Synthesis of a racemic nicotine-lobeline hybrid

Article abstract of DOI:10.1055/s-0029-1219966

The first synthesis of a racemic nicotine-lobeline hybrid is described based on a diastereoselective allylation and an RCM reaction as key steps. This synthetic route paves the way to the preparation of original potential ligands of nAChRs. Georg Thieme Verlag Stuttgart New York.

Full text of DOI:10.1055/s-0029-1219966

LETTER
1631
Synthesis of a Racemic Nicotine–Lobeline Hybrid
Synthesis of
a
R
s
a
cemic
N
i
icotine
r
–Lobelin
e
e
H
ybrid sy Ranaivondrambola,a Wilfried Hatton,a François-Xavier Felpin,a,1 Michel Evain,b
Monique Mathé-Allainmat,*a Jacques Lebreton*a
a
Université de Nantes, CNRS, Laboratoire CEISAM–UMR 6230, Faculté des Sciences et des Techniques,
2 Rue de la Houssinière, BP 92208, 44322 Nantes Cedex 3, France
Fax +33(2)51125562; E-mail: monique.mathe@univ-nantes.fr; E-mail: jacques.lebreton@univ-nantes.fr
Université de Nantes, CNRS, Institut des Matériaux Jean Rouxel UMR 6502,
b
2 Rue de la Houssinière, BP 32229, 44322 Nantes Cedex 3, France
Received 22 March 2010
Dedicated to the memory of Dr. Christian Marazano
tion structure of this family of receptors providing a great
opportunity to better understand the structure–activity re-
lationship of various nAChR agonists and therefore to
identify the structural factors that may contribute to im-
proving subtype selectivity.9
Abstract: The first synthesis of a racemic nicotine–lobeline hybrid
is described based on a diastereoselective allylation and an RCM
reaction as key steps. This synthetic route paves the way to the prep-
aration of original potential ligands of nAChRs.
Key words: alkaloids, ring-closing metathesis, total synthesis, neu-
ronal nicotinic acetylcholine receptors
N
Cl
H
N
N
Neuronal nicotinic acetylcholine receptors (nAChRs)
play important roles in the central nervous system (CNS)
and have multiple physiological functions.2 This family of
ligand-gated ion channels is built up from pentameric as-
semblies of a- and b-protein subunits. To date, nine a-sub-
units (a2–a10) and three b-subunits (b2–b4) have been
characterized in mammals, thus affording multiple popu-
lations of hetero- and homomeric receptor subtypes.3 For
many years, these nAChRs have been recognized as cru-
cial therapeutic targets for the treatment of several neuro-
logical and psychiatric disorders including Parkinson’s
and Alzheimer’s diseases, schizophrenia, Tourette’s syn-
drome, certain epilepsies, chronic pain, and nicotine ad-
diction.4 Various natural alkaloids, including (–)-nicotine
(1), (–)-epibatidine (2), (–)-anatoxin-a (3), and (–)-lo-
beline (4, see Figure 1), have been identified as potent but
nonselective agonists.5 To improve the pharmacological
properties of these natural compounds, a huge effort in
fundamental research as well as in drug development,
both in academia and in the pharmaceutical industry, has
been conducted for many years.6 Due to the diversity of
nAChRs, numerous analogues of this set of alkaloids have
been prepared and evaluated to find potent and subtype-
specific ligands with or without substantially reduced side
effects. Several analogues of these natural products ap-
pear promising and are in various stages of preclinical and
clinical evaluation although others have been abandoned.
In 2001, acetylcholine-binding protein (AChBP), a novel
homopentameric protein with high similarity to the extra-
cellular ligand-binding domain of the nAChR, was isolat-
ed from the freshwater snail Lymnaea stagnalis.7 The
crystal structure of AChBP8 offered the first high-resolu-
Me
N
(–)-nicotine (1)
(–)-epibatidine (2)
H
OH
O
N
O
N
Me
(–)-anatoxin-a (3)
(–)-lobeline (4)
Figure 1 Structures of (–)-nicotine (1), (–)-epibatidine (2), (–)-ana-
toxin-a (3), and (–)-lobeline (4)
In addition, combining the key structural features of these
alkaloids, a hybrid of (–)-anatoxin-a (3) and (–)-epibati-
dine (2), named UB-165 (5, see Figure 2), in which the
acetyl group of the first is replaced by a 5-(2-chloropy-
ridyl) moiety of the second, has been studied and used as
a lead compound for drug candidates.10 A few years ago,
we published the total synthesis of (–)-nicotine (1)11 and
(–)-lobeline (4).12 However, to our surprise and to the best
of our knowledge, no attempt to prepare a hybrid of these
two important alkaloids has since been reported in the lit-
erature. In order to find new ligands of nAChR with high-
er affinity and better subtype selectivity than their parent
alkaloids, we wish to describe in this paper the first syn-
thesis of nicotine–lobeline hybrid molecules 6 and 7.
Our synthetic strategy towards the nicotine–lobeline hy-
brid molecules 6 and 7 is outlined in Scheme 1. In consid-
ering a synthetic route to our target hybrids, we were
guided by previous investigations in our group on the di-
astereoselective allylation of imines such as intermediate
10.13 This latter compound, bearing an imine function and
a styryl moiety, offers the opportunity to add an allyl or-
ganometallic reagent to access to diene 11 with a stereo-
SYNLETT 2010, No. 11, pp 1631–1634
x
x
.x
x
.2
0
1
0
Advanced online publication: 10.06.2010
DOI: 10.1055/s-0029-1219966; Art ID: G09210ST
© Georg Thieme Verlag Stuttgart · New York

1632
T. Ranaivondrambola et al.
LETTER
The synthesis of the first intermediate ( )-9 commenced
with the BF3·OEt2-mediated Mukaiyama–aldol reaction
between the silylenolether 814 and benzaldehyde to fur-
nish the adduct 12 which was then protected as silyl ether
to give the intermediate 13 in 83% overall yield, as out-
lined in Scheme 2. The reduction of the racemic protected
aldol 13 with lithium aluminium hydride in THF–diethyl
ether at low temperature took place with a good diastereo-
selectivity (up to 9:1), affording the desired monoprotect-
ed syn-diol 14 as the major product in 91% yield on a 10
g scale. The observed diastereoselectivity was in agree-
ment with a six-membered chelate model and, according
to this model, the hydride was delivered by the less hin-
dered face opposite the phenyl ring. Attempts to increase
the diastereoselectivity by lowering the temperature from
–100 to –120 °C resulted in no significant improvement.
The reduction with zinc borohydride was found to be less
efficient: in the best conditions, an 8:2 ratio was obtained.
The relative configuration of the aldol adducts 14 was un-
ambiguously ascertained by removal of the TBS protect-
ing group and acetonide formation. Using the empirical
Rychnovsky’s acetonide 13C NMR spectra analysis15 in
the major diastereomer 14, the signals of the methyl
groups of the isopropylidene moiety occurred at d = 30.4
and 20.3 ppm, which indicated that the six-membered ac-
etonide ring was present in a chair form, in accordance
with a syn-l,3-diol system. In the minor diastereomer,
however, the anti-1,3-diol acetonide adopted a twist-boat
conformation, having both signals of the methyl groups at
d = 25.3 ppm.
N
Cl
OH
H
N
N
Ph
Me
N
(–)-UB-165 (5)
( )-6
O
N
Ph
Me
N
( )-7
Figure 2 Structures of (–)-UB-165 (5), racemic nicotine–lobeline
hybrid molecules 6 and 7
control of the protected alcohol function followed by an
RCM reaction to form the piperidine ring. Retrosynthetic
analysis for the first intermediate 9 suggested that the
Mukaiyama aldol reaction on the starting material 8 with
benzaldehyde followed by diastereoselective ketone re-
duction and further functional-group transformation
would give the desired amine intermediate 9. It was fur-
ther expected that condensation of this amine with nico-
tinaldehyde would lead to the formation of the
corresponding imine 10. It should be pointed out that
these two latter transformations in this strategy, in both
cases using an aldehyde, would be a way to implant diver-
sity on aromatic or/and heteroaromatic rings to prepare
analogues.
The resulting mixture of diol 14 was further converted
into the corresponding azide 15 in 86% yield with an inte-
gral switch to anti stereochemistry upon treatment with
diphenylphosphoryl azide in the presence of DBU.16 Sub-
jecting a mixture of the anti- and syn-azides 15 to Zn and
NH4Cl in EtOH–H2O (3:1) at reflux resulted in a clean re-
duction to yield, after purification on silica gel, the desired
pure anti-amino alcohol 9 in 66% yield (the minor isomer
was isolated in 12% yield).17 It should be noted that, in our
initial experiment, we performed this reduction by treat-
ment with SnCl2.11a However, with substrate 15 the work-
up was found to be rather tricky due to the formation of an
emulsion. Applying a one-pot two-step sequence that has
been successfully developed in our group,13 the amine
anti-9 was treated with the nicotinaldehyde in the pres-
ence of anhydrous MgSO4 in diethyl ether to form in situ
the corresponding imine 10. To this latter ethereal solu-
tion at –78 °C was added an excess of allyl Grignard re-
agent to provide an 85:15 mixture of dienes 11 and 16 in
85% yield. After chromatographic purification on silica
gel, the desired diene 11 was obtained in 57% overall
yield. Two arguments are worth mentioning to support the
unambiguous assignment of the relative configuration of
the major diastereomer 11. First, this latter transformation
has also been performed with other aromatic aldehydes
such as benzaldehyde leading to similar results, in terms
of yield and diastereoselectivity. Moreover, after an RCM
reaction with one of these dienes (vide infra), we were
CHO
OTBS
CHO
H2N
TMSO
N
( )-9
8
OTBS
N
N
( )-10
M
OTBS
RCM
O
N
H
N
N
( )-11
Me
N
( )-7
Scheme 1 Retrosynthetic plan for the preparation of racemic nico-
tine–lobeline hybrid 7
Synlett 2010, No. 11, 1631–1634 © Thieme Stuttgart · New York

LETTER
Synthesis of a Racemic Nicotine–Lobeline Hybrid
1633
O
RO
TMSO
OH OTBS
OH OTBS
a
c
Ph
Ph
+
Ph
Ph
e
Ph
Ph
Ph
(
)-anti-14
8
(
(
)-12 R = H
( )-syn-14
b
)-13 R = TBS
N3 OTBS
N3 OTBS
NH2 OTBS
d
f
+
Ph
Ph
Ph
Ph
Ph
Ph
(
)-syn-15
(
)-anti-15
(
)-anti-9
Ph
Ph
Ph
OR
Ph
OH
Ph
OTBS
Ph
OTBS
Ph
h
+
Pyr
N
H
Pyr
N
Pyr
N
H
Pyr
N
H
(
)-18
(
)-10
(
)-16
(
)-11 R = TBS
)-17 R = H
g
(
O
OH
j
OH
Ph
k
i
Pyr
N
Me
Ph
Pyr
N
Pyr
N
Me
Ph
Pyr
N
H
O
O
Ph
(
)-6
(
)-19
( )-7
(
)-20
Scheme 2 Reagents and conditions: (a) PhCHO, BF3·OEt2, CH2Cl2, –78 °C, 1 h (91%); (b) TBSCl, Et3N, DMAP, DMF, r.t., 8 h (92%); (c)
LiAlH4, Et2O–THF (9:1), –100 °C, 4 h (91%, ratio syn-14/anti-14 = 9:1); (d) DPPA, DBU, PhMe, r.t., 20 h (86%, ratio anti-15/syn-15 = 9:1);
(e) Zn, NH4Cl, EtOH–H2O (3:1), reflux, 2 h (for anti-9 66% and for syn-diastereomer 12%); (f) 3-Pyr-CHO, MgSO4, Et2O, r.t., 7 h, then allyl-
MgBr, –78 °C to r.t., 4 h (85%, ratio 11/16 = 85:15); (g) TBAF, 0 °C to r.t., 7 h (82%); (h) Grubbs II, PTSA, CH2Cl2–t-BuOH (20:1), reflux,
24 h (67%); (i) H2, Pd/C, EtOH, r.t., 48 h (83%); (j) NaH, CDI, THF, reflux, 3 h then LAH in excess, r.t., 12 h (82%); (k) ClCO–COCl, DMSO,
CH2Cl2, –78 °C, then ( )-6, 15 min, Et3N, –78 °C to r.t. (95%).
able to confirm the stereochemistry based on an X-ray the latter intermediate 17 as its ditosylate salt occurred
structure analysis (see Supporting Information). Further with the use of Grubbs second-generation catalyst in a
proof concerning the relative configuration of the major mixture of dichloromethane and tert-BuOH (20:1) to fur-
diastereomer 11 came after the RCM reaction step, as dis- nish the cyclized adduct 18 in 67% isolated yield. It is
cussed below. Contrary to our previous work,13 the forma- noteworthy that the use of t-BuOH as co-solvent proved to
tion of the diastereomer 11 could be rationalized by the be essential for the good solubility of the ditosylate salt
lack of internal chelation between the alcohol function, and therefore the success of this transformation. More-
protected in this substrate 10 as silylether, and the nitro- over, an NOE experiment on the RCM reaction product 18
gen atom of the imino function. Indeed, with a free-hy- revealed a cis configuration between the two hydrogen at-
droxy substrate, an internal chelation of the magnesium oms on C-2 and C-6 of the piperidine ring, allowing the
atom of the Grignard reagent with the corresponding alco- unambiguous assignment of the relative configuration of
holate and the imino nitrogen should form a six-mem- the diastereomer 11. As previously used by us11 in the to-
bered ring. This would favor an attack of the allyl reagent tal synthesis of (S)-N-methylanabasine, we attempted to
by the less hindered face to afford the adduct with the treat intermediate 18 under hydrogen in the presence of
same relative configuration as in the minor diastereomer palladium on carbon in a mixture of ethanol and formal-
16. Nevertheless, the origin of this diastereoselectivity dehyde to afford 6 directly. Under these modified
with compound 10 still remains unclear. Thus, some com- Eschweiler–Clarke conditions, the double bond of 18 was
plementary experiments are under way in order to eluci- hydrogenated, and the secondary amine was converted by
date the mechanism. As expected, this pyridinic substrate reaction with formaldehyde to the corresponding iminium
11 turned out to be problematic for the RCM reaction.18 which, instead of being reduced to the N-methylated de-
Attempts to perform the RCM reaction with Grubbs sec- rivative, reacted in an intramolecular fashion with the free
ond-generation catalyst on the corresponding dihydro- secondary alcohol function to give the undesired six-
chloride salt of the substrate 11 failed to provide the membered cyclic N,O-acetal in good yield. Owing to
desired compound in decent yield, due mainly to its insol- these difficulties, we decided to rely on a nice piece of
ubility in refluxing dichloromethane. We also observed chemistry described by Bates and co-workers in the route
partial cleavage of the silylether which afforded a mixture to sedamine.19 Hydrogenation of the double bond in 18
of compounds. So, at this point of our synthesis, we decid- under classical conditions gave the compound 19 in 83%
ed to remove the silyl protecting group by classical treat- yield. The amino alcohol 19 was treated with N,N¢-carbo-
ment with TBAF; the alcohol 17 was then isolated in 82% nyldiimidazole and sodium hydride to furnish the cyclic
yield after purification. Fortunately, the RCM reaction on carbamate intermediate 20 which was then reacted with an
Synlett 2010, No. 11, 1631–1634 © Thieme Stuttgart · New York

1634
T. Ranaivondrambola et al.
LETTER
(10) (a) Sutherland, A.; Gallagher, T.; Sharples, C. G. V.;
excess of lithium aluminium hydride leading to the de-
sired N-methylated compound 620 in 82% overall yield af-
ter purification by silica gel chromatography. Finally, this
latter intermediate 6 was converted into the target hybrid
Wonnacott, S. J. Org. Chem. 2003, 68, 2475. (b) Gohlke,
H.; Schwarz, S.; Gündisch, D.; Tilotta, M. C.; Weber, A.;
Wedge, T.; Seitz, G. J. Med. Chem. 2003, 46, 2031..
(11) (a) Felpin, F.-X.; Girard, S.; Vo Thanh, G.; Robins, R. J.;
Villiéras, J.; Lebreton, J. J. Org. Chem. 2001, 66, 6305.
(b) or a review on (–)-nicotine, see: Wagner, F. F.; Comins,
D. L. Tetrahedron 2007, 63, 8065. (c) For recent work on
the synthesis of (S)-nicotine analogues, see: Huang, K.;
Merced, F. G.; Ortiz-Marciales, M.; Meléndez, H. J.; Correa,
W.; De Jesús, M. J. Org. Chem. 2008, 73, 4017.
(12) (a) Compère, D.; Marazano, C.; Das, B. C. J. Org. Chem.
1999, 64, 4528. (b) Felpin, F.-X.; Lebreton, J. J. Org. Chem.
2002, 67, 9192. (c) BirmanV, B.; Jiang, H.; Li, X. Org. Lett.
2007, 17, 3237. (d) Krishnan, S.; Bagdanoff, J. T.; Ebner, D.
C.; Ramtohul, Y. K.; Tambar, U. K.; Stoltz, B. M. J. Am.
Chem. Soc. 2008, 130, 13745. (e) Chênevert, R.; Morin, P.
Bioorg. Med. Chem. 2009, 17, 1837. (f) For a review on
lobeline, see: Felpin, F.-X.; Lebreton, J. Tetrahedron 2004,
60, 10127.
(13) (a) Felpin, F.-X.; Lebreton, J. Tetrahedron Lett. 2003, 44,
527. (b) Felpin, F.-X.; Boubekeur, K.; Lebreton, J. Eur. J.
Org. Chem. 2003, 4518. (c) Felpin, F.-X.; Boubekeur, K.;
Lebreton, J. J. Org. Chem. 2004, 69, 1497. (d) Pearson, M.
S. M.; Evain, M.; Mathé-Allainmat, M.; Lebreton, J. Eur. J.
Org. Chem. 2007, 4888.
(14) This starting material was prepared on a 25 gram scale and
used without further purification according to a known
procedure, see: Ramachary, D. B.; Narayana, V. V.;
Ramakumar, K. Eur. J. Org. Chem. 2008, 3907.
7
21 in 95% yield upon Swern oxidation.
In conclusion, we have described the first synthesis of nic-
otine–lobeline hybrid molecules ( )-6 and ( )-7. Further-
more, application of this synthetic strategy to the
preparation of different analogues with various substitu-
ents on both aromatic and pyridine rings is currently in
progress and will be published in due course with their
binding results on nAChRs.
Supporting Information for this article is available online at
http://www.thieme-connect.com/ejournals/toc/synlett. Included are
experimental procedures and spectroscopic data for compounds:
anti-9, syn-9, 11, 16–19, 6, and 7. Copies of NMR data of com-
pounds 6 and 7, and single-crystal X-ray.
Acknowledgment
The authors are deeply indebted to the ‘Ministère de l’Enseigne-
ment Supérieur et de la Recherche’ (MESR), the ‘Université de
Nantes’ and the ‘Centre National de la Recherche Scientifique’
(CNRS) for their constant support. This program is supported in
part by the MESR (grant for F.-X.F.). We are also grateful to the
‘Société de Chimie Thérapeutique’ (SCT) and ‘Les Laboratoires
Servier’ for doctoral grants to W.H. and T.R., respectively.
(15) Rychnovsky, S. D.; Rogers, B.; Yang, G. J. Org. Chem.
1993, 58, 3511.
(16) (a) Thompson, A. S.; Humphrey, G. R.; DeMarco, A. M.;
Mathre, D. J.; Grabowski, E. J. J. J. Org. Chem. 1993, 58,
5886. (b) Thompson, A. S.; Hartner, F. W.; Grabowski, E. J.
J. Org. Synth. 1998, 75, 31.
References and Notes
(1) Present address: Université de Bordeaux, Institut des
Sciences Moléculaires, UMR-CNRS-5255, 351 Cours de la
Libération, 33405 Talence, France.
(2) Arneric, S. P.; Holladay, M.; Williams, M. Biochem.
Pharmacol. 2007, 74, 1092.
(17) Lin, W.; Zhang, X.; He, Z.; Jin, Y.; Gong, L.; Mi, A. Synth.
Commun. 2002, 32, 3279.
(18) It is well known that the presence of basic nitrogen in the
substrate molecule could strongly affect the efficiency of the
catalyst interfering with metal–alkylidene complexes of the
catalytic cycle; for leading references, see: (a) Felpin,
F.-X.; Lebreton, J. Eur. J. Org. Chem. 2003, 3693.
(b) Deiters, A.; Martin, S. F. Chem. Rev. 2004, 104, 2199.
(19) Bates, R. W.; Nemeth, J. A.; Snell, R. H. Synthesis 2008,
1033.
(3) Grutter, T.; Changeux, J.-P. Trends Biochem. Sci. 2001, 26,
459.
(4) (a) Jensen, A. A.; Frølund, B.; Liljefors, T.; Krogsgaard-
Larsen, P. J. Med. Chem. 2005, 48, 4705. (b) Holladay, M.
W.; Dart, M. J.; Lynch, J. K. J. Med. Chem. 1997, 40, 4169.
(5) Breining, S. R. Curr. Top. Med. Chem. 2004, 4, 609.
(6) (a) Romanelli, M. N.; Gratteri, P.; Guandalini, L.; Martini,
E.; Bonaccini, C.; Gualtieri, F. ChemMedChem 2007, 2,
746. (b) For a recent reference on this topic, see: Grazioso,
G.; Pomè, D. Y.; Matera, C.; Frigerio, F.; Pucci, L.; Gotti,
C.; Dallanoce, C.; De Amici, M. Bioorg. Med. Chem. Lett.
2009, 19, 6353.
(7) Smit, A. B.; Syed, N. I.; Schaap, D.; van Minnen, J.;
Klumperman, J.; Kits, K. S.; Lodder, H.; van der Schors,
R. C.; van Elk, R.; Sorgedrager, B.; Brejc, K.; Sixma, T. K.;
Geraerts, W. P. M. Nature (London) 2001, 411, 261.
(8) (a) Brejc, K.; van Dijk, W. J.; Klaassen, R. V.; Schuurmans,
M.; van Der Oost, J.; Smit, A. B.; Sixma, T. K. Nature
(London) 2001, 411, 269. (b) Celie, P. H. N.; van Rossum-
Fikkert, S. E.; van Dijk, W. J.; Brejc, K.; Smit, A. B.; Sixma,
T. K. Neuron 2004, 41, 907.
(20) Selected Physico-Chemical Data for Compound 6
1H NMR (300 MHz, CDCl3): d = 8.39–8.46 (m, 2 H), 7.57–
7.64 (m, 1 H), 7.15–7.22 (m, 1 H), 7.13–7.34 (m, 5 H), 5.18
(dd, 1 H, J = 3.0, 11.5 Hz), 2.99 (dd, 1 H, J = 3.0, 11.2 Hz),
2.46–2.57 (m, 1 H), 2.26 (ddd, 1 H, J = 3.6, 11.2, 15.1 Hz),
2.16 (s, 3 H), 2.04 (dq, 1 H, J = 3.2, 12.5 Hz), 1.65–1.87 (m,
3 H), 1.46–1.58 (m, 2 H), 1.35–1.49 (m, 1 H) ppm. 13C NMR
(75 MHz, CDCl3): d = 149.0, 148.8, 145.1, 140.1, 134.4,
128.3 (2×), 127.0, 125.5 (2×), 124.0, 71.9, 68.4, 63.7, 41.4,
39.9, 36.2, 29.9, 24.2 ppm. HRMS (CI): m/z calcd for
C19H25N2O [M + H+]: 297.1961; found: 297. 1967.
(21) Selected Physico-Chemical Data for Compound 7
1H NMR (300 MHz, CDCl3): d = 8.58 (br s, 1 H), 8.52 (br d,
1 H, J = 4.6 Hz), 8.02 (dm, 2 H, J = 7.4 Hz), 7.66–7.82 (m,
1 H), 7.56–7.65 (m, 1 H), 7.46–7.56 (m, 2 H), 7.22–7.31 (m,
1 H), 3.42–3.57 (m, 1 H), 3.10–3.22 (m, 1 H), 2.84–3.07 (m,
2 H), 2.01 (s, 3 H), 1.70–1.88 (m, 3 H), 1.47–1.66 (m, 3 H)
ppm. 13C NMR (75 MHz, CDCl3): d = 199.2, 149.3, 148.5,
148.4, 137.2, 134.8, 133.2, 128.7 (2 C), 128.2 (2 C), 123.6,
68.2, 60.5, 44.6, 40.9, 36.1, 32.9, 24.4 ppm. HRMS (CI):
m/z calcd for C19H23N2O [M + H+]: 295.1805; found:
295.1810.
(9) For leading references on these efforts, see: (a) Wei, Z.-L.;
Xiao, Y.; Yuan, H.; Baydyuk, M.; Petukhov, P. A.;
Musachio, J. L.; Kellar, K. J.; Kozikowski, A. P. J. Med.
Chem. 2005, 48, 1721. (b) Yuan, H.; Petukhov, P. A.
Bioorg. Med. Chem. 2006, 14, 7936.
Synlett 2010, No. 11, 1631–1634 © Thieme Stuttgart · New York