Accepted Manuscript Title: Structure elucidation of a process-related impurity of dapoxetine Author: Andr´as Darcsi Gerg˝o T´oth J´ozsef K¨ok¨osi Szabolcs B´eni PII: DOI: Reference:
S0731-7085(14)00184-8 http://dx.doi.org/doi:10.1016/j.jpba.2014.04.002 PBA 9528
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
18-2-2014 31-3-2014 3-4-2014
Please cite this article as: http://dx.doi.org/10.1016/j.jpba.2014.04.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Structure elucidation of a process-related impurity of dapoxetine
2 András Darcsi, Gergő Tóth, József Kökösi, Szabolcs Béni*
3 4
Department of Pharmaceutical Chemistry, Semmelweis University
6
H-1092 Hőgyes Endre street 9. Budapest, Hungary
7
cr
8 9
us
10 11
an
12 13 14
M
15 16
ip t
5
Highlights
d
17! ! A novel dapoxetine by-product has been identified using NMR and MS. 18! ! The mechanism for the impurity formation was proposed and also confirmed by synthesis. 20 21 22 23 24
Ac ce p
te
19! ! The separation of dapoxetine and the process-related impurities was accomplished by HPLC.
25
*
26
Dr. Szabolcs Béni
27
Tel.: +36 1 217 0891; fax: +36 1 217 0891; E-mail address:
28
[email protected]
Corresponding author:
29
1 Page 1 of 18
Abstract
30
Unknown by-product associated with the synthesis of dapoxetine was isolated. The structure
31
elucidation of this new compound using accurate mass data and NMR spectroscopy is
32
presented herein. The unambiguous resonance assignment concluded to the formation of a
33
tricyclic compound 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran, a new impurity of dapoxetine
34
which has never been reported previously. A proposed mechanism for the formation of the
35
new carbon-carbon bond is discussed. For the separation of dapoxetine and the process-
36
related impurities, a gradient HPLC method was developed.
cr
37
ip t
29
Keywords: Priligy, impurity profiling, NMR, HPLC-DAD, electrophilic aromatic
39
substitution, ring closure mechanism
us
38
41
an
40 Highlights
42! ! A novel dapoxetine by-product has been identified using NMR and MS.
M
43! ! The mechanism for the impurity formation was proposed and also confirmed by synthesis. 44! ! The separation of dapoxetine and the process-related impurities was accomplished by HPLC.
Ac ce p
te
d
45
2 Page 2 of 18
46 47 1.
1. Introduction
48
Dapoxetine
49
hydrochloride, Priligy®) is a novel short acting selective serotonin reuptake inhibitor (SSRI)
50
that is being developed specifically as an on-demand oral treatment of premature ejaculation
51
with a unique pharmacokinetic profile [1]. Dpx attains its peak plasma concentration in about
52
1.5 hours after dosing which is much faster than conventional SSRIs and by 24 hours the
53
plasma concentration decreases to approximately 5% of the peak concentration. These
54
pharmacokinetic properties make Dpx an excellent candidate for on-demand treatment of
55
premature ejaculation. The eutomer (S)-Dpx is 3.5 times more potent SSRI than (R)-Dpx, that
56
is why Dpx is marketed as a single enantiomer drug [2]. A wide range of synthetic procedures
57
were developed to synthesize racemic and enantiopure Dpx [3-7].
58
Our previous study reported a robust, sensitive and validated method for the chiral separation
59
of Dpx enantiomers via cyclodextrin-modified capillary electrophoresis [7] along with a
60
synthetic procedure for racemic Dpx following the main literature methods. In this
61
communication, the identification of a by-product is discussed, which was formed in the last
62
step of a common synthetic pathway using in situ mesylation for dapoxetine synthesis [6, 8-
63
15] (Figure 1.).
64
This well-characterized and widely accepted scheme for Dpx synthesis utilizes mesylate as an
65
excellent leaving group in nucleophilic substitution reactions. The mesylate intermediate (5)
66
is formed in situ and usually converted to Dpx without isolation. There is a single reference
67
reporting the characteristics of this intermediate [16], however the isolation of this compound
68
is missing. Several attempts failed to isolate compound 5, but provided an unknown by-
69
product.
70
The goal of this communication is the structure elucidation of this new by-product using
71
NMR spectroscopy and mass spectrometry. Following the structural characterization of this
72
impurity, a mechanism for the formation was also proposed. For the separation of Dpx and the
73
process-related impurities, a gradient HPLC method was developed.
(S)-N,N-dimethyl[3-(naphthalen-1-yloxy)-1-phenylpropyl]amine
Ac ce p
te
d
M
an
us
cr
ip t
(Dpx),
74 75
2.
Materials and methods
76
2.1.
Instrumentation
77
All NMR experiments were carried out on a 600 MHz Varian DDR NMR spectrometer
78
equipped with a 5 mm inverse-detection gradient (IDPFG) probehead. Standard pulse
79
sequences and processing routines available in VnmrJ 3.2 C/Chempack 5.1 were used for 3 Page 3 of 18
80
structure identifications. The complete resonance assignments were established from direct
81
1
82
gCOSY, 1H–13C gHSQCAD (J = 140 Hz), 1H–13C gHMBCAD (J = 8 Hz) experiments,
83
respectively. The probe temperature was maintained at 298 K and standard 5 mm NMR tubes
84
were used. The 1H chemical shifts were referenced to TMS (0.00 ppm) while
85
shifts were referenced to the applied NMR solvent CDCl3 (77.16 ppm).
86
The accurate mass of the products were determined with an Agilent 6230 time-of-flight mass
87
spectrometer. Samples were introduced by the Agilent 1260 Infinity LC system, the mass
88
spectrometer was operated in conjunction with a JetStream (ESI) ion source in positive ion
89
mode. Reference masses of m/z 121.050873 and 922.009798 were used to calibrate the mass
90
axis during analysis. Mass spectra were processed using Agilent MassHunter B.02.00
91
software.
92
HPLC-DAD analyses were performed using an Agilent 1260 Infinity LC apparatus. An
93
Agilent Zorbax Eclipse Plus C18, 100 mm x 4.6 mm i.d. (3.5 μm particle size) column was
94
applied. The mobile phase consisted of methanol and water with 0.1 v/v% formic acid using
95
the following gradient program: 0 min. 50% MeOH, 4 min. 50% MeOH, 6 min. 90% MeOH,
96
10.0 min. 90% MeOH, 11 min. 50% MeOH, 15 min 50% MeOH. The flow rate was set to 1
97
ml/min, while the column temperature was kept at 50°C. The UV spectra were recorded
98
ranging from 200-400 nm.
H–13C, long-range 1H–13C, and scalar spin–spin connectivities using 1D 1H,
13
C chemical
2.2.
101
All reagents for synthesis, HPLC grade solvents used for LC-MS analyses and CDCl3 (99.8
102
atom% D) for NMR were purchased from Sigma-Aldrich. Water was produced by a Millipore
103
Milli-Q Direct 8 water purifying system. TLC was performed using precoated Silica gel 60
104
F254 TLC plates and visualized with ultraviolet light at 254 nm. For column chromatography,
105
the 40-63 ! m silica was used.
Ac ce p
100
106
Chemicals
te
99
d
M
an
us
cr
ip t
13
C, 1H–1H
107
2.3.
Synthesis of 3-chloro-1-phenyl-1-propanol (3)
108
To the solution of 3-chloropropriophenone (1 g; 6 mmol) in THF (14 ml) and water (1 ml),
109
NaBH4 (0.27 g; 7 mmol) was added at 0°C. The mixture was stirred for 24 h at room
110
temperature.
111
After completion (checked by TLC using CH2Cl2, starting material Rf = 0.77, product Rf =
112
0.48), dilute acetic acid (12%) was added slowly to the reaction mixture with stirring at the
113
same temperature to set the pH to 4.5. The crude mixture was treated with water (10 ml), 4 Page 4 of 18
114
extracted with ethyl acetate (2 x 150 ml). The separated organic layers were combined and
115
washed with 5 % NaHCO3 solution. The organic layer was washed with water and dried over
116
sodium sulfate and filtered, followed by solvent evaporation.
117
The crude product was the racemic 3-chloro-1-phenyl-1-propanol (3, 1.02 g, 99%) as yellow
118
oil and subsequently crystallized overnight.
119
1
120
8.5, 4.7 Hz, 1H), 3.78 – 3.70 (m, 1H), 3.59 – 3.53 (m, 1H), 2.28 – 2.20 (m, 1H), 2.13 – 2.06
121
(m, 1H). 13C NMR (CDCl3, 151 MHz): δ(ppm): 143.83, 128.80, 128.05, 125.90, 71.47, 41.85,
122
41.58. HRMS: calc. [M+Na]+ 193.0391, found [M+Na]+ 193.0396.
us
123
cr
ip t
H NMR (CDCl3, 600 MHz): δ(ppm): 7.39 – 7.34 (m, 4H), 7.33 – 7.29 (m, 1H), 4.94 (dd, J =
2.4.
Synthesis of 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4)
125
Operating under inert atmosphere of nitrogen gas, to a mixture of DMF (2.3 ml) and 60%
126
sodium hydride in mineral oil (0.12 g, 2.9 mmol) at 0° C in an ice bath a solution of 1-
127
naphthol (0.43 g, 2.9 mmol) of DMF (2.3 ml) was added dropwise. The reaction mixture was
128
stirred at 0° C for two hours and a solution of 3 (0.53 g, 3.1 mmol) in DMF (2.1 ml) was
129
added. After stirring overnight at room temperature and completion (checked by TLC using n-
130
hexane and EtOAc (9:1, v/v), starting material Rf = 0.21, product Rf = 0.14), the reaction
131
mixture was poured into water (10 ml) and extracted with ethyl acetate (3 x 25 ml). The
132
extracts were combined and washed with water (30 ml) and sodium hydroxide solution (2 x
133
20 ml, 1 N), dried over anhydrous sodium sulfate and concentrated under reduced pressure.
134
The 0.84 g crude product was purified by column chromatography using n-hexane and EtOAc
135
(9:1, v/v) as an eluent, to afford purified 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4, 0.58 g,
136
70.2%).
137
1
138
(m, 2H), 7.47 – 7.42 (m, 3H), 7.41 – 7.35 (m, 3H), 7.31 (t, J = 7.3 Hz, 1H), 6.81 (d, J = 7.6
139
Hz, 1H), 5.14 (dd, J = 8.2, 4.9 Hz, 1H), 4.38 – 4.33 (m, 1H), 4.23 – 4.18 (m, 1H), 2.46 – 2.39
140
(m, 1H), 2.37 – 2.30 (m, 1H). 13C NMR (CDCl3, 151 MHz): δ(ppm): 154.57, 144.30, 134.63,
141
128.74, 127.84, 127.67, 126.54, 126.00, 125.96, 125.71, 125.40, 121.96, 120.55, 104.93,
142
72.28, 65.51, 38.65. HRMS: calc. [M+Na]+ 301.1199, found [M+Na]+ 301.1204.
Ac ce p
te
d
M
an
124
H NMR (CDCl3, 600 MHz): δ(ppm): 8.29 – 8.24 (m, 1H), 7.85 – 7.80 (m, 1H), 7.54 – 7.48
143 144
2.5.
Synthesis of 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran (6)
145
To a solution of 4 (0.54 g, 1.9 mmol) in CH2Cl2 (8 ml) triethylamine (0.29 g, 2.9 mmol) and
146
4-dimethylaminopyridine (DMAP) (1.02 mg, 8.2 μmol) were added under nitrogen
147
atmosphere. The reaction mixture was cooled to 0 °C. Methanesulfonyl chloride (0.29 g, 2.5 5 Page 5 of 18
mmol) was added dropwise to the reaction mixture and stirred at 0°C for 2 h. The reaction
149
mixture was stirred at room temperature for additional 21 hours. After completion (checked
150
by TLC using n-hexane and EtOAc (9:1, v/v), starting material Rf = 0.12, product Rf = 0.64),
151
the reaction mixture was added to CH2Cl2 (30 ml) and extracted with 0.5 M HCl (2 x 20 ml),
152
saturated sodium bicarbonate solution (2 x 20 ml), washed with water (2 x 20 ml), dried over
153
anhydrous sodium sulfate and concentrated under reduced pressure. The 0.65 g crude product
154
was purified by column chromatography using n-hexane and dichloromethane (9:1, v/v) as an
155
eluent, to afford purified 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran (6, 0.34 g, 67.3%).
156
HRMS: calc. [M+H]+ 261.1274, found [M+H]+ 261.1282. NMR spectra are shown in
157
Supplementary Materials, 1H and 13C chemical shifts are reported in Table 1.
us
cr
ip t
148
158 3.
Results and discussion
160
3.1.
Detection of by-product
161
The synthesis of dapoxetine usually follows the process route shown in Scheme 1. The
162
advantage of this procedure over other methods, is that it also offers the possibility to easily
163
prepare N-demethylated derivatives, the major metabolites of the parent compound [17].
164
Using methylamine or ammonia in the last step will result in the active metabolites
165
desmethyldapoxetine or didesmethyldapoxetine, respectively. Our aim was to isolate (for the
166
first time) the mesylate intermediate (5) for further synthetic purposes. Quenching the
167
mesylation and allowing the mixture to warm up to room temperature resulted in a less polar
168
product compared to the starting material (4) instead of the expected mesylate. The unknown
169
product was isolated by column chromatography and characterized by mass spectrometry and
170
various NMR techniques for structure identification.
171
Ac ce p
te
d
M
an
159
172
3.2.
Structure elucidation of the by-product
173
Following the isolation, accurate mass measurement was performed to determine the
174
elemental compositions of the by-product. High-resolution positive ion mode ESI-MS
175
spectrum of the unknown revealed [M+H]+ at m/z 261.1282 suggesting a molecular formula
176
[M] of C19H16O. The error between observed mass and theoretical mass of [M+H]+ was below
177
3 ppm. This formula also coincide with the main product ion of dapoxetine, indicating a
178
neutral loss of the dimethylamine moiety [M]+ [18, 19]. The MS data clearly show that
179
position 4 is crucial in the new compound, however it can not be excluded that the formula of
180
C19H17O [M]+ arose from the in-source dissociation of the methylsulfonyl moiety.
6 Page 6 of 18
In order to unequivocally prove the lack of the methylsulfonyl group and to elucidate the
182
structure of the by-product, NMR experiments were applied.
183
The 1H NMR spectrum of the isolated compound confirmed the lack of the methylsulfonyl
184
moiety as methyl singlet was not detected in the aliphatic region. The chemical shift of the
185
aliphatic methine hydrogen decreased by 0.84 ppm, presumably due to the loss of the
186
electron-withdrawing hydroxyl group. Instead of the twelve aromatic protons expected for
187
compound 5 only eleven were found, suggesting that the aromatic part is also involved in the
188
structural change.
189
The Attached Proton Test (APT) NMR experiment is a common way to assign CH
190
multiplicities in 13C NMR spectra as it provides the information on all sorts of carbons within
191
one experiment. The APT spectrum revealed an extra quaternary carbon at 117.78 ppm and
192
the lack of an aromatic CH carbon. All these spectral changes suggested a cyclization in
193
which a new carbon-carbon bond was formed between the alicyclic C-4 and the aromatic C-5
194
carbons. The hypothesized ring-closure was confirmed by HMBC experiment.
195
Figure 2 shows the key long-range correlations between the aliphatic H-3 methylene protons
196
and the aromatic C-5 carbon (3JC,H) as well as the aliphatic H-4 methine resonance and the
197
aromatic C-5 (2JC,H) and C-14 carbon (3JC,H) confirming the formation of a new carbon-
198
carbon bond. The complete resonance assignment of the by-product (see Table 1.)
199
unequivocally proved the formation of the tricyclic 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran
200
structure.
Ac ce p
201
te
d
M
an
us
cr
ip t
181
202
3.3.
Formation of the new by-product
203
The in situ prepared intermediate, phenyl-propanediol-naphthyl ether mesylate plays a critical
204
role in the formation of the tricyclic naphtopyrane derivative. At low temperature the rate of
205
the by-product formation is low, however the rate increases with rising temperature. The
206
preparation of fused aromatic pyran derivatives are often accomplished by various transition
207
or noble metal catalysts' assisted Friedel-Crafts reactions under vigorous reaction conditions
208
[20]. The mild reaction conditions of the cyclization process in our case can obviously be
209
explained with the chemical properties of compound 5. In the facile and effective ring closure
210
of phenyl-propanediol-naphthyl ether mesylate the lone electron pair of the ethereal oxygen is
211
crucial (Fig. 3).
212
The conjugation between the electron pairs of the ethereal oxygen and the naphthalene moiety
213
increases the electron density of the aromatic system, thereby increasing its nucleophilic 7 Page 7 of 18
reactivity. Concurrently, the electrophilicity of the benzyl methylene increases due to the
215
adjacent phenyl ring and the strongly polarized mesyl group, facilitating the direct interaction
216
with the naphthalene and the splitting of the mesylate. The electrophilic aromatic substitution
217
process results in a cyclization product in which the developed positive charge can not only
218
lead to a temporary high-energy carbocation intermediate, but forming an oxonium
219
intermediate and thereby opening the possibility of a relatively low energy transition for the
220
cyclization. The pyranium intermediate rearomatizes in the last step and yields the neutral 4-
221
phenyl-naphtopyran derivative. This proposed mechanism was further supported by
222
performing the same reaction using optically active alcohol as the starting material. The
223
isolated product was proved to be racemic by circular dichroism measurements corroborating
224
the proposed mechanism in Figure 3. The new by-product is not only generated in the course
225
of the isolation of the mesylate. This impurity is always present in raw dapoxetine batches as
226
the ring closure competes with the nucleophilic substitution especially at room temperature.
an
us
cr
ip t
214
227 3.4.
HPLC-DAD analysis of the new process-related impurity
229
HPLC is certainly the most popular method in impurity profiling of active pharmaceutical
230
ingredients (APIs). Recent advances in this technology allow fast and robust separation and
231
quantification of API impurities often hyphenated with sensitive and selective spectral
232
methods. The newly isolated and identified 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran is a
233
potential process-related impurity of Dpx. As Dpx is not yet official in any pharmacopoeia,
234
impurity profiling of the API is rather incomplete. There is a single literature describing an
235
HPLC method for the separation of process-related impurities of Dpx (compound 4 and 5)
236
[16]. To promote the impurity profiling of Dpx, a gradient HPLC-UV method was developed
237
for the separation of the new tricyclic compound (6) and the known impurity of 4 and
238
dapoxetine. Figure 4 shows the optimized separation of the impurities (for details see section
239
2.1)
240
The performance of the developed method was characterized as follows. Calibration curves
241
were constructed with 7 concentrations in the range of 1 μg/ml through 1000 μg/ml. Each
242
concentration was injected three times. The regression was calculated by the method of least
243
squares. The LOD of the compounds were determined as the concentration yielding a signal
244
three times the noise of the baseline, while the LOQ as ten times the noise. Table 2
245
summarizes the linearity and the LOD and LOQ values of the three compounds.
Ac ce p
te
d
M
228
246 247
4.
Conclusion 8 Page 8 of 18
248
A novel by-product has been identified in the synthetic pathway of dapoxetine. The new
249
compound was isolated, identified, and characterized using high-resolution MS and NMR
250
techniques. The 1H and
251
were completely assigned. Reaction mechanism for the formation of the new impurity was
252
proposed and also confirmed by synthesis. To advance the impurity profiling of dapoxetine, a
253
reversed-phase gradient HPLC method was also developed.
C NMR resonances of the new impurity, 4-phenyl-naphtopyran
ip t
13
254 255
5.
256
The financial support from OTKA PD109373 is highly appreciated. Sz. Béni thanks the
257
Hungarian Academy of Sciences for the financial support under the János Bolyai Research
258
Scholarship.
us
cr
Acknowledgment
an
259 2606. References
[1] N.B. Modi, M.J. Dresser, M. Simon, D. Lin, D. Desai, S. Gupta, Single- and multiple-dose
262
pharmacokinetics of dapoxetine hydrochloride, a novel agent for the treatment of premature
263
ejaculation, J. Clin. Pharm. 46 (2006) 301-309.
264
[2] D.W. Robertson, D.C. Thompson, D.T. Wong, 1-phenyl-3-naphthalenyloxypropanamines and
265
their use as selective serotonin reuptake inhibitors, Eli Lilly and Co., 1992, US5135947A.
266
[3] Q.-h. Jiang, Y.-x. Xu, Y. Yang, H.-p. Mu, R. Wan, The progress in the synthetic methods of
267
dapoxetine, Chinese J. Med. Chem. 23 (2013) 417-421.
268
[4] P. You, J. Qiu, E. Su, D. Wei, Carica papaya Lipase Catalysed Resolution of β-Amino Esters for the
269
Highly Enantioselective Synthesis of (S)-Dapoxetine, Eur. J. Org. Chem. 2013 (2013) 557-565.
270
[5] K. Lin, Process for synthesis of Dapoxetine from trans-cinnamaldehyde and N-Cbz-
271
hydroxylamine, in, Hunan Ouya Biological Co., Ltd., Peop. Rep. China . 2013, pp. 13.
272
[6] G.L. Khatik, R. Sharma, V. Kumar, M. Chouhan, V.A. Nair, Stereoselective synthesis of (S)-
273
dapoxetine: A chiral auxiliary mediated approach, Tetrahedron Lett. 54 (2013) 5991-5993.
274
[7] G. Neumajer, T. Sohajda, A. Darcsi, G. Tóth, L. Szente, B. Noszál, S. Béni, Chiral recognition of
275
dapoxetine
276
electrophoresis method, J. Pharm. Biomed. Anal. 62 (2012) 42-47.
277
[8] R.K. Rapolu, V.V.N. Kali, V.P. Raju, G.M. Reddy, J. Pasha, N.R. Nevuluri, R. Bandichhor, S.
278
Oruganti, Short enantioselective routes to (S)-Dapoxetine, Chem. Biol. Interface, 3 (2013) 50-60.
Ac ce p
te
d
M
261
enantiomers
with
methylated-gamma-cyclodextrin:
A
validated
capillary
9 Page 9 of 18
[9] K.P. Mohan Rao Dodda, Jithender Aadepu, Solid dapoxetine, in, 2013.
280
[10] M. Sasikumar, M.D. Nikalje, Simple and efficient synthesis of (S)-dapoxetine, Synth. Commun.
281
42 (2012) 3061-3067.
282
[11] S. Rakesh, S.M.A. Hussain, A. Roy, A method for preparing (S)-(+)-N,N-dimethyl-3-(naphthalen-
283
1-yloxy)-1-phenylpropan-1-amine or its salt and intermediate thereof, in, R L Fine Chem, India .
284
2012, pp. 43.
285
[12] S.P.R. Nalla, D.V.K.R. Dandu, A process for the preparation of S-dapoxetine hydrochloride, in,
286
India. 2012, pp. 51.
287
[13] A.M. Dave, D.J. Patel, R. Kumar, S.D. Dwivedi, A process for preparing (+)-dapoxetine and its
288
salts, in, Cadila Healthcare Limited, India . 2008, pp. 16.
289
[14] P. Chen, G. Zhang, F. Yu, Process for preparation of dapoxetine hydrochloride, in, Shanghai
290
Maryao Chemical Technique Limited Company, Peop. Rep. China . 2006, pp. 12.
291
[15] C.A. Alt, R.L. Robey, M.E.E. Van, Intermediates to 1-phenyl-3-naphthalenyloxy-propanamines,
292
Eli Lilly and Co., 1994, US5292962A
293
[16] T.A. Rohith, S., Development and validation of high performance liquid chromatography
294
method for the determination of process related impurities in dapoxetine hydrochloride., Internat.
295
J. Res. Pharm. Chem. 3 (2013) 74-82.
296
[17] C.L. Hamilton, J.D. Cornpropst, Determination of dapoxetine, an investigational agent with the
297
potential for treating depression, and its mono- and di-desmethyl metabolites in human plasma
298
using column-switching high-performance liquid chromatography, J. Chromatogr. B, 612 (1993)
299
253-261.
300
[18] L. Li, M.-Y. Low, X. Ge, B.C. Bloodworth, H.-L. Koh, Isolation and structural elucidation of
301
dapoxetine as an adulterant in a health supplement used for sexual performance enhancement, J.
302
Pharm. Biomed. Anal. 50 (2009) 724-728.
303
[19] T.K. Kim, I.S. Kim, S.H. Hong, Y.K. Choi, H. Kim, H.H. Yoo, Determination of dapoxetine in rat
304
plasma by ultra-performance liquid chromatography–tandem mass spectrometry, J. Chromatogr.
305
B, 926 (2013) 42-46.
306
[20] V.C. Pandurang, A.D. Dattatray, A. Sudalai, Process for the production of 4-substituted
307
chromanes via gold catalysis, in, WO 2013/088455, 2013.
Ac ce p
te
d
M
an
us
cr
ip t
279
308 309 10 Page 10 of 18
310 311
Figure captions
312
Figure 1. Synthetic scheme of dapoxetine involving in situ mesylation in the last step
313 Figure 2. Inset of the 1H–13C gHMBCAD spectrum of 4-phenyl-2H,3H,4H-naphtho[1,2-
315
b]pyran showing the key long-range correlations for complete structure elucidation
ip t
314
cr
316
Figure 3. The proposed reaction mechanism for the formation of 4-phenyl-2H,3H,4H-
318
naphtho[1,2-b]pyran during dapoxetine synthesis
us
317 319
Figure 4. Reversed-phase HPLC-UV chromatograms of dapoxetine (Rt = 2.67 min) and the
321
impurities: 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4) (Rt = 8.20 min) and 4-phenyl-
322
2H,3H,4H-naphtho[1,2-b]pyran (6) (Rt = 9.45 min) at 0.5% (top) and 0.1% (bottom)
323
concentration levels recorded at 310 nm..
324
Tables
325
Table 1. The numbering and the complete 1H and
326
product of dapoxetin. The 1H chemical shifts were referenced to TMS (0.00 ppm) while
327
chemical shifts were referenced to the applied NMR solvent CDCl3 (77.16 ppm).
329 330
M
d
C resonance assignment of the new by13
te
13
C
Ac ce p
328
an
320
Table 2. The linearity and the detection and quantitation limit of the developed LC method.
11 Page 11 of 18
2 3
O1
16 15 18
16
4
7
9
5 14 13
17
10
12 11
ip t
330 331
8
6
17
13 APT H, δ (ppm, multiplicity) C, δ (ppm) CH2 4.39-4.35 (m, 1H) 63.95 CH2 4.35-4.32 (m, 1H) 63.95 CH2 3 2.47-2.41 (m, 1H) 31.97 CH2 31.97 2.18-2.12 (m, 1H) CH 4 4.30 (t, J = 6.1 Hz, 1H) 41.05 C 5 117.78 C 6 150.50 C 7 125.37 CH 8 8.28 – 8.19 (m, 1H) 121.85 CH 9 7.50-7.44 (m, 1H) 125.44 CH 10 7.50-7.44 (m, 1H) 126.15 CH 11 7.78 – 7.71 (m, 1H) 127.57 C 12 133.61 CH 13 7.28 (d, J = 8.5 Hz, 1H) 119.79 CH 14 6.95 (d, J = 8.5 Hz, 1H) 128.45 C 15 146.04 2CH 16 7.15 (d, J = 7.2 Hz, 2H) 128.90 2CH 17 7.29 (t, J = 7.4 Hz, 2H) 128.57 CH 18 7.22 (t, J = 7.2 Hz, 1H) 126.58 1 13 H and C assignments: d, doublet; t, triplet; m, multiplet; J: coupling constants given in Hertz.
1
Ac ce p
332 333 334 335
te
d
M
an
us
cr
Position 2
12 Page 12 of 18
335 Dpx y = 0.2957x + 1.306 0.9998 0.51 1.7
compound 4 y = 0.4239x + 1.637 0.9999 0.47 1.56
compound 6 y = 0.5844x + 2.551 0.9997 0.44 1.46
ip t
regression determination coeff. LOD (! g/ml) LOQ (! g/ml)
Ac ce p
te
d
M
an
us
cr
336 337
13 Page 13 of 18
us
cr
i
*Graphical Abstract
O
MsCl Et3N
S
O
O
THF
ce pt
ed
0-5°C
Ac
O
CH3
M an
O
OH
(CH3)2NH
N
CH3
O
THF 0-5°C
Dapoxetine O
SEAr
SN2
H3C
-O S
CH3 O
rt °C !!! O
Dapoxetine impurity Page 14 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1.
Page 15 of 18
cr Ac
ce
pt
ed
M
an
us
Figure 2.
Page 16 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3.
Page 17 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4.
Page 18 of 18
S0731-7085(14)00184-8 http://dx.doi.org/doi:10.1016/j.jpba.2014.04.002 PBA 9528
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
18-2-2014 31-3-2014 3-4-2014
Please cite this article as: http://dx.doi.org/10.1016/j.jpba.2014.04.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Structure elucidation of a process-related impurity of dapoxetine
2 András Darcsi, Gergő Tóth, József Kökösi, Szabolcs Béni*
3 4
Department of Pharmaceutical Chemistry, Semmelweis University
6
H-1092 Hőgyes Endre street 9. Budapest, Hungary
7
cr
8 9
us
10 11
an
12 13 14
M
15 16
ip t
5
Highlights
d
17! ! A novel dapoxetine by-product has been identified using NMR and MS. 18! ! The mechanism for the impurity formation was proposed and also confirmed by synthesis. 20 21 22 23 24
Ac ce p
te
19! ! The separation of dapoxetine and the process-related impurities was accomplished by HPLC.
25
*
26
Dr. Szabolcs Béni
27
Tel.: +36 1 217 0891; fax: +36 1 217 0891; E-mail address:
28
[email protected]
Corresponding author:
29
1 Page 1 of 18
Abstract
30
Unknown by-product associated with the synthesis of dapoxetine was isolated. The structure
31
elucidation of this new compound using accurate mass data and NMR spectroscopy is
32
presented herein. The unambiguous resonance assignment concluded to the formation of a
33
tricyclic compound 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran, a new impurity of dapoxetine
34
which has never been reported previously. A proposed mechanism for the formation of the
35
new carbon-carbon bond is discussed. For the separation of dapoxetine and the process-
36
related impurities, a gradient HPLC method was developed.
cr
37
ip t
29
Keywords: Priligy, impurity profiling, NMR, HPLC-DAD, electrophilic aromatic
39
substitution, ring closure mechanism
us
38
41
an
40 Highlights
42! ! A novel dapoxetine by-product has been identified using NMR and MS.
M
43! ! The mechanism for the impurity formation was proposed and also confirmed by synthesis. 44! ! The separation of dapoxetine and the process-related impurities was accomplished by HPLC.
Ac ce p
te
d
45
2 Page 2 of 18
46 47 1.
1. Introduction
48
Dapoxetine
49
hydrochloride, Priligy®) is a novel short acting selective serotonin reuptake inhibitor (SSRI)
50
that is being developed specifically as an on-demand oral treatment of premature ejaculation
51
with a unique pharmacokinetic profile [1]. Dpx attains its peak plasma concentration in about
52
1.5 hours after dosing which is much faster than conventional SSRIs and by 24 hours the
53
plasma concentration decreases to approximately 5% of the peak concentration. These
54
pharmacokinetic properties make Dpx an excellent candidate for on-demand treatment of
55
premature ejaculation. The eutomer (S)-Dpx is 3.5 times more potent SSRI than (R)-Dpx, that
56
is why Dpx is marketed as a single enantiomer drug [2]. A wide range of synthetic procedures
57
were developed to synthesize racemic and enantiopure Dpx [3-7].
58
Our previous study reported a robust, sensitive and validated method for the chiral separation
59
of Dpx enantiomers via cyclodextrin-modified capillary electrophoresis [7] along with a
60
synthetic procedure for racemic Dpx following the main literature methods. In this
61
communication, the identification of a by-product is discussed, which was formed in the last
62
step of a common synthetic pathway using in situ mesylation for dapoxetine synthesis [6, 8-
63
15] (Figure 1.).
64
This well-characterized and widely accepted scheme for Dpx synthesis utilizes mesylate as an
65
excellent leaving group in nucleophilic substitution reactions. The mesylate intermediate (5)
66
is formed in situ and usually converted to Dpx without isolation. There is a single reference
67
reporting the characteristics of this intermediate [16], however the isolation of this compound
68
is missing. Several attempts failed to isolate compound 5, but provided an unknown by-
69
product.
70
The goal of this communication is the structure elucidation of this new by-product using
71
NMR spectroscopy and mass spectrometry. Following the structural characterization of this
72
impurity, a mechanism for the formation was also proposed. For the separation of Dpx and the
73
process-related impurities, a gradient HPLC method was developed.
(S)-N,N-dimethyl[3-(naphthalen-1-yloxy)-1-phenylpropyl]amine
Ac ce p
te
d
M
an
us
cr
ip t
(Dpx),
74 75
2.
Materials and methods
76
2.1.
Instrumentation
77
All NMR experiments were carried out on a 600 MHz Varian DDR NMR spectrometer
78
equipped with a 5 mm inverse-detection gradient (IDPFG) probehead. Standard pulse
79
sequences and processing routines available in VnmrJ 3.2 C/Chempack 5.1 were used for 3 Page 3 of 18
80
structure identifications. The complete resonance assignments were established from direct
81
1
82
gCOSY, 1H–13C gHSQCAD (J = 140 Hz), 1H–13C gHMBCAD (J = 8 Hz) experiments,
83
respectively. The probe temperature was maintained at 298 K and standard 5 mm NMR tubes
84
were used. The 1H chemical shifts were referenced to TMS (0.00 ppm) while
85
shifts were referenced to the applied NMR solvent CDCl3 (77.16 ppm).
86
The accurate mass of the products were determined with an Agilent 6230 time-of-flight mass
87
spectrometer. Samples were introduced by the Agilent 1260 Infinity LC system, the mass
88
spectrometer was operated in conjunction with a JetStream (ESI) ion source in positive ion
89
mode. Reference masses of m/z 121.050873 and 922.009798 were used to calibrate the mass
90
axis during analysis. Mass spectra were processed using Agilent MassHunter B.02.00
91
software.
92
HPLC-DAD analyses were performed using an Agilent 1260 Infinity LC apparatus. An
93
Agilent Zorbax Eclipse Plus C18, 100 mm x 4.6 mm i.d. (3.5 μm particle size) column was
94
applied. The mobile phase consisted of methanol and water with 0.1 v/v% formic acid using
95
the following gradient program: 0 min. 50% MeOH, 4 min. 50% MeOH, 6 min. 90% MeOH,
96
10.0 min. 90% MeOH, 11 min. 50% MeOH, 15 min 50% MeOH. The flow rate was set to 1
97
ml/min, while the column temperature was kept at 50°C. The UV spectra were recorded
98
ranging from 200-400 nm.
H–13C, long-range 1H–13C, and scalar spin–spin connectivities using 1D 1H,
13
C chemical
2.2.
101
All reagents for synthesis, HPLC grade solvents used for LC-MS analyses and CDCl3 (99.8
102
atom% D) for NMR were purchased from Sigma-Aldrich. Water was produced by a Millipore
103
Milli-Q Direct 8 water purifying system. TLC was performed using precoated Silica gel 60
104
F254 TLC plates and visualized with ultraviolet light at 254 nm. For column chromatography,
105
the 40-63 ! m silica was used.
Ac ce p
100
106
Chemicals
te
99
d
M
an
us
cr
ip t
13
C, 1H–1H
107
2.3.
Synthesis of 3-chloro-1-phenyl-1-propanol (3)
108
To the solution of 3-chloropropriophenone (1 g; 6 mmol) in THF (14 ml) and water (1 ml),
109
NaBH4 (0.27 g; 7 mmol) was added at 0°C. The mixture was stirred for 24 h at room
110
temperature.
111
After completion (checked by TLC using CH2Cl2, starting material Rf = 0.77, product Rf =
112
0.48), dilute acetic acid (12%) was added slowly to the reaction mixture with stirring at the
113
same temperature to set the pH to 4.5. The crude mixture was treated with water (10 ml), 4 Page 4 of 18
114
extracted with ethyl acetate (2 x 150 ml). The separated organic layers were combined and
115
washed with 5 % NaHCO3 solution. The organic layer was washed with water and dried over
116
sodium sulfate and filtered, followed by solvent evaporation.
117
The crude product was the racemic 3-chloro-1-phenyl-1-propanol (3, 1.02 g, 99%) as yellow
118
oil and subsequently crystallized overnight.
119
1
120
8.5, 4.7 Hz, 1H), 3.78 – 3.70 (m, 1H), 3.59 – 3.53 (m, 1H), 2.28 – 2.20 (m, 1H), 2.13 – 2.06
121
(m, 1H). 13C NMR (CDCl3, 151 MHz): δ(ppm): 143.83, 128.80, 128.05, 125.90, 71.47, 41.85,
122
41.58. HRMS: calc. [M+Na]+ 193.0391, found [M+Na]+ 193.0396.
us
123
cr
ip t
H NMR (CDCl3, 600 MHz): δ(ppm): 7.39 – 7.34 (m, 4H), 7.33 – 7.29 (m, 1H), 4.94 (dd, J =
2.4.
Synthesis of 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4)
125
Operating under inert atmosphere of nitrogen gas, to a mixture of DMF (2.3 ml) and 60%
126
sodium hydride in mineral oil (0.12 g, 2.9 mmol) at 0° C in an ice bath a solution of 1-
127
naphthol (0.43 g, 2.9 mmol) of DMF (2.3 ml) was added dropwise. The reaction mixture was
128
stirred at 0° C for two hours and a solution of 3 (0.53 g, 3.1 mmol) in DMF (2.1 ml) was
129
added. After stirring overnight at room temperature and completion (checked by TLC using n-
130
hexane and EtOAc (9:1, v/v), starting material Rf = 0.21, product Rf = 0.14), the reaction
131
mixture was poured into water (10 ml) and extracted with ethyl acetate (3 x 25 ml). The
132
extracts were combined and washed with water (30 ml) and sodium hydroxide solution (2 x
133
20 ml, 1 N), dried over anhydrous sodium sulfate and concentrated under reduced pressure.
134
The 0.84 g crude product was purified by column chromatography using n-hexane and EtOAc
135
(9:1, v/v) as an eluent, to afford purified 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4, 0.58 g,
136
70.2%).
137
1
138
(m, 2H), 7.47 – 7.42 (m, 3H), 7.41 – 7.35 (m, 3H), 7.31 (t, J = 7.3 Hz, 1H), 6.81 (d, J = 7.6
139
Hz, 1H), 5.14 (dd, J = 8.2, 4.9 Hz, 1H), 4.38 – 4.33 (m, 1H), 4.23 – 4.18 (m, 1H), 2.46 – 2.39
140
(m, 1H), 2.37 – 2.30 (m, 1H). 13C NMR (CDCl3, 151 MHz): δ(ppm): 154.57, 144.30, 134.63,
141
128.74, 127.84, 127.67, 126.54, 126.00, 125.96, 125.71, 125.40, 121.96, 120.55, 104.93,
142
72.28, 65.51, 38.65. HRMS: calc. [M+Na]+ 301.1199, found [M+Na]+ 301.1204.
Ac ce p
te
d
M
an
124
H NMR (CDCl3, 600 MHz): δ(ppm): 8.29 – 8.24 (m, 1H), 7.85 – 7.80 (m, 1H), 7.54 – 7.48
143 144
2.5.
Synthesis of 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran (6)
145
To a solution of 4 (0.54 g, 1.9 mmol) in CH2Cl2 (8 ml) triethylamine (0.29 g, 2.9 mmol) and
146
4-dimethylaminopyridine (DMAP) (1.02 mg, 8.2 μmol) were added under nitrogen
147
atmosphere. The reaction mixture was cooled to 0 °C. Methanesulfonyl chloride (0.29 g, 2.5 5 Page 5 of 18
mmol) was added dropwise to the reaction mixture and stirred at 0°C for 2 h. The reaction
149
mixture was stirred at room temperature for additional 21 hours. After completion (checked
150
by TLC using n-hexane and EtOAc (9:1, v/v), starting material Rf = 0.12, product Rf = 0.64),
151
the reaction mixture was added to CH2Cl2 (30 ml) and extracted with 0.5 M HCl (2 x 20 ml),
152
saturated sodium bicarbonate solution (2 x 20 ml), washed with water (2 x 20 ml), dried over
153
anhydrous sodium sulfate and concentrated under reduced pressure. The 0.65 g crude product
154
was purified by column chromatography using n-hexane and dichloromethane (9:1, v/v) as an
155
eluent, to afford purified 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran (6, 0.34 g, 67.3%).
156
HRMS: calc. [M+H]+ 261.1274, found [M+H]+ 261.1282. NMR spectra are shown in
157
Supplementary Materials, 1H and 13C chemical shifts are reported in Table 1.
us
cr
ip t
148
158 3.
Results and discussion
160
3.1.
Detection of by-product
161
The synthesis of dapoxetine usually follows the process route shown in Scheme 1. The
162
advantage of this procedure over other methods, is that it also offers the possibility to easily
163
prepare N-demethylated derivatives, the major metabolites of the parent compound [17].
164
Using methylamine or ammonia in the last step will result in the active metabolites
165
desmethyldapoxetine or didesmethyldapoxetine, respectively. Our aim was to isolate (for the
166
first time) the mesylate intermediate (5) for further synthetic purposes. Quenching the
167
mesylation and allowing the mixture to warm up to room temperature resulted in a less polar
168
product compared to the starting material (4) instead of the expected mesylate. The unknown
169
product was isolated by column chromatography and characterized by mass spectrometry and
170
various NMR techniques for structure identification.
171
Ac ce p
te
d
M
an
159
172
3.2.
Structure elucidation of the by-product
173
Following the isolation, accurate mass measurement was performed to determine the
174
elemental compositions of the by-product. High-resolution positive ion mode ESI-MS
175
spectrum of the unknown revealed [M+H]+ at m/z 261.1282 suggesting a molecular formula
176
[M] of C19H16O. The error between observed mass and theoretical mass of [M+H]+ was below
177
3 ppm. This formula also coincide with the main product ion of dapoxetine, indicating a
178
neutral loss of the dimethylamine moiety [M]+ [18, 19]. The MS data clearly show that
179
position 4 is crucial in the new compound, however it can not be excluded that the formula of
180
C19H17O [M]+ arose from the in-source dissociation of the methylsulfonyl moiety.
6 Page 6 of 18
In order to unequivocally prove the lack of the methylsulfonyl group and to elucidate the
182
structure of the by-product, NMR experiments were applied.
183
The 1H NMR spectrum of the isolated compound confirmed the lack of the methylsulfonyl
184
moiety as methyl singlet was not detected in the aliphatic region. The chemical shift of the
185
aliphatic methine hydrogen decreased by 0.84 ppm, presumably due to the loss of the
186
electron-withdrawing hydroxyl group. Instead of the twelve aromatic protons expected for
187
compound 5 only eleven were found, suggesting that the aromatic part is also involved in the
188
structural change.
189
The Attached Proton Test (APT) NMR experiment is a common way to assign CH
190
multiplicities in 13C NMR spectra as it provides the information on all sorts of carbons within
191
one experiment. The APT spectrum revealed an extra quaternary carbon at 117.78 ppm and
192
the lack of an aromatic CH carbon. All these spectral changes suggested a cyclization in
193
which a new carbon-carbon bond was formed between the alicyclic C-4 and the aromatic C-5
194
carbons. The hypothesized ring-closure was confirmed by HMBC experiment.
195
Figure 2 shows the key long-range correlations between the aliphatic H-3 methylene protons
196
and the aromatic C-5 carbon (3JC,H) as well as the aliphatic H-4 methine resonance and the
197
aromatic C-5 (2JC,H) and C-14 carbon (3JC,H) confirming the formation of a new carbon-
198
carbon bond. The complete resonance assignment of the by-product (see Table 1.)
199
unequivocally proved the formation of the tricyclic 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran
200
structure.
Ac ce p
201
te
d
M
an
us
cr
ip t
181
202
3.3.
Formation of the new by-product
203
The in situ prepared intermediate, phenyl-propanediol-naphthyl ether mesylate plays a critical
204
role in the formation of the tricyclic naphtopyrane derivative. At low temperature the rate of
205
the by-product formation is low, however the rate increases with rising temperature. The
206
preparation of fused aromatic pyran derivatives are often accomplished by various transition
207
or noble metal catalysts' assisted Friedel-Crafts reactions under vigorous reaction conditions
208
[20]. The mild reaction conditions of the cyclization process in our case can obviously be
209
explained with the chemical properties of compound 5. In the facile and effective ring closure
210
of phenyl-propanediol-naphthyl ether mesylate the lone electron pair of the ethereal oxygen is
211
crucial (Fig. 3).
212
The conjugation between the electron pairs of the ethereal oxygen and the naphthalene moiety
213
increases the electron density of the aromatic system, thereby increasing its nucleophilic 7 Page 7 of 18
reactivity. Concurrently, the electrophilicity of the benzyl methylene increases due to the
215
adjacent phenyl ring and the strongly polarized mesyl group, facilitating the direct interaction
216
with the naphthalene and the splitting of the mesylate. The electrophilic aromatic substitution
217
process results in a cyclization product in which the developed positive charge can not only
218
lead to a temporary high-energy carbocation intermediate, but forming an oxonium
219
intermediate and thereby opening the possibility of a relatively low energy transition for the
220
cyclization. The pyranium intermediate rearomatizes in the last step and yields the neutral 4-
221
phenyl-naphtopyran derivative. This proposed mechanism was further supported by
222
performing the same reaction using optically active alcohol as the starting material. The
223
isolated product was proved to be racemic by circular dichroism measurements corroborating
224
the proposed mechanism in Figure 3. The new by-product is not only generated in the course
225
of the isolation of the mesylate. This impurity is always present in raw dapoxetine batches as
226
the ring closure competes with the nucleophilic substitution especially at room temperature.
an
us
cr
ip t
214
227 3.4.
HPLC-DAD analysis of the new process-related impurity
229
HPLC is certainly the most popular method in impurity profiling of active pharmaceutical
230
ingredients (APIs). Recent advances in this technology allow fast and robust separation and
231
quantification of API impurities often hyphenated with sensitive and selective spectral
232
methods. The newly isolated and identified 4-phenyl-2H,3H,4H-naphtho[1,2-b]pyran is a
233
potential process-related impurity of Dpx. As Dpx is not yet official in any pharmacopoeia,
234
impurity profiling of the API is rather incomplete. There is a single literature describing an
235
HPLC method for the separation of process-related impurities of Dpx (compound 4 and 5)
236
[16]. To promote the impurity profiling of Dpx, a gradient HPLC-UV method was developed
237
for the separation of the new tricyclic compound (6) and the known impurity of 4 and
238
dapoxetine. Figure 4 shows the optimized separation of the impurities (for details see section
239
2.1)
240
The performance of the developed method was characterized as follows. Calibration curves
241
were constructed with 7 concentrations in the range of 1 μg/ml through 1000 μg/ml. Each
242
concentration was injected three times. The regression was calculated by the method of least
243
squares. The LOD of the compounds were determined as the concentration yielding a signal
244
three times the noise of the baseline, while the LOQ as ten times the noise. Table 2
245
summarizes the linearity and the LOD and LOQ values of the three compounds.
Ac ce p
te
d
M
228
246 247
4.
Conclusion 8 Page 8 of 18
248
A novel by-product has been identified in the synthetic pathway of dapoxetine. The new
249
compound was isolated, identified, and characterized using high-resolution MS and NMR
250
techniques. The 1H and
251
were completely assigned. Reaction mechanism for the formation of the new impurity was
252
proposed and also confirmed by synthesis. To advance the impurity profiling of dapoxetine, a
253
reversed-phase gradient HPLC method was also developed.
C NMR resonances of the new impurity, 4-phenyl-naphtopyran
ip t
13
254 255
5.
256
The financial support from OTKA PD109373 is highly appreciated. Sz. Béni thanks the
257
Hungarian Academy of Sciences for the financial support under the János Bolyai Research
258
Scholarship.
us
cr
Acknowledgment
an
259 2606. References
[1] N.B. Modi, M.J. Dresser, M. Simon, D. Lin, D. Desai, S. Gupta, Single- and multiple-dose
262
pharmacokinetics of dapoxetine hydrochloride, a novel agent for the treatment of premature
263
ejaculation, J. Clin. Pharm. 46 (2006) 301-309.
264
[2] D.W. Robertson, D.C. Thompson, D.T. Wong, 1-phenyl-3-naphthalenyloxypropanamines and
265
their use as selective serotonin reuptake inhibitors, Eli Lilly and Co., 1992, US5135947A.
266
[3] Q.-h. Jiang, Y.-x. Xu, Y. Yang, H.-p. Mu, R. Wan, The progress in the synthetic methods of
267
dapoxetine, Chinese J. Med. Chem. 23 (2013) 417-421.
268
[4] P. You, J. Qiu, E. Su, D. Wei, Carica papaya Lipase Catalysed Resolution of β-Amino Esters for the
269
Highly Enantioselective Synthesis of (S)-Dapoxetine, Eur. J. Org. Chem. 2013 (2013) 557-565.
270
[5] K. Lin, Process for synthesis of Dapoxetine from trans-cinnamaldehyde and N-Cbz-
271
hydroxylamine, in, Hunan Ouya Biological Co., Ltd., Peop. Rep. China . 2013, pp. 13.
272
[6] G.L. Khatik, R. Sharma, V. Kumar, M. Chouhan, V.A. Nair, Stereoselective synthesis of (S)-
273
dapoxetine: A chiral auxiliary mediated approach, Tetrahedron Lett. 54 (2013) 5991-5993.
274
[7] G. Neumajer, T. Sohajda, A. Darcsi, G. Tóth, L. Szente, B. Noszál, S. Béni, Chiral recognition of
275
dapoxetine
276
electrophoresis method, J. Pharm. Biomed. Anal. 62 (2012) 42-47.
277
[8] R.K. Rapolu, V.V.N. Kali, V.P. Raju, G.M. Reddy, J. Pasha, N.R. Nevuluri, R. Bandichhor, S.
278
Oruganti, Short enantioselective routes to (S)-Dapoxetine, Chem. Biol. Interface, 3 (2013) 50-60.
Ac ce p
te
d
M
261
enantiomers
with
methylated-gamma-cyclodextrin:
A
validated
capillary
9 Page 9 of 18
[9] K.P. Mohan Rao Dodda, Jithender Aadepu, Solid dapoxetine, in, 2013.
280
[10] M. Sasikumar, M.D. Nikalje, Simple and efficient synthesis of (S)-dapoxetine, Synth. Commun.
281
42 (2012) 3061-3067.
282
[11] S. Rakesh, S.M.A. Hussain, A. Roy, A method for preparing (S)-(+)-N,N-dimethyl-3-(naphthalen-
283
1-yloxy)-1-phenylpropan-1-amine or its salt and intermediate thereof, in, R L Fine Chem, India .
284
2012, pp. 43.
285
[12] S.P.R. Nalla, D.V.K.R. Dandu, A process for the preparation of S-dapoxetine hydrochloride, in,
286
India. 2012, pp. 51.
287
[13] A.M. Dave, D.J. Patel, R. Kumar, S.D. Dwivedi, A process for preparing (+)-dapoxetine and its
288
salts, in, Cadila Healthcare Limited, India . 2008, pp. 16.
289
[14] P. Chen, G. Zhang, F. Yu, Process for preparation of dapoxetine hydrochloride, in, Shanghai
290
Maryao Chemical Technique Limited Company, Peop. Rep. China . 2006, pp. 12.
291
[15] C.A. Alt, R.L. Robey, M.E.E. Van, Intermediates to 1-phenyl-3-naphthalenyloxy-propanamines,
292
Eli Lilly and Co., 1994, US5292962A
293
[16] T.A. Rohith, S., Development and validation of high performance liquid chromatography
294
method for the determination of process related impurities in dapoxetine hydrochloride., Internat.
295
J. Res. Pharm. Chem. 3 (2013) 74-82.
296
[17] C.L. Hamilton, J.D. Cornpropst, Determination of dapoxetine, an investigational agent with the
297
potential for treating depression, and its mono- and di-desmethyl metabolites in human plasma
298
using column-switching high-performance liquid chromatography, J. Chromatogr. B, 612 (1993)
299
253-261.
300
[18] L. Li, M.-Y. Low, X. Ge, B.C. Bloodworth, H.-L. Koh, Isolation and structural elucidation of
301
dapoxetine as an adulterant in a health supplement used for sexual performance enhancement, J.
302
Pharm. Biomed. Anal. 50 (2009) 724-728.
303
[19] T.K. Kim, I.S. Kim, S.H. Hong, Y.K. Choi, H. Kim, H.H. Yoo, Determination of dapoxetine in rat
304
plasma by ultra-performance liquid chromatography–tandem mass spectrometry, J. Chromatogr.
305
B, 926 (2013) 42-46.
306
[20] V.C. Pandurang, A.D. Dattatray, A. Sudalai, Process for the production of 4-substituted
307
chromanes via gold catalysis, in, WO 2013/088455, 2013.
Ac ce p
te
d
M
an
us
cr
ip t
279
308 309 10 Page 10 of 18
310 311
Figure captions
312
Figure 1. Synthetic scheme of dapoxetine involving in situ mesylation in the last step
313 Figure 2. Inset of the 1H–13C gHMBCAD spectrum of 4-phenyl-2H,3H,4H-naphtho[1,2-
315
b]pyran showing the key long-range correlations for complete structure elucidation
ip t
314
cr
316
Figure 3. The proposed reaction mechanism for the formation of 4-phenyl-2H,3H,4H-
318
naphtho[1,2-b]pyran during dapoxetine synthesis
us
317 319
Figure 4. Reversed-phase HPLC-UV chromatograms of dapoxetine (Rt = 2.67 min) and the
321
impurities: 3-(1-naphthalenyloxy)-1-pheny-l-propanol (4) (Rt = 8.20 min) and 4-phenyl-
322
2H,3H,4H-naphtho[1,2-b]pyran (6) (Rt = 9.45 min) at 0.5% (top) and 0.1% (bottom)
323
concentration levels recorded at 310 nm..
324
Tables
325
Table 1. The numbering and the complete 1H and
326
product of dapoxetin. The 1H chemical shifts were referenced to TMS (0.00 ppm) while
327
chemical shifts were referenced to the applied NMR solvent CDCl3 (77.16 ppm).
329 330
M
d
C resonance assignment of the new by13
te
13
C
Ac ce p
328
an
320
Table 2. The linearity and the detection and quantitation limit of the developed LC method.
11 Page 11 of 18
2 3
O1
16 15 18
16
4
7
9
5 14 13
17
10
12 11
ip t
330 331
8
6
17
13 APT H, δ (ppm, multiplicity) C, δ (ppm) CH2 4.39-4.35 (m, 1H) 63.95 CH2 4.35-4.32 (m, 1H) 63.95 CH2 3 2.47-2.41 (m, 1H) 31.97 CH2 31.97 2.18-2.12 (m, 1H) CH 4 4.30 (t, J = 6.1 Hz, 1H) 41.05 C 5 117.78 C 6 150.50 C 7 125.37 CH 8 8.28 – 8.19 (m, 1H) 121.85 CH 9 7.50-7.44 (m, 1H) 125.44 CH 10 7.50-7.44 (m, 1H) 126.15 CH 11 7.78 – 7.71 (m, 1H) 127.57 C 12 133.61 CH 13 7.28 (d, J = 8.5 Hz, 1H) 119.79 CH 14 6.95 (d, J = 8.5 Hz, 1H) 128.45 C 15 146.04 2CH 16 7.15 (d, J = 7.2 Hz, 2H) 128.90 2CH 17 7.29 (t, J = 7.4 Hz, 2H) 128.57 CH 18 7.22 (t, J = 7.2 Hz, 1H) 126.58 1 13 H and C assignments: d, doublet; t, triplet; m, multiplet; J: coupling constants given in Hertz.
1
Ac ce p
332 333 334 335
te
d
M
an
us
cr
Position 2
12 Page 12 of 18
335 Dpx y = 0.2957x + 1.306 0.9998 0.51 1.7
compound 4 y = 0.4239x + 1.637 0.9999 0.47 1.56
compound 6 y = 0.5844x + 2.551 0.9997 0.44 1.46
ip t
regression determination coeff. LOD (! g/ml) LOQ (! g/ml)
Ac ce p
te
d
M
an
us
cr
336 337
13 Page 13 of 18
us
cr
i
*Graphical Abstract
O
MsCl Et3N
S
O
O
THF
ce pt
ed
0-5°C
Ac
O
CH3
M an
O
OH
(CH3)2NH
N
CH3
O
THF 0-5°C
Dapoxetine O
SEAr
SN2
H3C
-O S
CH3 O
rt °C !!! O
Dapoxetine impurity Page 14 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1.
Page 15 of 18
cr Ac
ce
pt
ed
M
an
us
Figure 2.
Page 16 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3.
Page 17 of 18
Ac
ce
pt
ed
M
an
us
cr
i
Figure 4.
Page 18 of 18
