Organocatalyzed asymmetric synthesis of morphans

Article abstract of DOI:10.1021/ol400926p

A general effective organocatalyzed synthesis of enantioenriched morphans with up to 92% ee was developed. The morphan scaffold was constructed in a one-pot tandem asymmetric organocatalyzed Michael addition followed by a domino Robinson annulation/aza-Michael intramolecular reaction sequence from easily available starting materials.

Full text of DOI:10.1021/ol400926p

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Organocatalyzed Asymmetric Synthesis
of Morphans
Ben Bradshaw,* Claudio Parra, and Josep Bonjoch*
´
Laboratori de Quımica Organica, Facultat de Farmacia, IBUB,
Universitat de Barcelona, Av. Joan XXIII, s/n, 08028, Barcelona, Spain
ꢀ
ꢁ
benbradshaw@ub.edu; josep.bonjoch@ub.edu
Received April 4, 2013
ABSTRACT
A general effective organocatalyzed synthesis of enantioenriched morphans with up to 92% ee was developed. The morphan scaffold was
constructed in a one-pot tandem asymmetric organocatalyzed Michael addition followed by a domino Robinson annulation/aza-Michael
intramolecular reaction sequence from easily available starting materials.
The morphan ring system (2-azabicyclo[3.3.1]nonane)
features prominently in many complex and biologically
active natural products as well as medicinal compounds of
significant interest, including Strychnos,1 madangamines,2
some Daphniphyllum alkaloids3 and the immunosuppres-
sant FR9014834 (Figure 1).
strategy to this nucleus, based on a tandem Michael
addition/domino Robinson annulation and aza-Michael
intramolecular reaction.
Despite the plethora of methods that have been devel-
opedfor the formation of the morphan nucleus, to date5 no
generalcatalyticasymmetric strategy to polyfunctionalized
enantiopure morphan ring systems has been developed.6,7
Here, we outline an effective asymmetric organocatalyzed
(1) (a) Cannon, J. S.; Overman, L. E. Angew. Chem., Int. Ed. 2012, 51,
ꢀ
4288–4311. (b) Sole, D.; Bonjoch, J. Chem. Rev. 2000, 100, 3455–3482.
ꢀ
(2) (a) Amat, M.; Ballette, R.; Proto, S.; Perez, M..; Bosch, J. Chem.
Commun. 2013, 49, 3149–3151. (b) Quirante, J.; Paloma, L.; Diaba, F.;
Vila, X.; Bonjoch, J. J. Org. Chem. 2008, 73, 768–771.
(3) Yao, Y.; Liang, G. Org. Lett. 2012, 14, 5499–5501.
(4) (a) Huo, H.-H.; Xia, X.-E.; Zhang, H.-K.; Huang, P.-Q. J. Org.
Chem. 2013, 78, 455–465. (b) Bonjoch, J.; Diaba, F. Stud. Nat. Prod.
Chem. 2005, 32, 3–60.
(5) For a review of approaches to morphan-containing natural
products, see: Bonjoch, J.; Diaba, F.; Bradshaw, B. Synthesis 2011, 7,
993–1018. For a recently reported approach, see: Garrido, N. M.; Nieto,
C. T.; Dıez, D. Synlett 2013, 24, 169–172.
(6) For an organocatalyzed enantioselective method using a desym-
metrization reaction, see: Diaba, F.; Bonjoch, J. Org. Biomol. Chem.
2009, 7, 2517–2519.
Figure 1. Natural products containing the morphan (2-azabicyclo-
[3.3.1]nonane) nucleus.
(7) For synthesis of enantiopure morphans using a chiral auxiliary,
€
Bats, J. W.; Eschenmoser, A.; Quinkert, G. Helv. Chim. Acta 2000, 83,
1049–1078. (b) Quirante, J.; Diaba, F.; Vila, X.; Bonjoch, J.; Lago, E.;
Recently, we reported the synthesis of enantiopure
decahydroquinolines via an organocatalyzed Robinson
annulation/aza-Michael reaction, which constructed the
ring system in one pot, generating three stereogenic centers
see: (a) G. Karig, G.; Fuchs, A.; Busing, A.; Brandstetter, T.; Scherer, S.;
ꢀ
Molins, E. C. R. Acad. Sci. 2001, 4, 513–521. (c) Amat, M.; Perez, M.;
Minaglia, A. T.; Bosch, J. J. Org. Chem. 2008, 73, 6920–6923.
r
10.1021/ol400926p
XXXX American Chemical Society

the resultant alkene 2 with the HoveydaÀGrubbs second
generation catalyst and crotonaldehyde11 (Scheme 2).
With enal 1 in hand, we began our studies by investigat-
ing the coupling reaction in the nonasymmetric form.
Upon treatment of an equimolar mixture of 1 and the keto
Scheme 1. Synthesis of Nitrogen-Containing Heterocycles by
Forming Three Bonds in a One-Pot Reaction
ester 3a with LiOH H2O in iPrOH/H2O, we were pleased
3
to observe the formation of the desired morphan 4
(Scheme 3). A second, more polar compound 5 was also
isolated, which was identified as the same cyclized product
in its keto tautomeric form. Surprisingly, it was possible to
separate these two compounds by column chromatogra-
phy, and once isolated they did not undergo equilibration
in solution,12 allowing their NMR structures to be deter-
mined. In both cases, the allyl substituent was determined
to be equatorial.
Scheme 3. Evaluation of Feasibility of Robinson/Aza-Michael
Reaction in Racemic Form
in the process8 (Scheme 1a). We postulated that this
method could be adapted to access the morphan ring
system by moving the tethered nucleophilic N-Tosyl group
from the keto ester to the enal component (Scheme 1b).
However, it was not clear at the outset that the direct
application of the original method would be feasible, since
the resulting enal 1 would now bear both nucleo- and
electrophilic centers. Indeed, similar compounds separated
by additional carbons have been used in organocatalyzed
intramolecular aza-Michael reactions.9 We hoped that in
our case the formation of the azetidine ring by intramole-
cular cyclization would be sufficiently disfavored to allow
the intermolecular Michael reaction.
Access to the axially positioned substituent product 6
wasachieved by refluxing themixture of 4 and 5 withKFin
t-BuOH for 3 days.13 To rationalize the observed stereo-
chemistry of the allyl substituent, it was presumed that
upon attack of the N-Tosyl group, the resulting enolate
protonates from the less hindered top face to give 4 and 5,
with the allyl in the equatorial position (kinetic products).
The resulting steric compression suffered by the allyl sub-
stituent with the tosyl group14 was minimized in the kinetic
isomer 4 by nitrogen inversion. The axial orientation of the
N-tosyl group relieved the steric crowding with the equa-
torial side chain at C-8. This conformational change was
deduced from a shielding at C-4 observed when comparing
the 13C NMR spectra of 4 and the thermodynamic isomer
6. In the latter, the allyl group axially located at C-8 allowed
the N-tosyl to adopt an equatorial disposition (Figure 2).
Scheme 2. Synthesis of Enal 1
The required enal 1 was synthesized in a straightforward
manner from 3-buten-1-ol via Mitsonobu coupling
with tert-butyl tosylcarbamate10 and removal of the Boc
group withTFA, followedbya crossmetathesisreactionof
(11) For a related procedure, see: Chen, J.-R.; Li, C.-F.; An, X.-L.;
Zhang, J.-J.; Zhu, X.-Y.; Xaiao, W.-J. Angew. Chem., Int. Ed. 2008, 47,
2489–2492.
(8) Bradshaw, B.; Luque-Corredera, C.; Bonjoch, J. Org. Lett. 2013,
15, 326–329.
(9) For the use of this type of compound for the formation of
(12) It should be noted, however, that upon prolonged standing the
pure compounds would revert to mixtures of the enol/keto forms.
(13) For thermodynamic isomerization of R-substituted cycloalka-
nones using KF, see: Bradshaw, B.; Etxebarria-Jardı, G.; Bonjoch, J.
J. Am. Chem. Soc. 2010, 132, 5966–5967.
ꢀ
ꢀ
piperidines via organocatalysis, see: Fustero, S.; Jimenez, D.; Moscardo,
ꢀ
J.; Catalan, S.; del Pozo, C. Org. Lett. 2007, 9, 5283–5286. For a general
review on organocatalytic asymmetric aza-Michael reactions, see: Enders,
D.; Wang, C.; Liebich, J. X. Chem.;Eur. J. 2009, 15, 11058–11076.
(10) Teichert, J. F.; Zhang, S.; Zijl, A. W. v.; Slaa, J. W.; Minnaard,
A. J.; Feringa, B. L. Org. Lett. 2010, 12, 4658–4660.
(14) For a related 1,3-syn interaction within the N-substituent and
alkyl side chain in C-8 in morphan compounds, see: Bonjoch, J.;
Casamitjana, N.; Quirante, J.; Torrens, A.; Paniello, A.; Bosch, J.
Tetrahedron 1987, 43, 377–381.
B
Org. Lett., Vol. XX, No. XX, XXXX

intramolecular reaction, protection of the free NH of 1 as
an acetamide19 was investigated. Unfortunately, only trace
quantities of the coupled product were observed,20 indicat-
ing that the Michael reaction is generally unfavorable. By
carrying out the coupling reaction of 1 and 3 in the absence
of any solvent, the reaction was noticeably accelerated and
we were able to obtain moderate quantities of the coupled
product 7. Treatment of the crude mixture with LiOH
in iPrOH/H2O led to the cyclized product in 79% ee and
moderate 38% overall yield for the two steps (Table 1,
entry 1). Again, the product was isolated as a mixture of
enol/keto forms (∼3:1 ratio; only the enol form is shown
for clarity). To improve the reaction, a series of additives
(BzOH, TBAB,21 NaHCO3, LiOAc) were investigated,
with significantly improved yields being obtained with
the LiOAc (Table 1, entry 3). The use of water as an
additive (10 equiv) led to both an increase in yield and ee
(Table 1, entry 7). Carrying out the reaction at reduced
temperature, however, proved to be detrimental (Table 1,
entry 9). Switching to the more impeded catalyst 98,22 gave
Figure 2. Stereochemical assignment of compounds 4 and 6.
Having proof of concept, we then turned to investigate
the reaction in asymmetric form. Unfortunately, upon
treatment of keto ester 3a with enal 1 in the presence of
the HayashiÀJorgensen catalyst 815 in a range of organic
solvents,16 we observed only trace amounts of the Michael
addition product 7, which existed exclusively in its hemi-
aminal form (Scheme 4). While the keto ester 3a was stable
under the reaction conditions, enal 1 was slowly consumed
into an unidentified product. We believe that the side
reaction is a result of an isomerization of the double bond
of A via the dienamine B17 to form the cis isomer C, which
Table 1. Screening of Conditionsa for Organocatalyzed Synthesis
of Morphans
Scheme 4. Organocatalyzed Michael Reaction of 1 and 3a and
Loss of Enal 1 via a Parasitic Catalytic Cycle
additive
(equiv)
yield
(%)b
ee
entry
R
temp
(%)
1
À
TMS
TMS
TMS
TMS
TMS
TMS
TMS
TMS
TMS
SiPh3
SiPh3
SiPh3
SiPh3
SiPh3
rt
38
34
61
37
43
28
69
41
52
67
66
62
52
55
79
51
79
81
75
83
88
80
73
88
86
64
90
92
2
BzOH (0.1)
LiOAc (0.1)
NaHCO3 (0.1)
TBAB (0.1)
H2O (0.1)
rt
3
rt
4
rt
5
rt
6
rt
7
H2O (10)
rt
then cyclizes to form a six-membered ring, followed by
subsequent oxidation.18 It should be noted that this side
reaction did not occur in the absence of 3a, indicating that
the keto ester must somehow facilitate the isomerization
by proton transfer processes. To prevent the undesired
8c
9
H2O (10)
rt
H2O (10)
0 °C
rt
10
11
12
13
14d
H2O (10)
LiOAc (0.1)
LiOAc (0.1)
LiOAc(0.1)/H2O(10)
LiOAc(0.1)/H2O(10)
rt
0 °C
rt
rt
(15) (a) Hayashi, Y.; Gotoh, H.; Hayashi, T.; Shoji, M. Angew.
Chem., Int. Ed. 2005, 44, 4212–4215. (b) Marigo, M.; Wabnitz, T. C.;
Fielenbach, D.; Jorgensen, K. A. Angew. Chem., Int. Ed. 2005, 44, 794–
797.
(16) Solvents screened were toluene, CH2Cl2, and MeOH; all gave
equally poor results so a detailed solvent screen was not undertaken.
(17) For isomerization of enals to dienamines via organocatalysis,
a Conditions: 2 equiv of keto ester 3a, 1 equiv of enal 1, 10% catalyst,
3 h then add iPrOH (4 mL/mmol of enal), LiOH H2O (3 equiv), H2O
3
(10 equiv), 16 h. b Yield is given for enol and keto forms combined.
c 1:1 Ratio of keto ester and enal used. d Reaction time 24 h for the initial step.
ꢀ
see: Bertelsen, S.; Marigo, M.; Brandes, S.; Diner, P.; Jorgensen, K. A.
J. Am. Chem. Soc. 2006, 128, 12973–12980.
(18) For related intramolecular cyclization of N-Ts derivatives to
form pyridines, see: Donohoe, T. J.; Bower, J. F.; Baker, D. B.; Basutto,
J. A.; Chan, L. K. M.; Gallagher, P. Chem. Commun. 2011, 47, 10611–
10613.
(20) Presumably the formation of the hemiaminal (e.g., 7) is impor-
tant since it helps drive the reaction forward by removing the formed
aldehyde from competition with the organocatalyst, as well as prevent-
ing a possible retro-Michael reaction.
(19) Since N-Ts acetamides can be cleaved under mildly basic con-
ditions, it was planned to liberate the NH-Ts group under the cyclization
conditions, thus eliminating the need for a separate deprotection step.
(21) TBAB refers to tetrabutylammonium bromide. For its use, see:
Duce, S.; Mateo, A.; Alonso, I.; Garcia Ruano, J. L.; Cid, M. B. Chem.
Commun. 2012, 48, 5184–5186.
Org. Lett., Vol. XX, No. XX, XXXX
C

Scheme 5. Synthesis of C-8 Substituted Morphansa
Scheme 6. Synthesis of Indolomorphan 18
To explore the scope of the reaction, a number of varied
keto ester substrates (3bÀh)24 were used in the coupling
reaction. As can be seen in Scheme 5, the reaction works
with a wide range of substrates, including both aliphatic
and aromatic substituents, with some compounds being
isolated predominantly in the enol form. Although in
some cases the yields were moderate, we believe this
reflects difficulties in mixing the reagents under solvent-
free conditions.
To further expand the potential of the reaction, we
proposed it should be possible to add a third tandem
reaction using the C-8 substituent in an additional ring-
forming step. The 2-nitrophenyl keto ester 3i was coupled
with enal 1 under modified conditions25 to give morphan
17 possessing a latent indole moiety (Scheme 6). Treatment
with Zn, NH4Cl(aq)26 gave indole 18, which contains
the core structure of curan-type alkaloids,27 showing the
potential of this methodology to rapidly access complex
molecular scaffolds in a simple, straightforward manner.
In conclusion, the first effective catalytic route to poly-
functionalized enantiopure morphans has been developed
using a one-pot organocatalyzed Robinson/aza-Michael
reaction. The convergency and potential flexibility of
this method should open up new effective disconnection
approaches for the total synthesis of morphan-containing
natural products. Work in this direction is currently in
progress.
a Conditions: 2 equiv of ketoester 3, 1 equiv of enal, 10% catalyst, 24 h
then add iPrOH (4 mL/mmol of enal), LiOH H2O (3 equiv), H2O
3
b
(10 equiv), 24 h. Step ii required 48 h to reach completion. Forms
c
d
predominantly or exclusively as the enolic form. Forms as a mixture
of enol/keto forms; yields are given for both forms combined. For the
tautomer ratio, see Supporting Information.
similar results to catalyst 8 when water was used as the
additive (Table 1, entry 10), but it proved to be superior
when LiOAc was used (in Table 1, compare entry 3 vs 11).
Again, reducing the temperature resulted in significantly
inferior results (entry 12). The use of both additives
together, as opposed to individually, proved to be superior
(Table 1, entry 14), with the enantioselectivity increasing
to 92%.23
(22) Gomez-Bengoa, E.; Landa, A.; Lizarraga, A.; Mielgo, A.;
Oiarbide, M.; Palomo, C. Chem. Sci. 2011, 2, 353–357.
(23) The absolute configuration proposed for morphan 4 is based on
the accepted mechanism of organocatalyzed Michael addition of β-keto
esters upon enals: Jensen, K. L.; G. Dickmeiss, G.; Jiang, H.; Albrecht,
L.; Jørgensen, K. A. Acc. Chem. Res. 2012, 45, 248–264.
(24) Keto esters 3 were prepared by coupling the correpsponding
carboxylic acid with Meldrum’s acid using DCC, followed by refluxing
in t-BuOH. For an early use of this methdology, see: Li, B.; Franck,
R. W: Bioorg. Med. Chem. Lett. 1999, 9, 2629–2634.
Acknowledgment. Support of this research was provided
by the Spanish Ministry of Economy and Competitiveness
(Project CTQ2010-14846/BQU). C.P. is a recipient of a
predoctoral fellowship (CONICYT, Chile).
(25) Due to the highly crystalline nature of 17 and the coupled
product, mixing of the reagents proved difficult and the conditions of
the initial organocatalytic step had to be modified. Heating was used to
melt the substrates and the catalyst was switched to catalyst 8, which,
unlike catalyst 9, exists as an oil.
(26) Bradshaw, B.; Etxebarria-Jardı, G.; Bonjoch, J. Org. Biomol.
Chem. 2008, 6, 772–778.
Supporting Information Available. Experimental proce-
dures, spectroscopic and analytical data, and copies of NMR
spectra of the new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
ꢀ
(27) Bonjoch, J.; Sole, D.; Garcıa-Rubio, S.; Bosch, J. J. Am. Chem.
Soc. 1997, 119, 7230–7240.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX