- Email: [email protected]
MS method for the determination of phytolaccagenin in rat plasma and application to a pharmacokinetic study
Accepted Manuscript Title: Development and validation of a HPLC-MS/MS method for the determination of phytolaccagenin in rat plasma and application to a pharmacokinetic study Author: Fenghuan Wei Ravi Shankar Prasad Singh Matthias Fueth Steven Swarts Paul Okunieff Hartmut Derendorf PII: DOI: Reference:
S0731-7085(14)00631-1 http://dx.doi.org/doi:10.1016/j.jpba.2014.12.025 PBA 9863
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
29-9-2014 11-12-2014 14-12-2014
Please cite this article as: F. Wei, R.S.P. Singh, M. Fueth, S. Swarts, P. Okunieff, H. Derendorf, Development and validation of a HPLC-MS/MS method for the determination of phytolaccagenin in rat plasma and application to a pharmacokinetic study, Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.12.025 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
Graphical Abstract
2
ip t
COOCH 3
cr
COOH
HO HO
us
CH2 OH
an
3
Ac ce p
te
d
M
4
1
Page 1 of 29
4
Highlights
5
A novel validated LC-MS/MS method for the quantification of hESA in rat plasma
6
The hESA shows large volume of distribution in rats
7
The hESA is rapidly eliminated from systemic circulation in rats
Ac ce p
te
d
M
an
us
cr
ip t
8
2
Page 2 of 29
Development and validation of a HPLC-MS/MS method for the determination of
9
phytolaccagenin in rat plasma and application to a pharmacokinetic study
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
Fenghuan Wei1,3, Ravi Shankar Prasad Singh1, Matthias Fueth1, Steven Swarts2, Paul Okunieff2, Hartmut Derendorf1,* 1 Department of Pharmaceutics, College of Pharmacy, University of Florida, Gainesville, FL32610 2 Department of Radiation Oncology, University of Florida, Gainesville, Florida, United States of America 3 Department of Chinese Medicine Pharmaceutics, College of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China, 510515
Ac ce p
te
d
M
an
us
cr
ip t
8
*Corresponding Address: Hartmut Derendorf, Ph.D. Distinguished Professor and Chair V. Ravi Chandran Professor of Pharmaceutical Sciences Department of Pharmaceutics College of Pharmacy University of Florida 1345 Center Drive, P3-27 Gainesville, FL 32610-0494 Phone (352) 2737856 Fax (352) 3923249 E-mail: [email protected]
3
Page 3 of 29
ABSTRACT
50
Radix Phytolaccae (the dried root of Phytolacca acinosa Roxb. or Phytolacca americana L.) is
51
widely used in east Asian countries for the treatment of inflammation-related diseases. The
52
active component of Radix Phtolaccae is Phytolcaccagenin a triterpenoid saponin.
53
Phytolcaccagenin has anti-inflammatory activities that exceed those of Esculentoside A and its
54
derivatives regarding suppression of LPS-induced inflammation, and has a lower toxicity profile
55
with less hemolysis. To date, no information is available about analytical method and
56
pharmacokinetic studies of phytolaccagenin. To explore PK profile of this compound, a HPLC-
57
MS/MS assay of phytolaccagenin in rat plasma was developed and validated.
an
us
cr
ip t
49
The method was fully validated according to FDA Guidance for industry. The detection was
59
performed by a triple-quadrupole tanderm mass spectrometer with multiple reactions monitoring
60
(MRM) in positive ion mode via electrospray ionization. The monitored transitions were m/z
61
533.2>515.3 for Phytolcaccagenin, and 491.2>473.2 for I.S. The analysis was performed on a
62
Symmetry C18 column (4.6× 50mm, 3.5µm) using gradient elution with the mobile phase
63
consisting of acetonitrile and 0.1% formic acid water at a flow rate of 1ml/min with a 1:1 splitter
64
ratio. The method was validated with a LLOQ of 20 ng/ml and an ULOQ of 1000 ng/ml. The
65
response versus concentration data were fitted with 1/x weighting and the correlation coefficient
66
(r) were greater than 0.999. The average matrix effect and the average extraction recovery were
67
acceptable. This validation in rat plasma demonstrated that phytolaccagenin was stable for 30
68
days when stored below -20 ℃, for 6 h at room temperature (RT, 22 °C), for 12 h at RT for
69
prepared control samples in auto-sampler vials, and during three successive freeze/thaw cycles
70
results at -20 ℃.
71
Ac ce p
te
d
M
58
The validated method has been successfully applied to an intravenous bolus pharmacokinetic 4
Page 4 of 29
72
study of phytolaccagenin in male Sprague-Dawley rats (10 mg/kg,i.v.). Blood samples taken
73
from 0 to 24 h after injection were collected, and data analyzed with WinNonlin. The half-life
74
and clearance were 1.4±0.9 h and 2.1±1.1 L/h/kg, respectively.
75
Keywords: Phytolaccagenin, Anti-inflammatory, HPLC-MS/MS, Validation,
76
study
ip t
cr us
77 78
an
79
M
80
84 85 86
te Ac ce p
83
d
81 82
Pharmacokinetic
87 88 89
5
Page 5 of 29
90
1. Introduction Phytolaccagenin (hEsA) is a triterpenoid saponin aglycone and a secondary metabolite product
92
in Radix Phytolaccae (dried root of Phytolacca acinosa Roxb. and Phytolacca americana L).
93
Radix Phytolaccae is widely used in China (Shang lu, in Chinese and listed in Pharmacopeoia),
94
Japan and Korea for treating inflammation-related diseases such as edema, mastitis, lymphatic
95
congestion, and severe ascites of the abdomen [1,2]. Phytolaccagenin is a hydrolysis product of
96
Esculentoside A (EsA), 3-O-[b-D-glucopyranosyl-(1, 4)-b-D-xylopyranosyl] phytolaccagenin.
97
The latter is the main triterpenoid of total esculentosides and one of the major bioactive
98
compounds in Radix Phytolaccae [3]. EsA is also the first saponin found to have obvious anti-
99
inflammatory effects in several animal models of acute and chronic inflammation [4-6]. The IC50
100
of EsA for the inhibition of platelet activating factor was 1.5 µmol/l [7]. EsA at concentrations of
101
1-10 µmol/l decreased tumour necrosis factor (TNF) production by human monocytes induced
102
by LPS [8]. EsA also decreased the levels of TNF, IL-1 and IL-6 in the sera of passive Heymann
103
nephritis rats and suppressed inflammatory responses in LPS-induced acute lung injury [9,10].
104
However, the sugar moiety within the EsA structure is one of the causes of haemolysis [3,11].
105
Phytolaccagenin, as the aglycone of EsA (hEsA), showed higher inhibitory effects on LPS-
106
induced NO production, lower haemolytic activities than EsA and did not have any significant
107
toxic effect towards animals [3]. The better anti-inflammatory activity and lower toxicity makes
108
phytolaccagenin a better candidate for drug development than the parent EsA.
Ac ce p
te
d
M
an
us
cr
ip t
91
109
Pharmacokinetic profiling is a crucial step in drug development. Studies on the PK
110
characterization of EsA were reported in Beagle dogs [12]. To the best of our knowledge, there is
111
no published analytical method for the determination of hEsA and pharmacokinetic information
112
on this compound. 6
Page 6 of 29
The objective of this study was to develop a specific and sensitive bioanalytical method of
114
hEsA in rat plasma based on the Guidance for Industry Bioanalytical Method Validation (FDA)
115
[13-15], then to evaluate the pharmacokinetics of phytolaccagenin in rats after intravenous (i.v.)
116
injection.
117
2. Materials and methods
118
2.1. Chemicals and reagents
us
cr
ip t
113
Phytolaccagenin was provided by Steven G. Swarts, Ph.D., Department of Radiation
120
Oncology, University of Florida (purity 94.5%). Tween 80, Methanol, acetonitrile, formic acid
121
were all HPLC grade and were purchased from Fisher Scientific (Pittsburgh, PA, USA). Distilled
122
water was purified by a Barnstead Nanopure Diamond UV ultra-pure water system (Dubuque,
123
IA, USA).
124
2.2. Instrumentation and LC-MS/MS conditions
125
2.2.1. Chromatographic conditions
Ac ce p
te
d
M
an
119
126
Chromatographic separations were accomplished on a Symmetry C18 column (4.6×50mm,
127
3.5µm) (Waters, Dublin, Ireland) at room temperature. The mobile phase consisted of 0.1%
128
formic acid water (phase A) and acetonitrile (phase B) at 1.0 ml/min with 1:1 splitter ratio and a
129
gradient elution. The gradient change of 0.1% (v/v) formic acid (A) and acetonitrile (B) was 0-
130
4min, 80% A; 4-5 min, 80-45% A; 5-10 min, 45% A; 10-12 min, 45-80% A. The column was
131
maintained at room temperature (RT, 22°C); a sample injection volume of 50 µl was used.
132
2.2.2. Mass spectrometry conditions
7
Page 7 of 29
The LC-MS/MS system consisted of a PE200 auto-sampler (PerkinElmer, Waltham, MA), a
134
PE200 pump (PerkinElmer, Waltham, MA) and a Triple Quadrupole Mass Spectrometer
135
(API4000, AB Sciex Instruments). The instrument was controlled by Analyst® Software (version
136
1.4.2). The mass spectrometer was operated in the positive ion mode with a TurboIonSpray
137
source. The following instrument parameters were applied: source temperature of 300 °C, ion
138
spray voltage of 5.0 kV, curtain gas of 10 arbitrary units, collision gas of 5 arbitrary units, ion
139
source gas 1 of 35 arbitrary, ion source gas 2 of 10 arbitrary units, declustering potential of 50
140
eV and 60 eV, collision energy of 10 eV and 29 eV, collision cell exit potential of 15 eV and 24
141
eV for the phytolaccagenin, hEsA, and I.S, respectively. Quantification was carried out using
142
multi reaction mode of the transitions of m/z of 533.2 → 515.3 for hEsA and 491.2 → 473.2 for
143
I.S.
144
2.3. Sample preparation
145
2.3.1. Preparation of calibration standards and quality control samples
te
d
M
an
us
cr
ip t
133
The stock solution of hEsA and hydrocortisone hemiadipate (I.S) were prepared in methanol at
147
a concentration of 1.0 mg/ml. The I.S solution was prepared by diluting the I.S stock solution in
148
methanol: acetonitrile (50:50, v/v) to produce the final concentration of 100 ng/ml. Working
149
solutions of hEsA with concentrations in the range of 200-10000 ng/ml were obtained by diluting
150
the stock solution with the mixture solution of 80% of 0.1% formic acid in water and 20% of
151
acetonitrile (diluent). The calibration standard samples were prepared by spiking 90 µl of blank
152
rat plasma with the corresponding working solution (10 µl) to yield eight standards with
153
concentrations ranging from 20 to 1000 ng/ml. For validation, quality control (QC) samples were
154
prepared in the same way as the calibration standard samples at three concentrations (low quality
Ac ce p
146
8
Page 8 of 29
control/LQC=60 ng/ml, medium quality control/MQC=500 ng/ml, and high quality
156
control/HQC=900 ng/ml). All working solutions were stored at -20℃.
157
2.3.2. Plasma preparation
ip t
155
An aliquot of 300 µl I.S solution was added to 100 µl of plasma sample and vortex mixed
159
before centrifugation at 12000 rpm for 10 min at RT; the supernatant was then transferred to
160
auto-sampler vials and 50 µl of sample was injected into the LC-MS/MS system for analysis.
161
2.4. Method validation
an
us
cr
158
For all validation tests, precision is expressed by the CV (%), and accuracy is expressed by
163
bias (%) as deviation of mean from nominal values. The precision (CV %) must not exceed 15%
164
for all levels (20% for the LLOQ as exception), and the accuracy (Bias %) must be within ±15%
165
of the nominal value for all levels (±20% of the nominal value for the LLOQ as exception) (13-
166
14). The system suitability was established before each run by injecting at least six injections of
167
hEsA in diluent [16].
168
2.4.1. Linearity
Ac ce p
te
d
M
162
169
The calibration curve consisted of a double blank sample (rat plasma sample processed
170
without I.S.), a blank sample (rat plasma processed with I.S.), and eight calibration samples
171
covering the range from 20 to 1000 ng/ml. For the calibration and the run to be valid, the
172
coefficient of the determination (r2) should be greater than 0.99, and the accuracy of at least 75%
173
of calibration samples had to remain within ±15%.
174
2.4.2. Carry-over 9
Page 9 of 29
After an analytical HPLC run, two extracted double blank samples were injected immediately
176
after the highest concentration sample of calibration set. Double blank sample must not have
177
hEsA and I.S peak at a signal-to-noise (S/N) ratio of ≥ 3 [3,15].
178
2.4.3. Selectivity
ip t
175
Eight QC samples of hEsA at the LLOQ were prepared in plasma collected from individual
180
rats as described in section 2.3.2 and quantified using a valid calibration curve prepared in
181
pooled plasma from different source. The % bias and precision of these LLOQ should remain
182
within ±20%.
183
2.4.4. Accuracy and precision
M
an
us
cr
179
Eight replicates of QC samples at 60 ng/ml, 500 ng/ml and 900 ng/ml concentration levels
185
were processed as described in section 2.3.2 on three different days to determine intra-day and
186
inter-day accuracies and precision.
187
2.4.5. Recovery and Matrix effect
Ac ce p
te
d
184
188
The recoveries of hEsA from rat plasma was expressed as the mean of area ratios of extracted
189
QC plasma samples divided by the blank plasma samples spiked with hEsA after protein
190
precipitation at the same QC samples. The matrix effects were expressed as the mean of the peak
191
area ratios of the blank plasma samples spiked with hEsA after protein precipitation divided by
192
the injected working solution with hEsA at the same QC concentrations.
193
2.4.6. Dilution test
10
Page 10 of 29
In order to assess the reliability of the method at concentration levels outside the calibration
195
range (20-1000 ng/ml), six dilution samples were prepared by diluting 10,000 ng/ml plasma
196
samples with blank plasma at a 1:10 ratio.
197
2.4.7. Freeze and thaw stability
ip t
194
Six replicates at QC sample concentrations of 60 ng/ml, 500 ng/ml and 900 ng/ml were
199
subjected to three successive overnight freeze/thaw cycles (freezing temperature: at -20℃).
200
These samples were then processed and quantified with a set of calibration samples and QC
201
samples that were not subjected to the freeze/thaw cycles.
202
2.4.8. Biological sample stability on bench-top at room temperature
M
an
us
cr
198
Six replicates at QC sample concentrations of 60 ng/ml, 500 ng/ml and 900 ng/ml were
204
thawed and kept at RT for 6 h before processing as described in section 2.3.2 and quantifying
205
with a set of calibration samples that were processed immediately.
206
2.4.9. Processed sample stability in the auto-sampler at room temperature
Ac ce p
te
d
203
207
Eight replicates of QC samples were processed and quantified. After 12h of storage in the
208
auto-sampler (RT and protected from light), the run was re-injected and re-analyzed with freshly
209
prepared calibration and QCs.
210
2.4.10. Long-term stability at -20 °C
11
Page 11 of 29
Six replicates of hEsA rat plasma samples at three QC sample concentration levels (60
212
ng/ml,500 ng/ml, 900 ng/ml) were stored at -20 °C for one month. Afterwards, the samples were
213
processed and quantified with a set of freshly prepared calibration standards and QC samples
214
(from freshly prepared working solution and stocking solution).
215
2.5. Intravenous bolus pharmacokinetic study
216
2.5.1. Animals
us
an
M
217
cr
ip t
211
Male Sprague Dawley rats (weighing between 280 and 320g) were purchased from Charles River (Wilmington, MA, USA). The rats were single-housed in plastic cages and received a
219
standard chow and water ad libitum during the experiments. All the rats were maintained on a 12
220
h/12 h light/dark cycle. Non-fasted animals were used for the study. All animal experiments were
221
performed according to the policies and guidelines of the Institutional Animal Care and Use
222
Committee (IACUC) of the University of Florida, Gainesville, USA (NIH publication # 85-23).
223
2.5.2. Design of pharmacokinetics study in rats
224
The hEsA was dissolved in an aqueous solution containing 2% ethanol and 5 % (v/v) dimethyl
225
sulfoxide. The hEsA was intravenously administered to rats at the dose of 10 mg/kg. Blood
226
samples (approximately 300 μl) were collected from sublingual vein into BD Vacutainer
227
heparinized tubes at 0 (pre-dose), 0.25, 0.5, 1.0, 1.5, 2.0, 4.0, 6.0, 12.0, 24.0 h after dosing.
Ac ce p
te
d
218
12
Page 12 of 29
Approximately 500 μl of isotonic saline were replaced by intraperitoneal injection in order to
229
maintain the blood fluid after each bleeding. The blood samples were centrifuged at 2000 rpm
230
for 10 min, then the supernatant plasma were stored at -20 °C for analysis.
231
2.5.3. Sample analysis
an
us
cr
ip t
228
The plasma samples were processed using the extraction procedure described in section 2.3.2
233
and analyzed using LC-MS/MS method described in section 2.2. A calibration curve was fitted
234
by weighted linear regression as described in section 2.4.1. and the concentrations in plasma
235
samples were calculated.
236
2.5.4. Pharmacokinetic data analysis
d te
Ac ce p
237
M
232
The Non-compartmental analysis was performed using WinNonlin software (version 5.2.1,
238
Pharsight, St, MO, USA). The pharmacokinetic (PK) parameters determined were the
239
concentration at time 0 (C0), the terminal elimination rate constant, the terminal elimination half-
240
life (t1/2), the area under the curve (AUC), the area under the first moment curve (AUMC), the
241
mean residence time (MRT), the volume of distribution at terminal phase (Vz), and the clearance
242
(Cl). AUC 0→last was calculated using a linear trapezoidal method from time 0 to the last observed
243
concentration time point.
244
3. Results and Discussion 13
Page 13 of 29
245
3.1. Optimization of HPLC-MS/MS conditions The MS spectra of hEsA and I.S were analyzed in positive ion mode. The negative ion mode
247
was also tested, but the intensity obtained was very low for the hEsA. For optimizing the MS
248
parameters, a neat standard solution containing hEsA or I.S was directly infused into the mass
249
spectrometer. The two compounds were selected [M+H]+ because of their better response and
250
better stability than [M+Na]+, and the response of [M+H]+ was enough for the mass analysis. To
251
obtain chromatograms with satisfactory resolution and appropriate retention time, different
252
mobile phases were evaluated to optimize the analytical performance. Acetonitrile was found to
253
produce better resolution than methanol or the mixture of methanol with acetonitrile. With
254
addition of 0.1% formic acid to the mobile phase and gradient elution, the peak symmetry of the
255
two compounds were all greatly improved. The final acetonitrile－0.1% formic acid water
256
solution with gradient elution program was adopted as the mobile phase for hEsA and I.S. The
257
retention times were approximately 6.85min and 7.35min for I.S and hEsA, respectively. The
258
range of linear response was established by determining the accuracy and precision of calibration
259
curve samples. The lowest concentration with accuracy (% bias) and precision below ±20% was
260
selected as LLOQ.
261
3.1. Method validation
262
3.1.1. Linearity test
Ac ce p
te
d
M
an
us
cr
ip t
246
263
The response versus concentration data, in the range 20-1000 ng/ml for hEsA, was assessed by
264
analyzing the calibration curves using the peak area ratios of analyte to I.S versus the nominal
265
concentration of the calibration standard with a weighting factor (1/x). The average coefficient of
266
determination (r) was equal to 0.9994, and all calibration curves met the acceptance criteria. 14
Page 14 of 29
267 268
3.1.2. Assessment of carry-over The impact of the carry-over on blank rat samples following the highest calibration sample was assessed for both hEsA and I.S. The S/N ratios of hEsA and I.S were lower than 3,
270
indicating that the carry-over has no impact on the results.
271
3.1.3. Selectivity
cr
ip t
269
The selectivity precision (CV %) was 7.23%, and the accuracy (Bias %) was -0.69% (Table 1).
273
Since the precision was below 20%, and the accuracy was within ±20% at the LLOQ, this
274
analytical method for the quantification of hEsA in rat plasma was shown to be selective.
275
3.1.4. Accuracy and precision
M
an
us
272
The intra-day and inter-day precision CV % did not exceed 8.02% at any QC concentration
277
levels. The intra-day and inter-day accuracy bias % was between -10.82% and 3.73%, and
278
between -6.85% and -2.50%, respectively (Table 2). Since both intra-day and inter-day precision
279
was below 15%, and accuracy was within ±15%, this bioanalytical method was proved to be
280
precise and accurate.
281
3.1.5. Recovery and Matrix effect
te
Ac ce p
282
d
276
The matrix effect was between 128.46% and 106.42% for hEsA and 83.30% for I.S. (Table 3).
283
The recovery was between 113.01% and 114.88% for hEsA and 115.32% for I.S. (Table 3).
284
3.1.6. Dilution integrity
15
Page 15 of 29
The precision (CV %) and the accuracy (Bias %) of the diluted samples with blank plasma at a
286
1:10 ratio was 10.57% and 3.80%, respectively. Thus the dilution effect on the precision and
287
accuracy of the results was acceptable.
288
3.1.7. Freeze-thaw stability
ip t
285
After three successive overnight freeze/thaw cycles for the three concentration samples (60-
290
500-900 ng/ml), the precision (CV %) did not exceed 10.99%. The accuracy (Bias %) was
291
between -8.00% and -0.39% (Table 4). Hence, the hEsA in rat plasma proved to be stable after
292
three freeze/thaw cycles at -20 ℃.
293
3.1.8. Bench-top stability at room temperature
M
an
us
cr
289
After storage for 6h at room temperature, the precision (CV %) of the QC samples did not
295
exceed 6.16% and the accuracy (Bias %) was between -0.25% and 9.77% (Table 4). This
296
indicated that hEsA in rat plasma was stable after storing for 6 h at RT prior to processing.
297
3.1.9. Post-preparative stability
Ac ce p
te
d
294
298
Before and after 12h of storage for the QC samples, the precision (CV%) did not exceed
299
7.23%. The accuracy (Bias %) was between -2.08% and -0.02% (Table 4), demonstrating that
300
hEsA in rat plasma was stable up to 12 h in the auto-sampler (RT, protected from light).
301
3.1.10. Long-term stability at -20 °C
16
Page 16 of 29
After storage at -20℃ for one month, the precision of the samples didn’t exceed 3.46% and
303
the accuracy was between -5.17% and 1.19% (Table 4). The results demonstrated that hEsA in
304
rat plasma was stable for at least one month at -20 °C.
3.2. Pharmacokinetic data analysis
an
305
us
cr
ip t
302
This validated analytical method was applied to a pharmacokinetic study of hEsA in rats after
307
a single i.v. dose of 10 mg/kg. For all runs, calibration curves and QC samples met the required
308
acceptance criteria. Typical MRM chromatograms of blank rat plasma spiked with hEsA and I.S
309
is shown in Figure 2. Table 6 summarizes the main PK parameters of hEsA calculated by non-
310
compartmental analysis using WinNonlin software. The initial concentration (C0) was 8277
311
ng/ml, and the half time (t1/2) was 1.36 h. The area under the concentration-time curve (AUC0-last),
312
calculated based on the trapezoidal rule, was 5821 ng*h/ml and the clearance was 2.05 L/h/kg.
313
The concentration time profile (Figure 3) shows rapid decrease in systemic concentration of
314
hESA.
315
4. Conclusion
Ac ce p
te
d
M
306
316
The standard calibration curve of the phytolaccagenin, hEsA, between 20 and 1000 ng/ml with
317
the weighting 1/x was generated and the limit of quantification was 20 ng/ml. The rat plasma
318
samples containing hEsA could be diluted up to 10 fold without affecting the precision and
319
accuracy. The carryover had no effect on the results. The hEsA was found to be stable in rat 17
Page 17 of 29
plasma samples kept for 6 h on the bench at RT, after three successive freeze/thaw cycles, and
321
processed plasma samples stored at RT for 12 h in autosampler. The validation results
322
demonstrate that the analytical method is specific, selective, precise, accurate and capable of
323
producing reliable results. Hence, the method developed here is reliable for the quantification of
324
hEsA in rat plasma samples. The hESA has high volume of distribution and is rapidly eliminated
325
from systemic circulation in rats.
cr
ip t
320
us
326
Acknowledgements
328
We thank Yufei Tang (Department of Pharmaceutics, University of Florida) for the assistance in
329
the LC–MS/MS setup and maintenance.
331
M
te
d
330
an
327
References
333
[1] Editorial Committee of Pharmacopoeia of Ministry of Health PR China, The Pharmacopeoia
334
of People’s Republic of China (Part I), China, Chemical Industry Press, Beijing 2010, pp.304-
335
305.
336
[2] Y. Suga, Y. Maruyama, S. Kawanishi, J. Shoji, Studies on the structures of
337
phytolaccasaponin B, E and G from the roots of Phytolacca americana L, Chem. Pharm.
338
Bull.26(1978) 520-525.
339
[3]W. Gong , Z.H. Jiang, P. Sun, L. Li, Y.S. Jin, L.C. Shao, et al., Synthesis of Novel
340
Derivatives of Esculentoside A and Its Aglycone Phytolaccagenin, and Evaluation of Their
Ac ce p
332
18
Page 18 of 29
Haemolytic Activity and Inhibition of Lipopolysaccharide-Induced Nitric Oxide Production,
342
Chem. Biodivers. 8 (2011)1833-1852.
343
[4]Q.Y. Zheng, K. Mai, X.F. Pan , Y.H. Yi, Anti-inflammatory effects of esculentoside A, Chin.
344
J. Pharmacol. Toxicol. 6(1992)221-224.
345
[5] Z.Y. Xiao, Q.Y.Zheng, Y.Y.Jiang, B. Zhou, M.Yin, H.B.Wang, et al., Effects of
346
esculentoside A on production of interleukin-1, 2, and prostaglandin E2, Acta. Pharmacol. Sin.
347
25(2004) 817-821.
348
[6] Z. Xiao, Y. Su, S. Yang, L. Yin, W. Wang, Y. Yi, et al., Protective effect of esculentoside A
349
on radiation-induced dermatitis and fibrosis, Int. J.Radiat. Oncol. Biol. Phys. 65(2006)882–889.
350
[7] J. Fang, Q.Y. Zheng, Inhibitory effect of esculentoside A on platelet activating factor
351
released from calcimycin induced rat peritoneal macrophages, Acta. Pharm. Sin. 26(1991)721-
352
724.
353
[8] Q.Y. Zheng, J. Fang, H.B. Wang, Inhibitory effect of esculentoside A on TNF production by
354
human monocytes, Acad. J. Sec. Milit. Med. Univ. 18(1997) 415-417.
355
[9] D.W.Ju, Q.Y.Zheng, H.B.Wang, X.J. Guan, J. Fang, Y.H.Yi, Therapeutic effects of
356
esculentoside A on passive Heymann nephritis in rats and its inhibition on cytokine production,
357
Acta. Pharm. Sin. 34(1999) 9-12.
358
[10] W.F. Zhong, L.X.Jiang, J.Y.Wei, A.N.Qiao, M.M.Wei, L.W.Soromou, et al., Protective
359
effect of esculentoside A on lipopolysaccharide-induced acute lung injury in mice, J. Surg. Res.
360
185(2013) 364-372.
361
[11] F. Wu, Y.H.Yi, P. Sun, D.Z. Zhang, Synthesis, in vitro inhibitory activity towards COX-2
362
and haemolytic activity of derivatives of Esculentoside A, Bioorg. Med. Chem. Lett.
363
23(2007)6430-6433.
Ac ce p
te
d
M
an
us
cr
ip t
341
19
Page 19 of 29
[12] X.D. Guan, H.C. Chang, F.L. Sun, X.H. Chen, W. Zhang, G.R. Fan, Determination of
365
esculentoside A in dog plasma by LC–MS/MS method: Application to pre-clinical
366
pharmacokinetics, J. Pharm. Biomed. Anal. 72(2013)261-266.
367
[13] FDA,Guidance for Industry: Bioanalytical Method Validation, 2001,
368
http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/u
369
cm070107.pdf.
370
[14] FDA,Guidance for Industry (Draft Guidance): Bioanalytical Method Validation, 2013,
371
http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm
372
368107.pdf
373
[15]M. Whitmire, J. Ammerman, P. Lisio, J. Killmer, D. Kyle, E. Mainstone, et al., LC-MS/MS
374
Bioanalysis Method Development, Validation, and Sample Analysis: Points to Consider When
375
Conducting Nonclinical and Clinical Studies in Accordance with Current Regulatory Guidances.
376
J Anal Bioanal Techniques. S4. http://dx.doi.org/10.4172/2155-9872.S4-001
377
[16] S. Chandran, R.S.P. Singh, Comparison of various international guidelines for analytical
378
method validation, Pharmazie, 62(2007) 4-14.
Ac ce p
te
d
M
an
us
cr
ip t
364
20
Page 20 of 29
379
Figure captions:
381
Figure1: Chemical structures of the phytolaccagenin, hEsA (A) and Hydrocortisone hemiadipate
382
(B)
383
Figure 2: The extracted LC-MS/MS chromatograms of (A) blank rat plasma spiked with the
384
phytolaccagenin, hEsA, and I.S (400 ng/ml and 100 ng/ml, respectively), (B) plasma sample
385
from a rat obtained at 1.5h after i.v. administration of hEsA (786 ng/ml) with I.S (100 ng/ml)
386
Figure 3: Mean±SD plasma concentration and log mean concentration (inner panel)-time profile
387
of hEsA in male Sprague-Dawley rats (n=3) following i.v. administration (10 mg/kg)
M
an
us
cr
ip t
380
Ac ce p
te
d
388
21
Page 21 of 29
388 389
Figure 1
390
ip t
391 392 COOCH 3
cr
O
O
HO
us
COOH
HO
H
CH2 OH
A
400 401 402
te Ac ce p
399
B
d
396
398
H
M
395
397
O
O
an
394
OH
H
HO
393
O OH
403 404
22
Page 22 of 29
404 405
Figure 2 TIC 6.88
3.5e4
I.S
3.0e4
7.41
hEsA
ip t
Intensity
2.5e4 2.0e4 1.5e4
cr
1.0e4 5.0e3
1
2
3
4
5
6 Time, min
7
8
9
10
11
us
0.0e0
406 407 TIC 4.5e4
an
7.32
4.0e4
hEsA
3.5e4
2.5e4
I.S
2.0e4
6.84
1.5e4 1.0e4
M
Intensity
3.0e4
410 411 412 413
2
3
4
5
6 Time, min
7
8
9
10
11
te
409
1
Ac ce p
0.0e0
408
d
5.0e3
414
23
Page 23 of 29
Figure 3
418 419 420
te
417
Ac ce p
416
d
M
an
us
cr
ip t
414 415
421 422 423
24
Page 24 of 29
424
Table 1: Selectivity test at the LLOQ for hEsA. a
(ng/ml)
(ng/ml)
1
20
18.6
2
20
21.7
3
20
18.0
4
20
21.8
5
20
18.8
6
20
19.4
7
20
20.8
8
20
19.8
a
427
value.
429 430
-0.7
us an M
te
S.D.: standard deviation; CV: coefficient of variation; Bias: deviation of mean from nominal
Ac ce p
428
7.2
d
425 426
1.4
cr
19.9
Bias (%)
ip t
HPLC Run Nominal level Calculated values Mean S.D. CV(%)
431 432 433
25
Page 25 of 29
434
Table 2: Precision and accuracy for hEsA in rat plasma during validation. Accuracy and Precision of controls
900 ng/ml
435 436 437 438
(n=8)
(n=8)
(n=24)
Mean
53.8
55.5
58.4
55.9
S.D
3.2
2.4
2.0
3.1
CV (%)
5.9
4.4
3.4
Bias (%)
-10.8
-7.5
-2.8
Mean
504.4
471.0
518.6
498.0
S.D
30.2
20.1
18.1
30.3
CV (%)
6.0
4.3
3.5
6.1
Bias (%)
0.9
-5.8
3.7
-0.4
Mean
901.0
802.6
928.8
877.5
S.D
66.6
21.3
36.8
70.4
CV (%)
7.4
2.7
4.0
8.0
Bias (%)
0.1
-10.8
3.2
-2.5
cr
ip t
(n=8)
5.6
-6.9
us
an
M
ng/ml
Overall
d
500
Day 3
te
ng/ml
Day 2
Ac ce p
60
Day 1
439 440 441
26
Page 26 of 29
442
Table 3: Matrix effect and recovery of hEsA and I.S. Nominal concentration (ng/ml)
Mean extraction recovery (%)
Mean matrix effect (%)
60
113.0±5.6
128.5±4.2
500
113.6±5.2
119.4±7.1
114.9±2.6
106.4±3.7
cr
us
900
ip t
hEsA
100
an
hemiadipate-hydrocortisone 115.3±7.6
M
443
447 448 449
te Ac ce p
446
d
444 445
83.3±5.4
450 451 452
27
Page 27 of 29
Table 4: Short-term and long-term stabilities during storage in various conditions (n=8)
(ng/ml)
Remaining over nominal concentration (%) Post-preparative Three
Bench-top stability Long-term stability at
stability
at room temperature -20℃
freeze/thaw cycles at -20℃
98.0±5.9
92.0±11.0
99.8±5.3
500
99.1±3.6
98.3±3.2
109.8±6.2
900
98.8±5.4
99.6±3.7
104.4±3.4
461 462
te Ac ce p
460
d
456
459
101.2±2.9
M
455
458
96.6±3.5
an
454
457
94.8±2.3
us
60
ip t
Concentration
cr
453
463 464
28
Page 28 of 29
Table 5: The non-compartmental pharmacokinetic parameters of hEsA in rat plasma after i.v. (10
466
mg/kg) administration to SD rats (N=3).
Ke (1/h)
0.7 ±0.4
AUC0-last (ng*h/ml)
5821 ±3444
AUC 0 -∞ (ng*h/ml)
5999 ±3373
AUMClast (ng*h/ml)
4601±1581
AUMC 0 -∞ (ng*h/ml)
5993±1555 0.9±0.2
Vz (L/kg)
4.2±3.0
CL (L/h/kg)
2.1±1.1
te
MRT (hr)
cr
1.4 ±0.9
us
t1/2 (h)
an
8277± 6361
M
C0 (ng/ml)
d
Mean±SD
Ac ce p
467
Parameters
ip t
465
29
Page 29 of 29
S0731-7085(14)00631-1 http://dx.doi.org/doi:10.1016/j.jpba.2014.12.025 PBA 9863
To appear in:
Journal of Pharmaceutical and Biomedical Analysis
Received date: Revised date: Accepted date:
29-9-2014 11-12-2014 14-12-2014
Please cite this article as: F. Wei, R.S.P. Singh, M. Fueth, S. Swarts, P. Okunieff, H. Derendorf, Development and validation of a HPLC-MS/MS method for the determination of phytolaccagenin in rat plasma and application to a pharmacokinetic study, Journal of Pharmaceutical and Biomedical Analysis (2014), http://dx.doi.org/10.1016/j.jpba.2014.12.025 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
Graphical Abstract
2
ip t
COOCH 3
cr
COOH
HO HO
us
CH2 OH
an
3
Ac ce p
te
d
M
4
1
Page 1 of 29
4
Highlights
5
A novel validated LC-MS/MS method for the quantification of hESA in rat plasma
6
The hESA shows large volume of distribution in rats
7
The hESA is rapidly eliminated from systemic circulation in rats
Ac ce p
te
d
M
an
us
cr
ip t
8
2
Page 2 of 29
Development and validation of a HPLC-MS/MS method for the determination of
9
phytolaccagenin in rat plasma and application to a pharmacokinetic study
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
Fenghuan Wei1,3, Ravi Shankar Prasad Singh1, Matthias Fueth1, Steven Swarts2, Paul Okunieff2, Hartmut Derendorf1,* 1 Department of Pharmaceutics, College of Pharmacy, University of Florida, Gainesville, FL32610 2 Department of Radiation Oncology, University of Florida, Gainesville, Florida, United States of America 3 Department of Chinese Medicine Pharmaceutics, College of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China, 510515
Ac ce p
te
d
M
an
us
cr
ip t
8
*Corresponding Address: Hartmut Derendorf, Ph.D. Distinguished Professor and Chair V. Ravi Chandran Professor of Pharmaceutical Sciences Department of Pharmaceutics College of Pharmacy University of Florida 1345 Center Drive, P3-27 Gainesville, FL 32610-0494 Phone (352) 2737856 Fax (352) 3923249 E-mail: [email protected]
3
Page 3 of 29
ABSTRACT
50
Radix Phytolaccae (the dried root of Phytolacca acinosa Roxb. or Phytolacca americana L.) is
51
widely used in east Asian countries for the treatment of inflammation-related diseases. The
52
active component of Radix Phtolaccae is Phytolcaccagenin a triterpenoid saponin.
53
Phytolcaccagenin has anti-inflammatory activities that exceed those of Esculentoside A and its
54
derivatives regarding suppression of LPS-induced inflammation, and has a lower toxicity profile
55
with less hemolysis. To date, no information is available about analytical method and
56
pharmacokinetic studies of phytolaccagenin. To explore PK profile of this compound, a HPLC-
57
MS/MS assay of phytolaccagenin in rat plasma was developed and validated.
an
us
cr
ip t
49
The method was fully validated according to FDA Guidance for industry. The detection was
59
performed by a triple-quadrupole tanderm mass spectrometer with multiple reactions monitoring
60
(MRM) in positive ion mode via electrospray ionization. The monitored transitions were m/z
61
533.2>515.3 for Phytolcaccagenin, and 491.2>473.2 for I.S. The analysis was performed on a
62
Symmetry C18 column (4.6× 50mm, 3.5µm) using gradient elution with the mobile phase
63
consisting of acetonitrile and 0.1% formic acid water at a flow rate of 1ml/min with a 1:1 splitter
64
ratio. The method was validated with a LLOQ of 20 ng/ml and an ULOQ of 1000 ng/ml. The
65
response versus concentration data were fitted with 1/x weighting and the correlation coefficient
66
(r) were greater than 0.999. The average matrix effect and the average extraction recovery were
67
acceptable. This validation in rat plasma demonstrated that phytolaccagenin was stable for 30
68
days when stored below -20 ℃, for 6 h at room temperature (RT, 22 °C), for 12 h at RT for
69
prepared control samples in auto-sampler vials, and during three successive freeze/thaw cycles
70
results at -20 ℃.
71
Ac ce p
te
d
M
58
The validated method has been successfully applied to an intravenous bolus pharmacokinetic 4
Page 4 of 29
72
study of phytolaccagenin in male Sprague-Dawley rats (10 mg/kg,i.v.). Blood samples taken
73
from 0 to 24 h after injection were collected, and data analyzed with WinNonlin. The half-life
74
and clearance were 1.4±0.9 h and 2.1±1.1 L/h/kg, respectively.
75
Keywords: Phytolaccagenin, Anti-inflammatory, HPLC-MS/MS, Validation,
76
study
ip t
cr us
77 78
an
79
M
80
84 85 86
te Ac ce p
83
d
81 82
Pharmacokinetic
87 88 89
5
Page 5 of 29
90
1. Introduction Phytolaccagenin (hEsA) is a triterpenoid saponin aglycone and a secondary metabolite product
92
in Radix Phytolaccae (dried root of Phytolacca acinosa Roxb. and Phytolacca americana L).
93
Radix Phytolaccae is widely used in China (Shang lu, in Chinese and listed in Pharmacopeoia),
94
Japan and Korea for treating inflammation-related diseases such as edema, mastitis, lymphatic
95
congestion, and severe ascites of the abdomen [1,2]. Phytolaccagenin is a hydrolysis product of
96
Esculentoside A (EsA), 3-O-[b-D-glucopyranosyl-(1, 4)-b-D-xylopyranosyl] phytolaccagenin.
97
The latter is the main triterpenoid of total esculentosides and one of the major bioactive
98
compounds in Radix Phytolaccae [3]. EsA is also the first saponin found to have obvious anti-
99
inflammatory effects in several animal models of acute and chronic inflammation [4-6]. The IC50
100
of EsA for the inhibition of platelet activating factor was 1.5 µmol/l [7]. EsA at concentrations of
101
1-10 µmol/l decreased tumour necrosis factor (TNF) production by human monocytes induced
102
by LPS [8]. EsA also decreased the levels of TNF, IL-1 and IL-6 in the sera of passive Heymann
103
nephritis rats and suppressed inflammatory responses in LPS-induced acute lung injury [9,10].
104
However, the sugar moiety within the EsA structure is one of the causes of haemolysis [3,11].
105
Phytolaccagenin, as the aglycone of EsA (hEsA), showed higher inhibitory effects on LPS-
106
induced NO production, lower haemolytic activities than EsA and did not have any significant
107
toxic effect towards animals [3]. The better anti-inflammatory activity and lower toxicity makes
108
phytolaccagenin a better candidate for drug development than the parent EsA.
Ac ce p
te
d
M
an
us
cr
ip t
91
109
Pharmacokinetic profiling is a crucial step in drug development. Studies on the PK
110
characterization of EsA were reported in Beagle dogs [12]. To the best of our knowledge, there is
111
no published analytical method for the determination of hEsA and pharmacokinetic information
112
on this compound. 6
Page 6 of 29
The objective of this study was to develop a specific and sensitive bioanalytical method of
114
hEsA in rat plasma based on the Guidance for Industry Bioanalytical Method Validation (FDA)
115
[13-15], then to evaluate the pharmacokinetics of phytolaccagenin in rats after intravenous (i.v.)
116
injection.
117
2. Materials and methods
118
2.1. Chemicals and reagents
us
cr
ip t
113
Phytolaccagenin was provided by Steven G. Swarts, Ph.D., Department of Radiation
120
Oncology, University of Florida (purity 94.5%). Tween 80, Methanol, acetonitrile, formic acid
121
were all HPLC grade and were purchased from Fisher Scientific (Pittsburgh, PA, USA). Distilled
122
water was purified by a Barnstead Nanopure Diamond UV ultra-pure water system (Dubuque,
123
IA, USA).
124
2.2. Instrumentation and LC-MS/MS conditions
125
2.2.1. Chromatographic conditions
Ac ce p
te
d
M
an
119
126
Chromatographic separations were accomplished on a Symmetry C18 column (4.6×50mm,
127
3.5µm) (Waters, Dublin, Ireland) at room temperature. The mobile phase consisted of 0.1%
128
formic acid water (phase A) and acetonitrile (phase B) at 1.0 ml/min with 1:1 splitter ratio and a
129
gradient elution. The gradient change of 0.1% (v/v) formic acid (A) and acetonitrile (B) was 0-
130
4min, 80% A; 4-5 min, 80-45% A; 5-10 min, 45% A; 10-12 min, 45-80% A. The column was
131
maintained at room temperature (RT, 22°C); a sample injection volume of 50 µl was used.
132
2.2.2. Mass spectrometry conditions
7
Page 7 of 29
The LC-MS/MS system consisted of a PE200 auto-sampler (PerkinElmer, Waltham, MA), a
134
PE200 pump (PerkinElmer, Waltham, MA) and a Triple Quadrupole Mass Spectrometer
135
(API4000, AB Sciex Instruments). The instrument was controlled by Analyst® Software (version
136
1.4.2). The mass spectrometer was operated in the positive ion mode with a TurboIonSpray
137
source. The following instrument parameters were applied: source temperature of 300 °C, ion
138
spray voltage of 5.0 kV, curtain gas of 10 arbitrary units, collision gas of 5 arbitrary units, ion
139
source gas 1 of 35 arbitrary, ion source gas 2 of 10 arbitrary units, declustering potential of 50
140
eV and 60 eV, collision energy of 10 eV and 29 eV, collision cell exit potential of 15 eV and 24
141
eV for the phytolaccagenin, hEsA, and I.S, respectively. Quantification was carried out using
142
multi reaction mode of the transitions of m/z of 533.2 → 515.3 for hEsA and 491.2 → 473.2 for
143
I.S.
144
2.3. Sample preparation
145
2.3.1. Preparation of calibration standards and quality control samples
te
d
M
an
us
cr
ip t
133
The stock solution of hEsA and hydrocortisone hemiadipate (I.S) were prepared in methanol at
147
a concentration of 1.0 mg/ml. The I.S solution was prepared by diluting the I.S stock solution in
148
methanol: acetonitrile (50:50, v/v) to produce the final concentration of 100 ng/ml. Working
149
solutions of hEsA with concentrations in the range of 200-10000 ng/ml were obtained by diluting
150
the stock solution with the mixture solution of 80% of 0.1% formic acid in water and 20% of
151
acetonitrile (diluent). The calibration standard samples were prepared by spiking 90 µl of blank
152
rat plasma with the corresponding working solution (10 µl) to yield eight standards with
153
concentrations ranging from 20 to 1000 ng/ml. For validation, quality control (QC) samples were
154
prepared in the same way as the calibration standard samples at three concentrations (low quality
Ac ce p
146
8
Page 8 of 29
control/LQC=60 ng/ml, medium quality control/MQC=500 ng/ml, and high quality
156
control/HQC=900 ng/ml). All working solutions were stored at -20℃.
157
2.3.2. Plasma preparation
ip t
155
An aliquot of 300 µl I.S solution was added to 100 µl of plasma sample and vortex mixed
159
before centrifugation at 12000 rpm for 10 min at RT; the supernatant was then transferred to
160
auto-sampler vials and 50 µl of sample was injected into the LC-MS/MS system for analysis.
161
2.4. Method validation
an
us
cr
158
For all validation tests, precision is expressed by the CV (%), and accuracy is expressed by
163
bias (%) as deviation of mean from nominal values. The precision (CV %) must not exceed 15%
164
for all levels (20% for the LLOQ as exception), and the accuracy (Bias %) must be within ±15%
165
of the nominal value for all levels (±20% of the nominal value for the LLOQ as exception) (13-
166
14). The system suitability was established before each run by injecting at least six injections of
167
hEsA in diluent [16].
168
2.4.1. Linearity
Ac ce p
te
d
M
162
169
The calibration curve consisted of a double blank sample (rat plasma sample processed
170
without I.S.), a blank sample (rat plasma processed with I.S.), and eight calibration samples
171
covering the range from 20 to 1000 ng/ml. For the calibration and the run to be valid, the
172
coefficient of the determination (r2) should be greater than 0.99, and the accuracy of at least 75%
173
of calibration samples had to remain within ±15%.
174
2.4.2. Carry-over 9
Page 9 of 29
After an analytical HPLC run, two extracted double blank samples were injected immediately
176
after the highest concentration sample of calibration set. Double blank sample must not have
177
hEsA and I.S peak at a signal-to-noise (S/N) ratio of ≥ 3 [3,15].
178
2.4.3. Selectivity
ip t
175
Eight QC samples of hEsA at the LLOQ were prepared in plasma collected from individual
180
rats as described in section 2.3.2 and quantified using a valid calibration curve prepared in
181
pooled plasma from different source. The % bias and precision of these LLOQ should remain
182
within ±20%.
183
2.4.4. Accuracy and precision
M
an
us
cr
179
Eight replicates of QC samples at 60 ng/ml, 500 ng/ml and 900 ng/ml concentration levels
185
were processed as described in section 2.3.2 on three different days to determine intra-day and
186
inter-day accuracies and precision.
187
2.4.5. Recovery and Matrix effect
Ac ce p
te
d
184
188
The recoveries of hEsA from rat plasma was expressed as the mean of area ratios of extracted
189
QC plasma samples divided by the blank plasma samples spiked with hEsA after protein
190
precipitation at the same QC samples. The matrix effects were expressed as the mean of the peak
191
area ratios of the blank plasma samples spiked with hEsA after protein precipitation divided by
192
the injected working solution with hEsA at the same QC concentrations.
193
2.4.6. Dilution test
10
Page 10 of 29
In order to assess the reliability of the method at concentration levels outside the calibration
195
range (20-1000 ng/ml), six dilution samples were prepared by diluting 10,000 ng/ml plasma
196
samples with blank plasma at a 1:10 ratio.
197
2.4.7. Freeze and thaw stability
ip t
194
Six replicates at QC sample concentrations of 60 ng/ml, 500 ng/ml and 900 ng/ml were
199
subjected to three successive overnight freeze/thaw cycles (freezing temperature: at -20℃).
200
These samples were then processed and quantified with a set of calibration samples and QC
201
samples that were not subjected to the freeze/thaw cycles.
202
2.4.8. Biological sample stability on bench-top at room temperature
M
an
us
cr
198
Six replicates at QC sample concentrations of 60 ng/ml, 500 ng/ml and 900 ng/ml were
204
thawed and kept at RT for 6 h before processing as described in section 2.3.2 and quantifying
205
with a set of calibration samples that were processed immediately.
206
2.4.9. Processed sample stability in the auto-sampler at room temperature
Ac ce p
te
d
203
207
Eight replicates of QC samples were processed and quantified. After 12h of storage in the
208
auto-sampler (RT and protected from light), the run was re-injected and re-analyzed with freshly
209
prepared calibration and QCs.
210
2.4.10. Long-term stability at -20 °C
11
Page 11 of 29
Six replicates of hEsA rat plasma samples at three QC sample concentration levels (60
212
ng/ml,500 ng/ml, 900 ng/ml) were stored at -20 °C for one month. Afterwards, the samples were
213
processed and quantified with a set of freshly prepared calibration standards and QC samples
214
(from freshly prepared working solution and stocking solution).
215
2.5. Intravenous bolus pharmacokinetic study
216
2.5.1. Animals
us
an
M
217
cr
ip t
211
Male Sprague Dawley rats (weighing between 280 and 320g) were purchased from Charles River (Wilmington, MA, USA). The rats were single-housed in plastic cages and received a
219
standard chow and water ad libitum during the experiments. All the rats were maintained on a 12
220
h/12 h light/dark cycle. Non-fasted animals were used for the study. All animal experiments were
221
performed according to the policies and guidelines of the Institutional Animal Care and Use
222
Committee (IACUC) of the University of Florida, Gainesville, USA (NIH publication # 85-23).
223
2.5.2. Design of pharmacokinetics study in rats
224
The hEsA was dissolved in an aqueous solution containing 2% ethanol and 5 % (v/v) dimethyl
225
sulfoxide. The hEsA was intravenously administered to rats at the dose of 10 mg/kg. Blood
226
samples (approximately 300 μl) were collected from sublingual vein into BD Vacutainer
227
heparinized tubes at 0 (pre-dose), 0.25, 0.5, 1.0, 1.5, 2.0, 4.0, 6.0, 12.0, 24.0 h after dosing.
Ac ce p
te
d
218
12
Page 12 of 29
Approximately 500 μl of isotonic saline were replaced by intraperitoneal injection in order to
229
maintain the blood fluid after each bleeding. The blood samples were centrifuged at 2000 rpm
230
for 10 min, then the supernatant plasma were stored at -20 °C for analysis.
231
2.5.3. Sample analysis
an
us
cr
ip t
228
The plasma samples were processed using the extraction procedure described in section 2.3.2
233
and analyzed using LC-MS/MS method described in section 2.2. A calibration curve was fitted
234
by weighted linear regression as described in section 2.4.1. and the concentrations in plasma
235
samples were calculated.
236
2.5.4. Pharmacokinetic data analysis
d te
Ac ce p
237
M
232
The Non-compartmental analysis was performed using WinNonlin software (version 5.2.1,
238
Pharsight, St, MO, USA). The pharmacokinetic (PK) parameters determined were the
239
concentration at time 0 (C0), the terminal elimination rate constant, the terminal elimination half-
240
life (t1/2), the area under the curve (AUC), the area under the first moment curve (AUMC), the
241
mean residence time (MRT), the volume of distribution at terminal phase (Vz), and the clearance
242
(Cl). AUC 0→last was calculated using a linear trapezoidal method from time 0 to the last observed
243
concentration time point.
244
3. Results and Discussion 13
Page 13 of 29
245
3.1. Optimization of HPLC-MS/MS conditions The MS spectra of hEsA and I.S were analyzed in positive ion mode. The negative ion mode
247
was also tested, but the intensity obtained was very low for the hEsA. For optimizing the MS
248
parameters, a neat standard solution containing hEsA or I.S was directly infused into the mass
249
spectrometer. The two compounds were selected [M+H]+ because of their better response and
250
better stability than [M+Na]+, and the response of [M+H]+ was enough for the mass analysis. To
251
obtain chromatograms with satisfactory resolution and appropriate retention time, different
252
mobile phases were evaluated to optimize the analytical performance. Acetonitrile was found to
253
produce better resolution than methanol or the mixture of methanol with acetonitrile. With
254
addition of 0.1% formic acid to the mobile phase and gradient elution, the peak symmetry of the
255
two compounds were all greatly improved. The final acetonitrile－0.1% formic acid water
256
solution with gradient elution program was adopted as the mobile phase for hEsA and I.S. The
257
retention times were approximately 6.85min and 7.35min for I.S and hEsA, respectively. The
258
range of linear response was established by determining the accuracy and precision of calibration
259
curve samples. The lowest concentration with accuracy (% bias) and precision below ±20% was
260
selected as LLOQ.
261
3.1. Method validation
262
3.1.1. Linearity test
Ac ce p
te
d
M
an
us
cr
ip t
246
263
The response versus concentration data, in the range 20-1000 ng/ml for hEsA, was assessed by
264
analyzing the calibration curves using the peak area ratios of analyte to I.S versus the nominal
265
concentration of the calibration standard with a weighting factor (1/x). The average coefficient of
266
determination (r) was equal to 0.9994, and all calibration curves met the acceptance criteria. 14
Page 14 of 29
267 268
3.1.2. Assessment of carry-over The impact of the carry-over on blank rat samples following the highest calibration sample was assessed for both hEsA and I.S. The S/N ratios of hEsA and I.S were lower than 3,
270
indicating that the carry-over has no impact on the results.
271
3.1.3. Selectivity
cr
ip t
269
The selectivity precision (CV %) was 7.23%, and the accuracy (Bias %) was -0.69% (Table 1).
273
Since the precision was below 20%, and the accuracy was within ±20% at the LLOQ, this
274
analytical method for the quantification of hEsA in rat plasma was shown to be selective.
275
3.1.4. Accuracy and precision
M
an
us
272
The intra-day and inter-day precision CV % did not exceed 8.02% at any QC concentration
277
levels. The intra-day and inter-day accuracy bias % was between -10.82% and 3.73%, and
278
between -6.85% and -2.50%, respectively (Table 2). Since both intra-day and inter-day precision
279
was below 15%, and accuracy was within ±15%, this bioanalytical method was proved to be
280
precise and accurate.
281
3.1.5. Recovery and Matrix effect
te
Ac ce p
282
d
276
The matrix effect was between 128.46% and 106.42% for hEsA and 83.30% for I.S. (Table 3).
283
The recovery was between 113.01% and 114.88% for hEsA and 115.32% for I.S. (Table 3).
284
3.1.6. Dilution integrity
15
Page 15 of 29
The precision (CV %) and the accuracy (Bias %) of the diluted samples with blank plasma at a
286
1:10 ratio was 10.57% and 3.80%, respectively. Thus the dilution effect on the precision and
287
accuracy of the results was acceptable.
288
3.1.7. Freeze-thaw stability
ip t
285
After three successive overnight freeze/thaw cycles for the three concentration samples (60-
290
500-900 ng/ml), the precision (CV %) did not exceed 10.99%. The accuracy (Bias %) was
291
between -8.00% and -0.39% (Table 4). Hence, the hEsA in rat plasma proved to be stable after
292
three freeze/thaw cycles at -20 ℃.
293
3.1.8. Bench-top stability at room temperature
M
an
us
cr
289
After storage for 6h at room temperature, the precision (CV %) of the QC samples did not
295
exceed 6.16% and the accuracy (Bias %) was between -0.25% and 9.77% (Table 4). This
296
indicated that hEsA in rat plasma was stable after storing for 6 h at RT prior to processing.
297
3.1.9. Post-preparative stability
Ac ce p
te
d
294
298
Before and after 12h of storage for the QC samples, the precision (CV%) did not exceed
299
7.23%. The accuracy (Bias %) was between -2.08% and -0.02% (Table 4), demonstrating that
300
hEsA in rat plasma was stable up to 12 h in the auto-sampler (RT, protected from light).
301
3.1.10. Long-term stability at -20 °C
16
Page 16 of 29
After storage at -20℃ for one month, the precision of the samples didn’t exceed 3.46% and
303
the accuracy was between -5.17% and 1.19% (Table 4). The results demonstrated that hEsA in
304
rat plasma was stable for at least one month at -20 °C.
3.2. Pharmacokinetic data analysis
an
305
us
cr
ip t
302
This validated analytical method was applied to a pharmacokinetic study of hEsA in rats after
307
a single i.v. dose of 10 mg/kg. For all runs, calibration curves and QC samples met the required
308
acceptance criteria. Typical MRM chromatograms of blank rat plasma spiked with hEsA and I.S
309
is shown in Figure 2. Table 6 summarizes the main PK parameters of hEsA calculated by non-
310
compartmental analysis using WinNonlin software. The initial concentration (C0) was 8277
311
ng/ml, and the half time (t1/2) was 1.36 h. The area under the concentration-time curve (AUC0-last),
312
calculated based on the trapezoidal rule, was 5821 ng*h/ml and the clearance was 2.05 L/h/kg.
313
The concentration time profile (Figure 3) shows rapid decrease in systemic concentration of
314
hESA.
315
4. Conclusion
Ac ce p
te
d
M
306
316
The standard calibration curve of the phytolaccagenin, hEsA, between 20 and 1000 ng/ml with
317
the weighting 1/x was generated and the limit of quantification was 20 ng/ml. The rat plasma
318
samples containing hEsA could be diluted up to 10 fold without affecting the precision and
319
accuracy. The carryover had no effect on the results. The hEsA was found to be stable in rat 17
Page 17 of 29
plasma samples kept for 6 h on the bench at RT, after three successive freeze/thaw cycles, and
321
processed plasma samples stored at RT for 12 h in autosampler. The validation results
322
demonstrate that the analytical method is specific, selective, precise, accurate and capable of
323
producing reliable results. Hence, the method developed here is reliable for the quantification of
324
hEsA in rat plasma samples. The hESA has high volume of distribution and is rapidly eliminated
325
from systemic circulation in rats.
cr
ip t
320
us
326
Acknowledgements
328
We thank Yufei Tang (Department of Pharmaceutics, University of Florida) for the assistance in
329
the LC–MS/MS setup and maintenance.
331
M
te
d
330
an
327
References
333
[1] Editorial Committee of Pharmacopoeia of Ministry of Health PR China, The Pharmacopeoia
334
of People’s Republic of China (Part I), China, Chemical Industry Press, Beijing 2010, pp.304-
335
305.
336
[2] Y. Suga, Y. Maruyama, S. Kawanishi, J. Shoji, Studies on the structures of
337
phytolaccasaponin B, E and G from the roots of Phytolacca americana L, Chem. Pharm.
338
Bull.26(1978) 520-525.
339
[3]W. Gong , Z.H. Jiang, P. Sun, L. Li, Y.S. Jin, L.C. Shao, et al., Synthesis of Novel
340
Derivatives of Esculentoside A and Its Aglycone Phytolaccagenin, and Evaluation of Their
Ac ce p
332
18
Page 18 of 29
Haemolytic Activity and Inhibition of Lipopolysaccharide-Induced Nitric Oxide Production,
342
Chem. Biodivers. 8 (2011)1833-1852.
343
[4]Q.Y. Zheng, K. Mai, X.F. Pan , Y.H. Yi, Anti-inflammatory effects of esculentoside A, Chin.
344
J. Pharmacol. Toxicol. 6(1992)221-224.
345
[5] Z.Y. Xiao, Q.Y.Zheng, Y.Y.Jiang, B. Zhou, M.Yin, H.B.Wang, et al., Effects of
346
esculentoside A on production of interleukin-1, 2, and prostaglandin E2, Acta. Pharmacol. Sin.
347
25(2004) 817-821.
348
[6] Z. Xiao, Y. Su, S. Yang, L. Yin, W. Wang, Y. Yi, et al., Protective effect of esculentoside A
349
on radiation-induced dermatitis and fibrosis, Int. J.Radiat. Oncol. Biol. Phys. 65(2006)882–889.
350
[7] J. Fang, Q.Y. Zheng, Inhibitory effect of esculentoside A on platelet activating factor
351
released from calcimycin induced rat peritoneal macrophages, Acta. Pharm. Sin. 26(1991)721-
352
724.
353
[8] Q.Y. Zheng, J. Fang, H.B. Wang, Inhibitory effect of esculentoside A on TNF production by
354
human monocytes, Acad. J. Sec. Milit. Med. Univ. 18(1997) 415-417.
355
[9] D.W.Ju, Q.Y.Zheng, H.B.Wang, X.J. Guan, J. Fang, Y.H.Yi, Therapeutic effects of
356
esculentoside A on passive Heymann nephritis in rats and its inhibition on cytokine production,
357
Acta. Pharm. Sin. 34(1999) 9-12.
358
[10] W.F. Zhong, L.X.Jiang, J.Y.Wei, A.N.Qiao, M.M.Wei, L.W.Soromou, et al., Protective
359
effect of esculentoside A on lipopolysaccharide-induced acute lung injury in mice, J. Surg. Res.
360
185(2013) 364-372.
361
[11] F. Wu, Y.H.Yi, P. Sun, D.Z. Zhang, Synthesis, in vitro inhibitory activity towards COX-2
362
and haemolytic activity of derivatives of Esculentoside A, Bioorg. Med. Chem. Lett.
363
23(2007)6430-6433.
Ac ce p
te
d
M
an
us
cr
ip t
341
19
Page 19 of 29
[12] X.D. Guan, H.C. Chang, F.L. Sun, X.H. Chen, W. Zhang, G.R. Fan, Determination of
365
esculentoside A in dog plasma by LC–MS/MS method: Application to pre-clinical
366
pharmacokinetics, J. Pharm. Biomed. Anal. 72(2013)261-266.
367
[13] FDA,Guidance for Industry: Bioanalytical Method Validation, 2001,
368
http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/u
369
cm070107.pdf.
370
[14] FDA,Guidance for Industry (Draft Guidance): Bioanalytical Method Validation, 2013,
371
http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm
372
368107.pdf
373
[15]M. Whitmire, J. Ammerman, P. Lisio, J. Killmer, D. Kyle, E. Mainstone, et al., LC-MS/MS
374
Bioanalysis Method Development, Validation, and Sample Analysis: Points to Consider When
375
Conducting Nonclinical and Clinical Studies in Accordance with Current Regulatory Guidances.
376
J Anal Bioanal Techniques. S4. http://dx.doi.org/10.4172/2155-9872.S4-001
377
[16] S. Chandran, R.S.P. Singh, Comparison of various international guidelines for analytical
378
method validation, Pharmazie, 62(2007) 4-14.
Ac ce p
te
d
M
an
us
cr
ip t
364
20
Page 20 of 29
379
Figure captions:
381
Figure1: Chemical structures of the phytolaccagenin, hEsA (A) and Hydrocortisone hemiadipate
382
(B)
383
Figure 2: The extracted LC-MS/MS chromatograms of (A) blank rat plasma spiked with the
384
phytolaccagenin, hEsA, and I.S (400 ng/ml and 100 ng/ml, respectively), (B) plasma sample
385
from a rat obtained at 1.5h after i.v. administration of hEsA (786 ng/ml) with I.S (100 ng/ml)
386
Figure 3: Mean±SD plasma concentration and log mean concentration (inner panel)-time profile
387
of hEsA in male Sprague-Dawley rats (n=3) following i.v. administration (10 mg/kg)
M
an
us
cr
ip t
380
Ac ce p
te
d
388
21
Page 21 of 29
388 389
Figure 1
390
ip t
391 392 COOCH 3
cr
O
O
HO
us
COOH
HO
H
CH2 OH
A
400 401 402
te Ac ce p
399
B
d
396
398
H
M
395
397
O
O
an
394
OH
H
HO
393
O OH
403 404
22
Page 22 of 29
404 405
Figure 2 TIC 6.88
3.5e4
I.S
3.0e4
7.41
hEsA
ip t
Intensity
2.5e4 2.0e4 1.5e4
cr
1.0e4 5.0e3
1
2
3
4
5
6 Time, min
7
8
9
10
11
us
0.0e0
406 407 TIC 4.5e4
an
7.32
4.0e4
hEsA
3.5e4
2.5e4
I.S
2.0e4
6.84
1.5e4 1.0e4
M
Intensity
3.0e4
410 411 412 413
2
3
4
5
6 Time, min
7
8
9
10
11
te
409
1
Ac ce p
0.0e0
408
d
5.0e3
414
23
Page 23 of 29
Figure 3
418 419 420
te
417
Ac ce p
416
d
M
an
us
cr
ip t
414 415
421 422 423
24
Page 24 of 29
424
Table 1: Selectivity test at the LLOQ for hEsA. a
(ng/ml)
(ng/ml)
1
20
18.6
2
20
21.7
3
20
18.0
4
20
21.8
5
20
18.8
6
20
19.4
7
20
20.8
8
20
19.8
a
427
value.
429 430
-0.7
us an M
te
S.D.: standard deviation; CV: coefficient of variation; Bias: deviation of mean from nominal
Ac ce p
428
7.2
d
425 426
1.4
cr
19.9
Bias (%)
ip t
HPLC Run Nominal level Calculated values Mean S.D. CV(%)
431 432 433
25
Page 25 of 29
434
Table 2: Precision and accuracy for hEsA in rat plasma during validation. Accuracy and Precision of controls
900 ng/ml
435 436 437 438
(n=8)
(n=8)
(n=24)
Mean
53.8
55.5
58.4
55.9
S.D
3.2
2.4
2.0
3.1
CV (%)
5.9
4.4
3.4
Bias (%)
-10.8
-7.5
-2.8
Mean
504.4
471.0
518.6
498.0
S.D
30.2
20.1
18.1
30.3
CV (%)
6.0
4.3
3.5
6.1
Bias (%)
0.9
-5.8
3.7
-0.4
Mean
901.0
802.6
928.8
877.5
S.D
66.6
21.3
36.8
70.4
CV (%)
7.4
2.7
4.0
8.0
Bias (%)
0.1
-10.8
3.2
-2.5
cr
ip t
(n=8)
5.6
-6.9
us
an
M
ng/ml
Overall
d
500
Day 3
te
ng/ml
Day 2
Ac ce p
60
Day 1
439 440 441
26
Page 26 of 29
442
Table 3: Matrix effect and recovery of hEsA and I.S. Nominal concentration (ng/ml)
Mean extraction recovery (%)
Mean matrix effect (%)
60
113.0±5.6
128.5±4.2
500
113.6±5.2
119.4±7.1
114.9±2.6
106.4±3.7
cr
us
900
ip t
hEsA
100
an
hemiadipate-hydrocortisone 115.3±7.6
M
443
447 448 449
te Ac ce p
446
d
444 445
83.3±5.4
450 451 452
27
Page 27 of 29
Table 4: Short-term and long-term stabilities during storage in various conditions (n=8)
(ng/ml)
Remaining over nominal concentration (%) Post-preparative Three
Bench-top stability Long-term stability at
stability
at room temperature -20℃
freeze/thaw cycles at -20℃
98.0±5.9
92.0±11.0
99.8±5.3
500
99.1±3.6
98.3±3.2
109.8±6.2
900
98.8±5.4
99.6±3.7
104.4±3.4
461 462
te Ac ce p
460
d
456
459
101.2±2.9
M
455
458
96.6±3.5
an
454
457
94.8±2.3
us
60
ip t
Concentration
cr
453
463 464
28
Page 28 of 29
Table 5: The non-compartmental pharmacokinetic parameters of hEsA in rat plasma after i.v. (10
466
mg/kg) administration to SD rats (N=3).
Ke (1/h)
0.7 ±0.4
AUC0-last (ng*h/ml)
5821 ±3444
AUC 0 -∞ (ng*h/ml)
5999 ±3373
AUMClast (ng*h/ml)
4601±1581
AUMC 0 -∞ (ng*h/ml)
5993±1555 0.9±0.2
Vz (L/kg)
4.2±3.0
CL (L/h/kg)
2.1±1.1
te
MRT (hr)
cr
1.4 ±0.9
us
t1/2 (h)
an
8277± 6361
M
C0 (ng/ml)
d
Mean±SD
Ac ce p
467
Parameters
ip t
465
29
Page 29 of 29