A new synthesis of (R)-(-)-sumanirole (PNU-95666E)

Article abstract of DOI:10.1055/s-0029-1219942

A twelve-step synthesis of (R)-(-)-sumanirole hydrochloride, starting from quinoline, has been achieved. The key reaction features selective epoxidation of the C3-C4 double bond of a 1,2-dihydroquinoline bearing a chiral auxiliary at N-1.

Full text of DOI:10.1055/s-0029-1219942

LETTER
1473
A New Synthesis of (R)-(–)-Sumanirole (PNU-95666E)
New
Synthesis
u
(
R
)-(–)-Su
d
manirole ivine Jean-Gérard,a Frédéric Macé,a Hélène Dentel,a Anh Ngoc Ngo,a Sylvain Collet,*a André Guingant,*a
Michel Evainb
a
Université de Nantes, CNRS, Chimie et Interdisciplinarité: Synthèse, Analyse, Modélisation (CEISAM), UMR CNRS n° 6230,
2 Rue de la Houssinière, BP 92208, 44322 Nantes Cedex 3, France
Fax +33(2)51125402; E-mail: andre.guingant@univ-nantes.fr; E-mail: sylvain.collet@univ-nantes.fr
Institut des Matériaux Jean Rouxel, 2 Rue de la Houssinière, BP 92208, 44322 Nantes Cedex 03, France
b
Received 19 March 2010
Dedicated to Professor Jacques Lebreton on the occasion of his 50th birthday
cessible, via regioselective nitration and stereoselective
hydroxyl group substitution, from alcohol 5, which in turn
should be derived from 6 by regioselective opening of its
epoxide functionality. Finally, the synthesis of 6 could be
envisaged through an asymmetric epoxidation reaction of
chiral 1,2-dihydroquinoline 7, whose preparation should
be readily achieved from quinoline.
Abstract: A twelve-step synthesis of (R)-(–)-sumanirole hydro-
chloride, starting from quinoline, has been achieved. The key reac-
tion features selective epoxidation of the C3–C4 double bond of a
1,2-dihydroquinoline bearing a chiral auxiliary at N-1.
Key words: quinoline, reduction-acylation, asymmetric epoxida-
tion, (R)-(–)-sumanirole
Parkinson’s disease is a progressive neurodegenerative
disease characterized by deterioration of motor control.
The symptoms are caused by loss of cells in the brain that
secrete the neurotransmitter dopamine formed by decar-
boxylation of L-DOPA under the action of L-DOPA de-
carboxylase. Current treatment consists of administration
of L-DOPA which is converted to dopamine in situ. How-
ever, loss of L-DOPA effectiveness and apparition of side
effects are very common after a period of time. In order to
minimize these adverse effects, selective stimulation of
D2 receptors, which are believed to be responsible for the
restoration of motor function, is highly desirable. In con-
trast to currently available drugs [e.g., pergolide (1), cab-
ergoline (2)], which are not selective and show affinity for
other receptor subtypes, sumanirole (3, Figure 1) has been
shown to display high in vitro and in vivo selectivity for
the D2 receptor subtype.1
N
O
S
H
O
NMe2
N
H
N
H
H
N
H
HN
pergolide (1)
HN
cabergoline (2)
NHMe
N
HN
O
sumanirole (3)
Figure 1
Owing to these remarkable properties, sumanirole was in-
vestigated as a potential candidate for the treatment of
Parkinson disease but the phase III clinical development
was ceased in 2004 because the molecule failed to provide
sufficient distinction from currently available therapies.
Nevertheless, we became interested in the synthesis of
sumanirole2 because it represented for us a good model to
evaluate the feasibility of new methods aimed at preparing
variously substituted chiral N-substituted 1,2,3,4-tetrahy-
droquinolines in view of further applications in the field
of natural products.
NH2
NHMe
OH
N
N
N
NH2 CO2Me
HN
CO2Me
O
4
5
3
O
N
N
N
O
quinoline
CO2Me
O
N
6
O
Our retrosynthetic approach to sumanirole, as illustrated
in Scheme 1, relies on initial disconnection of 3 across the
NHCO bond of the imidazolone ring to give N-protected
1,2,3,4-tetrahydroquinoline 4. Compound 4 should be ac-
Ph
7
Scheme 1 Retrosynthetic analysis of sumanirole
Our synthetic sequence began with attachment of the (S)-
4-benzyloxazolidin-2-one-3-carbonyl and 1,2-dihydro-
quinoline fragments to obtain 7. This could be realized in
a one-pot operation following the method of Minter and
SYNLETT 2010, No. 10, pp 1473–1476
Advanced online publication: 19.05.2010
DOI: 10.1055/s-0029-1219942; Art ID: D07310ST
© Georg Thieme Verlag Stuttgart · New York
x
x
.x
x
.2
0
1
0

1474
L. Jean-Gérard et al.
LETTER
Cl
O
O
N
O
Ph
N
O
9
i
7
61%
O
N
N
N
O
Ph
Al
R
R
8
10
Scheme 2 Reagents and conditions: DIBAL-H, (1 equiv), CH2Cl2, 1 h, r.t.; then red solution of 8 added to a solution of 9 (0.5 equiv) in CH2Cl2
at 0 °C; then stirring for 5 h at r.t.
Stotter3 who reported the 1,2-reduction of quinoline and
O
O
trapping of the aminoalane intermediate 8 by use of a
large excess of methyl chloroformate. By analogy, cap-
i
N
N
ture of 8 with the chiral carbamoyl chloride 94 should lead
to the formation of 7 (Scheme 2). However, and for obvi-
ous reasons, we searched for conditions avoiding the use
of elaborated 9 in excess. The best results were obtained
through addition of two equivalents of intermediate 8 onto
one equivalent of 9 at 0 °C and subsequent stirring for five
hours at 20 °C. Following these conditions, chiral N-acyl-
1,2-dihydroquinoline 7 was isolated in 61% yield.5 It
should also be noted that these conditions suppress the
formation of isomeric 2,3-dihydroquinoline 10, a byprod-
uct frequently formed in such a reduction–acylation se-
quence.3
O
O
+
7
O
N
O
N
O
O
Ph
Ph
11
12
OH
O
i, ii
N
7
54%
O
N
O
Ph
13
Scheme 3 Reagents and conditions: i) MCPBA (1 equiv), NaHCO3
(1.6 equiv), CH2Cl2, 18 h, 20 °C; ii) H2 (8 bar), 10% Pd/C, CH2Cl2–
EtOAc (1:1), 18 h, 20 °C.
Having prepared chiral 1,2-dihydroquinoline 7 we were
now in a position to examine the crucial epoxidation step.
Although the chiral moiety of the molecule was apparent-
ly far remote from the electrophilic double bond, we were
pleased to observe that reaction of 7 with MCPBA was
significantly diastereoselective, leading to a mixture of
epoxides 11 and 12 in a 4:1 to 9:1 ratio. A pure sample of
the major epoxide could be obtained after chromatograph-
ic separation of the crude mixture and subsequent recrys-
tallization.6 Its relative and absolute stereochemistry was
demonstrated to be that shown in structure 11 by a single-
crystal X-ray analysis7 (Scheme 3 and Figure 2). As 11is
not very stable to silica gel, we found it to be experimen-
tally preferable to hydrogenate the crude mixture of ep-
oxides and effect the separation of the resulting alcohols.
Following this protocol,8 the required alcohol 13 was iso-
lated in 54% yield from 7 (Scheme 3).
Figure 2 X-ray structure of 11
Having 13 successfully in hand as a single isomer, we
judged preferable to postpone the OH → NHMe transfor-
mation to a later stage and to concentrate first on elabora-
tion of the annelated imidazolone ring. In this direction,
installation of a NH2 group at C8 was envisaged via reduc-
tion of a nitro-group precursor. Because C6 was the ex-
pected most electrophilic center, this latter had to be
protected first with a removable group. We thus examined
a bromination–nitration sequence9 as a means of introduc-
ing a nitro group cleanly at C8. With the preliminary work
having shown the incompatibility of the oxazolidin-2-one
moiety with this sequence of reaction, this motif was re-
moved by action of MeOH in the presence of samarium
triflate10 to give the N-carbamate-protected 1,2,3,4-tet-
rahydroquinoline 14. Bromination of 14 proceeded to give
its expected 6-bromo derivative 15. Subsequent regiose-
lective nitration of 15 was effected by action of NaNO3 in
TFA to give compound 16 in good yield (Scheme 4).
At this stage we returned to the installation of the amino
substituent at C3. Treatment of 16 with mesyl chloride af-
forded the corresponding mesylate 17 whose reaction
with sodium azide led to azido compound 18 with yields
ranging from 40–65%.11,12 Subsequent hydrogenation of
Synlett 2010, No. 10, 1473–1476 © Thieme Stuttgart · New York

LETTER
New Synthesis of (R)-(–)-Sumanirole
1475
OH
Br
OH
Br
OH
i
iii
ii
13
66%
84%
83%
N
N
N
O2N
O
OMe
O
OMe
O
OMe
14
15
16
Scheme 4 Reagents and conditions: i) Sm(OTf)3 (0.25 equiv), MeOH, 3 h, 80 °C; ii) Br2 (1 equiv), AcOH, NaOAc, 1 h, 20 °C; iii) NaNO3
(1 equiv), TFA, 30 min, 20 °C.
OMs
N3
NH2
Br
Br
i
ii
iii
16
40–65%
99%
92%
N
N
N
O2N
O2N
H2N
O
OMe
O
OMe
O
OMe
17
18
19
H
NHCHO
N
NH2
Me
iv
v
vi
64%
•HCl
98%
77%
N
N
N
HN
HN
HN
O
O
O
20
21
3⋅HCl
Scheme 5 Reagents and conditions: i) MsCl (2 equiv), Et3N (3 equiv), CH2Cl2, 45 min, 20 °C; ii) NaN3 (5 equiv), DMF, 80 °C, 1h; iii) H2,
Pd(OH)2, EtOH, 18 h, 20 °C; iv) KOt-Bu (5 equiv), THF, 3.5 h, 20 °C; v) AcOCHO, MeCN, 1 h, 20 °C; vi) BH3·SMe2 (2.3 equiv), THF, reflux,
3 h, then 2 N HCl in MeOH, reflux, 2 h.
18 in the presence of Pearlman’catalyst effected three References and Notes
chemical transformations, that is, hydrogenolysis of the
C–Br bond as well as reduction of the nitro and azido
Lipton, M. F.; Martin, I. J.; Mauragis, M. A.; Piercey, M. F.;
(1) (a) Heier, R. F.; Dolak, L. A.; Duncan, J. N.; Hyslop, D. K.;
groups, to afford compound 19 in almost quantitative
yield (Scheme 5). Final transformation of 19 into
sumanirole was achieved in three additional steps. Clo-
sure of the imidazolone ring, to give the tricyclic com-
pound 20 (nor-Me sumanirole), was realized in excellent
yield under the action of KOt-Bu in THF. Finally,
monomethylation of the amino group at C3 was accom-
plished via formation of a formamide intermediate13 21
whose reduction led to (–)-sumanirole hydrochloride
(3·HCl).14
Nichols, N. F.; Schreur, P. J. K. D.; Smith, M. W.; Moon,
M. W. J. Med. Chem. 1997, 40, 639. (b) McCall, R. B.;
Lookingland, K. L.; Bédard, P. J.; Huff, R. M. J. Pharmacol.
Exp. Ther. 2005, 314, 1248.
(2) For a review of the synthetic routes to sumanirole, see:
(a) Wuts, P. G. M. Curr. Opin. Drug Discovery Dev. 1999,
2, 557. For more recent syntheses, see: (b) Wuts, P. G. M.;
Gu, R. L.; Northuis, J. M.; Kwan, T. A.; Beck, D. M.; White,
M. J. Pure Appl. Chem. 2002, 74, 1359. (c) Gala, D.;
Dahanukar, V. H.; Eckert, J.; Lucas, B. S.; Schumacher,
D. P.; Zavialov, I. A. Org. Process Res. Dev. 2004, 8, 754.
(d) Jagdale, A. R.; Reddy, R. S.; Sudalai, A. Org. Lett. 2009,
11, 803.
In conclusion, we have achieved a twelve-step synthesis
of (R)-(–)-sumanirole hydrochloride 3·HCl from quino-
line. Our work also demonstrates that attachment of a
chiral 4-benzyloxazolidin-2-one-3-carbonyl moiety at N-
1 of 1,2-dihydroquinoline may control the selectivity of
the C3–C4 double bond epoxidation. Further manipula-
tions of the epoxide functionality should thus allow the
preparation of chiral diversely N-substituted 1,2,3,4-tet-
rahydroquinolines. Work is in progress to find a coherent
explanation of the efficiency of the chiral auxiliary in ori-
enting the sense of epoxidation. We are also currently ap-
plying the results of this study in the field of natural
product synthesis.
(3) Minter, D. E.; Stotter, P. L. J. Org. Chem. 1981, 46, 3965.
(4) Pirkle, W. H.; Simmons, K. A. J. Org. Chem. 1983, 48,
2520.
(5) Procedure for the Preparation of 7
To a solution of quinoline (1.6 mL, 13.6 mmol) in CH2Cl2
(20 mL), maintained at 0 °C under an argon atmosphere, was
added a 1 M solution of DIBAL-H in hexane (13.6 mL, 13.6
mmol). After stirring at 0 °C for 1 h, the red solution was
cannulated into a solution of carbamoyl chloride 9 (1.63 g,
6.8 mmol) in 2mL of CH2Cl2. The temperature was slowly
raised to 20 °C, and stirring was continued for 5 h at this
temperature. The reactive mixture was then cannulated into
150 mL of cold H2O. The resulting emulsion was stirred for
30 min then aq 6 N HCl was added until the aqueous phase
reached pH 4. After phase separation, the aqueous phase was
extracted with CH2Cl2 (4 × 60 mL). The combined organic
layers were dried over MgSO4, filtered, and concentrated in
vacuo. The crude product was purified by silica gel chroma-
tography (toluene–PE–Et2O = 2:1:1, Rf = 0.26) to afford 1,2-
dihydroquinoline 7 as a slightly yellow oil (1.39 g, 61%,
calculated based on carbamoyl chloride 9); [a]D20 +7.7 (c
0.5, CHCl3). 1H NMR (300 MHz, CDCl3): d = 7.12–7.40 (m,
9 H), 6.59 (dd, J = 9.6, 2.4 Hz, 1 H), 5.99–6.04 (m, 1 H),
Acknowledgment
We thank Adeline Lectard for preliminary experiments.
Synlett 2010, No. 10, 1473–1476 © Thieme Stuttgart · New York

1476
L. Jean-Gérard et al.
LETTER
4.73–4.78 (m, 1 H), 4.40 (dd, J = 17.0, 5.5 Hz, 1 H), 4.08–
4.31 (m, 3 H), 2.92 and 3.26 (ABX system, J = 13.5, 8.7, 3.3
Hz, 2 H). 13C NMR (75 MHz, CDCl3): d = 153.3, 152.4,
136.0, 134.9, 129.5 (2 C), 129.2, 129.0 (2 C), 128.5, 128.3,
127.5, 126.7, 126.4, 125.5, 122.9, 67.0, 56.5, 45.9, 38.1. IR
(KBr): 3060–3028, 3003, 2923–2853, 1722, 1685 cm–1. MS:
m/z (%) = 334 (<1) [M+], 130 (100), 91 (19). HRMS (EI): m/
z calcd for C20H18N2O3: 334.1317; found: 334.1313 [M]+.
(668 mg, 9%mol). The mixture was stirred for 18 h at 20 °C
under a hydrogen pressure of 8 bar then filtered through a
pad of Celite. Celite was rinsed with a 1:1 mixture of
CH2Cl2–EtOAc, then solvents were concentrated in vacuo.
The residue was purified by silica gel chromatography (PE–
EtOAc = 1:1, Rf = 0.39) to afford alcohol 13 as a white
powder (0.9 g, 54%); [a]D20 –191.0 (c 0.34, CHCl3). 1H
NMR (300 MHz, CDCl3): d = 7.00–7.40 (m, 9 H), 4.80 (m,
1 H), 4.28 (m, 1 H), 4.11 and 4.24 (ABX system, J = 8.4, 8.4,
7.5 Hz, 2 H), 3.38 and 4.41 (ABX system, J = 12.6, <1, <1
Hz, 2 H), 2.99 and 3.15 (ABX system, J = 18.0, 5.4, <1 Hz,
2 H), 2.85 and 3.60 (ABX system, J = 12.9, 10.2, 3.6 Hz, 2
H). 13C NMR (75 MHz, CDCl3): d = 154.1, 152.9, 138.0,
133.9, 129.4 (2 C), 128.1 (2 C), 128.0, 126.5, 125.2, 124.2,
120.3, 67.2, 62.9, 55.2, 50.6, 37.0, 34.7. IR (film): 3452,
3050, 2932, 1772, 1683 cm–1. MS: m/z (%) = 323 (9) [M+],
132 (38), 130 (38), 118 (20), 117 (16), 91 (100). HRMS (EI):
m/z calcd for C20H20N2O4: 352.1423; found: 352.1427 [M]+.
(9) Moon, M. W.; Morris, J. K.; Heier, R. F.; Chidester, C. G.;
Hoffmann, W. E.; Piercey, M. F.; Althaus, J. S.;
(6) Epoxide 11: recrystallization from CHCl3–PE (1:1); mp
204–203 °C; [a]D20 +54.2 (c 0.5, CHCl3).
(7) Crystal Structure Data for C20H18N2O4
Mw = 350.4, colorless block, 0.48 × 0.42 × 0.38 mm3,
orthorhombic, P212121, a = 9.7121 (9)Å, b = 12.6679 (16)
Å, c = 13.6030 (8) Å, V = 1673.6 (3) Å3, Z = 4, Dx = 1.390 g
cm–3, m = 0.10 mm–1. 33596 reflections were measured on a
Nonius-Kappa CCD diffractometer (graphite monochro-
mator, l = 0.71073 Å) up to a resolution of (sin q/l)max = 0.7
Å–1 at r.t.; 4820 reflections were unique (Rint = 0.038). The
structure was solved by direct methods15 and refined with
JANA2006 program16 against F2 for all reflections.
Nonhydrogen atoms were refined with aniso-tropic
displacement parameters. All H atoms were introduced in
geometrically optimized positions and refined with a riding
model, except for two H atoms (attached to C3 and C4)
which positions were refined under constrains. Altogether,
241 parameters were refined. R1/wR2 [I ≥ 2s(I)] = 0.0473/
0.1072. R1/wR2 [all reflections] = 0.0682/0.1189, S = 1.74.
Residual electron density is between 0.14 and –0.13 e Å–3.
CCDC 769777 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Von Voiglander, P. F.; Evans, D. L.; Figur, L. M.; Lahti,
R. A. J. Med. Chem. 1992, 35, 1076.
(10) Pauvert, M.; Collet, S.; Bertrand, M.-J.; Guingant, A.; Evain,
M. Tetrahedron Lett. 2005, 46, 2983.
(11) Azido compound 18 was isolated along with variable
amounts of an elimination product (1,2-dihydroquinoline).
Because mesylate 17 is stable in DMF at 80 °C, formation of
the elimination byproduct is thus induced by NaN3.
Noteworthy also is the formation of the sole elimination
product by treatment of alcohol 16 with DPPA under
Mitsunobu conditions.
(12) HPLC profiles showed that azide substitution took place
20
(8) Procedure for the Preparation of 13
without erosion of enantioselectivity. Mesylate 17: [a]D
Epoxidation
+34.0 (c 0.2, CHCl3; 96% ee determined by chiral HPLC);
azido compound 18: [a]D20 +58.4 (c 0.51, CHCl3; 96% ee
determined by chiral HPLC).
To a solution of 1,2-dihydroquinoline 7 (1.6g, 4.78 mmol) in
CH2Cl2 (250 mL) were added NaHCO3 (644 mg, 6.22 mmol)
and 70–75% MCPBA (1.5g, 6.22 mmol). The mixture was
stirred at 20 °C for 18 h under argon, then washed with a sat.
NaHCO3 solution (250 mL). After phase separation, the
aqueous phase was extracted with CH2Cl2 (3 × 160 mL). The
combined organic layers were dried over MgSO4, filtered,
and concentrated in vacuo. The crude product (white
powder) was engaged in the reduction process without
further purification.
(13) This two-step N-methylation protocol has already been
reported for transforming sumanirole into its N,N-dimethyl
analogue, see ref. 1a.
(14) Hydrochloride of (R)-3: [a]D20 –29.1 (c 0.25, MeOH; 98% ee
determined by chiral HPLC), lit.1a [a]D20 –30.3 (c 1, MeOH).
(15) Burla, M. C.; Caliandro, R.; Camalli, M.; Carrozzini, B.;
Cascarano, G. L.; De Caro, L.; Giacovazzo, C.; Polidori, G.;
Spagna, R. J. Appl. Crystallogr. 2005, 38, 381.
(16) Petricek, V.; Dusek, M.; Palatinus, L. Jana2006. The
Crystallographic Computing System; Institute of Physics:
Praha, 2006.
Reduction
To a solution of the above crude epoxide product in a 1:1
mixture of CH2Cl2–EtOAc (80 mL) was added 10% Pd/C
Synlett 2010, No. 10, 1473–1476 © Thieme Stuttgart · New York

Copyright of Synlett is the property of Georg Thieme Verlag Stuttgart and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.