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

Quantitative analysis of quazepam and its metabolites in human blood, urine, and bile by liquid chromatography-tandem mass spectrometry.

Quazepam (QZP), which is a long-acting benzodiazepine-type hypnotic, and its 4 metabolites, 2-oxoquazepam, N-desalkyl-2-oxoquazepam (DOQ), 3-hydroxy-2...
298KB Sizes 0 Downloads 0 Views