Asymmetric synthesis of 2-azabicyclo[3.3.1]nonanes by a microwave-assisted organocatalysed tandem desymmetrisation and intramolecular aldolisation

Article abstract of DOI:10.1039/b906835j

The six-membered nitrogen-containing ring of the morphan scaffold, ubiquitous in natural products, is formed by an intramolecular aldol process of an aza-tethered dicarbonyl compound, leading to the first asymmetric synthesis of a morphan derivative using

Full text of DOI:10.1039/b906835j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
Asymmetric synthesis of 2-azabicyclo[3.3.1]nonanes by a microwave-assisted
organocatalysed tandem desymmetrisation and intramolecular aldolisation†
Fa¨ıza Diaba* and Josep Bonjoch*
Received 6th April 2009, Accepted 6th May 2009
First published as an Advance Article on the web 18th May 2009
DOI: 10.1039/b906835j
The six-membered nitrogen-containing ring of the morphan
scaffold, ubiquitous in natural products, is formed by an
intramolecular aldol process of an aza-tethered dicarbonyl
compound, leading to the first asymmetric synthesis of a
morphan derivative using organocatalysis.
Reductive amination of 1,4-cyclohexanedione monoethylene ac-
etal with aminoacetaldehyde diethyl acetal gave the secondary
amine 2, which was converted to the carbamate 3 (85% overall
yield, Scheme 1). After several hydrolytic studies, we found that
the treatment of 3 with 5% aqueous HCl in THF for 10 min at
room temperature furnished the key intermediate keto aldehyde 1
(95%). If the acid treatment was performed (10% HCl) for a period
of 4 h concomitant to the hydrolytic process the aldol cyclisation
stereoselectively gave the bridged hydroxy ketone rac-4 (87%).
However, if the acid treatment was shortened to 25 min, the major
compound isolated after the basic work-up was the isoquinuclidine
rac-5. These results a priori suggested that 5 was produced by a
kinetic-controlled process, while 4 derived from a thermodynamic
control. Yet when 5 was submitted to acid treatment no reversion
was observed and the isoquinuclidine was recovered. Thus, the
formation of rac-5 in a short reaction time was likely to be the
result of the basic treatment in the work-up of the unreacted keto
aldehyde. In fact, a few minutes of exposure of 1 to an aqueous
K2CO3 solution gave compound 5 in good yield (88%).
Many biologically active natural products such as the immuno-
suppressant FR901483,1 the cytotoxic agents aspernomine2 and
madangamine A,3 the serine protease inhibitor suomilide,4 as well
as the classical alkaloids morphine,5 strychnine,6 and Daphniphyl-
lum7 contain the 2-azabicyclo[3.3.1]nonane (morphan) scaffold
within their frameworks (Fig. 1). Nevertheless, very few methods
are currently available to synthesise enantiopure morphans.8 Of
these, only the Eschenmoser approach relies on an asymmetric
procedure, which involves a Diels–Alder catalysed by TADDOL in
the key step.8b Thus, the development of catalytic enantioselective
methods in the field of morphan synthesis remains challenging.
Fig. 1 Challenging morphan-containing alkaloids.
Here we report a direct, catalytic, highly diastereoselective
and enantioselective approach to morphans using a proline-type
organocatalysed intramolecular aldol reaction.9–12 Although the
bridged azabicyclic nucleus, when embedded in a more complex
structure (e.g. FR90148313,14 and dasycarpidan skeleton15), has
been achieved by an aldol process, to date no general protocol has
been developed to synthesise its scaffold using a classical aldol
reaction upon an aminotethered keto aldehyde, even for racemic
compounds. This could be partly attributed to the lability of the
required 2-aminoacetaldehydes.16,17
In order to proceed with our studies, we required an achiral
keto aldehyde of type 1, which is unprecedented in the literature.18
Scheme 1 The regiodivergent synthesis of racemic 4 and 5.
Laboratori de Qu´ımica Orga`nica, Facultat de Farma`cia, Institut de
Biomedicina (IBUB), Universitat de Barcelona, Av. Joan XXIII s/n, 08028-
Barcelona, Spain. E-mail: josep.bonjoch@ub.edu, faiza.diaba@ub.edu;
Fax: +34 934024539
† Electronic supplementary information (ESI) available: Experimental and
NMR data of all compounds and copies of 1H and 13C NMR spectra. See
DOI: 10.1039/b906835j
With the achiral keto aldehyde 1 and the racemic azabicyclic
compounds rac-4 and rac-5 in hand, we turned our attention to
the organocatalysed cyclisation of 1 and the deracemisation of
the bridged b-hydroxy carbonyl compounds 4 and 5 to obtain the
targeted enantioenriched morphan 4. Starting from rac-5 the best
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2517–2519 | 2517
©

View Article Online
Table 1 Catalyst screeninga
Table 2 Results of the microwave-assisted aldol reaction of 1 in the
presence of L-4-(tert-butyldiphenylsilyloxy)proline Ca
Entry
Time/min
Cat. mol%
Yield (%)b
ee (%)c
5
1
15
15
15
15
15
15
15
5
25
25
20
20
10
5
70
78
70
66
77
74
50
65
70
54
65
62
62
54
39
63
—
—
—
—
—
10
38
13
2d
3
4e
5
6
7
8
2.5
10
a All reactions were carried out in sealed 10 mL tubes. Maximal irradiated
power 300 W. Maximal temperature 100 ◦C. CH3CN was used as the
solvent and H2O (10 equiv) was added unless noted. b Yields of isolated
compounds. c Determined by HPLC with a DAICEL Chiralpack IC
column. d H2O was omitted. e BINOL (3%) was used as an additive
Entry Catalyst t/h
Cat. mol% Yield (%)b ee (%)c
5 (%)
1
2
3
4
5
6
7
8
Ad
Ae
Bf
48
48
48
48
48
48
48
50
50
50
30
50
25
20
10
56
38
18
70
46
20
41
8
18
-2
40
72
70
3
35
16
out without water, the ee clearly diminished (entry 2). Decreasing
the catalyst loading from 25% to 20% had practically no effect
(entry 3), but further reductions to 10%, 5%, and 2.5% led to a
drop in ee (entries 5–7). The use of L-BINOL (3%) as an additive
did not improve the results (66% yield, 62% ee). Finally, it is
also worth mentioning that when using catalyst C (10%) under
microwave conditions, isoquinuclidine rac-5 was converted to (-)-4
(74% yield, 60% ee) while, as expected, rac-4 and enantioenriched
4 remained unchanged.
A remaining task was to determine the absolute configura-
tion of the major heterocyclic ring formed. Thus, a sample of
enantioenriched (-)-4 was converted to the corresponding OAc-
mandelic derivatives22 (see ESI†), which were separated by column
chromatography. NMR analysis of both diastereomers of 4.AcMA
allowed us to assign the configuration 1R,4S,5S to the major
enantiomer.
Cg
Ch
Ci
24
50
Dj
Ck
30 min 20
70
a Unless otherwise noted, the reaction was carried out from 100 mg of keto
aldehyde 1 at room temperature. b Yield of isolated product 4. c Determined
by chiral HPLC with a DAICEL Chiralpak IC column. d DMSO as the
solvent. e CH3CN as the solvent. f Trifluoroacetate salt of B was used. g H2O
as the solvent. h CH3CN (2 mL), 10 equiv H2O. i 50 mg scale. j CH2Cl2;
benzoic acid (30%). k 100 ◦C, sealed tube.
result was observed using L-Pro (50%) in CH3CN, compound (-)-4
being obtained in 52% yield after one week, although with a low
ee (33%). On the contrary, compound rac-4 was found to be inert
under the same reaction conditions.
The proposed mechanism for the synthesis of enantioenriched
aldol 4 through reaction of the prochiral keto aldehyde 1 is
illustrated in Scheme 2. The kinetically formed isoquinuclidine 5
evolves to morphan 4, since the first aldol process (1 → 5), in which
the aldehyde and ketone act as a nucleophile and electrophile,
respectively, is reversible. The transformation (1 → 4) involves
two successive steps: an asymmetric desymmetrisation for the
enamine formation, and a diastereoselective and enantioselective
aldol cyclisation of the generated enamine. The results suggest
that the asymmetric induction in the initial enamine formation
The organocatalysed intramolecular reaction of 1 was studied
with four proline derivatives (A–D) as shown in Table 1. L-Proline
(A) catalysed the formation of 4 in very poor yield in DMSO,
but in CH3CN the conversion clearly improved (56%), although
once again the ee was very low (entries 1,2). The bifunctional
catalyst diamine B/TFA was also ineffective (entry 3). Treatment
of keto aldehyde 1 with the proline-substituted catalyst C, using
CH3CN as a solvent in the presence of water,19 gave the best
results, producing (-)-4 in 70% overall yield and 72% ee with
total diastereoselectivity (entry 5). Catalyst D gave poor results in
this aldol process, isoquinuclidine 5 being isolated in 50% yield
(entry 7).
In order to assess if the rotamer ratio of the keto aldehyde 1 (2:1
at room temperature) had any effect upon the enantioselectivity of
the process, the aldol reaction was carried out at 100 ◦C for 30 min
using 20 mol% of catalyst C. Although the chemical yield of 4 was
low (entry 8), the enantioselectivity was maintained, which ruled
out the influence of the ratio rotamers.20
Since the best reaction conditions at room temperature were
lengthy and required high catalyst loading (Table 1, entry 5),
and considering that the enantioselectivity was maintained when
working at 100 ◦C (entry 8), we decided to accelerate the process
using microwave activation.21 Indeed, as can be seen in Table 2,
(-)-4 was obtained in 70% yield, with total diastereoselectivity
and the same enantioselectivity, after only 15 min and a lower
amount of catalyst (entry 1).‡ When the reaction was carried
Scheme 2 The organocatalyzed synthesis of (-)-4.
2518 | Org. Biomol. Chem., 2009, 7, 2517–2519
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
was not very high, while the diastereoselectivity of the second
aldol process was excellent. Since the treatment under proline-type
catalysis of rac-4 and (-)-4 gave the starting material unchanged,
we can conclude that the reverse aldolisation was not operative
in these reaction conditions. Thus, the observed enantioselectivity
was determined by an irreversible kinetically-controlled step in
which the ketone acts as a nucleophile (Scheme 2). Additionally,
it is worth mentioning that water plays an effective role in the
process,23 probably with a favourable hydrophilic interaction in
the transition state leading to (-)-4 (compare entries 1 and 2 in
Table 2).
8 (a) J. Quirante, M. Torra, F. Diaba, C. Escolano and J. Bonjoch,
Tetrahedron: Asymmetry, 1999, 10, 2399–2410; (b) G. Karig, A. Fuchs,
A. Bu¨sing, T. Brandstetter, S. Scherer, J. W. Bats, A. Eschenmoser and
G. Quinkert, Helv. Chim. Acta, 2000, 83, 1049–1078; (c) J. Quirante, F.
Diaba, X. Vila, J. Bonjoch, E. Lago and E. Molins, C. R. Acad. Sci,
2001, 4, 513–521; (d) M. Amat, M. Pe´rez, A. T. Minaglia and J. Bosch,
J. Org. Chem., 2008, 73, 6920–6923.
9 For some reviews about asymmetric aminocatalysis of aldol processes,
see: (a) C. Allemann, R. Gordillo, F. R. Clemente, P. H.-Y. Cheong
and K. N. Houk, Acc. Chem. Res., 2004, 37, 558–569; (b) G. Guillena,
C. Na´jera and D. J. Ramon, Tetrahedron: Asymmetry, 2007, 18, 2249–
2293; (c) C. F. Barbas III, Angew. Chem., Int. Ed., 2008, 47, 42–47; (d) P.
Melchiorre, M. Marigo, A. Carlone and G. Bartoli, Angew. Chem., Int.
Ed., 2008, 47, 6138–6171.
10 Apart from the studies in the Wieland–Miescher ketone and analogs
synthesis,11 few examples of asymmetric intramolecular aldol processes
using organocatalysis have been reported12.
In summary, we have described the first nitrogen-containing
ring synthesis by means of an asymmetric aldol reaction us-
ing organocatalysis. The desymmetrisation of a prochiral 4-
N-protected aminocyclohexanone24 through its intramolecular
aldolisation, as reported here, still has some outstanding enan-
tioselectivity issues to be resolved. With the aim of improving the
enantioselectivity of the process and applying it to natural product
synthesis, studies with additional chiral amines are ongoing.
11 (a) T. Bui and C. F. Barbas III, Tetrahedron Lett., 2000, 41, 6951–6954;
(b) B. List, L. Hoang and H. J. Martin, Proc. Natl. Acad. Sci. U. S.
A., 2004, 101, 5839–5842 and references therein; (c) S. G. Davies, A.
J. Russell, R. L. Sheppard, A. D. Smith and J. E. Thompson, Org.
Biomol. Chem., 2007, 3190–3200; (d) D. B. Ramachary and M. Kishor,
J. Org. Chem., 2007, 72, 5056–5068; (e) T. Kanger, K. Kriis, M. Laars, T.
Kailas, A.-M. Mu¨u¨risepp, T. Pehk and M. Lopp, J. Org. Chem., 2007,
72, 5168–5173; (f) G. Guillena, C. Na´jera and S. F. Vio´zquez, Synlett,
2008, 3031–3035.
12 (a) C. Pidathala, L. Hoang, N. Vignola and B. List, Angew. Chem.,
Int. Ed., 2003, 42, 2785–2788; (b) N. Itagaki, M. Kimura, T. Sugahara
and Y. Y. Iwabuchi, Org. Lett., 2005, 7, 4185–4188; (c) Y. Hayashi, H.
Sekizawa, J. Yamaguchi and H. Gotoh, J. Org. Chem., 2007, 72, 6493–
6499; (d) J. Zhou, V. Wackchaure, P. Kraft and B. List, Angew. Chem.,
Int. Ed., 2008, 47, 7656–7658.
13 For racemic compounds, see: (a) J.-H. Maeng and R. L. Funk, Org.
Lett., 2001, 3, 1125–1128; (b) T. Kan, T. Fujimoto, S. Ieda, Y. Asoh, H.
Kitaoka and T. Fukuyama, Org. Lett., 2004, 6, 2729–2731.
14 For enantiopure compounds by diastereoselective processes, see: (a) B.
B. Snider and H. Lin, J. Am. Chem. Soc., 1999, 121, 7778–7786; (b) G.
Scheffler, H. Seike and E. J. Sorensen, Angew. Chem., Int. Ed., 2000,
39, 4593–4596; (c) M. Ousmer, N. A. Braun, C. Bavoux, M. Perin and
M. A. Ciufolini, J. Am. Chem. Soc., 2001, 123, 7534–7538; (d) S. Kaden
and H.-U. Reissig, Org. Lett., 2006, 8, 4763–4766.
15 S. Patir, P. Rosenmund and P. H. Goetz, Heterocycles, 1996, 43, 15–22.
16 For N-dealkylations of aminoacetaldehyde derivatives, see: D. Sole´, J.
Bosch and J. Bonjoch, Tetrahedron, 1996, 52, 4013–4028 and references
therein.
17 For the use of a protected aminoacetaldehyde in organocatalytic
asymmetric intermolecular aldol reactions, see: R. Thayumanavan, F.
Tanaka and C. F. Barbas III, Org. Lett., 2004, 6, 3541–3544.
18 A search in SciFinder reveals that the structure of 4-(2-
oxoethyl)aminocyclohexanone (1) has not been reported for any N-
substituent-type.
19 Y. Hayashi, Angew. Chem., Int. Ed., 2006, 45, 8103–8104.
20 The influence of the protecting group on the nitrogen atom was
unexplored. In any case, the use of a phthalimido protecting group,
which was the key in Barbas’ work (ref. 17), is structurally impossible
in the synthesis of nitrogen-containing rings.
21 For some organocatalyzed asymmetric reactions via microwave acti-
vation: (a) B. Westermann and C. Neuhaus, Angew. Chem., Int. Ed.,
2005, 44, 4077–4079; (b) B. Rodr´ıguez and C. Bolm, J. Org. Chem.,
2006, 71, 2888–2891; (c) S. Mosse´ and A. Alexakis, Org. Lett., 2006, 8,
3577–3580; (d) M. Hosseini, N. Stiasni, V. Barbieri and C. O. Kappe,
J. Org. Chem., 2007, 72, 1417–1424.
Acknowledgements
This research was supported by the Ministerio de Ciencia e Inno-
vacio´n (Spain)-FEDER through project CTQ2007-61338/BQU.
Thanks are also due to the Comissionat per a Universitats i
Recerca (Catalonia) for Grant 2005SGR-00042.
Notes and references
‡ Reaction procedure: in a 10 mL vessel were placed keto aldehyde 1
(100 mg, 0.47 mmol), catalyst C (43 mg, 0.12 mmol, 25%), acetonitrile
(1 mL) and water (0.08 mL, 4.7 mmol). The mixture was heated
with stirring to 100 ◦C using microwave irradiation for 15 min. After
concentration, the reaction mixture was purified by chromatography
(dichloromethane/ethyl acetate 1:1) to give 4 (70 mg, 70%) as a viscous
colourless oil: HPLC (Daicel Chiralpak IC, dichloromethane/methanol
99:1, 1 mL min-1, l = 290 nm; major isomer t = 8.48 min; minor isomer
9.63 min); IR (NaCl, neat): 3413, 1698 cm-1; 1H NMR (CDCl3, 400 MHz,
gCOSY) 1.90–2.24 (m, 4H), 2.46 (dt, 1H, J = 18, 8.4 Hz, H-7ax), 2.59
(ddd, 1H, J = 18, 9.2, 4.8 Hz, H-7eq), 2.82 (br s, 1H, H-5), 2.89 (t, 1H,
J = 12.4 Hz, H-3ax), 3.24 and 3.27 (2d, 1H, J = 5.6 Hz, OH), 3.73 (s, 3H,
CH3), 3.98 (br s, 1H, H-4), 4.23 and 4.37 (2dd, 1H, J = 13.2, 6 Hz, H-3eq),
4.45 and 4.61 (2brs, 1H, H-1); 13C NMR (CDCl3, 100 MHz, gHSQC) 27.7
and 28.6 (C-9), 29.4 and 29.9 (C-8), 38.3 (C-7), 43.4 and 43.7 (C-1), 45.9
and 46.1 (C-3), 49.4 (C-5), 52.8 (CH3), 67.4 and 67.7 (C-4), 156.0 (CO),
212.3 and 213.2 (C-6). HRMS (ESI-TOF) Calcd for C10H16NO4: 214.1073
(M + H+), Found 214.1074.
1 (a) J. Bonjoch and F. Diaba, Studies in Natural Products Chemistry,
2005, 32, 3–60; (b) C. A. Carson and M. A. Kerr, Org. Lett., 2009, 11,
777–779 and references therein.
2 G. M. Staub, J. B. Gloer, D. T. Wicklow and P. F. Dowd, J. Am. Chem.
Soc., 1992, 114, 1015–1017.
3 J. Quirante, L. Paloma, F. Diaba, X. Vila and J. Bonjoch, J. Org. Chem.,
2008, 73, 768–771 and references therein.
22 (a) I. Chataigner, J. Lebreton, D. Durand, A. Guingant and J. Villie´ras,
Tetrahedron Lett., 1998, 39, 1759–1762; (b) J. M. Seco, E. Quin˜oa´ and
R. Riguera, Chem. Rev., 2004, 104, 17–117.
23 For a review about water in stereoselective organocatalytic reactions,
see: M. Gruttadauria, F. Giacalone and R. Notoa, Adv. Synth. Catal.,
2009, 351, 33–57.
24 The only precedent of a 4-aminocyclohexanone asymmetric desym-
metrization is upon the N(Boc)2 derivative and using Simpkins’ base:
V. K. Aggarwal and B. Olofsson, Angew. Chem., Int. Ed., 2005, 44,
5516–5519.
4 C. S. Schindler, C. R. J. Stephenson and E. M. Carreira, Angew. Chem.,
Int. Ed., 2008, 47, 8852–8855.
5 (a) J. Zezula and T. Hudlicky, Synlett, 2005, 388–405; (b) H. Tanimoto,
R. Saito and N. Chida, Tetrahedron Lett., 2008, 49, 358–361 and
references therein.
6 (a) J. Bonjoch and D. Sole´, Chem. Rev., 2000, 100, 3455–3482; (b) J.
Boonsombat, H. Zhang, M. J. Chughtai, J. Hartung and A. Padwa, J.
Org. Chem., 2008, 73, 3539–3550 and references therein.
7 D. Sole´, X. Urbaneja and J. Bonjoch, Org. Lett., 2005, 7, 5461–5464
and references therein.
This journal is
The Royal Society of Chemistry 2009
Org. Biomol. Chem., 2009, 7, 2517–2519 | 2519
©