Enantioselective route to ferrugine and its methyl analogue via aldol deoxygenation

Article abstract of DOI:10.1016/j.tetlet.2009.10.045

A simple enantioselective approach to ferrugine (2α-benzoyltropane) and its methyl analogue (2-acetyltropane) is reported. The four-step sequence uses an enantioselective aldol reaction of tropinone with benzaldehyde or acetaldehyde, combined with an aldo

Full text of DOI:10.1016/j.tetlet.2009.10.045

Tetrahedron Letters 50 (2009) 7196–7198
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Enantioselective route to ferrugine and its methyl analogue via
aldol deoxygenation
*
Michal Sienkiewicz, Urszula Wilkaniec, Ryszard Lazny
Institute of Chemistry, University of Bialystok, ul. Hurtowa 1, 15-399 Bialystok, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
A simple enantioselective approach to ferrugine (2a-benzoyltropane) and its methyl analogue (2-acetyl-
Received 2 September 2009
Revised 29 September 2009
Accepted 9 October 2009
Available online 14 October 2009
tropane) is reported. The four-step sequence uses an enantioselective aldol reaction of tropinone with
benzaldehyde or acetaldehyde, combined with an aldol deoxygenation via tosylhydrazone reduction
and oxidation of the side-chain hydroxy group. The final products, ferrugine and its methyl analogue,
are prepared in 35% and 23% overall yields, respectively. Both enantiomers of the products (ee 90–
99%) are accessible via the same route using either enantiomer of N,N-bis(1-phenylethyl)amine hydro-
chloride as the chiral reagent.
Ó 2009 Elsevier Ltd. All rights reserved.
Tropane alkaloids are a group of structurally related natural
products, with examples including cocaine, atropine and baogont-
eng A, many of which are known for their potent biological activ-
ity.1 Ferrugine is one of the alkaloids isolated from the extracts
of the Australian arboreal species Darlingiana ferruginea.2 Ferrugine
and its close relative, ferruginine, are nicotinic receptor antago-
nists, being potentially useful in the treatment of Alzheimer’s
disease.3
Unlike ferruginine, which has been the target of many synthe-
ses,4 ferrugine has attracted little attention. To the best of our
knowledge, only two approaches to ferrugine have been reported.
The first synthesis served as confirmation of the assigned structure
of the newly isolated alkaloid.4e Bick’s synthesis adopted the chiral
pool approach, utilizing (ꢀ)-cocaine (1) as the starting material
and anhydroecgonine5 2 as an intermediate. As a result, the syn-
thesis gave the unnatural enantiomer, (ꢀ)-ent-ferrugine (Scheme
1). Preparation of the natural isomer via such a route is not
possible.
Very recently, a new route to racemic ferrugine, using cycloh-
ept-4-enecarboxylic acid as starting material (Scheme 2), was re-
ported.6 The sequence of nine synthetic steps including Curtius
rearrangement, LAH reduction of an amide, formation of an acid
chloride with triphosgene, light-induced radical cyclization and a
Chugaev-like thermal elimination followed by PhLi or MeLi addi-
tion gave ( )-ferrugine (3) or its methyl analogue 6 in 11.8% and
13.6% overall yields, respectively (Scheme 2).
same reaction sequence. The simple, four-step synthesis employs
selective introduction of the acyl group to the tropane skeleton
using, as the key transformation, the proven enantioselective aldol
reaction of tropinone, combined with a novel aldol deoxygenation.
Synthesis of functionalized tropanes can, in principle, be based on
either of the three strategies: (i) manipulation of the available
functionalized tropane derivatives (e.g., a chiral pool approach
using natural cocaine), (ii) assembling the tropane nucleus from
suitably functionalized building blocks and (iii) functionalization
and further transformation of the tropane skeleton. The third strat-
egy is the most versatile and conceptually simple. It allows for the
preparation of either enantiomer and has already been used to ac-
cess various tropane derivatives successfully.7 Consequently, the
simplest way to construct ferrugine and similar structures would
OCOPh
MeOOC
HOOC
ref. 5
N
N
Me
Me
2
cocaine (1)
five steps
O
Ph
N
Herein, we report a new, expedient synthetic approach to ferru-
gine and its methyl analogue that allows for the preparation of all
the three forms: racemic and either enantiomeric form, via the
Me
ent-ferrugine
−
3
[( )-ent- ]
* Corresponding author. Tel.: +48 85 745 7834; fax: +48 85 747 0113.
E-mail address: lazny@uwb.edu.pl (R. Lazny).
Scheme 1. Bick’s synthesis of ent-ferrugine (ent-3).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.10.045

M. Sienkiewicz et al. / Tetrahedron Letters 50 (2009) 7196–7198
7197
were prepared in virtually diastereomerically pure form (Scheme
3). To obtain the ferrugine structure, the keto group of the aldol
products needs to be removed, and the secondary hydroxy group
oxidized to a carbonyl. Finding suitable and mild enough condi-
tions for these transformations was the main challenge of our ap-
proach. Based on our experience with the stability of tropinone
aldols, a suitable method for removal of the carbonyl group could
be via reduction of an intermediate tosylhydrazone. The first obsta-
cle was the conversion of the aldol products into tosylhydrazones
11 and 12. The use of typical literature procedures proved unsuc-
cessful: refluxing in MeOH or EtOH with p-toluenesulfonyl hydra-
zide; with 3 Å molecular sieves and p-toluenesulfonic acid,9 or
reaction in benzene with BF3ꢁEt2O.10 Under these conditions, the
aldols decomposed giving none of the desired products. Finally,
the reaction promoted by titanium(IV) isopropanolate, which was
previously used for the syntheses of imines,11 gave good results.
Investigation of the reaction at room temperature in three different
solvents (Table 1, Scheme 4) indicated absolute EtOH as the med-
ium of choice.
Hydrazones 11 and 12 were synthesized under the optimized
reaction conditions in 82% and 62% yield, respectively. Work-up
of the titanium-promoted reaction was complicated by copious
amounts of titanium hydroxides that were difficult to be sepa-
rated. The crude tosylhydrazones were eventually purified by
fast dry-column flash chromatography or crystallization from
ethanol. Attempts to oxidize the hydroxy group at this stage with
DMP,12 PDC, SO3ꢁPy,13 Swern oxidation14 or IBX15 gave negligible
O
COOH
Me
N
seven
steps
two
steps
O
R
N
Me
Et2N
S
4
( )-3, R = Ph
( )-6, R = Me
5
S
Scheme 2. Synthesis of ( )-ferrugine (3) via thermal elimination of
a
diethyldithiocarbamate.
be via installation of an acyl group at C-2 of the tropane skeleton.
Plans based on direct tropinone acylation, although theoretically
feasible, were hard to achieve in practice owing to the problems
with chemoselective deoxygenation of one of the carbonyls of
the resulting 1,3-diketone or their synthetic equivalents. We envis-
aged, however, an aldol reaction, followed by functional group
manipulation, might be a better and synthetically equivalent trans-
formation sequence. This could be realized provided that the sen-
sitive aldol product (prone to retroaldolization, dehydration and
isomerization) was stable under the subsequent reaction condi-
tions and could be deoxygenated. To the best of our knowledge
such an approach to enantioselective acylation of the tropane scaf-
fold has not been studied.
Consequently, starting from tropinone, which is a relatively
cheap and readily available scaffold, suitable for enantioselective
deprotonation,8 and using as the key transformation, a highly ster-
eoselective aldol reaction (a-deprotonation with LDA or a chiral
lithium amide, followed by reaction with benzaldehyde or acetal-
Table 1
dehyde), the known product 8 and its novel methyl analogue 9
Studies on the preparation of tosylhydrazone 11
Entry
Reagent and
solvent
Conversion of 8
Isolated yield
into productsa (%)
of 11b (%)
3
6
( )-8 R = Ph (91%),
( )-9 R = Me (87%)
( )-
( )-
1
2
3
4
5
MeOH
abs EtOH
—
86
—
80
100
—
15
—
20
82
1. LDA, THF, 0 °C
2. RCHO
BF3ꢁEt2O
Ti(Oi-Pr)4, AcOEt
Ti(Oi-Pr)4, abs EtOH
O
N
Calculated from the 1H NMR spectra of crude reaction mixtures.
Product purified by chromatography on silica gel using CH2Cl2 saturated with
NH3 as eluent.
a
b
Me
tropinone (7)
1. (R,R)-10,
n-BuLi
1. (S,S)-10,
n-BuLi
NHTs
HO
HO
O
N
2. RCHO
2. RCHO
H
H
Me Me
TsNHNH2,
Ti(OiPr)4,
HO
OH
O
O
H
R
R
H
*
Ph *
N
Ph
N
N
H · HCl
R
R
EtOH
10
N
N
Me
Me
( )-8, ( )-9
( )-11, R = Ph (82%),
Me
Me
−
−
−
≥
( )-8, ( )-9
( )-11, (55%,ee 99%)
( )-12, R = Me (62%),
(−
8
)- R = Ph (80%,
8
(+)- R = Ph (79%,
−
≥
( )-12, (60%, ee 99%)
≥
ee 99%),
≥
ee 99%),
−
( )-9 R = Me (65%,
(+)-9 R = Me (66%,
≥
ee 99%)
≥
ee 99%)
DIBAL,
CH2Cl2, 0 °C
HO
H
O
O
O
R
R
DMP, Py
CH2Cl2
N
N
R
R
N
N
Me
Me
Me
Me
( )-13, R = Ph (83%),
( )-3, R = Ph (56%),
ent-ferrugine
−
≥
−
≥
( )-13, (74%, ee 99%)
( )-14, R = Me (82%),
ferrugine
( )-3, (56%, ee 99%)
( )-6, R = Me (53%),
−
3
[( )-ent- ]
+
3
[( )- ]
− ≥
( )-14, (71%, ee 90%)
−
≥
( )-6, (59%, ee 90%)
Scheme 3. The aldol reaction of tropinone (7) as the key step in the synthesis of
ferrugine and its methyl analogue.
Scheme 4. Synthesis of ( )-ferrugine (3) and ent-3 by aldol deoxygenation.

7198
M. Sienkiewicz et al. / Tetrahedron Letters 50 (2009) 7196–7198
amounts of the desired product. It is thought that the oxidation
may be hindered either by hydrogen bond formation between
the hydroxy group and the nitrogen atom of the tosylhydrazone
group or by steric congestion in the structure. Therefore, reduc-
tive deoxygenation of the tosylhydrazone group was investigated
at this stage. The use of NaBH4 in absolute ethanol16 or in DMF
with and without heating, as well as LAH,17 failed to give the de-
sired product. However, reduction with NaBH4 in a mixture of
acetic acid and MeOH gave the expected product 13 in 55% yield.
A better result was achieved using DIBAL-H.18
enantioselective deprotonation and the aldol reaction of tropinone.
Thus, by using our approach, the racemate or either enantiomer of
the 2-acyltropane derivative can be synthesized with high enantio-
meric purity (P90–99%) using LDA or the commercially available
chiral reagents, (R,R)- or (S,S)-N,N-bis(1-phenylethyl)amine hydro-
chloride.21 The natural enantiomer of ferrugine, which is not avail-
able using Bick’s chiral pool strategy4e starting from (ꢀ)-cocaine,
was synthesized for the first time.
Acknowledgements
When this reaction was carried out at 0 °C for 72 h the product
was obtained in 83% yield after purification on deactivated silica
gel (elution with dichloromethane saturated with NH3). Represen-
tative results on tosylhydrazone reductions are shown in Table 2.
Applying the optimized reduction method to the methyl analogue
12 provided exo-2-(1-hydroxyethyl)tropane 14 in 82% yield. The fi-
nal step was oxidation of the side-chain hydroxy group. Attempted
Swern oxidation gave poor results, but the use of typical Dess–
Martin oxidation (in dry dichloromethane with pyridine at rt for
22 h) provided ferrugine 3 and the methyl analogue 6 (Scheme
4), in 56% and 53% yields, respectively, after chromatography. Ana-
lytical data was in agreement with those reported in the litera-
ture.4e,6 Having devised a synthetic route to racemic ferrugine
the enantioselective synthesis of either enantiomer was realized
using the commercially available hydrochloride of chiral amine
10 (Scheme 3). Thus (ꢀ)-exo-2-(1-hydroxybenzyl)tropinone (ꢀ)-8
was prepared using (R,R)-N,N-bis(1-phenylethyl)amine hydrochlo-
ride8a and two equivalents of n-butyllithium. Following the litera-
ture procedure,8a the aldol product (ꢀ)-8 was isolated by
precipitation and crystallization in 80% yield and with an
ee P 99%. The tosylhydrazone (ꢀ)-11 was obtained in the next
step in 55% yield and with an ee P 99%.19 Reduction of (ꢀ)-11 with
DIBAL-H gave the chiral alcohol (ꢀ)-13 in 74% yield (ee P 99%) and
oxidation of the hydroxy group gave ent-ferrugine [(ꢀ)-3] in 56%
yield. Measurement of the enantiomeric excess of the final product
by 1H NMR in the presence of (+)-TFAE proved the high enantio-
meric purity, ee P 99%. However, the optical rotations of the syn-
We are grateful to the University of Bialystok (BST-125 and BW-
173) for financial support. We also thank Dr. L. Siergiejczyk for
assistance in recording NMR spectra.
References and notes
1. (a) Lounasmaa, M.. In The Alkaliods; Brossi, A., Ed.; Academic Press: New York,
1988; Vol. 33, (b) Clarke, R. L.. In The Alkaloids; Manske, R. H. F., Ed.; Academic
Press: New York, 1977; Vol. 16, (c) Fodor, G.. In The Alkaloids; Manske, R. H. F.,
Ed.; Academic Press: New York, 1971; Vol. 13, (d) Fodor, G.; Dharanipragada, R.
Nat. Prod. Rep. 1993, 9, 199–206; (e) Fodor, G.; Dharanipragada, R. Nat. Prod.
Rep. 1994, 11, 443–456.
2. Bick, I. R. C.; Gillard, J. W.; Woodruff, M. Chem. Ind. 1975, 794.
3. Swanson, K. L.; Albuquerque, E. X. Handbook Exp. Pharmacol. 1992, 102, 620–
621.
4. Selected papers: (a) Jonsson, S. Y.; Löfström, C. M. G.; Bäckvall, J.-E. J. Org. Chem.
2000, 65, 8454–8457; (b) Aggarwal, V. K.; Astle, C. J.; Rogers-Evans, M. Org. Lett.
2004, 6, 1469–1471; (c) Katoh, T.; Kakiya, K.; Nakai, T.; Nakamura, S.; Nishide,
K.; Node, M. Tetrahedron: Asymmetry 2002, 13, 2351–2358; (d) Gauthier, I.;
Royer, J.; Husson, H.-P. J. Org. Chem. 1997, 62, 6704–6705; (e) Bick, I. R. C.;
Gillard, J. W.; Leow, H.-M. Aust. J. Chem. 1979, 32, 2537–2543; (f) Hernández, A.
S.; Thaler, A.; Castells, J.; Rapoport, H. J. Org. Chem. 1996, 61, 314–323; (g)
Davies, H. M. L.; Saikali, E.; Young, W. B. J. Org. Chem. 1991, 56, 5696–5700.
5. (a) Campbell, H. F.; Edwards, O. E.; Kolt, R. Can. J. Chem. 1977, 55, 1372–1379;
(b) Einhorn, A. Ber. Dtsch. Chem. Ges. 1887, 20, 1221–1229.
6. Ahmed, S.; Baker, L. A.; Grainger, R. S.; Innocenti, P.; Quevedo, C. E. J. Org. Chem.
2008, 73, 8116–8119.
7. Pollini, G. P.; Benetti, S.; De Risi, C.; Zanirato, V. Chem. Rev. 2006, 106, 2434–
2454.
8. (a) Majewski, M.; Lazny, R. J. Org. Chem. 1995, 60, 5825–5830; (b) Bunn, B. J.;
Simpkins, N. S. J. Org. Chem. 2002, 58, 533–534; (c) Bunn, B. J.; Simpkins, N. S.;
Spavold, Z.; Crimrnin, J. J. Chem. Soc., Perkin Trans. 1 1993, 3113–3116; (d) Lee, J.
C.; Lee, K.; Cha, J. K. J. Org. Chem. 2000, 65, 4773–4775; (e) Majewski, M.; Zheng,
G.-Z. Can. J. Chem. 1992, 70, 2618–2626.
9. (a) Yamamoto, H.; Nakamura, Y.; Inokawa, S.; Yamashita, M.; Armour, M. A.;
Nakashima, T. T. J. Org. Chem. 2002, 49, 1364–1370; (b) Kabalka, G. W.;
Summers, S. T. J. Org. Chem. 2002, 46, 1217–1218; (c) Miller, V. P.; Yang, D. Y.;
Weigel, T. M.; Han, O.; Liu, H. W. J. Org. Chem. 2002, 54, 4175–4188.
10. Grieco, P. A.; Nishizawa, M. J. Org. Chem. 2002, 42, 1717–1720.
11. Alexakis, A.; Gille, S.; Prian, F.; Rosset, S.; Ditrich, K. Tetrahedron Lett. 2004, 45,
1449–1451.
12. (a) Meyer, S. D.; Schreiber, S. L. J. Org. Chem. 2002, 59, 7549–7552; (b) Dess, D.
B.; Martin, J. C. J. Org. Chem. 2002, 48, 4155–4156.
13. Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 2002, 89, 5505–5507.
14. (a) Liu, Y.; Vederas, J. C. J. Org. Chem. 1996, 61, 7856–7859; (b) Marx, M.;
Tidwell, T. T. J. Org. Chem. 2002, 49, 788–793; (c) Chrisman, W.; Singaram, B.
Tetrahedron Lett. 1997, 38, 2053–2056.
thesized product measured at two different concentrations (½a D
ꢀ100, c 1, CHCl3 and ½a D ꢀ29, c 0.3, CHCl3) did not compare well
with the literature value for natural ferrugine (½a D
+55, CHCl3),4e
ꢂ
ꢂ
ꢂ
however, the concentration at which the optical rotation of the
natural product was recorded was not given.20 The natural enan-
tiomer of ferrugine was also obtained via the same reaction se-
quence using (S,S)-N,N-bis(1-phenylethyl)amine hydrochloride
(S,S-10) for the tropinone deprotonation. The enantiomers of the
methyl analogues were equally accessible from enantiomers of
aldols 9 (Scheme 4). The acetyl tropane 6 was, however, present
admixed with a small amount of its exo diastereomer (2b-acetyl-
tropane, ca. 15%).
15. More, J. D.; Finney, N. S. Org. Lett. 2002, 4, 3001–3003.
16. (a) Tavernier, D.; Hosten, N.; Anteunis, M. Synthesis 1979, 613–614; (b)
Hutchins, R. O.; Maryanoff, B.; Milewski, C. J. Am. Chem. Soc. 2002, 93, 1793–
1794.
In summary, we have developed a simple, four-step, route to
2-acyl tropanes based on aldol deoxygenation via tosylhydrazone
reduction. The method was used for the first enantioselective syn-
thesis of ferrugine and its methyl analogue (2-acetyltropane) using
17. Kutney, J. P.; Singh, A. K. Can. J. Chem. 1983, 61, 1111–1114.
18. Lightner, D. A.; Paquette, L. A.; Chayangkoon, P.; Lin, H. S.; Peterson, J. R. J. Org.
Chem. 2002, 53, 1969–1973.
19. All enantiomeric excesses were measured by 1H NMR in the presence of (+)-
2,2,2-trifluoro-1-(9-anthranyl)ethanol, (+)-TFAE.
20. The substantial difference in the values recorded at c = 1 and c = 0.3 clearly
showed that the specific rotation of (ꢀ)-ferrugine is dependent on
Table 2
Selected results on the deoxygenative reduction of the tosylhydrazone group in 11
concentration and that the literature value must have been taken at
concentration between 0.3 and 1.0.
a
Entry
Reaction conditions
Yield (%)
21. All new compounds gave satisfactory analytical data: (ꢀ)-2
a-Benzoyltropane,
1
2
3
4
5
6
7
NaBH4, EtOH, 35 °C
NaBH4, DMF, 80 °C
NaBH4, MeOH, AcOH, 80 °C
LAH, THF, rt
LAH, THF, reflux
DIBAL, CH2Cl2, 0 °C
NaH, PhH
—
—
[(ꢀ)-ent-ferrugine], (ꢀ)-3. Purification using preparative thin layer
chromatography (PTLC) (5% MeOH/CH2Cl2+NH3) gave a yellow oil (0.110 g,
56%). Rf = 0.5 (10% MeOH/CH2Cl2); 1H NMR (CDCl3, 400 MHz): 7.98–7.92 (m,
2H), 7.57–7.51 (m, 1H), 7.49–7.43 (m, 2H), 3.82–3.76 (m, 1H), 3.35 (d,
J = 6.0 Hz, 1H), 3.20–3.15 (m, 1H), 2.35 (s, 3H), 2.05–1.69 (m, 5H), 1.62–1.45
(m, 3H); 13C NMR (CDCl3, 100 MHz) 201.5, 136.3, 132.8, 128.6, 128.4, 63.6,
61.1, 47.7, 40.3, 29.8, 26.0, 22.7, 18.5; mmax (CHCl3) 1675 (C@O) cmꢀ1 HRMS
55
—
33a
83
—
(EI): M+, found 229.1460, C15H19NO requires 229.1467; ee P 99%;½a D20
ꢀ100 (c
ꢂ
1, CHCl3), ꢀ29 (c 0.3, CHCl3); lit.4e data for (+)-enantiomer ½a 1D9
ꢂ
+55, (CHCl3).
a
Calculated from the 1H NMR spectra of crude reaction mixtures.