Accepted Manuscript Title: Quantitative analysis of quazepam and its metabolites in human blood, urine, and bile by liquid chromatography–tandem mass spectrometry Author: Jing Zhou Koji Yamaguchi Youkichi Ohno PII: DOI: Reference:
S0379-0738(14)00180-7 http://dx.doi.org/doi:10.1016/j.forsciint.2014.04.027 FSI 7586
To appear in:
FSI
Received date: Revised date: Accepted date:
4-2-2014 17-4-2014 27-4-2014
Please cite this article as: Jing Zhou, Koji Yamaguchi, Youkichi Ohno, Quantitative analysis of quazepam and its metabolites in human blood, urine, and bile by liquid chromatographyndashtandem mass spectrometry, Forensic Science International http://dx.doi.org/10.1016/j.forsciint.2014.04.027 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 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Case Study
19
*Corresponding author at: Department of Legal Medicine, Graduate School of Medicine, Nippon Medical
20
School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8602, Japan
21
Tel.: +81 3 3822 2131; fax: +81 3 5814 5680.
22 23 24 25
E-mail address:
[email protected] (K. Yamaguchi)
27 28 29 30 31 32 33 34 35 36 37 38 39
1
ip t
Department of Forensic Science, Zhejiang Police College, Hangzhou, PR China. Department of Legal Medicine, Graduate School of Medicine, Nippon Medical School, Tokyo, Japan
M
an
2
us
cr
Jing Zhou1, 2, **, Koji Yamaguchi2, *, **, Youkichi Ohno2
d
**these authors contributed equally to the work
Ac ce pt e
26
Quantitative analysis of quazepam and its metabolites in human blood, urine, and bile by liquid chromatography-tandem mass spectrometry
Highlights
► Quazepam and its metabolites in blood, urine, and bile were quantified by LC-MS/MS. ► Protein precipitation-solid phase extraction was applied to blood. ► Liquid-liquid extraction followed by PSA cleanup was applied to urine and bile. ► The concentration of total 3-hydroxy-2-oxoquazepam in bile was extremely high. ► The usefulness of the concentrations of drug metabolites was evaluated. Abstract Quazepam (QZP), which is a long-acting benzodiazepine-type hypnotic, and its 4 metabolites,
40
2-oxoquazepam,
N-desalkyl-2-oxoquazepam
(DOQ),
3-hydroxy-2-oxoquazepam
(HOQ),
and
41
3-hydroxy-N-desalkyl-2-oxoquazepam, in human blood, urine, and bile were quantitatively analyzed by
42
liquid chromatography-tandem mass spectrometry. The analytes were extracted from blood by protein
1
Page 1 of 28
precipitation followed by solid phase extraction, and from urine and bile by liquid-liquid extraction and
44
cleanup using a PSA solid phase extraction cartridge. This method was applied to a medico-legal autopsy
45
case, in which the deceased had been prescribed QZP approximately 3 weeks before his death. In blood, the
46
concentrations of free DOQ (160 ± 7 ng/mL for heart blood and 181 ± 12 ng/mL for femoral blood) were the
47
highest of all the analytes and in agreement with the concentration at a steady state. This indicates that the
48
deceased consecutively received QZP for at least several days until the concentrations reached approximately
49
the same level as that in the steady state. An extremely high concentration of total HOQ (the sum of
50
conjugated and free HOQ) in bile was also found (56,200 ± 1900 ng/mL). This accumulation of HOQ in bile
51
is probably due to enterohepatic circulation. This study demonstrates that the combination of the
52
concentrations of QZP and its metabolites in biological matrices can provide more information about the
53
amount and frequency of QZP administration.
an
us
cr
ip t
43
54 55
Keywords: Quazepam; Metabolites; Quantification; Liquid chromatography; Mass spectrometry
M
56
58
1. Introduction
d
57
Quantitative analysis of pharmaceuticals in biological matrices provides useful information about the
60
amount and frequency of drug intake. In a medico-legal autopsy, when the blood concentration of a parent
61
drug is at a lethal level, drug poisoning is considered, and when it is in the therapeutic range, other causes of
62
death should be considered. Drug metabolite concentrations, however, are not commonly quantified. In some
63
cases, the concentration of a metabolite is used as an alternative to its parent drug that has been extensively
64
metabolized or totally eliminated. However, the usefulness of the concentrations of drug metabolites in
65
forensic toxicology has not been evaluated fully.
Ac ce pt e
59
66
Quazepam (QZP, Fig. 1) is a long-acting, trifluorinated benzodiazepine with a hypnotic effect used in
67
the treatment of insomnia. It is sold under the brand name Doral by Mitsubishi Tanabe Pharma and ranked
68
third in terms of the sales volume of hypnotics in Japan [1]. We consider QZP to be an ideal drug to
69
demonstrate the usefulness of the concentration of drug metabolites because: (a) it belongs to
70
benzodiazepine-type hypnotics, which are often used in crimes such as drug-facilitated sexual assault; (b) it
71
is converted to several metabolites; and (c) pharmacokinetic parameters for some of the metabolites are
2
Page 2 of 28
72
available. The metabolism of QZP has been studied extensively since the 1980s [2–6]. As illustrated in Fig. 1,
73
QZP is first metabolized by substitution of oxygen for sulfur to give 2-oxoquazepam (OQ), which is then
74
transformed along two main pathways: (a) hydroxylation to give 3-hydroxy-2-oxoquazepam (HOQ) and (b)
75
N-dealkylation
76
3-hydroxy-N-desalkyl-2-oxoquazepam (HDOQ). A pharmacokinetic study of QZP, OQ, and DOQ after a
77
single oral administration of 14C-labeled QZP was reported by Zampaglione et al. [4]. QZP and OQ appeared
78
rapidly in plasma after dosing, whereas DOQ and HOQ glucuronide accounted for the majority of plasma
79
radioactivity at later times after dosing. In urine, HOQ glucuronide was the most abundant metabolite, and
80
the other metabolites present were glucuronides of DOQ and HDOQ. OQ and DOQ are active metabolites of
81
QZP, whereas HOQ glucuronide is pharmacologically inactive [6].
form
N-desalkyl-2-oxoquazepam
(DOQ),
which
is
further
hydroxylated
to
us
cr
ip t
to
A reliable quantitative method is required to evaluate the usefulness of drug metabolite concentrations in
83
various body fluids. However, to date, there are few reports on the quantitative analysis of QZP and its
84
metabolites in human plasma [2–5, 7–9] and only two studies in urine [4, 10], with no reports in bile. In the
85
present study, we developed extraction methods and quantitative analysis for QZP and its metabolites, OQ,
86
DOQ, HOQ, and HDOQ, in human blood, urine, and bile by liquid chromatography coupled with tandem
87
mass spectrometry (LC-MS/MS). This method was applied to a medico-legal autopsy case.
89
M
d
Ac ce pt e
88
an
82
2. Case history
90
A man in his early sixties was found dead at home. The cadaver was kept in a refrigerator and an
91
autopsy was held at 2.5 days after its discovery. The cause of death was thought to be ischemic heart disease,
92
and the interval between death and the autopsy was estimated at approximately 4 days according to the
93
autopsy findings. Heart blood, femoral blood, urine, and bile were collected for toxicological analysis. The
94
heart blood was obtained from the right atrium. All samples were stored in a freezer (-30ºC) until analysis.
95
No preservative agent was added to these samples. At approximately 3 weeks before death, the man was
96
prescribed central nervous system drugs including QZP, triazolam, levomepromazine, nitrazepam, etizolam,
97
and flunitrazepam. QZP was prescribed at one 20 mg tablet daily. All drugs prescribed were detected in the
98
deceased’s blood by LC-MS/MS and the concentrations were within therapeutic levels.
99 100
3. Materials and methods
3
Page 3 of 28
101
3.1 Reagents QZP, OQ, DOQ, and HOQ were donated by Hisamitsu Pharmaceutical (Tokyo, Japan). HDOQ was
103
synthesized according to the literature [11]. Diazepam-d5, used as an internal standard (I.S.), and
104
β-glucuronidase (type HP-2, from Helix pomatia, 152,900 U/mL β-glucuronidase, 714 U/mL sulfatase) were
105
purchased from Sigma-Aldrich (St. Louis, MO, USA). Water was purified using a Milli-Q water purification
106
system (Millipore, Billerica, MA, USA). Oasis® HLB (1 cc, 30 mg) and Sep-Pak Vac® PSA (1 cc, 50 mg)
107
cartridges were purchased from Waters (Milford, MA, USA). Other reagents were purchased from Wako
108
Pure Chemical (Osaka, Japan). Acetonitrile, methanol, and ethyl acetate were of analytical grade. Human
109
whole blood and bile were purchased from Biopredic International (Rennes, France). Drug-free human urine
110
was donated by healthy volunteers. Blank blood, bile, and urine were obtained from autopsies for use in
111
investigating the specificity of the method.
an
us
cr
ip t
102
112 113
3.2 Preparation of standard solutions
Mixed standard stock solutions of QZP and its metabolites (10 µg/mL) were prepared in acetonitrile.
115
Mixed working solutions were prepared by serial dilution of the stock solutions with methanol-water (3:1,
116
v/v). Diazepam-d5 in methanol-water (3:1, v/v) was prepared at a concentration of 500 ng/mL for blood and
117
urine and 2 µg/mL for bile. All solutions were stored at 4ºC. Calibrators were prepared by spiking whole
118
blood, urine, and bile with the appropriate volume of the mixed working solution. The concentrations of the
119
calibrators were 0.5, 1, 2.5, 5, 10, 25, 50, 100, and 200 ng/mL for blood and urine and 4, 8, 20, 40, 80, 200,
120
400, 800, and 1600 ng/mL for bile.
122
d
Ac ce pt e
121
M
114
3.3 Extraction
123
Blood samples were extracted by protein precipitation followed by solid phase extraction (SPE). An
124
aliquot of the sample (50 µL) was placed in a plastic tube. For the total concentration (the sum of the free and
125
conjugated drug) of the analytes, 10 µL sodium acetate buffer (1 mol/L, pH 5) and 10 µL β-glucuronidase
126
were added and incubated at 37°C for 2 h. Subsequently, the samples were spiked with 10 µL I.S. (500
127
ng/mL). Proteins were precipitated by adding 0.4 mL acetonitrile followed by ultrasound treatment and
128
centrifugation at 7200 × g for 2 min. The upper layer was transferred to another plastic tube and 1.2 mL
129
water was added. The mixture was subsequently loaded onto an Oasis® HLB SPE cartridge (1 cc, 30 mg),
4
Page 4 of 28
130
which was pre-conditioned with methanol (1 mL) and purified water (1 mL). The cartridge was washed with
131
1 mL of 20% acetonitrile in water, and the analytes were eluted with 1.5 mL acetonitrile. The eluate was
132
evaporated to dryness under a stream of nitrogen at 45ºC. The residue was reconstituted in 50 µL
133
methanol-water (3:1, v/v) followed by ultrasound treatment and centrifugation. Finally, 15 µL was injected
134
into the LC-MS/MS system. For urine samples, liquid-liquid extraction followed by cleanup using a PSA SPE cartridge (Sep-Pak
136
Vac® 1 cc, 50 mg) was employed. A urine sample (50 µL) was placed in a plastic tube. To obtain the total
137
concentration, the sample was treated with β-glucuronidase as described above. Subsequently, 10 µL I.S.
138
(500 ng/mL), 0.5 mL sodium bicarbonate buffer (0.2 mol/L, pH 10), and 1 mL ethyl acetate were added. The
139
resulting mixture was vortexed for 2 min and centrifuged at 7200 × g for 2 min. The upper layer was loaded
140
onto the PSA cartridge, which was pre-conditioned with acetonitrile (1 mL). Acetonitrile (0.5 mL) was then
141
added to the cartridge, and the solution that passed through the cartridge was collected and evaporated to
142
dryness under a stream of nitrogen at 45ºC. The residue was reconstituted in 50 µL methanol-water (3:1, v/v).
143
After ultrasound treatment and centrifugation, 10 µL was injected into the LC-MS/MS system.
M
an
us
cr
ip t
135
For the bile samples, the same method as for the urine samples was used, except that the sample aliquot
145
was 25 µL, the I.S. concentration was 2 µg/mL, the solvent volume for reconstitution was 200 µL, and the
146
injection volume was 15 µL.
Ac ce pt e
d
144
147
When necessary, the samples were diluted with blank matrices before extraction in order to obtain
148
concentrations in the range of the calibration curve. The amount of femoral blood was insufficient for the
149
analysis; therefore, the sample was mixed with blank blood (1:1, v/v) to obtain the total concentrations.
150 151
3.4 Optimization of the incubation time
152
To optimize the incubation time for the hydrolysis of the conjugates of HOQ and HDOQ, the mixture of
153
the urine sample and blank urine (39:1, v/v) were mixed with β-glucuronidase using the method described
154
above and incubated for 1, 2, and 4 h (n = 3) at 37ºC. After incubation, the mixture was extracted and
155
analyzed.
156 157 158
3.5 LC-MS/MS conditions LC-MS/MS was performed using an Agilent 1100 series high-performance liquid chromatograph (Palo
5
Page 5 of 28
Alto, CA, USA) equipped with a Thermo-Fisher Scientific LCQ Deca XP Plus mass spectrometer (San Jose,
160
CA, USA). Chromatographic separation was achieved on an L-column2 ODS column (internal diameter 150
161
× 2.1 mm, particle size 3 µm; Chemical Evaluation and Research Institute, Tokyo, Japan) at 40°C. The
162
mobile phase consisted of 0.05% formic acid in water (A) and acetonitrile (B) at a flow rate of 0.2 mL/min.
163
The gradient program was a linear gradient from 25% to 95% B over 16 min and then isocratic elution at
164
95% B for 14 min. It was changed to 25% B and kept for 8 min for re-equilibration of the column. The total
165
runtime was 38 min.
ip t
159
The conditions for electrospray ionization were as follows: capillary temperature, 300ºC; sheath gas
167
flow rate, 45 arbitrary units; auxiliary gas flow rate, 5 arbitrary units; and source voltage, 5 kV. The mass
168
spectrometer was operated in the MS/MS scan mode. The precursor ions, collision energies, and scan range
169
are listed in Table 1. For HDOQ, Wideband Activation® was employed. All experiments were conducted in
170
positive mode.
an
us
cr
166
172
3.6 Method validation
M
171
Peak area was determined by using product ions of m/z 259, 341, 343, 354, and 262 for HDOQ, HOQ,
174
OQ, QZP, and I.S., respectively. For DOQ, the sum of product ions of m/z 140, 226, and 261 was used.
175
Calibration curves were constructed by plotting the peak area ratio (analyte/I.S., y) against the nominal
176
concentration of the calibration standards (x), fitted by weighted least-squares linear regression with a
177
weighting factor of 1/x. Selectivity was investigated by analyzing 5 blank samples. The limit of detection
178
(LOD) was determined as the lowest concentration yielding signal-to-noise (S/N) ratios of at least 3 with an
179
acceptable chromatographic peak shape. The limit of quantification (LOQ) was the lowest concentration with
180
an accuracy below 20% of the target concentration and an S/N ratio greater than 10.
Ac ce pt e
d
173
181
The recovery and matrix effect were determined by analyzing 3 replicates at 3 concentrations (shown in
182
Tables 2–4). The recovery was calculated by dividing the average peak area of the analyte spiked before
183
extraction by that spiked after extraction. The matrix effect was estimated by dividing the average peak area
184
of the analyte spiked after extraction by that derived from the neat standards.
185
Inaccuracy was determined as the percentage deviation of the average of the results from the
186
corresponding nominal value. Imprecision was expressed as the percent relative standard deviation (%RSD).
187
Intraday and interday assays were determined by analyzing 3 concentrations (listed in Tables 2–4) on the
6
Page 6 of 28
188
same day (n = 5) and on 5 different days (n = 15), respectively. Autosampler stability was determined by analyzing the spiked matrices immediately after preparation
190
and 2 days after preparation at 2 concentrations for each analyte shown in Table 6 (n = 5). The extracts were
191
kept on the autosampler at room temperature between the analyses. Freeze-thaw stability was determined by
192
analyzing the spiked matrices after 3 freeze-thaw cycles and the freshly spiked matrices (n = 5). Long-term
193
stability was investigated after the storage of the spiked matrices at -30ºC for 4 weeks (n = 5).
ip t
189
4. Results
196
4.1 Extraction method and LC-MS/MS
us
195
cr
194
The recovery of QZP in blood was 40–60% when only SPE was applied. It was increased to higher than
198
77.0% (Table 2) by adopting protein precipitation using acetonitrile before SPE. More than 95% of QZP in
199
the plasma is bound to plasma proteins [12]; therefore, the addition of acetonitrile probably dissociated this
200
binding. For bile, yellow bile pigment was extracted along with the analytes by ethyl acetate. This pigment
201
was effectively removed by cleanup using the PSA cartridge, while keeping the recovery of all analytes at
202
more than 82.6%.
M
an
197
Initially, a shorter LC method was used: a linear gradient from 25% to 90% acetonitrile over 15 min and
204
isocratic elution at 90% acetonitrile for 4 min. However, the peak areas of QZP decreased during consecutive
205
analysis of blood extracts. This may have resulted from the matrix effect caused by residual substances on
206
the column from the previous run. Therefore, we used a method with a longer elution time (14 min at 95%
207
acetonitrile), which prevented the decrease in the QZP peak area.
Ac ce pt e
d
203
208
Fig. 2 shows the chromatograms of QZP, its metabolites, and I.S. obtained from the extraction of heart
209
blood and spiked blood (50 ng/mL) samples. The retention times were 10.8, 12.2, 14.1, 15.6, 17.6, and 13.7
210
min for HDOQ, DOQ, HOQ, OQ, QZP, and I.S., respectively. Fig. 3 shows the mass spectra of peaks A–E
211
and I.S. shown in Fig. 2.
212
To determine the peak areas, the most intense product ions were used for OQ, QZP, and I.S. For HOQ,
213
the second most intense product ion (m/z 341) was used because the most intense one (m/z 369) was assigned
214
to the elimination of an H2O fragment. For HDOQ, the only intense product ion was m/z 287, and it was also
215
assigned to H2O elimination. Therefore, Wideband Activation® was employed, and m/z 259, which was the
216
most intense product ion under this condition, was used. On the mass spectrum of DOQ, 3 intense product
7
Page 7 of 28
217
ions (m/z 261, 226, and 140) were observed, so their sum was used to determine the peak area.
218
To optimize the incubation time, changes in peak areas in the urine sample were checked after
219
incubating for 1, 2, and 4 h (Fig. 4). The peak areas of HOQ and HDOQ were increased from 0 to 2 h and the
220
areas were almost identical at 4 h. Therefore, we set the incubation time to 2 h and the same time was applied
221
to all samples.
225
Selectivity experiments were carried out by assessing 5 blank samples obtained from autopsies. It was
cr
224
4.2 Method validation
confirmed that there were no interfering peaks at the retention times of the analytes.
us
223
ip t
222
The results of method validation are shown in Tables 2–5. The calibration curves of each analyte
227
exhibited good linearity (r2 > 0.991) over the calibration range. The recoveries and matrix effects for all
228
analytes ranged from 74.7–101.0% and 80.1–116.3%, respectively. The inaccuracies and imprecisions were
229
within 15%, except for the intraday precision of low concentration DOQ (19.3%).
an
226
For stability, QZP in the blood and urine was decreased slightly during the storage at -30ºC for 4 weeks;
231
the peak areas of QZP decreased to about 80% in blood at high (40 ng/mL) and low concentration (4 ng/mL)
232
and in urine at low concentration (4 ng/mL).
234
d
Ac ce pt e
233
M
230
4.3 Concentrations in autopsy samples
235
All analytes were detected in the heart blood, femoral blood, urine, and bile. The presence of all
236
analytes was confirmed by the retention time and mass spectra. The errors of the relative retention time were
237
within 1%, and at least two fragment ions were detected for each analyte and the differences in relative
238
intensities (% of base peak) to those of standards were within ±20%. The results of the quantification are
239
summarized in Table 6.
240
There was no large difference in the concentrations of analytes between heart and femoral blood. The
241
concentrations of DOQ were the highest of all the analytes and it mainly existed as the free form. The
242
concentrations of free DOQ were 8.3 and 9.7 times higher than those of QZP in heart and femoral blood,
243
respectively. Meanwhile, the concentrations of QZP and OQ were very similar (13.7–17.7 ng/mL).
244
In urine, conjugated HOQ and HDOQ were the major metabolites. The total concentrations of HOQ and
245
HDOQ were 5080 ± 350 and 1600 ± 80 ng/mL, respectively. They were much higher than the corresponding
8
Page 8 of 28
246 247 248
free concentrations. Conversely, the concentration of the unchanged drug was less than 1 ng/mL. In bile, a surprisingly high concentration of total HOQ was found (56200 ± 1900 ng/mL). It mainly existed as the conjugated form. Interestingly, the concentration of free QZP was higher than that of the total QZP in all samples. This is
250
probably caused by the instability of QZP in the matrices during the incubation step. The instability of QZP
251
was also observed in the stability test during the method validation.
252 5. Discussion
cr
253
ip t
249
According to a previous pharmacokinetic study, after a single-dose administration of QZP, the plasma
255
elimination half-lives (T1/2β) of QZP and OQ are 25–40 h, whereas DOQ is 70–75 h [3, 4]. The long T1/2β of
256
DOQ resulted in its accumulation in blood after multiple-dose administration. When QZP was orally
257
administered at 15 mg daily for 14 days, the concentration of DOQ reached the steady state after 13 days [5].
258
The concentrations of QZP, OQ, and DOQ in the steady state were 11.1 ± 5.5, 8.0 ± 3.2, and 92 ± 31 ng/mL,
259
respectively. The maximum concentrations of QZP, OQ, and DOQ in the steady state were 30 ± 12, 17 ± 6,
260
and 157 ± 53 ng/mL, respectively, whereas those for a single dose were 31 ± 22, 18 ± 8, and 26 ± 12 ng/mL,
261
respectively. In the present case, the concentrations of free DOQ in blood (160 ± 7 ng/mL for heart blood and
262
181 ± 12 ng/mL for femoral blood) were much higher than those of QZP and OQ (14–19 ng/mL).
263
Considering that the deceased was prescribed one 20 mg QZP tablet daily, the blood concentrations of DOQ
264
were comparable with its steady-state concentration. This result indicates that the deceased had been taking
265
QZP for at least several days before death, until the concentration of DOQ reached approximately the same
266
level as that in the steady state.
Ac ce pt e
d
M
an
us
254
267
It is noteworthy that DOQ existed as mainly the free form in urine, which differs from previous findings
268
[4]. One possible explanation is that the DOQ glucuronide is less stable and its concentration decreased
269
during the interval between death and the autopsy (approximately 4 days). The glucuronic acid moiety of the
270
DOQ glucuronide should be bound to the amide group and this bond seems to be less stable compared with
271
those in the HOQ and HDOQ glucuronides.
272
This study quantified QZP and its metabolites in human bile for the first time, and revealed that HOQ
273
accumulated in bile. The total concentration of HOQ was more than 50 times higher than that of any other
274
analyte, indicating that HOQ accumulated selectively in bile. Although the mechanism of this accumulation
9
Page 9 of 28
275
is not clear, one possible explanation is enterohepatic circulation. This process can lead to a prolonged T1/2β
276
and cumulative amount of HOQ, and consequently, the unexpectedly high level of HOQ in bile. The high
277
concentration of HOQ also reflected the deceased’s daily administration of QZP. The existence of HOQ in bile can be used to prove QZP administration, particularly if a deceased
279
individual took QZP days before death and the parent drug has already disappeared from the blood. However,
280
further investigation of the elimination of HOQ in bile is required.
ip t
278
In a previous study, bile to blood drug concentration ratios were investigated for a variety of drugs [13].
282
For benzodiazepines, the ratio of the bile after the β-glucuronidase treatment to blood without the treatment
283
was 3.58 for diazepam, 7.3 for nordiazepam, 360 for chlorodiazepam, and 386.3 for lorazepam. In the
284
present study, the ratios of total analyte in bile to free analyte in heart blood were 4.6 for DOQ, 32 for HDOQ,
285
and 776 for HOQ. The ratio appeared to increase for analytes with a hydroxyl group, indicating that
286
O-glucuronidation is one of the factors that cause enterohepatic circulation.
an
us
cr
281
The findings of this study clarified that each metabolite of QZP has their own characteristics: the
288
concentration of OQ in blood is close to that of QZP; DOQ accumulates in blood at multiple doses; and
289
conjugated HOQ is the main metabolite in urine and accumulated selectively in bile. By combining this
290
information, it can be possible to estimate roughly when or how a deceased individual took QZP before death.
291
For example, in an acute intoxication case, the blood concentrations of QZP and OQ are much higher than
292
that of DOQ. Meanwhile, in a case in which the deceased individual had been taking QZP daily before death,
293
the concentration of DOQ in blood is higher than that of QZP and OQ, and the accumulation of HOQ in bile
294
is also observed. If the deceased individual had stopped taking QZP for a period before death, only DOQ in
295
blood and HOQ in bile would be detected. These findings need to be confirmed by analyzing various cases in
296
the future. In particular, various factors that affect drug metabolism, such as enzyme or transporter
297
polymorphisms, aging, drug–drug interaction, and diseases such as liver failure must be evaluated.
299
d
Ac ce pt e
298
M
287
6. Conclusions
300
In this study, the concentrations of QZP and its metabolites, OQ, DOQ, HOQ, and HDOQ, were
301
determined in blood, urine, and bile using LC-MS/MS. A protein precipitation method followed by SPE was
302
employed to extract the analytes from blood. For urine and bile, liquid-liquid extraction together with
303
cleanup using an SPE cartridge was applied. The high concentration of DOQ in the cadaver’s blood reflects
10
Page 10 of 28
the daily administration of QZP before death. The total concentration of HOQ in bile was extremely high.
305
This accumulation of HOQ in bile also supports the daily administration of QZP. Conjugated HOQ in bile
306
can be used to prove QZP administration when a deceased individual took QZP days before death because it
307
remains for an extended period. This study demonstrates that more information about dose and frequency of
308
QZP administration can be provided by the combination of the concentrations of QZP and its metabolites in
309
biological matrices.
ip t
304
310 References
312
[1] Ethical Drug Data Book 2012 Vol. 2 (Japanese), Tokyo Marketing Division, Fuji Keizai, Tokyo, 2012.
313
[2] J.M. Hilbert, J.M. Ning, G. Murphy, A. Jimenez, N. Zampaglione, Gas chromatographic determination
318 319 320 321 322 323 324 325 326
us
an
317
quazepam kinetics, Clin. Pharmacol. Ther. 36 (1984) 99-104.
[4] N. Zampaglione, J.M. Hilbert, J. Ning, M. Chung, R. Gural, S. Symchowicz, Disposition and metabolic
M
316
[3] J.M. Hilbert, M. Chung, G. Maier, R. Gural, S. Symchowicz, N. Zampaglione, Effect of sleep on
fate of 14C-quazepam in man, Drug Metab. Dispos. 13 (1985) 25-9. [5] M. Chung, J.M. Hilbert, R.P. Gural, E. Radwanski, S. Symchowicz, N. Zampaglione, Multiple-dose
d
315
of quazepam and two major metabolites in human plasma, J. Pharm. Sci. 73 (1984) 516-9.
quazepam kinetics, Clin. Pharmacol. Ther. 35 (1984) 520-4.
Ac ce pt e
314
cr
311
[6] J.M.
Hilbert,
D.
Battista,
Quazepam
and
flurazepam:
differential
pharmacokinetic
and
pharmacodynamic characteristics. J. Clin. Psychiatry 52 Suppl (1991) 21-26. [7] S.K. Gupta, E.H. Ellinwood, Liquid chromatographic assay and pharmacokinetics of quazepam and its metabolites following sublingual administration of quazepam. Pharm. Res. 5 (1988) 365-368. [8] N. Yasui-Furukori, T. Kondo, T. Takahata, K. Mihara, S. Ono, S. Kaneko, T. Tateishi, Effect of dietary fat content in meals on pharmacokinetics of quazepam, J. Clin. Pharmacol. 42 (2002) 1335-40.
327
[9] A. Yamazaki, Y. Kumagai, T. Fujita, T. Hasunuma, S. Yokota, M. Maeda, Y. Otani, M. Majima,
328
Different effects of light food on pharmacokinetics and pharmacodynamics of three benzodiazepines,
329
quazepam, nitrazepam and diazepam, J. Clin. Pharm. Ther. 32 (2007) 31-9.
330
[10] M. Terada, T. Shinozuka, C. Hasegawa, E. Tanaka, M. Hayashida, Y. Ohno, K. Kurosaki, Analysis of
331
quazepam and its metabolites in human urine by gas chromatography-mass spectrometry: application to
332
a forensic case, Forensic Sci. Int. 227 (2013) 95-99.
11
Page 11 of 28
333
[11] J.V. Earley, R.I. Fryer, D. Winter, L.H. Sternbach, Quinazolines and 1,4-benzodiazepines. XL. The synthesis
334
of
335
7-chloro-1-(2-diethylaminoethyl)-5-(2-fluorophenyl)-1,3-dihydro-2H-1,4-benzodiazepin-2-one,
336
Chem. 11 (1968) 774-777.
339 340
properties, and therapeutic efficacy in insomnia, Drugs 35 (1988) 42-62. [13] R. Vanbinst, J. Koenig, V. Di Fazio, A. Hassoun, Bile analysis of drugs in postmortem cases, Forensic Sci. Int. 128 (2002) 35-40.
us
342 343
an
344
d Ac ce pt e
348
M
345
347
Med.
[12] S.I. Ankier, K.L. Goa, Quazepam: A preliminary review of its pharmacodynamic and pharmacokinetic
341
346
J.
ip t
338
of
cr
337
metabolites
12
Page 12 of 28
cr us an
Figure Captions
M
Fig. 1. Metabolic pathways of quazepam.
Fig. 2. Chromatograms of heart blood extract. A: 3-Hydroxy-N-desalkyl-2-oxoquazepam (HDOQ), B: N-desalkyl-2-oxoquazepam
ed
(DOQ), C: 3-hydroxy-2-oxoquazepam (HOQ), D: 2-oxoquazepam (OQ), E: quazepam (QZP), and internal standard (I.S.).
pt
Fig. 3. Mass spectra of peaks A–E and of internal standard (I.S.) shown in Fig. 2. A: 3-hydroxy-N-desalkyl-2-oxoquazepam (HDOQ),
ce
B: N-desalkyl-2-oxoquazepam (DOQ), C: 3-hydroxy-2-oxoquazepam (HOQ), D: 2-oxoquazepam (OQ), E: quazepam (QZP).
Fig. 4. Changes of peak areas of 3-hydroxy-2-oxoquazepam (HOQ, ▲), and 3-hydroxy-N-desalkyl-2-oxoquazepam (HDOQ,
) in
100%.
Ac
urine with 1, 2, and 4 h of incubation at 37°C. The peak areas are expressed as relative values, with the maximum peak area being
Table 1. Precursor ion, collision energy, and scan range for each analyte.
13
Page 13 of 28
cr QZP OQ DOQ HOQ HDOQ DZP-d5
387 371 289 387 305 290
us
Precursor ion (m/z)
Collision energy (%)
Scan range (m/z)
42 40 40 42 37 40
105–400 100–390 75–300 105–400 80–320 75–300
M
ce
pt
ed
a
an
Analyte
a
Ac
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam, 3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
14
Page 14 of 28
us
cr OQ
1.0
2.5
Ac
DOQ
0.25
HOQ
HDOQ
0.5
2.5
2.5
2.5
Calibration range (ng/mL)
5
5
5
Spiked concentration (ng/mL)
Recovery (%, n = 3)
Matrix effect (%, n = 3)
2
78.0
20
M
LOQ (ng/mL)
0.5–50
ed
QZP
LOD (ng/mL)
2.5–100
pt
a
ce
Analyte
an
Table 2. Validation parameters for blood.
5–200
5–200
5–200
Intraday (n = 5) Inaccuracy (%)
Imprecision (%RSD)
Inaccuracy (%)
Imprecision (%RSD)
95.7
-9.5
5.8
2.9
11.7
82.9
87.8
-9.2
5.7
5.0
7.8
40
77.0
90.1
-6.1
4.3
5.0
7.9
4
74.7
99.2
-2.6
5.0
8.3
13.5
40
89.8
94.9
-2.3
2.6
2.9
5.0
80
81.2
93.7
-5.1
3.7
2.1
7.6
8
75.9
107.3
5.1
19.3
2.8
14.4
80
88.6
104.1
8.1
5.4
4.6
5.3
160
85.8
101.6
5.9
5.3
5.4
6.5
8
84.1
105.3
-4.1
14.0
0.2
10.0
80
90.2
102.5
2.3
6.1
3.9
5.4
160
83.6
105.4
0.8
2.5
2.8
7.4
8
82.0
106.7
-12.8
8.5
6.2
12.9
80
88.0
100.5
3.0
5.3
7.7
5.5
160
81.9
100.8
3.3
2.0
4.7
6.5
a
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam,
15
Interday (5 days, n = 15)
Page 15 of 28
cr us
Ac
ce
pt
ed
M
an
3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
16
Page 16 of 28
us
cr OQ
1.0
2.5
Ac
DOQ
0.25
HOQ
HDOQ
17
0.5
2.5
2.5
Calibration range (ng/mL)
2.5
5
5
5
Spiked concentration (ng/mL)
Recovery (%, n = 3)
Matrix effect (%, n = 3)
2
78.1
20
M
LOQ (ng/mL)
ed
QZP
LOD (ng/mL)
0.5–50
pt
a
ce
Analyte
an
Table 3. Validation parameters for urine.
2.5–100
5–200
5–200
5–200
Intraday (n = 5)
Interday (5 days, n = 15)
Inaccuracy (%)
Imprecision (%RSD)
Inaccuracy (%)
Imprecision (%RSD)
96.1
5.4
14.7
-1.0
11.0
82.8
87.6
-3.9
6.1
-1.6
9.2
40
78.5
94.3
-3.3
7.5
-2.8
7.3
4
97.2
92.6
-5.1
5.5
-4.9
12.4
40
92.9
96.1
-4.6
5.2
3.0
7.7
80
89.8
102.0
-7.0
5.1
3.2
8.6
8
101.0
116.3
-8.0
10.5
-2.4
11.7
80
90.9
99.8
5.0
6.9
4.1
6.8
160
96.8
96.7
6.1
4.4
6.6
6.8
8
94.3
106.0
3.2
6.2
6.3
10.2
80
92.1
97.2
2.6
2.4
3.0
8.0
160
93.5
97.9
0.6
2.2
3.6
5.8
8
96.8
94.3
7.7
4.2
-7.1
14.9
80
94.7
95.4
1.3
3.7
3.3
7.9
160
96.0
94.4
2.4
5.8
5.7
5.5
Page 17 of 28
cr us
a
Ac
ce
pt
ed
M
an
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam, 3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
18
Page 18 of 28
cr us
4
4–400
8
HOQ
HDOQ
19
20
8
20
8
40
40
40
Spiked concentration (ng/mL)
Recovery (%, n = 3)
Matrix effect (%, n = 3)
16
97.8
160 320 32
M
2
Ac
DOQ
Calibration range (ng/mL)
ed
OQ
LOQ (ng/mL)
20–800
pt
QZP
LOD (ng/mL)
ce
Analyte
a
an
Table 4. Validation parameters for bile.
40–1600
40–1600
40–1600
Intraday (n = 5)
Interday (5 days, n = 15)
Inaccuracy (%)
Imprecision (%RSD)
Inaccuracy (%)
Imprecision (%RSD)
94.6
5.5
12.0
-4.0
13.2
83.1
94.6
-5.2
4.2
-2.1
12.6
82.6
100.0
-9.6
5.5
-5.5
9.3
91.8
90.0
-5.9
12.2
-0.1
11.5
320
93.5
98.9
-4.0
5.6
5.2
9.3
640
98.2
100.6
-9.3
4.4
0.3
8.9
64
99.7
80.1
0.6
14.6
-5.7
10.0
640
97.0
95.3
5.7
6.2
3.0
11.0
1280
95.8
100.5
1.8
4.1
1.7
7.5
64
92.8
88.7
-6.0
4.5
-11.5
10.6
640
88.8
99.1
0.0
0.9
4.2
8.3
1280
94.7
101.5
-0.6
4.7
2.6
8.3
64
98.3
96.4
3.3
9.2
3.8
10.3
640
97.5
98.0
1.6
5.8
2.8
11.6
1280
93.5
98.4
-4.2
4.0
-1.4
8.4
Page 19 of 28
cr us
a
Ac
ce
pt
ed
M
an
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam, 3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
20
Page 20 of 28
cr us
a
Concentrations (ng/mL)
Blood Three freeze-thaw cycles (%)
Four weeks at -30ººC (%)
4 40
98.9 ± 100.1 ±
4.5 1.5
101.5 ± 93.6 ±
5.2 3.1
82.6 81.3
± ±
2.9 3.7
OQ
8 80
98.9 ± 102.0 ±
8.9 4.2
96.4 ± 95.9 ±
4.2 3.0
91.8 96.9
± ±
2.4 1.8
DOQ
16 160
95.9 ± 102.7 ±
2.5 1.9
100.6 ± 94.6 ±
9.7 2.0
98.0 98.3
± ±
7.2 2.6
16 160
106.0 ± 102.9 ±
9.5 3.8
94.1 ± 96.1 ±
2.2 3.4
96.8 95.5
± ±
6.1 4.6
16 160
102.4 ± 101.0 ±
9.2 1.8
95.2 ± 97.6 ±
7.8 2.8
95.1 99.2
± ±
6.6 2.5
pt
ed
QZP
HDOQ
a
Ac
Analyte
ce
HOQ
21
Autosampler 2days (%)
M
Analyte
an
Table 5. Stabilities of analytes after extraction (2 days at room temperature), three freeze-thaw cycles, and 4 weeks at -30ºC in blood, urine, and bile (n = 5).
Concentrations (ng/mL)
Urine Autosampler 2days (%)
Three freeze-thaw cycles (%)
Four weeks at -30ººC (%)
QZP
4 40
94.3 103.3
± ±
5.8 2.0
111.5 89.2
± ±
9.2 7.0
78.9 96.6
± ±
8.2 3.2
OQ
8 80
97.7 104.7
± ±
7.8 3.0
93.5 98.3
± ±
5.5 3.6
101.4 99.1
± ±
9.9 6.9
DOQ
16 160
104.2 105.4
± ±
3.2 1.6
96.7 93.3
± 13.3 ± 4.0
101.9 99.6
± ±
6.7 2.3
HOQ
16 160
100.0 98.6
± ±
6.2 2.9
103.5 99.6
± ±
8.2 3.8
108.5 98.6
± ±
9.8 2.1
HDOQ
16 160
97.6 99.1
± ±
6.6 1.2
109.1 99.2
± ±
8.5 4.6
104.2 97.4
± ±
5.2 1.8
Page 21 of 28
cr 32 320
98.2 ± 93.1 ±
103.0 110.4
± ±
7.4 6.9
101.6 ± 103.7 ±
2.7 6.0
OQ
64 640
96.9 ± 95.8 ±
3.3 1.4
111.6 103.8
± ±
5.8 3.2
103.6 ± 98.6 ±
3.8 8.5
DOQ
128 1280
91.0 ± 95.1 ±
8.9 5.0
98.4 99.3
± ±
7.4 1.2
82.5 ± 89.8 ±
7.4 9.2
HOQ
128 1280
89.3 ± 94.1 ±
6.3 2.2
114.1 103.0
± ±
5.1 3.1
101.5 ± 96.6 ±
10.3 7.3
HDOQ
128 1280
99.7 ± 95.6 ±
3.0 0.7
103.0 100.9
± ±
3.5 2.9
104.2 ± 99.4 ±
3.7 6.2
ed
pt
5.3 4.5
M
QZP
Four weeks at -30ººC (%)
ce
a
Autosampler 2days (%)
us
Concentrations (ng/mL)
Bile Three freeze-thaw cycles (%)
an
Analyte
a
22
Ac
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam, 3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
Page 22 of 28
cr us
Analyte
an
Table 6. Concentrations of QZP and its metabolites in the forensic autopsy case.
Concentration (mean ± SD, n = 5, ng/mL)
a
DOQ
total
17.5 ± 1.2
free
16.7 ± 1.0
17.7 ± 1.2
13.7 ± 1.6
free
160 ± 7
181 ± 12
free
19.9 ± 0.8
27.9 ± 0.9
total
38.1 ± 0.9
72.4 ± 4.8
14.1 ± 1.1
20.5 ± 1.4
ce free
total
22.3 ± 0.8
b
b
212 ± 10
172 ± 3
b
14.4 ± 1.7
15.2 ± 0.6
Ac
HDOQ
18.6 ± 1.3
total
total
HOQ
19.3 ± 0.8
ed
OQ
free
pt
QZP
Femoral blood
M
Heart blood
b
32.5 ± 3.7
b
Urine
Bile
0.94 ± 0.09
27.2 ± 1.7
0.64 ± 0.09
24.1 ± 1.8
4.69 ± 0.43
509 ± 21
4.20 ± 0.55
510 ± 28
47.6 ± 1.5
832 ± 61
67.6 ± 6.4
805 ± 36
102 ± 5
7,720 ± 300
5080 ± 350 60.7 ± 2.1
b
1,600 ± 80
b
b
56,200 ± 1,900 178 ± 12 1,040 ± 50
a
QZP, OQ, DOQ, HOQ, HDOQ, and DZP-d5 represent quazepam, 2-oxoquazepam, N-desalkyl-2-oxoquazepam, 3-hydroxy-2-oxoquazepam, 3-hydroxy-N-desalkyl-2-oxoquazepam, and diazepam-d5 (I.S.), respectively.
23
b
Page 23 of 28
b
cr us
The concentrations are approximate values because they were obtained by diluting the sample with blank matrices.
Ac
ce
pt
ed
M
an
b
24
Page 24 of 28
ip t cr us an
Ac ce pt e
d
M
06 Fig1 .
25
Page 25 of 28
ip t cr us an M
Ac ce pt e
d
07 Fig2 .
26
Page 26 of 28
ip t cr
Ac ce pt e
d
M
an
us
08 Fig3 .
27
Page 27 of 28
ip t cr us
Ac ce pt e
d
M
an
09 Fig4 .
28
Page 28 of 28