 equiv of Cs2CO3, 10% Xantphos,
1,4-dioxane, 110 °C, 3 h
inseparable
mixture of
13a and 17
53% of 13a
Scheme 5. Synthesis of Trihalo Precursors 15a-c
68% of 13b
38% of 18
a Reactions were run on a 0.3-0.8 mmol scale.
took place and amino compound 18 was formed in 38% yield
(entry 5). Because of this result, we decided to investigate
additional cross-coupling reactions at C-4 of derivatives 7
and 20 (Scheme 6). Unfortunately, compounds 3 and 21 were
formed in low yield (20% and 24%, respectively). It was
thought that, once formed, compound 21 might ring close
at C-5 under the same reaction conditions, but the desired
product was not observed. The low yield in the Buchwald
amination step was attributed to the activated C-6 chlorine
that could participate in a subsequent cross-coupling reaction
decreasing the yield of the desired product. We therefore
decided to install the C-6 methyl of (S)-brevicolline (1)
before attempting the Suzuki cross-coupling reaction at C-4
(Scheme 7).
Stille and Kumada couplings with tetramethyltin and
methyl Grignard, respectively, were carried out but only
afforded no or low yields of (S)-5-chloro-6-methylnicotine
(22). After extensive optimization, compound 22 was formed
in good yield from (S)-5,6-dichloronicotine (4) via a Suzuki
cross-coupling reaction using trimethylboroxine.14
The amination at C-5 went smoothly for a number of
different halogenated nicotine derivatives (15a-c) using 5%
Pd2dba3, 1.5 equiv of Cs2CO3, 2-bromoaniline, and Xantphos
as a ligand in 1,4-dioxane (Table 1). When the reaction was
stirred overnight at 110 °C (entry 2), disubstitution occurred
at the activated C-5 and C-6 positions of 15b to form
compound 17. With the secondary amines (13a-c) in hand,
several attempts to close the ring at C-4, using different types
of cross-coupling reactions (Ullman, Suzuki, or Stille) or
radical-mediated cyclization, failed as decomposition oc-
curred or starting material was recovered. However, to our
surprise, under standard Buchwald amination conditions, the
amino boronate ester 10 and 15c did not afford the desired
boronate product. Instead, a Suzuki cross-coupling reaction
The iodination at C-4 of 22 constituted a tricky step due
to the competitive deprotonation of the pseudo benzylic
Org. Lett., Vol. 8, No. 16, 2006
3551

Scheme 6. Attempted Suzuki Reaction at C-4
Scheme 8. Suzuki Cross-Coupling Reaction at C-4 and
Completion of the Total Synthesis
position at C-6. Although addition of 22 to n-BuLi (inverse
addition) afforded inconsistent results, addition of n-BuLi
to 22 (normal addition) gave a 53% yield of the desired
4-iodo derivative 23b with 12% of the dinicotine adduct 25
(Scheme 7).
Finally, an intramolecular Buchwald amination was chosen
to close the indole ring to form 1. Standard Buchwald ami-
nation conditions on 26 did not affect the desired transforma-
tion. However, the use of Pd2dba3, Cs2CO3, and PCy2(o-biph)
as the ligand in 1,4-dioxane yielded (S)-brevicolline (1) in
80% isolated yield. The spectral properties of our (-)-1
are in agreement with reported data.1,2,10 Because the ring-
closure conditions only differ from the usual Buchwald
amination conditions by the ligand, we thought that a one-
pot, two-step process from 23b could be used; however, in
all attempts, only 26 was formed and brevicolline was not
observed.
Scheme 7. Methylation at C-6 and Iodination at C-4
In conclusion, after extensive investigation and optimiza-
tion, (S)-brevicolline was synthesized in six steps from (S)-
nicotine in a 17% overall yield. This practical synthesis was
carried out with retention of configuration on the pyrrolidine
ring and constitutes the shortest synthesis of this natural
product to date.
Acknowledgment. NMR and mass spectra were obtained
at NCSU instrumentation laboratories, which were estab-
lished by grants from the North Carolina Biotechnology
Center and the National Science Foundation (Grants CHE-
0078253 and CHE-9509532). FFW thanks GlaxoSmithKline
for the Burroughs-Wellcome Research Fellowship for a
second-year graduate student and Eli Lilly for the Eli Lilly
Research Fellowship for a third-year graduate student.
The cross-coupling reaction between 23b and amino
boronate ester 10 went smoothly to give 26 (Scheme 8). It
is noteworthy to point out that the use of a sealed tube in
this reaction slightly increased the yield but more importantly
gave a cleaner product.
Supporting Information Available: Experimental pro-
cedures, characterization, and NMR data for 1, 3, 7-9,
13a-c, 15b,c, 18, 20-22, 23b, 25, and 26. This material is
available free of charge via the Internet at http://pubs.acs.org.
(14) Gray, M.; Andrews, I. P.; Hook, D. F.; Kitteringham, J.; Voyle,
M.; Tetrahedron Lett. 2000, 41, 6237.
OL061334H
3552
Org. Lett., Vol. 8, No. 16, 2006