Synthesis, resolution and absolute configuration of a tolperisone metabolite

Article abstract of DOI:10.1016/S0957-4166(01)00088-X

1-(4′-Hydroxymethyl-phenyl)-2-methyl-3-(piperidine-1-yl)-propane-1-one M2, a metabolite of tolperisone, was synthesised in a solvent-free Mannich reaction. The optical resolution was carried out by diastereoisomeric salt formation and separation, for which three resolving agents ((2R,3R)-O,O′-dibenzoyl tartaric acid, (2R,3R)-O,O′-di-p-toluoyl tartaric acid and (R)-2-hydroxy-4-(2-methoxyphenyl)-5,5-dimethyl-1,3,2-dioxaphosphorinane-2- oxide (anicyphos)) were found. The absolute configuration of M2 was determined by the single-crystal X-ray diffraction method.

Full text of DOI:10.1016/S0957-4166(01)00088-X

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 719–724
Synthesis, resolution and absolute configuration of a tolperisone
metabolite
Jo´zsef Ba´lint,a Imre Markovits,a,* Gabriella Egri,a Zsuzsanna Tuza,a La´szlo´ Pa´rka´nyib and
Eleme´r Fogassya
aDepartment of Organic Chemical Technology, Budapest University of Technology and Economics, H-1521 Budapest,
PO Box 91, Hungary
bInstitute of Chemistry, Chemical Research Centre, H.A.S., H-1525 Budapest, PO Box 17, Hungary
Received 13 December 2000; accepted 22 February 2001
Abstract—1-(4%-Hydroxymethyl-phenyl)-2-methyl-3-(piperidine-1-yl)-propane-1-one M2, a metabolite of tolperisone, was synthe-
sised in a solvent-free Mannich reaction. The optical resolution was carried out by diastereoisomeric salt formation and
separation, for which three resolving agents ((2R,3R)-O,O%-dibenzoyl tartaric acid, (2R,3R)-O,O%-di-p-toluoyl tartaric acid and
(R)-2-hydroxy-4-(2-methoxyphenyl)-5,5-dimethyl-1,3,2-dioxaphosphorinane-2-oxide (anicyphos)) were found. The absolute
configuration of M2 was determined by the single-crystal X-ray diffraction method. © 2001 Elsevier Science Ltd. All rights
reserved.
1. Introduction
momethyl-propiophenone 2 to 4%-hydroxymethyl-pro-
piophenone 3 was unsuccessful both in aqueous and
methanolic NaOH solution, due to extensive dimer or
methyl ether formation. 4-Hydroxymethyl-propiophe-
none 3 was then obtained by converting the bromide to
an acetyl group followed by hydrolysis. Mannich reac-
tion without solvent yielded M2 (Fig. 1).
According to
a
previous investigation of the
metabolism of tolperisone,1 the main metabolic path-
way involves oxidation of the aromatic methyl group to
a hydroxymethyl and then to a carboxylic acid group.
d-Tolperisone is a classic muscle relaxant agent, while
the l-enantiomer has bronchodilatory and peripheral
vasodilatory activity.2 As the metabolites may also
carry different biological activities, we decided to syn-
thesise the major metabolites and their enantiomers.
Herein, our work3 on the tolperisone metabolic path-
way is continued: synthesis and resolution of 1-(4%-
hydroxymethyl-phenyl)-2-methyl-3-(piperidine-1-yl)-
propane-1-one (M2, Fig. 1) is presented.
2.2. Resolution of M2
Searching the literature we found some amino ketones
structurally analogous to M2 that had been resolved
using dibenzoyl tartaric acid.5–7 Following several
unsuccessful attempts to resolve M2 we finally found
that two tartaric acid derivatives, (2R,3R)-O,O%-diben-
zoyl-tartaric acid (DBTA) and (2R,3R)-O,O%-di-p-tolu-
oyl-tartaric acid (DPTTA), were suitable for
enantioseparation. With DBTA the diastereoisomeric
salt containing the (S)-enantiomer precipitated with
low e.e. (Table 1, entry 1). Resolution experiments
using DPTTA, however, gave unexpected results: the
first experiments yielded the (S)-enantiomer with rea-
sonable optical purity (Table 1, entry 2), but, after this,
the crystallised diastereoisomeric salt always contained
the (R)-isomer and the e.e. was considerably lower than
in the initial resolution experiments. As precipitation of
the (R)-isomer became consistent and reproducible, we
optimised the reaction conditions and obtained the
highest e.e. values of around 80% (Table 1, entry 3)
2. Results and discussion
2.1. Synthesis of M2
To form the desired molecule, 4%-methyl-propiophenone
1 was first brominated with N-bromosuccinimide.4 The
attempted direct hydrolysis of the resulting 4%-bro-
* Corresponding author. E-mail: markovits.oct@chem.bme.hu
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00088-X

720
J. Ba´lint et al. / Tetrahedron: Asymmetry 12 (2001) 719–724
Figure 1. Synthesis of M2.
when the reaction was completed in acetone/ethyl
acetate mixture by 0.5 molar equivalents of resolving
agent and using about half an equivalent of achiral
additive (acetic acid). Table 1 contains the detailed
results for the resolution (see Fig. 2).
from the mixture as while the racemic mixture is pow-
dery, the pure enantiomer is a needle-like crystalline
material.
2.3. Determination of the absolute configuration
The enantiomeric purity of the resulting mixture is not
critical as it can be quite effectively enriched by recrys-
tallisation. The pure enantiomer is physically different
The diastereroisomeric salts formed with DBTA or
DPPTA were unsuitable for X-ray analysis so it was
necessary to search for another resolving agent. After
Table 1. Optical resolution of M2: conditions and results
Entry
Resolving agent
Solvent
Diastereoisomeric salt
Mother liquor
OP (%)
Config. of M2
OP (%)
Yield (%)a
Config. of M2
Yield (%)a
1
2
3
4
DBTA
DPTTA
DPTTA/AcOH
Anicyphos
Acetone
Acetone
Acetone/EtOAc
Acetone
S
S
R
R
11
38
79
86
86
40
92
70
R
R
S
S
6
17
80
51
104
138
85
110
a Yields refer to 50% of the racemate.
Figure 2. Resolution of M2 with di-p-toluoyl tartaric acid.

J. Ba´lint et al. / Tetrahedron: Asymmetry 12 (2001) 719–724
721
Table 2. Crystal data, data collection and refinement
diastereoisomeric salt with M2. The yield and optical
purity values are good (Table 1, entry 4). Although this
resolving agent (Fig. 3) was not the immediate first
choice for large scale resolution due to the price of the
anicyphos, the diastereoisomeric salt proved to be
excellent for single-crystal X-ray diffraction analysis.
parameters
Empirical formula
Formula weight
Temperature (K)
C28H40NO7P·0.81(C3H6O)
580.69
293(2)
,
Radiation and wavelength (A) Mo-Ka, u=0.710730
Crystal system
Space group
Monoclinic
X-Ray diffraction analysis showed the (−)-enantiomer
to have (R) absolute configuration (Fig. 4, Table 2
contains the crystallographic data).
P21
Unit cell dimensions
,
a (A)
10.410(2)
8.576(1)
18.533(1)
95.93(1)
1645.7(4)
2
,
b (A)
,
c (A)
i (°)
3. Experimental
3
,
Volume (A )
Z
3.1. Materials and methods
Dcalcd (mg/m3)
Absorption coefficient, v
1.172
0.129
1
The H NMR spectra were recorded at 250 MHz on
(mm−1
F(000)
)
a Bruker WM250 spectrometer. Chemical shift values
are expressed in ppm values on the l scale. IR spec-
tra of thin film and KBr samples were taken on a
Perkin–Elmer 1600 Series FT-IR spectrophotometer.
Optical rotations were determined on a Perkin–Elmer
241 polarimeter. Thin-layer chromatography was car-
ried out using POLYGRAM® SIL G/UV254 sheets.
Spots were visualised by UV light or by treatment
with 5% ethanolic phosphomolybdic acid solution and
heating of the dried plates. DPTTA and DBTA were
purchased from Aldrich. All solvents used were
freshly distilled.
624
Crystal colour
Colourless
Prism
0.55×0.30×0.25
Psi-scan
0.9894 and 0.9659
2.625q529.96
Crystal description
Crystal size (mm)
Absorption correction
Max. and min. transmission
q-Range for data collection
(°)
Index ranges
−145h514; −125k512;
−265l526
11003
0.985
Reflections collected
Completeness to 2q
Number of standard
reflections
3
Decay (%)
8
The enantiomeric purity of the samples having the
highest optical rotatory power was justified by the H
1
Independent reflections
Reflections I\2|(I)
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Absolute structure parameter
Final R indices [I\2|(I)]
R indices (all data)
Max. and mean shift/esd
9408 [Rint=0.0156]
5091
NMR method as described. Those samples contained
no detectable signals for the other enantiomer so they
were assumed to be homochiral. Throughout the
paper the optical purity was determined by relating
the o.r.p. to the o.r.p. of the above mentioned enan-
tiopure sample ([h]2D0=28.1, c=1, methanol).
Full-matrix least-squares on F2
9408/11/388
0.888
0.01(7)
R1=0.0467, wR2=0.1004
R1=0.1035, wR2=0.1108
0.001; 0.000
Largest diff. peak and hole
0.275 and−0.158
3.2. Preparation of 4%-bromomethyl-propiophenone 2
−3
,
(e A
)
To 4%-methyl-propiophenone 1 (80 g, 0.54 mol) in
CCl4 (600 mL) N-bromosuccinimide (96 g, 0.54 mol)
and benzoyl-peroxide (1 g) were added and the mix-
ture was heated slowly to its boiling point during
vigorous stirring. N.B. In case of contaminated N-
screening several acids we found that (R)-(+)-2-
hydroxy-4-(2-methoxyphenyl)-5,5-dimethyl-1,3,2-diox-
aphosphorinane-2-oxide (anicyphos) forms a crystalline
Figure 3. Resolution of M2 by anicyphos.

722
J. Ba´lint et al. / Tetrahedron: Asymmetry 12 (2001) 719–724
Figure 4. Structure of the diastereoisomeric salt.
bromosuccinimide the reaction mixture will foam heav-
ily. After the reaction mixture lost its colour (1.5–6 h of
boiling), it was cooled to 5°C, the precipitated succin-
imide was removed by filtration and washed with CCl4.
Evaporation of the filtrate gave a thick oil residue from
which precipitation began at low temperature. The
crystals were filtered and washed with hexane. The
crude product was used directly in further conversions.
The colour of the product can vary between colourless
and purple depending on the starting materials, but it
does not influence the further reactions.
mmol) in methanol (14 mL) a solution of NaOH (5.4 g,
13.5 mmol) in water (6 mL) was added. The solution
was stirred under reflux for 15 min, and the solvent was
evaporated. To the residue, water (50 mL) was added,
the mixture was extracted with diethyl ether, and the
organic extract dried over Na2SO4 and evaporated. The
crystalline residue was triturated with hexane and dried
completely to afford white crystals of 4 (10.0 g, 91%),
1
mp 52–53°C. H NMR (CDCl3) 1.21 (t, 3H, H3, 7.27
Hz), 1.98 (s, 1H, OH), 2.99 (q, 2H, H2, 7.27 Hz), 4.76
(s, 2H, Ar-CH2-O), 7.43 (d, 2H, Ar-H, 7.68 Hz), 7.95
(d, 2H, Ar-H, 8.21 Hz). FT-IR (KBr, cm−1): 3456, 2863,
1674, 1608, 1228, 1059, 793. Calcd for C10H12O2: C,
73.15; H, 7.37. Found: C, 73.48; H, 7.39%.
1
Yield: 64.8 g (52.9%), mp 57–58°C. H NMR (CDCl3)
1.20 (t, 3H, H3, 7.23 Hz), 2.97 (q, 2H, H2, 7.19 Hz),
4.47 (s, 2H, Ar-CH2-Br), 7.45 (m, 2H, Ar-H), 7.91 (m,
2H, Ar-H). FT-IR (KBr, cm−1): 3440, 2971, 1683, 1222.
Calcd for C10H11BrO: C, 52.89; H, 4.88; Br, 35.18.
Found: C, 52.64; H, 4.86; Br, 35.29%.
3.5. Preparation of racemic M2
To a mixture of 4%-hydroxymethyl-propiophenone 4 (15
g, 0.09 mol), piperidine hydrochloride (12 g, 0.099 mol),
paraformaldehyde (6 g, 0.2 mol) and ethanolic HCl (2
mL) was added and the reaction mixture was stirred at
100–110°C for 2 h. At this temperature the reaction
mixture melted. After cooling the reaction mixture to
room temperature, water (100 mL) was added, the
mixture was extracted with dichloromethane (3×15
mL), then it was treated with NaOH (6 g) in water (40
mL) to alkaline pH. The yellow oil was extracted with
dichloromethane (3×40 mL) and the organic phase was
dried and evaporated. The product was triturated with
hexane and filtered. The crude product was recrys-
tallised from a mixture of ethyl acetate and hexane to
3.3. Preparation of 4%-acetoxymethyl-propiophenone 3
A mixture of 4%-bromomethyl-propiophenone 2, (18 g,
0.08 mol), anhydrous NaOAc (13 g, 0.158 mol) and
glacial acetic acid (36 mL) was stirred under reflux for
2 h. After cooling, water (300 mL) was added, and the
mixture was extracted by diethyl ether. The organic
phase was washed with NaOH solution, dried over
Mg2SO4 and evaporated. The product was isolated as a
yellow oil (13.8 g, 84.4%) that needs no further purifica-
1
tion. Bp 126°C (0.8 mmHg). H NMR (CDCl3) 1.20 (t,
3H, H3, 7.27 Hz), 2.11 (s, 3H, acetic CH3), 2.98 (q, 2H,
H2, 7.26 Hz), 5.14 (s, 2H, Ar-CH2-O), 7.42 (d, 2H,
Ar-H, 8.15 Hz), 7.93 (d, 2H, Ar-H, 8.24 Hz). FT-IR
(film, cm−1): 2978, 1740, 1646, 1224, 1044. Calcd for
C12H14O3: C, 69.89; H 6.84. Found: C, 69.59; H, 6.86%.
afford M2 as a white solid (10.3 g, 39.4 mmol, 43.2%),
1
mp 103–104°C. H NMR (CDCl3) 1.19 (d, 3H, H2Me
,
6.96 Hz), 1.39 (m, 2H), 1.51 (m, 4H), 2.42 (m, 6H), 2.89
(m, 1H), 3.76 (m, 1H, Hkir), 4.78 (s, 2H, H4%Me), 7.46 (d,
2H, HAr, 8.09 Hz), 7.97 (d, 2H, HAr, 8.06 Hz). FT-IR
(KBr, cm−1): 3427, 3112, 2941, 2804, 1672, 1231, 1062.
Calcd for C16H23NO2: C, 73.53; H, 8.87; N, 5.36.
Found: C, 73.88; H, 8.87; N, 5.38%.
3.4. Preparation of 4%-hydroxymethyl-propiophenone 4
To 4%-acetoxymethyl-propiophenone 3 (13.8 g, 6.7

J. Ba´lint et al. / Tetrahedron: Asymmetry 12 (2001) 719–724
723
3.6. General procedure for resolution of M2
The mother liquor of the resolution was evaporated.
After addition of water it was acidified with cc. HCl
and the precipitated anicyphos was filtered and washed
with water. The mother liquor was treated with base
until alkaline and the precipitated base collected by
filtration and washed with water then dried to afford a
white solid (5.48 g, 21.0 mmol, 110%), [h]D=+14.5
[c=1, methanol], enantiomeric purity=51.6%, mp 122–
123°C.
Racemic M2 (14.49 g, 55.4 mmol), DPTTA (9.64 g,
24.9 mmol) and acetic acid (1.74 mL, 30.5 mmol) were
dissolved in a mixture of EtOAc (22 mL) and acetone
(11 mL). The mixture was stirred for 2 h and the
resultant precipitate was collected by filtration and
washed with ethyl acetate. The diastereomeric salt was
suspended in water (32 mL), cc. HCl (3 mL) was added,
and the mixture was extracted with ethyl acetate (66
mL). The aqueous phase was treated with NaOH to
alkaline pH. After filtration the precipitate was washed
with water and hexane, and a yellowish crystalline
material was obtained (6.68 g, 25.6 mmol). [h]D= −22.2
[c=1, methanol].
3.10. X-Ray crystallography8–13
Crystals were obtained as follows: the diastereoisomeric
salt of (R)-(−)-M2 (0.3 g, 1.14 mmol, [h]D= −28.0
(c=1, methanol)) and anicyphos (0.31 g, 1.15 mmol)
were dissolved in hot acetone–ethanol (4.5–0.5 mL).
The solution was cooled to room temperature, inocu-
lated and left to crystallise overnight.
The crystallisation mother liquor was acidified with cc.
HCl and extracted with ethyl acetate (30 mL). After
treating the aqueous phase with base until; alkaline the
solid was filtered, washed with water and hexane and
dried to afford of yellowish crystals, (6.14 g, 23.3
mmol). [h]D=+22.6 [c=1, methanol].
Crystal data: C28H40NO7P·0.81(C3H6O), Fwt: 580.69,
colourless, prism, size: 0.55×0.30×0.25 mm, monoclinic,
space group P21, a=10.410(2), b=8.576(1), c=
,
18.533(1) A, h=90.00, i=95.93(1), k=90.00°, V=
3
,
1645.7(4) A , T=293(2) K, Z=2, F(000)=624,
Dx=1.172 Mg/m3, m=0.129 mm−1.
3.7. Enantiomeric enrichment
(a) Enantiomeric mixture of M2 (6.14 g, [h]D=+22.6
[c=1, methanol]) was recrystallised three times from a
small amount of methanol to yield white crystals (3.63
g, 59%), [h]D=+28.1 [c=1, methanol], mp 121–124°C.
Cell parameters were determined by least-squares of the
setting angles of 25 (16.075q517.17°) reflections.
Intensity data were collected on an Enraf–Nonius
CAD4 diffractometer (graphite monochromator; Mo-
(b) Enantiomeric mixture of M2 (6.68 g, [h]D=−22.2
[c=1, methanol]) was recrystallised three times from a
small amount of methanol to yield of white crystals
(3.61 g, 54%), [h]D=−28.0 [c=1, methanol], mp 122–
125°C.
,
Ka radiation, u=0.710730 A) at 293(2) K in the range
2.625q529.96° using ꢀ/2q scans. Backgrounds were
measured half the total time of peak scans. The intensi-
ties of three standard reflections were monitored regu-
larly (every 60 min). The intensities of the standard
reflections indicated a crystal decay of 8% (the data
were corrected for decay).
3.8. Determination of the enantiomeric excess
A total of 11003 reflections were collected of which
9408 were unique [Rint=0.0156, R(|) =0.0691]; inten-
sities of 5091 reflections were greater than 2|(I). Com-
pleteness to q=0.985.
The e.e. of the above samples was determined with the
aid of (R)-(−)-1-(9-anthryl)-2,2,2-trifluoroethanol. At
the 1H NMR spectrum of rac-M2 and the chiral solvat-
ing agent (molar ratio: 1:1.96) the H2Me signal was split
into two doublets: 1.07 ppm, 1.12 ppm, D=27.8 Hz. In
the spectra of the two enantiomeric mixtures ([h]D=+
28.1 and −28.0 [c=1, methanol], respectively) with the
chiral solvating agent only one enantiomeric peak was
detectable, so we can assume them enantiomerically
pure (e.e.>97%).
A psi-scan absorption correction was applied to the
data (the minimum and maximum transmission factors
were 0.9659 and 0.9894).
The structure was solved by direct methods (and subse-
quent difference syntheses).
Anisotropic full-matrix least-squares refinement on F2
for all non-hydrogen atoms yielded R1=0.0467 and
wR2=0.1004 for 5091 [I>2|( I)] and R1=0.1035 and
wR2=0.1108 for all (9408) intensity data (number of
3.9. Resolution of M2 by anicyphos
Racemic M2 (10 g, 38 mmol) and anicyphos (5.21 g,
19.1 mmol) was dissolved in acetone (100 mL) and the
solution was allowed to crystallise. The crystals were
filtered and washed with acetone then dried to afford
product (8.08 g). The diastereoisomeric salt was dis-
solved in water, acidified, and the precipitate was col-
lected by filtration (3.76 g, 13.8 mmol, 72%). The
filtrate was treated with base to render it alkaline and
the precipitated base was collected by filtration (3.50 g,
13.4 mmol, 70%), [h]D= −24.3 [c=1, methanol], enan-
tiomeric purity=86% mp 121–124°C.
parameters=388,
goodness-of-fit=0.888,
absolute
structure parameter x=0.01(7), the maximum and
mean shift/esd is 0.001 and 0.000).
The maximum and minimum residual electron density
−3
,
in the final difference map was 0.275 and −0.158 e A .
The weighting scheme applied was w=1/[|2(Fo )+
2
(0.0603P)2] where P=(Fo +2Fc )/3.
2
2

724
J. Ba´lint et al. / Tetrahedron: Asymmetry 12 (2001) 719–724
Hydrogen atomic positions were calculated from
assumed geometries (except H19 and H37 that were
located in difference maps). Hydrogen atoms were
included in structure factor calculations but they were
not refined. The isotropic displacement parameters of
the hydrogen atoms were approximated from the Ueq
value of the atom they were bonded to.
E. Tetrahedron: Asymmetry 2000, 11, 1323–1329.
4. Shiozawa, A.; Narita, K.; Izumi, G; Kurashige, S.; Saki-
tama, K.; Ishikawa, M. Eur. J. Med. Chem. Chim. Ther.
1995, 30, 85–94.
5. Newman, P. Optical Resolution Procedures for Chemical
Compounds; Optical Resolution Information Center:
Manhattan College, Riverdale, New York, 1978; Vol. 1,
p. 326.
6. MacConaill, R. J.; Scott, F. L. Tetrahedron Lett. 1970,
2993.
Acknowledgements
7. Takamatsu, H. H.; Minaki, Y. Yakugaku Zasshi 1956,
76, 1230.
The Hungarian OTKA Foundation (Project No.
T29251) is gratefully acknowledged for financial sup-
port. J.B. thanks the Zolta´n Magyary Foundation and
OTKA (D 32705) for a postdoctoral fellowship. G.E.
thanks the OTKA (D 29445) for a postdoctoral fellow-
ship. Gedeon Richter Ltd is acknowledged for financial
support.
8. Harms, K. XCAD4 Data Reduction Program for CAD4
Diffractometers, Philipps-Universita¨t Marburg, Ger-
many, 1996.
9. North, A. C.; Philips, D. C.; Matthews, F. Acta Crystal-
logr., Sect. A 1968, 24, 350–359.
10. SHELXS: Sheldrick, G. M. SHELXS-97 Program for
Crystal Structure Solution, University of Go¨ttingen, Ger-
many, 1997.
11. Sheldrick, G. M. SHELXL-97 Program for Crystal Struc-
ture Refinement, University of Go¨ttingen, Germany,
1997.
References
1. Miyazaki, H.; Ishibashi, M.; Takayama, H.; Abuki, H.;
Idzu, G.; Morishita, N.; Ando, M. Proceedings of the 4th
Symposium on Drug Metabolism and Action; 1972 (Pub
1973); pp. 154–164.
2. Furuta, Y.; Nakamura, K.; Tashiro, Y.; Taka, S.;
Nagashima, T.; Japan Kokai 78 40,779; CA 89, 10918m,
1978.
12. Neutral atomic scattering factors and anomalous scatter-
ing factors were taken from: International Tables for
X-ray Crystallography; Wilson, A. J. C., Ed.; Kluwer
Academic Publishers: Dordrecht, 1992; Vol. C, Tables
6.1.1.4 (pp. 500–502), 4.2.6.8 (pp. 219–222) and 4.2.4.2
(pp. 193–199).
13. Flack, H. D. Acta Crystallogr., Sect. A 1983, 39, 876–
881.
3. Ba´lint, J.; Hell, Z.; Markovits, I.; Pa´rka´nyi, L.; Fogassy,
.