Structural elucidation of novel degradation product of brotizolam

Article abstract of DOI:10.1016/j.bmcl.2013.04.044

When the drug product of brotizolam (1) is decomposed according to the interview form and patent, hydrolysate (2) is obtained, but its physicochemical data are still missing. To elucidate the structure based on a spectroscopic approach, the above-mentione

Full text of DOI:10.1016/j.bmcl.2013.04.044

Bioorganic & Medicinal Chemistry Letters 23 (2013) 3515–3518
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structural elucidation of novel degradation product of brotizolam
⇑
Yoshinori Sugimoto , Aya Kugimiya, Yoshifumi Yoshimura, Hideo Naoki
Sanyo Chemical Laboratory Co., Ltd, Tajii, Mihara-ku, Sakai-shi, Osaka 587-0012, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 February 2013
Revised 8 April 2013
Accepted 17 April 2013
Available online 28 April 2013
When the drug product of brotizolam (1) is decomposed according to the interview form and patent,
hydrolysate (2) is obtained, but its physicochemical data are still missing. To elucidate the structure
based on a spectroscopic approach, the above-mentioned degradation product was isolated and applied
to the structural analysis. As a result, the structure of decomposed product was found to be different from
that of 2. In this report, its isolation and structure elucidation by NMR and MS spectra are described.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Brotizolam
Degradation
Hydrolyzation
LC/MS
NMR
Brotizolam (1),1 2-bromo-4-(2-chlorophenyl)-9-methyl-6H-
thieno[3,2-f][1,2,4]triazolo[4,3-a][1,4]benzodiazepine, is a sleep-
inducing benzodiazepine drug, and is licensed as LendorminÒ by
Boehringer Ingelheim GmbH.2
spectrum showed a few peaks besides that of 1. Among them, we
focused on the LC/MS peak at m/z 410.9679 which showed the
isotope pattern including one Br and one Cl elements same as 2.
Because its production was at 0.45% area5 and the amount was
The hydrolysate (2) derived from 1 was reported as an impurity
of the drug product on the Interview Form (IF),3 and was produced
at ca. 0.8% by storage of the tablet at room temperature for 3 years,
and also according to the different patent, at 1.7–3.2% by the tablet
degradation at 60 °C and 75% relative humidity for 15 days4
(Fig. 1). However we were not able to find the physicochemical
data of 2 in the IF or patent literature.
Therefore, we attempted to decompose the drug product
according to the patent, to obtain 2, and to elucidate the structure
of the decomposed impurity based on a spectroscopic approach.
When the LendorminÒ tablet had been degraded for ca. 2 weeks
according to the above mentioned patent manner,4 LC/MS
Figure 1. Chemical structures of brotizolam (1), the hydrolysate (2), and the novel
degradation product (3).
Figure 2. HPLC chromatogram of (a) the degraded drug product under 60 °C and
75% RH, (b) the degraded drug product under 86 °C and ca. 100% RH, (c) the target
component isolated from the drug product degraded under 86 °C and 100% RH, (d)
the target component synthesized from BRT and (e) mixture of the isolated and
synthetic sample.
⇑
Corresponding author. Tel.: +81 72 361 6633; fax: +81 72 362 0670.
E-mail address: sugimoto@sanyokagaku.co.jp (Y. Sugimoto).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.04.044

3516
Y. Sugimoto et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3515–3518
Scheme 1. Preparation of 3 from BRT (1) via an equilibrium mixture of the hydrochloride (4) and 5.
Table 1
NMR assignments for 1, 3 and benzoylthiophene (6)
No.
1
3
6
1Ha
13Ca
1Hb
13Cb
1Hb
13Cb
1
2
3
3a
4
5
108.6
128.5
129.8
163.4
93.3
93.1
6.89
6.51
129.5
114.7
187.2
6.45
7.10
129.0
116.0
188.4
9.97
4.54
6
4.87
46.3
45.1
6a
7
152.4
158.2
8
9
149.5
154.8
10
10a
10
20
30
40
50
60
13.4
136.0
137.5
131.3
127.3
131.2
129.6
130.7
168.6
139.7
130.5
130.1
130.6
126.8
128.4
166.4
139.7
130.5
130.0
130.6
126.7
128.2
7.44
7.44
ꢀ
ꢁ
7:54 ꢀ 7:42
ð4HÞ
ꢀ
ꢁ
ꢀ
ꢁ
7:41 ꢀ 7:30
ð3HÞ
7:39 ꢀ 7:32
ð3HÞ
CH3
2.62
11.3
2.28
11.9
a
Both 1H and 13C chemical shifts (ppm) in DMSO-d6 were referenced to the tetramethylsilane (TMS) signal (0.00 ppm).
b
Both 1H and 13C chemical shifts (ppm) in CDCl3 were referenced to the tetramethylsilane (TMS) signal (0.00 ppm).
not sufficient to elucidate its chemical structure by NMR experi-
ment, we explored conditions for better yield (Fig. 2).
to be exchangeable via hydrogen–deuterium exchange experiment
with D2O. Meanwhile, in 13C NMR experiment, a total of 15 carbon
signals were observed as shown in Table 1.
The COSY spectrum indicated that the proton at 9.97 ppm (t,
1H) coupled with the methylene proton at 4.54 ppm (d, 2H), which
suggested that they were in the vicinal position. The protons at
7.44 (dd, 1H) and 7.41–7.30 (m, 3H) were assigned to the same
aromatic ring protons. The singlet proton at 6.51 ppm remained
to be determined.
When the degradation was carried out under the condition of
86 °C and ca. 100% RH for 40 h, the yield of the target component
significantly-increased (at ca. 8.6% area). Accordingly, 500 tablets
of LendorminÒ, containing 125 mg of 1, were degraded under the
same condition, extracted with ethyl acetate, and purified by
silica-column chromatography to give ca. 8 mg of the target
component (yield: ca. 6%; LC purity: ca. 95% area).6
Concurrently, we examined the synthesis of the target compo-
nent from BRT (1). Gallo et al. reported that the hydrolysis of 1 un-
der aqueous HCl solution gave an equilibrium mixture of the
hydrochlorides (4) and 5 inseparable from each other.7 Surpris-
ingly, when triethylamine (TEA) as a base affected the equilibrium
mixture to deprotonate the ammonium of 5, the target component
was obtained (yield: 52%) in a quantity sufficient for NMR mea-
surements8 (Scheme 1).
The HSQC spectrum indicated that all the carbon signals ob-
served at the following chemical shift values were classified as
The 1D and 2D NMR,9 and MS/MS spectra of the synthetic sam-
ple were confirmed to be identical to those of the isolated sample
from the degraded LendorminÒ tablet. Therefore, we conducted the
structural elucidation of the target component using the synthetic
sample obtained from BRT (1).
First, in 1H NMR experiment in CDCl3, a total of 12 proton sig-
nals were observed at 13.4 (br s, 1H), 9.97 (t, 1H), 7.44 (dd, 1H),
7.41–7.30 (m, 3H), 6.51 (s, 1H), 4.54 (d, 2H) and 2.28 ppm (s,
3H). Among those, the two protons at 13.4 and 9.97 ppm proved
Figure 3. The HMBC of 3.

Y. Sugimoto et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3515–3518
3517
However, because there was no HMBC related to C-2 carbon at
93.3 ppm, the INADEQUATE experiment was conducted. The corre-
lation shown in Figure 4 was observed, and C-2 carbon was found
to be adjacent to C-3 carbon (see Fig. 5).
Finally, the chemical shift value 93.3 ppm of C-2 carbon was
comparatively shifted to upfield from the corresponding signal of
BRT at 108.6 ppm. To confirm the validity of this upfield shift,
the benzoylthiophene (6) was prepared10 and its NMR spectra
were compared with those of 3 (Fig. 6). As a result, 1D NMR spectra
and 2D NMR correlations of 6 were considerably consistent with
those of the corresponding part of 3, which suggests that the up-
field chemical shift value of C-2 carbon is valid (Table 1).
Figure 4. INADEQUATE correlation of 3.
the quaternary carbons: 187.2, 168.6, 158.2, 154.8, 139.7, 130.5,
114.7 and 93.3 ppm. Moreover, the singlet proton at 6.51 ppm
was assigned to the carbon at 129.5 ppm, and the methyl and
methylene protons were assigned to the carbons at 11.9 and
45.1 ppm, respectively.
The cross peaks were observed in HMBC experiment as shown
in Figure 3. The H-10 in CDCl3 had a broadened peak on HMBC
spectrum, while, in DMSO-d6, the cross peaks were observed be-
tween H-10 and C-9, and C-6a. A series of the connectivity from
the methyl proton to H-6 was also confirmed.
Figure 6. The location numbers in 3 and benzoylthiophene (6).
Figure 5. 1H and 13C NMR spectra of compound (3) prepared from BRT (1) (in CDCl3).

3518
Y. Sugimoto et al. / Bioorg. Med. Chem. Lett. 23 (2013) 3515–3518
3. See the revised 10th edition published from Japanese Society of Hospital
Pharmacists.
4. Yamaguchi, T.; Shimada, T.; Kume, H.; Kitada, A.; Utigaki, M.; Okuda, N. JP
Patent 2,011,063,567, 2011.
5. LC–mass spectrometry: LC–MS system used was a Shimadzu LCMS–IT-TOF
equipped with Prominence UFLCXR (Shimadzu, Japan). Waters SunFire™ C18
(4.6 ꢂ 150 mm, 5
lm) column was used with the following gradient condition
of T (min)/%B (v/v): 0/40, 25/95, 35/95. The gradient involves two mobile
phases consisting of AcONH4 (10 mM) as solvent A and CH3CN as solvent B. The
LC–MS spectrum of the degradation product was performed with positive
electro spray ionization (ESI+) setting interface voltage at 4.5 kV and the
capillary temperature at 200 °C. The collision gas for MS/MS experiment was
used Ar gas.
6. Isolation of 3 from the degraded drug product: 500 Tablets of LendorminÒ 0.25-
mg made in Boehringer Ingelheim were grinded and powdered with a mortar,
and then set in the container and kept at 86 °C and ca. 100% RH for sum 40 h
(on this operation, you must pay attention to the sample, and, in some cases,
stir it with a spatula, lest it becomes a dark-brown hard lump). The resultant
milky-white powder was suspended in and extracted with ethyl acetate
(600 mL) twice. After filtered to remove insoluble matters, the ethyl acetate
layer was concentrated and purified by silica column chromatography (eluent:
hexane/ethyl acetate) to give 8 mg of 3.
Figure 7. MS/MS fragment pattern of 3.
The LC/MS spectra of 3 were indicated at m/z 410.9679, [M+H]+
and m/z 408.9541, [MꢁH]ꢁ respectively, and the positive ion MS/
MS spectrum of molecular related ion at m/z 410.9679 showed
these product ions at m/z 331.0443, 297.0780 and 138.9939,
respectively. The proposed fragmentation mechanism of these
product ions is given in Figure 7.
In conclusion, we succeeded in the isolation, synthesis, and
structural elucidation of the novel degradation product (3) that
had been known as one of the impurity in the degraded BRT. The
chemical structure of 3 was identified as 2-(5-methyl-4H-1,2,4-
triazol-3-yl)methylamino-3-(2-chlorobenzoyl)-5-bromothio-
phene, which is different from the structure of hydrolysate (2).
7. Gallo, B.; Alonso, R. M.; Lete, E.; Badía, M. D.; Patriarche, G. J.; Gelbcke, M. J.
Heterocycl. Chem. 1988, 25, 867.
8. Preparation of
3 via HCl-hydrolyzed solution of BRT: To BRT (1) (6.00 g,
15.2 mmol) in THF (60 mL), 1.3 N HCl aq (60 mL, 76.0 mmol) was added, and
then stirred at 60 °C for ca. 5 h. After cooled to rt, TEA (7.69 g, 76.0 mmol) was
added to the resultant solution and then stirred at rt overnight. The reaction
solution was concentrated under reduced pressure and extracted with ethyl
acetate. The organic layer was washed with water and satd. NaCl soln, dried
over Na2SO4, concentrated under reduced pressure and purified by silica
column chromatography (eluent: hexane/ethyl acetate) to give 3.27 g (52%) of
3 as greenish yellow amorphous. The 1H and 13C NMR spectra in CDCl3 were
shown in Figure 5.
9. NMR spectrometry: NMR measurements were performed at rt on Varian
400 MHz NMR spectrometer using CDCl3 and DMSO-d6 as solvent. Sample
concentration was 1 mg in 0.6 mL for 1H and 13C, and 300 mg in 0.7 mL for
inadequate experiment. The chemical shifts were referenced to TMS. All pulse
sequences were applied by using the standard spectrometer software package.
10. Preparation of 4: To (E)-7-bromo-5-(2-chlorophenyl)-1H-thieno[2,3-
e][1,4]diazepin-2(3H)-one (14.2 g, 40.0 mmol) in THF (142 mL) and H2O
(14.4 mL), p-toluenesulfonic acid-mono hydrate (6.89 g, 40.0 mmol) was
added at rt, and then stirred at 65 °C for 30 h. After cooled to 0 °C, the
reaction solution was quenched with TEA (8.10 g, 80.0 mol) and allowed to rt.
After that, the quenched solution was concentrated and purified by silica
column chromatography (eluent: hexane/ethyl acetate) to give 5.77 g (46%) of
benzoylthiophene 4 as brown solid.
Acknowledgment
We thank Professor Y. Ohfune (Osaka City University) for the
valuable discussions about the degradation mechanism.
References and notes
1. Weber, K.-H.; Bauer, A.; Langbein, A.; Daniel, H. Liebigs Ann. Chem. 1978, 1257.
2. Wikipedia page: Brotizolam. http://en.wikipedia.org/wiki/.