Asymmetric synthesis of chiral piperazinylpropylisoxazoline ligands for dopamine receptors

Article abstract of DOI:10.1016/j.ejmech.2006.12.030

The asymmetric synthesis of chiral piperazinylpropylisoxazoline analogues, (R)-(+)-1, 2 and (S)-(-)-1, 2 was accomplished through a seven-step sequence of reactions, which involved asymmetric 1,3-dipolar cycloaddition, alkyl chain extension, and reductive

Full text of DOI:10.1016/j.ejmech.2006.12.030

European Journal of Medicinal Chemistry 42 (2007) 1044e1048
http://www.elsevier.com/locate/ejmech
Short communication
Asymmetric synthesis of chiral piperazinylpropylisoxazoline
ligands for dopamine receptors
b
Ji Young Jung a, Sun Ho Jung a, , Hun Yeong Koh
*
a Department of Chemistry and Institute of Basic Science, Sungshin Women’s University, Seoul 136-742, South Korea
b Department of Chemistry, Inha University, Inchon 402-751, South Korea
Received 16 October 2006; received in revised form 18 December 2006; accepted 18 December 2006
Available online 13 January 2007
Abstract
The asymmetric synthesis of chiral piperazinylpropylisoxazoline analogues, (R)-(þ)-1, 2 and (S)-(ꢀ)-1, 2 was accomplished through a seven-
step sequence of reactions, which involved asymmetric 1,3-dipolar cycloaddition, alkyl chain extension, and reductive amination as key reac-
tions. Chiral ligands (R)-(þ)-1, 2 exhibited the higher binding affinity and selectivity for the D3 receptor over the D4 receptor than (S)-(ꢀ)-1, 2
ligands.
Ó 2007 Elsevier Masson SAS. All rights reserved.
Keywords: Asymmetric synthesis; Piperazinylalkylisoxazolines; Chiral sultam; Dopamine receptors; 1,3-Dipolar cycloaddition
1. Introduction
2. Results and discussion
In recent years, extensive efforts have been made to explore
potent ligands for dopamine D3 [1] or D4 [2] receptor [3] for
the discovery of antipsychotic drugs. In this connection, we
have recently reported [4] a simple and efficient way of con-
structing libraries of isoxazol(in)es and a library of piperaziny-
lalkylisoxazoles, of which some ligands were found to exhibit
high binding affinity and selectivity for the D3 and D4 recep-
tors over the D2 receptor. In the course of continuation of this
research program, some piperazinylalkylisoxazolines [4d,5]
such as 1 and 2 were also found to possess good binding affin-
ity for the D3 receptor (Fig. 1). Since piperazinylalkylisoxazo-
line analogues have stereogenic centers, it was deemed
worthwhile to investigate asymmetric synthetic route to both
enantiomers and evaluate their binding affinities for dopamine
receptor subtypes. Herein, we wish to report asymmetric syn-
thesis of (þ)- and (ꢀ)-piperazinylpropylisoxazoline.
Our approach to (þ)- and (ꢀ)-piperazinylpropylisoxazoline
analogues was based upon diastereofacial selective 1,3-dipolar
cycloaddition [6] of nitrile oxides and alkyl chain extension
strategy. First, in order to find suitable chiral auxiliary for
asymmetric induction, the asymmetric nitrile oxide cycloaddi-
tion was examined by employing acryloyl derivatives 3ae3e
of five chiral auxiliaries, i.e. (1S )-(ꢀ)-2,10-camphorsultam
(a), (1R)-(þ)-2,10-camphorsultam (b), (S)-(ꢀ)-4-(diphenyl-
methyl)-2-oxazolidinone (c), (1R)-(þ)-benzyl-2-oxazolidinone
(d), and (4S,5R)-(ꢀ)-4-methyl-5-phenyl-2-oxazolidinone (e)
(Xc, Scheme 1) [7,8]. Upon treating 3ae3e and 3,4-dimethox-
ybenzaldehyde oxime 4 with NaOCl solution, diastereomeric
mixtures of cycloadducts 5ae5e were obtained in 58e65%
yields.
Although the diastereomeric ratios of cycloadducts 5ae5e
could not be measured by either HPLC analysis or 1H NMR in-
tegration, the degree of diastereoselectivity in the cycloaddition
reactions could successfully be determined by 19F NMR inte-
gration of diastereomeric MTPA esters 7, which were obtained
through a two-step sequence of reactions: (1) L-Selectride
* Corresponding author. Tel.: þ82 2 920 7192; fax: þ82 2 920 7192.
E-mail address: shjung@sungshin.ac.kr (S. Ho Jung).
0223-5234/$ - see front matter Ó 2007 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.ejmech.2006.12.030

J.Y. Jung et al. / European Journal of Medicinal Chemistry 42 (2007) 1044e1048
1045
OCH3
OCH
3
OCH3
OCH
3
F
N
OEt
N
O
N
N
O
N
N
1
2
F
Fig. 1.
reduction of 5ae5e to alcohols 6 and (2) subsequent esterifica-
tion of alcohols 6 with (þ)-methoxy(trifluoromethyl)phenyla-
cetyl chloride [9] (Scheme 1). Diastereoselectivities observed
in this way are shown in Table 1. The asymmetric induction
was far better with the use of camphorsultams (entries 1 and
2) as chiral auxiliaries than with the use of oxazolidinones (en-
tries 3e5). The observed diastereoselectivities of 85:15 and
70:30 (entries 1 and 2), respectively, in favor of (R)-configura-
tion and (S)-configuration at the newly generated stereogenic
center (vide infra), was acceptable for our purpose. Thus, fur-
ther elaboration was embarked to obtain chiral piperazinylpro-
pylisoxazoline analogues from cycloadducts 5a and 5b
obtained via the asymmetric nitrile oxide cycloaddition of
acryloyl derivatives of chiral camphorsultams, 3a and 3b.
Since the diastereomeric mixture of cycloadducts 5a was
chromatographically inseparable, other means of separation
were explored. Fortunately, the mixture could be separated
by recrystallization in the solvent mixture comprising
CH2Cl2, EtOH and diethyl ether (3:1:1) to give (R)-5a in
pure form. The structure and absolute stereochemistry of
(R)-5a were proven by an X-ray analysis on it (Fig. 2) [10].
The pure cycloadduct (S)-5b could also be separated in the
same way from a mixture of cycloadducts 5b. In this way,
pure (R)-5a and (S)-5b were obtained in 70% and 61% yields,
respectively, from the cycloaddition of dipolarophiles 3a and
3b with 4. Subsequent reduction of (R)-5a and (S)-5b with
L-Selectride gave (R)-6 and (S)-6 in 80% and 82% yields, re-
spectively. The alcohol (R)-6 showed an optical rotation of
[a]2D0 ¼ ꢀ132ꢁ (c ¼ 1, CHCl3), which was compared well to
both reported value of [a]2D5 ¼ ꢀ118ꢁ (c ¼ 0.11, MeOH) [11]
and the rotation of its enantiomer (S)-6, [a]2D0 ¼ þ133ꢁ
(c ¼ 1, CHCl3). The alcohol (R)-6 was then converted to its tri-
flate (R)-8 for alkyl chain extension (Scheme 2).
The (R)-6 underwent triflation smoothly with triflic anhy-
dride in the presence of Et3N to give the triflate (R)-8. Exten-
sion of alkyl chain was then achieved by reacting (R)-8 with
lithium enolate of t-butyl acetate in THFeHMPA (4:1) at
ꢀ78 ꢁC. The use of HMPA as a cosolvent was necessary for
the success of alkylation. Without HMPA the yield was very
low. It should also be noted that the use of enolate of t-butyl
acetate gave much better results than that of enolate of either
ethyl or i-propyl acetate. L-Selectride reduction and the fol-
lowing PCC oxidation gave the aldehyde (S)-10. The enantio-
meric aldehyde (R)-10 could also be obtained from the alcohol
(S)-6 in the same manner as (R)-6 was converted to (S)-10.
Final assembly to targets, chiral piperazinylpropylisoxazo-
line analogues, was accomplished by the reductive amination
of aldehydes and selected amines using NaBH(OAc)3 (Scheme 3).
O
O
i)
*
ii)
R1
HO
OH
*
R1
+
Xc
R1
N
Xc
N
O
O N
5a-e
3a-e
4
6
R1=
O
iii)
*
R1
O
H3CO
F3C OCH3
O N
OCH3
7
O
Ph
N
O
O
Ph
O
N
N
Xc=
N
O
N
S
S
Ph
Me
Ph
O
O
O
O
O
a
b
c
d
e
Scheme 1. Reagents and reaction conditions: (i) 0.54 M NaOCl, 0 ꢁC, CH2Cl2, 58e65%; (ii) 1 M L-Selectride, 0 ꢁC, THF, 80e85%; (iii) (S)-(þ)-MTPA-Cl,
DMAP, 0 ꢁC, toluene, 90e92%.

1046
J.Y. Jung et al. / European Journal of Medicinal Chemistry 42 (2007) 1044e1048
Fig. 2. X-ray crystal structure of (R)-5a.
O
R1
R1
R
R1
a
b
TfO
HO
N
O N
O N
S
O N
O
O
(R)-5a
(R)-6
(R)-8
O
O
d, e
c
R1
R1
H
t-BuO
O N
O N
(S)-9
(S)-10
Scheme 2. Reaction conditions and [a]2D0 values: (a) 1 M L-Selectride, THF, 0 ꢁC, for (R)-6, 80%, [a]2D0 ¼ꢀ132ꢁ (c ¼ 1, CHCl3); for (S)-6, 82%, [a]2D0 ¼ þ133ꢁ
(c ¼ 1, CHCl3). (b) (CF3SO2)2O, Et3N, ꢀ30 ꢁC, CH2Cl2, for (R)-8, 89%, [a]D20 ¼ ꢀ84.7ꢁ (c ¼ 1, CHCl3); for (S)-8, 87%, [a]D20 ¼ þ85.6ꢁ (c ¼ 1, CHCl3). (c) t-Butyl
acetate, LDA, HMPA, ꢀ78 ꢁC, THF, for (S)-9, 78%, [a]2D0 ¼ ꢀ85.8ꢁ (c ¼ 1, CHCl3); for (R)-9, [a]D20 ¼ þ86.0ꢁ (c ¼ 1, CHCl3). (d) L-Selectride (1 M), 0 ꢁC, THF
(e) PCC, SiO2, CH2Cl2 ; for (S)-10, 66% for two steps, [a]D20 ¼ ꢀ104ꢁ (c ¼ 1, CHCl3); for (R)-10, 60% for two steps, [a]D20 ¼ þ103ꢁ (c ¼ 1, CHCl3).
R1
N
O N
N
N
R2
(S)-1 (R2= R21
(S)-2 (R2= R22
)
)
R2
N
(S)-10
a
or
+
or
N
H
(R)-10
R1
N
11
O N
R2
(R)-1 (R2= R21
(R)-2 (R2= R22
)
)
R2=
R1=
F
F
OCH3
OCH3
OCH2CH3
R21
R22
Scheme 3. Reaction conditions and [a]2D0 values: NaBH(OAc)3 (3 equiv), molecular sieves (3 beads), CH2Cl2, 3e4 h, rt, 92e96%. (S)-1, [a]2D0 ¼ ꢀ71.4ꢁ (c ¼ 1,
CHCl3); (S)-2, [a]2D0 ¼ ꢀ61.2ꢁ (c ¼ 1, CHCl3); (R)-1, [a]D20 ¼ þ72.7ꢁ (c ¼ 1, CHCl3); (R)-2, [a]D20 ¼ þ62.0ꢁ (c ¼ 1, CHCl3).

J.Y. Jung et al. / European Journal of Medicinal Chemistry 42 (2007) 1044e1048
1047
Table 1
Diastreoselectivity of asymmetric cycloaddition
D3 receptor over the D4 receptor than (S)-(ꢀ)-1, 2 ligands. Fur-
ther studies on the synthesis of other chiral piperazinylpropyli-
soxazoline compounds employing this route are in progress.
Entry
Dipolarophile
Configurationa
Deb of 7 (%)
1
2
3
4
5
a
3a
3b
3c
3d
3e
R
S
70
40
4
Acknowledgments
R
R
S
16
2
This work was supported by the grant from Sungshin Wom-
en’s University in 2006. We thank Dr. Jaekyun Lee and Dr.
Yong Seo Cho at the Korea Institute of Science and Technol-
ogy for the X-ray structure determination.
Configuration of the major cycloadduct at the new stereogenic center.
Determined by 19F NMR analysis of MTPA esters 7.
b
Combination of (S)-10 and (R)-10 isomer with two amines 11
under the well established reaction conditions [4] gave four
enantiomerically pure isomers, i.e. (S)-(ꢀ)-1, (S)-(ꢀ)-2, (R)-
(þ)-1, and (R)-(þ)-2 in 92e96% yields. All enantiomers
were obtained with a high optical purity (>99% ee), which
was determined by HPLC (Chiral Pak AD column, 1 mL/
min, 2-propanol:hexane ¼ 2:8, 254 nm).
References
[1] P. Sokoloff, B. Giros, M.-P. Martres, M.-L. Bouthenet, J.-C. Schwartz,
Nature 347 (1990) 146.
[2] H.M. Van Tol, J.R. Bunzow, H.C. Guan, R.K. Sunahara, P. Seeman,
H.B. Niznik, O. Civelli, Nature 30 (1991) 610.
[3] (a) C. Enguehard-Gueiffier, H. Hubner, A. El Hakmaoui, H. Allouchi,
¨
P. Gmeiner, A. Argiolas, M.R. Melis, A. Gueiffier, J. Med. Chem. 49
(2006) 3938;
The prepared chiral isomers were evaluated in vitro for
dopamine D2eD4 receptors binding affinity by measuring
their ability to displace radioligands ([3H]spiperone for D2
and D4, [3H]YM-09151-2 for D3) from the cloned human
dopamine receptors D2long, D3 and D4.2 which were expressed
in CHO cells. Table 2 shows the binding data of the prepared
target chiral compounds.
(b) S. Lober, H. Hubner, P. Gmeiner, Bioorg. Med. Chem. Lett. 16 (2006)
2955;
¨
¨
(c) P. Mohr, M. Decker, C. Enzensperger, J. Lehmann, J. Med. Chem. 49
(2006) 2110;
(d) M. Leopoldo, E. Lacivita, N.A. Colabufo, F. Beradi, R. Perrone, J.
Pharm. Pharmacol. 58 (2006) 209;
(e) J. Chen, K. Ding, B. Levant, S. Wang, Bioorg. Med. Chem. Lett. 16
(2006) 443;
Of the chiral isomers, (R)-(þ)-1 also exhibited high binding
affinity with Ki value of 2.1 nM and high selectivity for D3
receptor over D4 receptor. In addition, it exhibited moderate
(38-fold higher) selectivity for D3 receptor versus D2 receptor.
Similarly, for (R)-(þ)-2, the high binding affinity with Ki value
of 2.1 nM for D3 receptor was also observed. In comparison
with (R)-enantiomers, (S)-(ꢀ)-1 and (S)-(ꢀ)-2, showed lower
binding affinity with the Ki values of 20 nM and 5.1 nM,
respectively, for the D3 receptor. In addition, the (S)-(ꢀ)-iso-
mers exhibited lower selectivity for the D3 receptor over the
D4 receptor and for the D3 receptor versus the D2 receptor.
(f) D. Lentz, F. Boeckler, H. Hubner, P. Gmeiner, Bioorg. Med. Chem. 13
¨
(2005) 4434;
(g) X. Wang, P.A. Bhatia, J.F. Daanen, S.P. Latsaw, J. Rhode, T. Kolsa,
A.A. Hakeem, M.A. Matulenko, M. Nakane, M.E. Uchic, L.N. Miller,
R. Chang, R.B. Moreland, J.D. Brioni, A.O. Stewart, Bioorg. Med.
Chem. 13 (2005) 4667;
(h) K. Ding, J. Chen, M. Ji, X. Wu, J. Varady, C.Y. Yang, Y. Lu,
J.R. Deschamps, B. Levant, S. Wang, J. Med. Chem. Lett. 48 (2005)
3171;
(i) L. Bettinetti, S. Lober, H. Hubner, P. Gmeiner, J. Comb. Chem. 7
¨
¨
(2005) 309;
(j) M. Ji, J. Chen, K. Ding, X. Wu, J. Varady, B. levant, S. Wang, Bioorg.
Med. Chem. Lett. 15 (2005) 1701;
(k) W. Chu, Z. Tu, E. McElveen, J. Xu, M. Taylor, R.R. Luedtke,
R.H. March, Bioorg. Med. Chem. 13 (2005) 77.
3. Conclusions
[4] (a) K.H. Kang, A.N. Pae, K.I. Choi, Y.S. Cho, B.Y. Chung, J.E. Lee,
S.H. Jung, H.Y. Koh, H.-Y. Lee, Tetrahedron Lett. 42 (2001) 1057;
(b) M.Y. Cha, B.C. Choi, K.H. Kang, A.N. Pae, K.I. Choi, Y.S. Cho,
H.Y. Koh, H.-Y. Lee, D. Jung, J.Y. Kong, Bioorg. Med. Chem. Lett. 12
(2002) 1327;
In conclusion, the synthesis of chiral piperazinylpropyli-
soxazoline analogues, (R)-(þ)-1, 2 and (S)-(ꢀ)-1, 2 was ac-
complished through a seven-step sequence of reactions. Key
steps involved (1) asymmetric 1,3-dipolar cycloaddition using
Oppolzer’s chiral sultams as chiral auxiliaries, (2) alkyl chain
extension of triflates 8 to esters 9, and (3) reductive amination
of aldehydes 10 with piperazines 11. Chiral ligands (R)-(þ)-1,
2 exhibited the higher binding affinity and selectivity for the
(c) For in vitro metabolic stability and identification of metabolites, see
J. Lee, J. Son, H.Y. Koh, A.N. Pae, D.H. Kim, Anal. Biochem. 313
(2003) 292;
(d) H.Y. Koh, K. Choi, Y.S. Cho, A.N. Pae, J.Y. Kong, D.Y. Jeong, S.H.
Jung, H.Y. Lee, Brit. UK Pat. Appl., GB 2369617 A1, 2002, 0605.
[5] J.Y. Jung, S.H. Jung, H.Y. Koh, A.N. Pae, W.K. Park, J.Y. Kong, Bull.
Korean Chem. Soc. 27 (2006) 1861.
Table 2
Binding affinity (Ki, nM) [12]
[6] (a) A. Padwa, 1, 3-Dipolar Cycloaddition Chemistry, vols. 1 and 2, John-
Wiley & Sons, New York, 1984;
(b) D.P. Curran, Advances in Cycloaddition, vol. 1, 1988, pp. 129;
Compounds
D2 (Ki, nM) D3 (Ki, nM) D4 (Ki, nM) D2/D3 D4/D3
(c) V. Jager, I. Muller, Tetrahedron 41 (1985) 3519;
¨
¨
(R)-(þ)-1
(S )-(ꢀ)-1
(R)-(þ)-2
(S )-(ꢀ)-2
Haloperidola
a
80
400
46
2.1
20
426
483
507
225
103
38
20
22
14
203
24
(d) A. Studer, D.P. Curran, Tetrahedron 53 (1997) 6681;
(e) For a review on asymmetric 1,3-dipolar cycloaddition, see: K.V.
Gothelf, L.A. Jorgensen Chem. Rev. 98 (1998) 863.
2.1
5.1
6.4
241
44
70
[7] (a) D.P. Curran, T.A. Heffner, J. Org. Chem. 55 (1990) 4585;
(b) K.S. Kim, B.H. Kim, W.M. Park, S.J. Cho, B.J. Mhin, J. Am. Chem.
Soc. 115 (1993) 7472;
3.3
0.5
16
Reference drug.

1048
J.Y. Jung et al. / European Journal of Medicinal Chemistry 42 (2007) 1044e1048
(c) D.P. Curran, B.H. Kim, H.P. Pyyasena, R.J. Loncharich, K.N. Houk, J. [12] Assay protocol: Sf-9 membranes expressing either dopamine hD2L
,
Org. Chem. 52 (1987) 2137;
(d) D.P. Curran, B.H. Kim, J. Daugherty, T.A. Heffner, Tetrahedron Lett.
29 (1988) 3555.
dopamine rD3 or hD4.2 receptors were purchased from PerkinElmer Life
Sciences (Boston, MA). Radioligands used were [3H]spiperone (D2 and
D3 dopamine binding assays, 1 nM), and [3H]YM-09151-2 (D4 dopamine
binding assays, 0.06 nM). [3H]Spiperone and [3H]YM-09151-2 bindings
were performed by the protocol provided by supplier of Sf-9 membranes
for both hD2L and rD3, and hD4.2 receptors, respectively. Briefly, the buffer
used in hD2L or rD3 receptor binding assay was 50 mM TriseHCl (pH
7.4), 10 mM MgCl2, 1 mL EDTA, or 50 mM TriseHCl (pH 7.4), 5 mM
MgCl2, 5 mM EDTA, 5 mM KCl, 1.5 mM CaCl2, 120 mM NaCl, respec-
tively. In [3H]YM-09151-2 receptor binding assays, the buffer containing
50 mM TriseHCl (pH 7.4), 5 mM MgCl2, 5 mM EDTA, 5 mM KCl and
1.5 mM CaCl2 was used. Non-specific binding was determined with halo-
peridol (10 mM) and clozapine (10 mM) for D2 and D3, and D4 receptors,
respectively. Competition binding studies were carried out with 8 concen-
trations of the test compound run in duplicate tubes, and isotherms from
3e5 assays were calculated by computerized nonlinear regression analysis
(GraphPad PrismProgram, San Diego, CA) to yield inhibitionconstant(Ki)
values.
[8] (a) For a general review on chiral oxazolidinones in asymmetric
synthesis, see: D.J. Ager, I. Prakash, D.R. Schaad Aldrichim. Acta. 30
(1997) 3;
(b) T. Akiba, O. Tanura, S. Terashima, Org. Synth. 75 (1997) 45.
[9] J.A. Dale, D.L. Dull, H.S. Mosher, J. Org. Chem. 34 (1969) 2543.
[10] Crystallographic data for 5a, as a CIF file, have been deposited with the
Cambridge Crystallographic Data Center as supplementary publication
numbers CCDC 619105. Copies of the data can be obtained, free of
charge, on application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK [fax: þ44(0) 1223 336033 or e-mail: deposit@ccdc.cam.
ac.uk].
[11] For an enantioselective synthesis of (R)-6 by asymmetric 1,3-dipolar
cycloaddition using diisopropyl (R,R)-tartarate as a chiral auxiliary,
see: Y. Ukaji, M. Ima, T. Yamada, K. Inomata Heterocycles 52
(2000) 563.