Accepted Manuscript Title: Quantification of levornidazole and its metabolites in human plasma and urine by ultra-performance liquid chromatography - mass spectrometry Author: Yuran Cao Miao Zhao Xiaojie Wu Beining Guo Yuancheng Chen Jicheng Yu Guoying Cao Jing Zhang Yaoguo Shi Yingyuan Zhang PII: DOI: Reference:
S1570-0232(14)00375-4 http://dx.doi.org/doi:10.1016/j.jchromb.2014.05.058 CHROMB 18983
To appear in:
Journal of Chromatography B
Received date: Revised date: Accepted date:
6-1-2014 24-5-2014 28-5-2014
Please cite this article as: Y. Cao, M. Zhao, X. Wu, B. Guo, Y. Chen, J. Yu, G. Cao, J. Zhang, Y. Shi, Y. Zhang, Quantification of levornidazole and its metabolites in human plasma and urine by ultra-performance liquid chromatography - mass spectrometry, Journal of Chromatography B (2014), http://dx.doi.org/10.1016/j.jchromb.2014.05.058 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.
Quantification of levornidazole and its metabolites in human plasma and urine by ultra-performance liquid chromatography - mass spectrometry
ip t
Yuran Caoa,b,1, Miao Zhaoa,b,1, Xiaojie Wua,b, Beining Guoa,b, Yuancheng Chena,b, Jicheng Yua,b, Guoying Caoa,b, Jing Zhanga,b,*, Yaoguo Shia,b, and Yingyuan Zhanga,b Institute of Antibiotics, Huashan Hospital, Fudan University, 12 Middle Wulumuqi
cr
a
b
us
Road, Shanghai 200040, China
Key Laboratory of Clinical Pharmacology of Antibiotics, National Health and
an
Family Planning Commission, 12 Middle Wulumuqi Road, Shanghai 200040, China These authors contributed equally to this work.
*
Corresponding author. Institute of Antibiotics, Huashan Hospital, 12 Middle
M
1
te
02162484347.
d
Wulumuqi Road, Shanghai 200040, China.Tel:+86 02152888190; Fax: +86
Ac ce p
E-mail:
[email protected] 1
Page 1 of 29
an
us
cr
Ultra-performance liquid chromatography Mass spectrometry Multiple reaction monitoring Atmospheric pressure chemical ionization High performance liquid chromatography Lower limit of quantification Internal standard Quality control Relative standard error Area under the plasma concentration-time curve Half-life Maximum plasma concentration Metabolic ratio Cumulative urinary excretion
Ac ce p
te
d
M
UPLC MS MRM APCI HPLC LLOQ IS QC RSD AUC T1/2 Cmax MR Ae
ip t
Abbreviations
2
Page 2 of 29
Abstract We developed and validated an ultra-performance liquid chromatographic (UPLC) method coupled with atmospheric pressure chemical ionization (APCI) mass
ip t
spectrometry for simultaneous determination of levornidazole and its first-pass metabolites, l-chloro-3-(2-hydroxymethyl-5-nitro-l-imidazolyl)-2-propanol (M l), 2-
methyl-5-nitroimidazole (M2) and 3-(2-methyl-5-nitro-1-imidazolyl)-1,2-propanediol
cr
(M4), in human plasma and urine. The biological samples were pretreated by protein
us
precipitation and liquid-liquid extraction and analyzed using an ACQUITY UPLC CSH C18 column (2.1 ×50 mm, 1.7 μm) and a QTRAP mass spectrometer in multiple reaction monitoring mode via APCI. Acetonitrile and 0.1% formic acid in water was
an
used as the mobile phase in gradient elution at a flow rate of 0.6 mL/min. The lower limit of quantification of this method was 0.0100, 0.00500, 0.0200 and 0.00250
M
μg/mL for levornidazole, M1, M2 and M4, respectively. The linear calibration curves were obtained for levornidazole, M1, M2, and M4 over the range of 0.0100-5.00,
d
0.00500-2.50, 0.0200- 10.0 and 0.00250-1.25 μg/mL, respectively. The intra- and
te
inter-batch precision was less than 12.2% in plasma and less than 10.8% in urine. The intra- and inter-batch accuracy was 87.8-105.7% in plasma and 92.8-109.2% in urine.
Ac ce p
The mean recovery of levornidazole, M1, M2 and M4 was 91.1-105.1%, 95.8-103.8%, 87.8-96.8%, 96.8-100.6% from plasma and 96.0-100.9%, 96.9-107.9%, 95.1-102.7%, 103.7-105.9% from urine respectively. This method was validated under various conditions, including room temperature, freeze-thaw cycles, long-term storage at -40 ± 5℃,after pretreatment in the autosampler (at 10℃), and 10- and 100-fold dilution.
This newly established analytical method was successfully applied in a pharmacokinetic study following single intravenous infusion of levornidazole in 24 healthy Chinese subjects. Keywords: levornidazole; metabolite; LC-MS/MS; pharmacokinetics
3
Page 3 of 29
1.
Introduction Ornidazole, 1-(3-chloro-2-hydroxypropyl)-2-methyl-5-nitroimidazole, was
developed by Roche (Switzerland) in 1970s and has already been on market in China.
ip t
Ornidazole is soluble in chloroform (solubility> 50%), its solubility in water is 12% and pKa is 2.3. Ornidazole has obvious antimicrobial activity on anaerobes,
cr
Trichomonas vaginalis and Amoebae. Levornidazole is the levo isomer of ornidazole, which is a new 5-nitroimidazole antimicrobial drug. Levornidazole has shown
us
anti-anaerobic activity, which is similar to ornidazole in general, but slightly stronger than ornidazole on some strains. Levornidazole has significantly lower central
an
neurotoxicity than ornidazole. It has been reported that ornidazole is metabolized to five phase I metabolites in human urine [1]: M1, l-chloro-3-(2-hydroxymethyl
M
-5-nitro-l-imidazolyl)-2-propanol; M2, 2-methyl-5-nitroimidazole; M3, N-(3-chloro-2-hydroxypropyl)acetamide; M4, 3-(2-methyl-5-nitro-l-imidazolyl)-l, 2-propanediol, and M5, acetamide (Fig. 1).
d
A number of different methods have been reported for detection of
te
nitroimidazoles in biological fluids [2-11]. However, no method is available for simultaneous determination of levornidazole and its metabolites. The purpose of this
Ac ce p
study was to develop and validate a method that can be used for the determination of levornidazole and its metabolites (M1, M2 and M4) in real human plasma and urine samples. In this paper, a simple, sensitive, precise, accurate and specific ultraperformance liquid chromatographic-tandem mass spectrometric (UPLC-MS/MS) method is described. The newly established method was successfully applied to a pharmacokinetic study, which required high sensitivity and selectivity. 2.
Materials and methods
2.1. Chemicals and reagents Levornidazole, M1 and M4 were all supplied by Sanhome Pharmaceutical Co., Ltd (Nanjing, China). M2 and internal standard (IS, metronidazole) were obtained from the National Institute for the Control of Pharmaceutical and Biological Products. Acetonitrile and methanol, both of HPLC grade were purchased from Sigma (Sigma, 4
Page 4 of 29
USA). Water was Milli-Q grade. All the other chemicals and solvents were of analytical grade. 2.2. Instrument and LC-MS/MS conditions Determination was performed using a Waters ACQUITY UPLC system (Waters,
ip t
USA) coupled with an API4000 QTRAP mass spectrometer (AB SCIEX, USA). Chromatographic separation was carried out on an ACQUITY UPLC CSH C18
cr
column (2.1 ×50 mm, 1.7 μm) using a mobile phase composed of 0.1% formic acid in water and acetonitrile in gradient elution (Table 1) at a flow rate of 0.6 mL/min. The
us
injection volume was 3 μL. Both analytes and IS were determined by atmospheric pressure chemical ionization (APCI) in positive ion mode. Quantification was
an
performed using multiple reaction monitoring (MRM) of the transitions of m/z 220.0 → m/z 127.5 for levornidazole, m/z 236.2→ m/z 170.9 for M1, m/z 127.9→ m/z
M
110.7 for M2, m/z 202.5→ m/z 127.7 for M4 and m/z 172.1→ m/z 127.8 for IS, respectively. The curtain gas and ion source gas 1 were set at 20 and 40 psi. The
d
nebuilzer current was 4.0. The temperature was 500℃. The optimized collision
respectively.
te
energies of 25,19, 20, 22 and 20 eV were used for levornidazole, M1, M2, M4 and IS,
Ac ce p
Data acquisition was performed using Analyst 3.13 software (AB SCIEX, USA). 2.3. Preparation for standard and quality control samples The stock solutions of levornidazole, M1, M2, M4 and IS were prepared by
dissolving appropriate amount of the compounds in methanol-water (1:1) to give final concentration of 1000 μg/mL for all compounds. The stock solutions of levornidazole, M1, M2 and M4 were mixed in proportion and then diluted successively with 0.1% formic acid in water to achieve the desired concentrations. Plasma (heparin lithium as anticoagulant) or urine samples were spiked with the above solutions to obtain the calibration standard samples and quality control (QC) samples. The nominal plasma or urine concentrations (7 points) of calibration standards contained 0.0100, 0.0250, 0.0500, 0.250, 0.500, 2.50 and 5.00 μg/mL of levornidazole; 0.00500, 0.0125, 0.0250, 0.125, 0.250, 1.25 and 2.50 μg/mL of M1; 0.0200, 0.0500, 0.100, 0.500, 1.00, 5.00 and 10.0 μg/mL of M2; 0.00250, 0.00625, 0.0125, 0.0625, 5
Page 5 of 29
0.125, 0.625 and 1.25 μg/mL of M4, respectively. QC samples contained 0.0100, 0.0300, 0.200 and 4.00 μg/mL of levornidazole; 0.00500, 0.0150, 0.100 and 2.00 μg/mL of M1; 0.0200, 0.0600, 0.400 and 8.00 μg/mL of M2; 0.00250, 0.00750, 0.0500 and 1.00 μg/mL of M4, respectively (Table 2).
ip t
The stock solution of IS was diluted with 0.1% formic acid in water to prepare
the working solution of internal standard (containing 1.00 μg/mL of metronidazole).
cr
2.4. Extraction from plasma or urine
The same extraction method was used for plasma and urine samples. IS (20 μL)
us
and acetonitrile (600 μL) were added to 200 μL plasma or urine sample in a
polypropylene tube. Following vortex mixing, the mixture was centrifuged at 13800
an
g for 15 minutes. An aliquot of supernatant (600 μL) was transferred to a clean polypropylene tube and 600 μL of ethyl acetate was added. The mixture was mixed
M
by rotation for 10 minutes and then centrifuged at 13800 g for 10 minutes. The supernatant (800 μL) was transferred to a clean polypropylene tube and evaporated
d
to dryness under a stream of nitrogen at 40℃. The pellet was reconstituted with 200 μL of acetonitrile-0.1% formic acid in water (5:95, v/v) solution and centrifuged at
te
13800 g for 5 minutes. Three microliters of the reconstituted solution was injected
Ac ce p
into the UPLC system. 2.5. Assay validation
The method was validated in terms of selectivity, linearity, precision and
accuracy, matrix effect and recovery, dilution integrity, stability of stock solutions, plasma or urine samples at room temperature, long-term stability at -40 ± 5℃,
stability after freeze-thaw cycles, and stability of the pretreated samples in the autosampler (at 10℃).
Selectivity was assayed by using 6 sources of the blank plasma (or urine) which were individually analyzed and evaluated for the interference from the endogenous components in matrices at the retention times of levornidazole, M1, M2 , M4 and IS . Additionally, the impacts of haemolized (containing 2% whole blood) and lipemic plasma were evaluated. Six sources of plasma (or urine) samples containing only one analyte were individually analyzed and evaluated for the cross-interference. 6
Page 6 of 29
Linearity of the assay was examined for levornidazole, M1, M2 and M4 over the concentration range from 0.0100 to 5.00 μg/mL, from 0.00500 to 2.50μg/mL, from 0.0200 to 10.0μg/mL and from 0.00250 to 1.25μg/mL by assaying plasma or urine standards at seven concentrations on six separate days. A weighted (1/x2) linear
ip t
regression was used to perform calibration curve fitting through plotting peak area ratios of each analyte to IS against the concentrations of the spiked analyte.
cr
The inter-batch precision and accuracy were determined in conjunction with the linearity studies by assaying on six separate days using QC samples at each of four
us
concentrations. The intra-batch precision and accuracy were determined by assaying QC samples six times on one day. Concentration of levornidazole, M1, M2 and M4 in
an
QC samples was determined by application of the appropriate standard curve obtained on that day.
M
Recovery of levornidazole, M1, M2, M4 and IS was assessed by direct comparison of peak areas from extracted versus non-extracted samples by using six
d
replicate QC samples at each of four concentrations plus the appropriate amount of IS. To evaluate the matrix effect, six blank plasma or urine samples from six untreated
te
volunteers were extracted and reconstituted with each analyte at four QCs or IS
Ac ce p
concentrations in mobile phase. The corresponding peak areas then were compared with those of the analyte in mobile phase. The plasma and urine QC samples were diluted by blank matrix (plasma or urine)
10- or 100-fold in six replicate for testing the dilution integrity. Three QC samples at each of four concentrations were left at room temperature
for 24 hours to test the short-term stability of levornidazole in plasma or urine samples. The QC samples were extracted, loaded onto the autosampler, and kept in the autosampler at 10℃ until 48 h and injected at the time points of 0, 3, 6, 9, 24, and 48 hours. The QC samples were exposed to 3 freeze-thaw cycles, and the samples were then processed and assayed. The plasma and urine QC samples were stored at -40 ± 5℃ for 3 and 4 months respectively and then processed and assayed to test the long-term stability. 2.6. Pharmacokinetic study 7
Page 7 of 29
The UPLC-MS/MS method we developed in this report was applied in a singlecenter, open, and parallel pharmacokinetic study following single intravenous infusion of levornidazole sodium chloride injection in 24 (12 males and 12 females) healthy Chinese subjects. Following written informed consent, the subjects were
ip t
randomized to receive a single dose of 500 mg or 750 mg levornidazole after
breakfast. All volunteers were provided with standard meal. Water intake was
cr
controlled before intravenous infusion. The infusion time was 60 ± 10 min for 500 mg dose group, and 90 ± 10 min for 750 mg dose group. The age, body weight and height
us
of the subjects in 500 mg and 750 mg group were 24.25 ± 2.53 and 23.75 ± 1.71 years, 60.68 ± 9.02 and 63.04 ± 6.37 kg, 1.66 ± 0.08 and 1.70 ± 0.07 meters, respectively.
an
Blood samples (4 mL) were drawn immediately before (0), 0.33 hours after the start of infusion, immediately and 0.5, 1, 2, 4, 8, 12, 24, 36, 48, 60 and 72 hours after the
M
end of intravenous infusion. The blood samples were collected into heparin lithium test tubes, and placed on ice immediately after collection and centrifuged at 3000 rpm
d
for 10 minutes at 4℃. The plasma was separated and stored in polypropylene tubes at -40 ± 5℃until analysis. The collection period of urine samples was -12 to 0 hours
te
before the administration, 0-4 hours,4-8 hours, 8-12 hours, 12-24 hours, 24-48 hours
Ac ce p
and 48-72 hours after intravenous infusion. After the measurement of urine volume, 4 mL urine sample was immediately stored in polypropylene tubes at -40 ± 5℃ until analysis. The pharmacokinetic parameters were calculated using a Phoenix WinNonlin 6.0 software (Pharsight, USA). Differences were considered statistically at p < 0.05 (independent sample t-test). 3.
Results
3.1. Selectivity
The retention time of levornidazole, M1, M2, M4 and IS in plasma and urine samples was 3.34 and 3.41 min, 3.10 and 2.99 min, 1.42 and 1.49 min, 1.65 and 1.57 min, 1.61 and 1.71 min, respectively. They were well separated from plasma and urine components within 3.6 min (Fig.2 and Fig. 3). For both the drug and IS, the chromatograms were free of interfering peaks at their respective retention times. The impact on leavornidazole, M1, M2, M4 and IS was ranging from 80.1 % to 8
Page 8 of 29
116.3% in haemolized plasma and 96.4% to 118.1% in lipemic plasma. The cross-interference result showed that levornidazole, M1, M2, M4 had small (< 2.0%) MS signal influence with each other. 3.2. Linearity
ip t
The calibration curves of levornidazole, M1, M2 and M4 were linear for
concentration ranges 0.0100-5.00 μg/mL, 0.00500-2.50 μg/mL, 0.0200-10.0 μg/mL
cr
and 0.00250-1.25 μg/mL, respectively. The typical calibration curves of levornidazole, M1, M2 and M4 in plasma and urine were represented by the following regression
us
equations: Y = 0.000178 + 1.14 X and Y = 0.00218 + 1.1 X, Y = 0.000381 + 1.71 X and Y = -0.0000269 +1.9 X, Y = 0.000175 + 0.555 X and Y = 0.00106 + 0.375 X, Y =
an
0.000552 + 1.77 X and Y = 0.000234 + 2.14 X. The correlation coefficients (r) were greater than 0.999 for all curves.
M
The lower limit of quantification (LLOQ), defined as the lowest concentration on the calibration curve which signal-to-noise was great than 5:1, was 0.0100, 0.00500,
3.3. Accuracy and precision
d
0.0200 and 0.00250 μg/mL for levornidazole, M1, M2 and M4, respectively.
te
The following validation criteria for accuracy and precision were used to assess
Ac ce p
the suitability of the method: accuracy should be within 85.0 to 115.0% except at the LLOQ where it should be within 80.0 to 120.0%; RSD (relative standard deviation) should not exceed 15.0% except at the LLOQ where it should not exceed 20.0%. The intra- and inter-batch precision for levornidazole in plasma and urine were less than 5.1% and 9.3%, 10.8% and 6.4%, respectively. Assay accuracy, defined as the percentage of the estimated concentration to the nominal concentration, was better than 91.1% and 94.8%, 96.0% and 92.8%, respectively. The intra- and inter-batch accuracy and precision of M1, M2 and M4 were all within the acceptance criteria (Table 3). 3.4. Recovery and matrix effect The mean recovery of levornidazole, M1, M2 and M4 was 91.1-105.1%, 95.8-103.8%, 87.8-96.8%, 96.8-100.6% from plasma and 96.0-100.9%, 96.9-107.9%, 95.1-102.7%, 103.7-105.9% from urine respectively (Table 4). The recovery of IS 9
Page 9 of 29
was113.2 ± 12.4% from plasma and 101.1 ± 2.5% from urine. The matrix effect factor of levornidazole in plasma was 101.2 ± 9.4%, 96.6 ± 3.8%, 96.2 ± 2.9% and 94.8 ± 1.9% at concentration of 0.0100, 0.0300, 0.200 and 4.00 μg/mL, respectively (Table 4). The matrix effects factors of M1, M2 and M4
ip t
were ranged from 91.5 % to 105.7% in plasma and from 95.9% to 109.2% in urine samples respectively. Plasma and urine matrix had little effect on the detection of
cr
levornidazole, M1, M2 and M4. 3.5. Dilution integrity
us
After 10-fold dilution, the accuracy and precision of levornidazole, M1, M2 and M4 was 110.5% and 4.6%, 103.7% and 7.1%, 104.9% and 4.3%, and 106.8% and
an
1.4% for the plasma samples; 103.5 and 1.8%, 118.9 and 6.1%, 93.9 and 3.2%, and 111.0 and 2.3% for the urine samples. For the urine samples after100-fold dilution,
M
the accuracy and precision of levornidazole, M1 and M4 was 98.2% and 3.5%, 101.3% and 3.2%, and 97.6% and 1.7%, respectively .The100-fold dilution of M2
3.6. Stability
d
wasn’t conducted due to its lower concentration in urine.
te
The recovery of the stock solution of levornidazole, M1, M2, M4 and IS was
Ac ce p
(106.2 ± 1.5)%, (111.1 ± 5.7)%, (109.1 ± 6.7)%, (100.9 ± 2.4)% and (99.5 ± 3.2)% after 5-month storage at -40 ± 5℃. The mean recovery of levornidazole were 85.7%-101.7%, 85.5%-95.1% and
94.9%-109.4% after 24 hours placement at room temperature, in the autosampler up
to 48 hours, and after 3 freeze–thaw cycles in human plasma. The recovery in human urine was 97.5%-105.6%, 99.1%-103.1% and 96.4%-113.9% respectively. No degradation of the compound was observed during the long- term stability test at -40 ± 5℃ for 3 months in plasma (recovery, 103.8%- 112.0%) and 4 months in urine
(recovery, 93.8%-112.5%) (Table 5). Just like levornidazole, the plasma and urine samples of M1, M2 and M4 were found stable during the exposure period of 24 hours at room temperature, in the autosampler up to 48 hours, after 3 freeze–thaw cycles and for at least 3 and 4 months, 10
Page 10 of 29
respectively, at -40 ± 5℃. 3.7. Pharmacokinetic evaluation The above UPLC-MS/MS method was applied to a clinical study. The concentration-time course of levornidazole was fairly slow in both 500 mg and 750
ip t
mg group, evidenced by half-life (T1/2) of (11.6 ± 0.8) h and (11.9 ±1.3) h. The
maximum plasma concentration (Cmax) and the area under plasma concentration-time
cr
curve (AUC) were (12.5 ± 3.4) μg/mL and (176 ± 41) μg·h/mL in 500 mg group,
(17.3 ± 2.2) μg/mL and (265 ± 42) μg·h/mL in 750 mg group. Following dosing of
us
500 mg, the Cmax of M1 and M4 was (0.118 ± 0.043) μg/mL and (0.131 ± 0.027) μg/mL. The metabolic ratio (MR=AUC0-∞,metabolite/AUC0-∞,parent drug×100%) was (2.06
an
± 0.55)% and (2.58 ± 0.13)%, respectively. In 750 mg group, the Cmax of M1 and M4 was (0.150 ± 0.060) μg/mL and (0.184 ± 0.030) μg/mL. The MR was (1.63 ± 0.45)%
M
and (2.47 ± 0.39)%. The plasma concentration of M2 was all lower than the LLOQ (0.0200 μg/mL) in either dose group (Table 6).
d
The cumulative urinary excretion rate during 0-72 hours (Ae0-72h) of levornidazole, M1, M2 and M4 was (9.56 ± 2.15)%, (1.05 ± 0.28)%, (0.421 ±
te
0.863)% and (2.62 ± 0.27)% in 500 mg group, and (9.32 ± 3.51)%,(1.03 ± 0.50)%,
Ac ce p
(0.124 ± 0.036)% and (1.97 ± 0.99)% in 750 mg group. 4.
Discussion
Levornidazole is a new drug developed by Sanhome Pharmaceutical Co., Ltd
(Nanjing, China). So far, only a few studies [5,12,13] have examined the metabolites and pharmacokinetics of levornidazole. Here we report the development and validation of a new UPLC-MS/MS method for simultaneous determination of levornidazole and its metabolites in human plasma and urine. In the method development, we preferred APCI rather than ESI due to its smaller
matrix effects and good peak shape of the analytes. Deproteinization or liquid-liquid extraction alone was excluded in preliminary experiments. Deproteinization plus liquid-liquid extraction with acetonitrile is adopted in our study for its faster, cleaner extraction and better sensitivity. Our established LC-MS/MS method is successful in simultaneous determination 11
Page 11 of 29
of levornidazole and its three metabolites M1, M2 and M4, except for M3 and M5. Under this LC-MS condition, the signals of M3 and M5 were influenced significantly by the matrix of plasma and urine, leading to bad result of M3 and M5. M3 signal was low and variable, which makes the detection of M3 unreliable. We did not find out the
ip t
appropriate channel for detection of M5 in our pilot studies due to its smaller molecular weight.
cr
The newly developed UPLC-MS/MS method was applied in a pharmacokinetic study. The results revealed that the pharmacokinetic characteristics of levornidazole
us
were similar after intravenous infusion of single dose of 500 mg and 750 mg
levornidazole. The urinary excretion rates of levornidazole and its phase I metabolites,
an
M1, M2 and M4, were all lower than 10.0%, suggesting that levornidazole did not reach mass balance in human body. M2 was undetectable because its concentration
M
was two low in both plasma and urine samples. In most cases, M2 was below the limit of detection (0.0200 μg/mL). This can be explained by the possibility that minimal
high for M2.
d
levornidazole is metabolized to M2 or the LLOQ of this established method is set too
te
It is also reported that levornidazole has other phase II metabolites in human bile
Ac ce p
[13]. The fecal excretion rate was 22.1% after a single oral dose of 750 mg ornidazole in healthy volunteers [14]. Therefore, we need to further determine the concentration of levornidazole and the other phase I and phase II metabolites in urine and feces to confirm if levornidazole reach mass balance in human body. 5.
Conclusions
We successfully developed and validated an UPLC-MS/MS method for
simultaneous determination of levornidazole and its three phase I metabolites, M1, M2 and M4, in plasma and urine. The established method was also proved useful and reliable in a clinical pharmacokinetic study. Levornidazole, M1, M2 and M4 are detectable with this method in the range of 0.0100-5.00, 0.00500-2.50, 0.0200-10.0 and 0.00250-1.25 μg/mL. It is useful for clinical study in terms of good reproducibility and accuracy, and lower limits of quantification and detection. 12
Page 12 of 29
Acknowledgments We would like to thank Nanjing Sanhome Pharmaceutical Co., Ltd for their help
ip t
in this study. This work was supported by New Drug Creation and Manufacturing Program of the Ministry of Science and Technology of China (2012ZX09303004-001),
cr
and New Drug Creation and Manufacturing Program of the Ministry of Science and
us
Technology of China for phase IV clinical study of the levornidazole and sodium
Ac ce p
te
d
M
an
chloride injection (2012ZX09104101).
13
Page 13 of 29
References [1]
D.E. Schwartz, J.C. Jordan, W. Vetter, G. Oesterhelt, Metabolic studies of ornidazole in the rat, in the dog and in man, Xenobiot. 9 (1979) 571-581. M. Agudelo, O. Vesga, Therapeutic equivalence requires pharmaceutical,
ip t
[2]
pharmacokinetic, and pharmacodynamic identities: true bioequivalence of a generic
cr
product of intravenous metronidazole, Antimicrob. Agent. Chemother. 52 (2012) 2659-2665.
K.C. Lamp, C.D. Freeman, N.E. Klutman, Pharmacokinetics and pharmacodynamics of
us
[3]
the nitroimidazole antimicrobials, Clin. Pharmacokinet. 36 (1999) 353-373. C. Mark, B. Patrice, F. Barry, M. Edward, R. Liam, Development and validation of a
an
[4]
rapid method for the determination and confirmation of 10 nitroimidazoles in animal
M
plasma using liquid chromatography tandem mass spectrometry, J. Chromatogr. B. 877 (2009) 1494-1500. [5]
L.L. Mu, Z.N. Cheng, X. Guo, X. Luo, P. Yu, Investigation of chiral inversion and
d
pharmacokinetics of leavornidazole by high-performance liquid chromatography, J. Clin.
[6]
te
Pharm. Ther. 38(2013) 31-35.
H.W. Sun, F.C. Wang, L.F. Ai, Simultaneous determination of seven nitroimidazole
Ac ce p
residues in meat by using HPLC-UV detection with solid-phase extraction, J.Chromatogr. B. 857 (2007) 296-300.
[7]
H.W. Sun, F.C. Wang, L.F. Ai, Validated method for determination of eight banned
nitroimidazole residues in natural casings by LC/MS/MS with solid-phase extraction, J. AOAC. Int.92 (2009) 612-621.
[8]
J.Q. Huang, G.Y. Cao, X. Hu, C.H. Sun, J.R. Zhang, Chiral separation of rac-ornidazole
and detection of the impurity of (R)-ornidazole in (S)-ornidazole injection and raw material, Chirality. 18 (2006) 587-591. [9]
Y.S.R. Krishnaiah, Y.I. Muzib, P. Bhaskar, G.S. Rao, Reverse-phase HPLC method for the estimation of ornidazole in human plasma, Asian. J. Chem. 15(2003)941-944.
[10]
M. Cronly, P. Behan, B. Foley, E. Maloneb, S. Martin, M. Doyle, L. Regan, Rapidmulti-class multi-residue method for the confirmation of chloramphenicol and 14
Page 14 of 29
eleven nitroimidazoles in milk and honey by liquid chromatography-tandem mass spectrometry (LC-MS), Food. Addit. Contam. 27 (2010)1233-1246. [11]
W.D.S. Solomon, K.V. Gowda, P.S. Selvan, HPLC method for quantification of ornidazole in human plasma, Asian. J. Chem. 20 (2008) 4361-4368. J. Du, Z. Ma, Y. Zhang, T. Wang, X. Chen, D.F Zhong, Enantioselective determination of
ip t
[12]
ornidazole in human plasma by liquid chromatography-tandem mass spectrometry on a
[13]
cr
Chiral-AGP column,J. Pharm. Biomed. Anal. 86 (2013) 182-188.
J.B. Du, P. Deng, X.Y. Chen, H.D. Wang, T.G. You, D.F. Zhong, Characterization of
us
ornidazole metabolites in human bile after intraveneous doses by ultra performance liquid chromatography/quadrupole time-of-flight mass spectrometry, Act. Pharm. Sin. B.
[14]
an
2 (2012) 159-167.
D.E. Schwartz, F. Jeunet, Comparative pharmacokinetic studies of ornidazole and
Ac ce p
te
d
M
metronidazole in man, Chemother. 22 (1976) 19-29.
15
Page 15 of 29
Figure Legends
ip t
Fig. 1. Schematic diagram illustrating the metabolism of ornidazole. *14C labelled, [ ]
Ac ce p
te
d
M
an
us
cr
= postulated intermediates
16
Page 16 of 29
Fig.2. Representative chromatograms of drug-free plasma and the plasma spiked with 0.0100 μg/mL levornidazole, 0.00500 μg/mL M1, 0.0200 μg/mL M2, 0.00250 μg/mL
ip t
M4 and 1.00 μg/mL IS. The red and blue lines represent the chromatograms of the
Ac ce p
te
d
M
an
us
cr
drug-free plasma and the plasma spiked with the specified analyte, respectively.
17
Page 17 of 29
Fig. 3. Representative chromatograms of drug-free urine and the urine spiked with 0.0100 μg/mL levornidazole, 0.00500 μg/mL M1, 0.0200 μg/mL M2, 0.00250 μg/mL
Ac ce p
te
d
M
an
us
cr
drug-free urine sample and the urine spiked with the specified analyte.
ip t
M4 and 1.00 μg/mL IS. The red and blue lines represent the chromatograms of the
18
Page 18 of 29
Table 1 Gradient elution procedure 0.1%Formic acid in water (%)
Acetonitrile (%)
Cure
(A)
0
5
95
0.1
5
95
10
2.5
70
30
10
3.6
5
95
1
Ac ce p
te
d
M
an
us
cr
(B)
ip t
Time(min)
19
Page 19 of 29
Table 2. Concentrations of the standard and QC samples of leavornidazole, M1, M2 and M4 in human plasma and urine The range of calibration standards (μg/mL)
Concentration of QC samples (μg/mL)
Leavornidazole
0.0100~5.00
0.0100,
0.0300,
0.200,
4.00
M1
0.00500~2.50
0.00500,
0.01500,
0.100,
2.00
M2
0.0200~10.0
0.0200,
0.0600,
0.400,
8.00
M4
0.00250~1.25
0.00250,
0.00750,
0.0500,
1.00
cr
ip t
Analyte
Ac ce p
te
d
M
an
us
20
Page 20 of 29
Table 3 Intra- and inter-bath accuracy and precision of levornidazole, M1, M2 and M4 in human plasma and urine samples Intra-batch (n=6) RSD (%)
4.00
91.1 ±4.6
5.1
3.79
0.186
92.8 ±3.7
4.0
0.192
0.0285
94.8 ±3.9
4.1
0.0290
0.0100
0.0105
105.1 ±5.2
4.9
0.0101
2.00
1.92
95.8 ±4.6
4.8
2.09
0.100
0.0966
96.6 ±4.4
4.5
0.0150
0.0150
100.0 ±4.5
4.5
Plasma
0.00500
0.00519
103.8 ±9.0
8.7
8.00
7.41
92.7 ±4.3
4.7
2.0
96.2 ±2.9
3.0
96.6 ±3.8
4.0
101.2 ±9.4
9.3
104.6 ±4.3
4.1
0.104
104.4 ±6.4
6.1
0.0156
104.2 ±6.5
6.2
0.00529
105.7 ±9.0
8.6
7.65
95.6 ±4.4
4.7
0.400
0.351
87.8 ±2.6
3.0
0.366
91.5 ±4.6
5.0
0.0600
0.0558
93.0 ±6.8
7.3
0.0583
97.1 ±6.7
6.9
0.0200
0.0194
1.00
1.01
0.0500
0.0497
0.00750 0.00250
12.2
M
0.0206
103.2 ±10.0 9.7
100.6 ±2.1
2.1
1.00
100.1 ±5.4
5.4
99.4 ±1.8
1.8
0.0494
97.8 ±5.8
5.9
0.00749
99.9 ±5.5
5.5
0.00774
103.2 ±7.3
7.1
0.00242
96.8 ±8.6
8.9
0.00241
96.5 ±6.9
7.2
3.84
96.0 ± 2.7
2.8
3.71
92.8 ± 4.9
4.9
0.196
98.1 ± 3.3
3.4
0.200
100.0 ± 4.9 4.9
0.0295
98.4 ± 3.1
3.2
0.0298
99.3 ± 4.3
0.0100
0.0101
100.9 ± 10.9 10.8
0.0106
105.6 ± 6.7 6.4
2.00
1.94
96.9 ± 2.0
1.98
99.1 ± 3.9
0.100
0.101
100.7 ± 3.5 3.5
0.107
106.8 ± 2.6 2.5
0.0150
0.0154
102.8 ± 7.9 7.7
0.0156
103.7 ± 7.3 7.1
0.00500
0.00540
107.9 ± 8.8 8.2
0.00546
109.2 ± 5.4 5.0
8.00
7.84
98.0 ± 3.6
7.87
98.3 ± 3.6
3.7
0.400
0.4.7
101.7 ± 4.7 4.7
0.384
95.9 ± 4.7
4.9
0.0600
0.0616
102.7 ± 3.3 3.3
0.0599
99.9 ± 9.4
9.4
0.0200
0.0190
95.1 ± 5.5
0.0208
104.0 ± 7.4 7.1
1.00
1.04
104.0 ± 2.8 2.7
0.979
97.9 ± 4.3
0.0500
0.530
105.9 ± 3.6 3.4
0.0515
102.9 ± 2.7 2.6
0.00750
0.00782
104.2 ± 6.7 6.5
0.00792
105.5 ± 4.5 4.2
0.00250
0.00259
103.7 ± 8.4 8.1
0.00263
105.1 ± 7.3 7.0
4.00 Levornida 0.200 zole 0.0300
M1
Urine
M2
M4
te
M4
d
96.8 ±11.8
Ac ce p
M2
RSD (%)
94.8 ±1.9
cr
3.64
Levornida 0.200 zole 0.0300
M1
Determined Recovery concentration (%) (μg/mL)
ip t
Nominal Determined Recovery concentration concentration (%) (μg/mL) (μg/mL)
us
Analyte
an
Matrix
Inter-bath (n=6)
2.1
3.7
5.8
4.3 3.9
4.4
21
Page 21 of 29
Table 4 Absolute recovery and matrix effect of levornidazole, M1, M2 and M4 in human plasma and urine samples
M2
M4
RSD (%) Recovery (%) RSD (%)
4.00
91.1 ±4.6
2.1
94.8 ±1.9
2.0
0.200
92.8 ±3.7
1.8
96.2 ±2.9
3.0
0.0300
94.8 ±3.9
5.5
96.6 ±3.8
4.0
0.0100
105.1 ±5.2
8.9
101.2 ±9.4
9.3
2.00
95.8 ±4.6
4.8
104.6 ±4.3
4.1
0.100
96.6 ±4.4
4.5
104.4 ±6.4
6.1
0.0150
100.0 ±4.5
4.5
104.2 ±6.5
6.2
0.00500
103.8 ±9.0
8.7
105.7 ±9.0
8.6
8.00
92.7 ±4.3
4.7
95.6 ±4.4
4.4
0.400
87.8 ±2.6
3.0
91.5 ±4.6
4.6
Urine
M2
M4
cr
0.0600
93.0 ±6.8
7.3
97.1 ±6.7
6.7
0.0200
96.8 ±11.8
12.2
103.2 ±10.0
10.0
1.00
100.6 ±2.1
5.4
100.1 ±5.4
5.4
0.0500
99.4 ±1.8
5.9
97.8 ±5.8
5.9
0.00750
99.9 ±5.5
7.1
103.2 ±7.3
7.1
0.00250
96.8 ±8.6
7.2
96.5 ±6.9
7.2
96.0 ± 2.7
2.8
92.8 ± 4.9
4.9
98.1 ± 3.3
3.4
100.0 ± 4.9
4.9
0.0300
98.4 ± 3.1
3.2
99.3 ± 4.3
4.3
0.0100
100.9 ± 10.9
10.8
105.6 ± 6.7
6.4
2.00
96.9 ± 2.0
2.1
99.1 ± 3.9
3.9
0.100
100.7 ± 3.5
3.5
106.8 ± 2.6
2.5
0.0150
102.8 ± 7.9
7.7
103.7 ± 7.3
7.1
0.00500
107.9 ± 8.8
8.2
109.2 ± 5.4
5.0
8.00
98.0 ± 3.6
3.7
98.3 ± 3.6
3.7
0.400
101.7 ± 4.7
4.7
95.9 ± 4.7
4.9
0.0600
102.7 ± 3.3
3.3
99.9 ± 9.4
9.4
0.0200
95.1 ± 5.5
5.8
104.0 ± 7.4
7.1
1.00
104.0 ± 2.8
2.7
97.9 ± 4.3
4.4
0.0500
105.9 ± 3.6
3.4
102.9 ± 2.7
2.6
0.00750
104.2 ± 6.7
6.5
105.5 ± 4.5
4.2
0.00250
103.7 ± 8.4
8.1
105.1 ± 7.3
7.0
te
0.200
Ac ce p M1
ip t
Recovery (%)
4.00 Levornidazole
Matrix effect (n=6)
us
Plasma
(μg/mL)
an
M1
Absolute recovery (n=6)
M
Levornidazole
Concentration
d
Matrix Analyte
22
Page 22 of 29
Table 5 Stability data of levornidazole, M1, M2 and M4 under various conditions
3.48
0.200
0.194
101. 0
0.0300
0.0276
85.7
0.0100
0.0095
95.7
1.82
102. 8
M4
Uri ne
Lev orn
3.67 0.196 0.0297
108.2
0.0103
2.06
112.0
108.6
103.8
97.0
Recov ery (%)
ip t
Deter mined concen tration (μg/m L)
3.69
0.173
Deter mined concen tration (μg/m L)
cr
Recov ery (%)
3 freeze-thaw cycles (n=3) Reco very (%)
95.1
3.88
107. 6
85.5
0.202
107. 3
0.0264
88.7
0.0297
109. 4
0.0089
93.9
0.0094 8
94.9
1.93
102. 1
1.75
90.8
105. 7
0.105
109.1
0.0904
86.6
0.104
102. 5
0.100
0.0993
0.0150
0.0145
95.4
0.0158
99.4
0.0144
96.2
0.015
98.0
0.0048 3
89.9
0.0049 1
93.0
0.0046 9
92.0
0.0051
109. 2
8.00
8.50
101. 4
7.67
99.5
6.83
85.9
7.95
108. 9
0.400
0.422
100. 3
0.395
105.8
0.404
94.6
0.427
110. 4
0.00500
M2
Deter mined concen tration (μg/m L)
Postextraction, 48 h at 10℃ (n=3)
an
4.00
Ac ce p
Plas ma
Reco very (%)
101. 7
2.00 M1
Determ ined concent ration (μg/mL )
M
Lev orn ida zol e
Concentr ation (μg/mL)
d
An alyt e
te
Mat rix
storage at -40℃ * (n=3)
us
Long-term
24 h at room temperature (n=3)
0.0600
0.0608
95.8
0.0592
104.6
0.0583
86.3
0.0676
113. 3
0.0200
0.0213
108. 7
0.0195
108.3
0.0223
114.7
0.0195
111. 9
1.00
0.895
99.3
0.989
98.3
0.945
97.3
0.971
98.3
0.0500
0.045
96.1
0.0498
100.1
0.0469
93.4
0.0502
98.0
0.00750
0.0689
92.0
0.0076 5
98.1
0.0061 6
86.1
0.0071 5
87.4
0.00250
0.0024 4
87.1
0.0027 2
102.8
0.0022 4
87.7
0.0023 6
102. 2
3.81
104. 5
4.00
3.7
97.5
3.92
108.6
3.76
103.1
23
Page 23 of 29
0.201
106.7
0.192
101.6
0.212
0.0300
0.0285
105. 6
0.0306
112.5
0.0288
100.5
0.0327
113. 9
93.8
0.0101
99.1
0.0098
96.4
103.4
2.08
99.7
98.4
2.00
1.89
92.2
2.11
0.100
0.0972
96.7
0.106
98.1
0.107
0.0150
0.0151
103. 8
0.0155
94.1
0.0169
0.00500
0.0044 7
81.4
0.0047 6
82
0.0052 4
8.00
7.2
85.7
6.86
96.1
0.400
0.36
85.7
0.357
96.1
0.0600
0.059
90.8
0.0517
0.0200
0.0236
119. 8
0.0169
101.9
2.12
101. 4
0.11
105. 1
0.0169
113. 5 109. 4
ip t
0.0905
cr
0.0100
0.0093 7
113.9
8.26
108.3
6.89
90.3
0.421
104
0.397
98.3
87.0
0.0595
93.5
0.0608
95.4
87.1
0.0172
82.6
0.0218
105. 0
an
us 113.2
0.0050 7
0.966
100.6
0.967
94.2
1.07
103. 9
91.6
0.0492
97.0
0.049
94.2
0.0534
102. 7
100. 6
0.0069 2
85.5
0.0075 9
104.3
0.0078 8
108. 3
107. 7
0.0023 6
85.8
0.0022 6
97.5
0.0021 3
87.8
1.00
0.977
92.1
0.0500
0.0481
0.00750
0.0072 6
0.00250
0.0024 6
Ac ce p
M4
98.6
M
M2
0.185
d
M1
0.200
112. 5
te
ida zol e
* Plasma and urine samples were placed at -40℃ for 3 and 4 months, respectively.
24
Page 24 of 29
Table 6 Pharmacokinetic parameters of levornidazole, M1 and M4 after single dose intravenous infusion of 500 mg and 750 mg levornidazole to healthy Chinese subjects
500mg M1
750mg
500mg 750mg
MRT0-∞
CLt
Vd
MR
Unit
µg/mL
h·µg/mL
h·µg/mL
h
h
L/h
L
%
Mean
12.5
88.7
176
11.6
SD
3.4
16.8
41
0.8
Mean
17.3
201
265
11.9
SD
2.2
32
42
1.3
ip t
T1/2
2.96
49.3
1.7
0.56
9.1
16.6
2.90
50.0
0.51
8.6
cr
16.7
1.8
t*
-4.10
-10.9
-5.22
-0.818
0.120
0.275
0.096
P
0.000
0.000
0.000
0.429
0.905
0.786
0.924
Mean
0.118
1.14
3.85
11.8
23.5
SD
0.043
0.41
1.60
1.5
1.7
/
/
Mean
0.150
2.86
4.53
12.5
23.7
SD
0.060
1.14
1.71
2.4
2.8
/
/
t*
-1.50
-4.94
-0.984
-0.310
-0.226
P
0.147
0.000
0.336
0.759
0.823
/
/
Mean
0.131
1.44
4.52
12.1
23.5
SD
0.027
0.34
1.03
1.2
1.26
/
/
Mean
0.184
3.86
6.44
14.2
25.0
/
/
/
/
t*
-4.56
-12.5
-5.05
-2.22
-1.24
P
0.000
0.000
0.000
0.037
0.238
SD
0.030
0.58
0.82
3.1
3.9
te
M4
AUC0-∞
us
750mg
AUC0-τ
an
Levornidazole
Cmax
M
500mg
Parameter
d
Group
/ / / 2.06 0.55 1.63 0.45 2.06 0.052 2.58 0.13 2.47 0.39 0.896 0.386
Ac ce p
Cmax , maximum plasma concentration; AUC0-∞, area under the plasma concentration versus time curve extrapolated to infinity; T 1 / 2, half life; MRT, mean residence time; CL, creatinine clearance; Vd, apparent volume of distribution; MR, metabolic rate.*independent sample t-test; /, not available.
25
Page 25 of 29
Highlights We developed an UPLC-MS/MS method for simultaneous determination of
•
levornidazole and its three metabolites M1, M2 and M4 in human plasma and urine.
cr
accuracy, recovery and stability under various conditions.
ip t
The method is fully validated in terms of linearity, specificity, precision and
•
The method is successfully applied in a clinical pharmacokinetic study.
•
Ac ce p
te
d
M
an
us
26
Page 26 of 29
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1
Page 27 of 29
Ac ce p
te
d
M
an
us
cr
ip t
Figure 2
Page 28 of 29
Ac ce p
te
d
M
an
us
cr
ip t
Figure 3
Page 29 of 29