Liquid chromatography/electrospray ionization mass spectrometry method for the determination of the...

Copyright © 2007 John Wiley & Sons, Ltd. Biomed. Chromatogr. 21: 1297–1302 (2007)DOI: 10.1002/bmc

Determination of the active metabolite M-1 of suplatast tosilate 1297ORIGINAL RESEARCHORIGINAL RESEARCH

Copyright © 2007 John Wiley & Sons, Ltd.

BIOMEDICAL CHROMATOGRAPHYBiomed. Chromatogr. 21: 1297–1302 (2007)Published online 17 July 2007 in Wiley InterScience(www.interscience.wiley.com) DOI: 10.1002/bmc.894

Liquid chromatography/electrospray ionization massspectrometry method for the determination of the activemetabolite M-1 of suplatast tosilate in human plasma

Li Ding,1* Likun Ding,1 Xia Zhou,1 Lin Yang2 and Aidong Wen2

1Department of Pharmaceutical Analysis, China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009, People’s Republic of China2Xijing Hospital of the Fourth Military Medical University, Xi’an 710032, People’s Republic of China

Received 10 April 2007; revised 26 May 2007; accepted 29 May 2007

ABSTRACT: A liquid chromatography/electrospray ionization mass spectrometry (LC/ESIMS) method for the determination of4-(3-ethoxy-2-hydroxypropoxy) acrylanilide (M-1), the active metabolite of suplatast tosilate, in human plasma was established.Plasma samples were extracted with diethyl ether, separated on a C18 column with a mobile phase of acetonitrile–10 mm ammo-nium acetate solution containing 0.1% formic acid (28:72, v/v) and detected by ESIMS. The method was linear over the con-centration range 0.15–60.0 ng/mL. The lowest limit of quantification was 0.15 ng/mL. The intra- and inter-run relative standarddeviations obtained from three validation runs were all less than 8.6%, and the intra- and inter-run relative errors were all lessthan 3.1%. The method was successfully applied for the evaluation of pharmacokinetic profiles of M-1 in healthy volunteers.Copyright © 2007 John Wiley & Sons, Ltd.

KEYWORDS: suplatast tosilate; 4-(3-ethoxy-2-hydroxypropoxy) acrylanilide; LC/ESIMS; pharmacokinetics

*Correspondence to: Li Ding, Department of Pharmaceutical Analysis,China Pharmaceutical University, 24 Tongjiaxiang, Nanjing 210009,People’s Republic of China.E-mail: [email protected]

Abbreviations used: LLOQ, lowest limit of quantification; M-1, 4-(3-ethoxy-2-hydroxypropoxy) acrylanilide.

INTRODUCTION

Suplatast tosilate (see Fig. 1), {2-[4-(3-ethoxy-2-hydroxypropoxy)-phenylcarbamoyl]-ethyl} dimethylsul-forium p-toluenesulfonate, is an anti-allergic drug (Ushioand Yamamoto, 1994). It shows effectiveness for the treat-ment of type I allergy-related diseases such as bronchialasthma, allergic rhinitis urticaria and similar maladies(Satoh et al., 2003; Taniguchi et al., 1996). After oraladministration, suplatast tosilate is metabolized to4-(3-ethoxy-2-hydroxypropoxy) acrylanilide (M-1) byalkaline hydrolysis, and the in vivo effect of this drugis mainly due to the action of this metabolite (Keizoand Dotai, 1992). So far, only one article has reportedthe determination of M-1 in human plasma, in which aGC/MS method with a lowest limit of quantification(LLOQ) of 0.2 ng/mL and run time of 9 min wasdescribed (Tei et al., 1992). In the present study, asensitive and simple LC/ESIMS method with an LLOQof 0.15 ng/mL and a shorter run time was developed

Figure 1. Chemical structures of suplatast tsoilate (A), M-1(B) and the IS (C).

for the determination of M-1 in the human plasma.The method was successfully applied to study thepharmacokinetics of M-1 in Chinese volunteers.

EXPERIMENTAL

Materials. M-1 (99.3% purity) and the suplatast tosilategranules (50 mg/g) were provided by Chongqing KeruiPharmaceutical Co. Ltd (Chongqing, China). Phenacetin(see Fig. 1), the internal standard (IS), was obtained fromLiangshan Lantian Pharmaceutical Co. Ltd (Shangdong,China). Acetonitrile of HPLC grade was purchased fromMerck KGaA. Formic acid, ammonium acetate and diethylether were of analytical-grade purity and purchased fromNanjing Chemical Reagent Co. Ltd (Nanjing, China).


1298 L. Ding et al.ORIGINAL RESEARCH

Instrumentation. LC/ESIMS analyses were performed usingan Agilent Technologies Series 1100 LC/MSD SL system(Agilent Technologies, Palo Alto, CA) with a Zorbax Extend-C18 Aglient column (Narrow-Bore dp 5 µm, 2.1 × 150 mmi.d.). The LC/ESIMS was controlled by a computer employingthe Aglient ChemStation software supplied by Aglient.

Conditions. The mobile phase was acetonitrile–10 mM

ammonium acetate solution containing 0.1% formic acid(28:72, v/v) at a flow rate of 0.25 mL/min. The columntemperature was maintained at 28°C. LC/ESIMS was carriedout using nitrogen to assist nebulization. The quadrupolemass spectrometer equipped with an electrospray ionizationsource was set with a drying gas (N2) flow rate of 10 L/min,a nebulizer pressure of 40 psig, a drying gas temperature of350°C, a capillary voltage of 3 kV and positive ion mode. Thefragmentor voltage was 120 V. LC/ESIMS was performed inselected-ion monitoring mode using the protonated moleculeat m/z 266.1 for M-1 and m/z 180.1 for the IS as the targetions.

Sample preparation. A 1 mL aliquot plasma sample wasadded with 5 mL diethyl ether after addition of 40 µL ISsolution (100 ng/mL). After vortex-mixing for 3 min, themixture was centrifuged for 10 min at 4000 rpm. The organicphase was separated and evaporated to dryness under agentle stream of nitrogen in a water bath of 30°C. Theresidue was reconstituted with 120 µL of the mobile phase,and a 5 µL aliquot of the reconstituted solution was injectedonto the LC/ESIMS for analysis.

Preparation of the stock and standard solutions. Thestock solutions of M-1 (1 mg/mL) and the internal standard(1 mg/mL) were prepared in acetonitrile and stored at −20°C.Standard solutions of M-1 with concentrations of 100, 10 and1 µg/mL, and 100 and 10 ng/mL were prepared by serialdilution of the M-1 stock solution with acetonitrile in separate10 mL volumetric flasks. A solution containing 100 ng/mLinternal standard was also obtained by dilution of the ISstock solution with acetonitrile.

Preparation of calibration curves and quality controlsamples. Calibration standards of M-1 were prepared atconcentration levels of 0.15, 0.3, 1.0, 3.0, 10.0, 20.0, 40.0 and60.0 ng/mL by spiking appropriate amounts of the standardsolutions in 1 mL blank plasma. The calibration curve wasprepared and assayed along with the quality control (QC)samples. The QC samples were prepared in 1.0 mL blankplasma at concentrations of 0.3, 5.0 and 50.0 ng/mL, respec-tively, and stored at −20°C.

Method validation. Calibration standards of eight M-1concentration levels at 0.15, 0.3, 1.0, 3.0, 10.0, 20.0, 40.0 and60.0 ng/mL were extracted and assayed. To evaluate thelinearity, the calibration standards were prepared and assayedon three separate runs. The calibration curve was constructedusing the peak area ratios of M-1 to the IS vs the concentra-tion of M-1, using weighed (1/C) least squares linear regres-sion. The LLOQ was defined as the lowest concentration onthe calibration curve at which precision was within 20% andaccuracy was within 20%, and it was established using fivesamples independent of standards (CDER, 2001).

Validation samples were prepared and analyzed on threeseparate runs to evaluate the accuracy and the intra-run andinter-run precision of the analytical method. The accuracy aswell as the intra-run and inter-run precision were determinedby analyzing five replicates at 0.3, 5.0, and 50.0 ng/mL of M-1along with one standard curve in each run. Assay precisionwas calculated using the relative standard deviation (RSD%).The accuracy is the degree of closeness of the determinedvalue to the nominal true value under prescribed conditions.The accuracy is defined as the relative deviation in thecalculated value (E) of a standard from its true value (T)expressed as a percentage (RE%). It was calculated usingthe formula: RE% = (E − T)/T × 100. The results (seeTable 1) demonstrate that the method is accurate and precise.

The extraction recoveries of M-1 at three QC levels weredetermined by comparison of the peak areas of M-1 extractedfrom plasma samples with that of M-1 dissolved in the blankplasma sample’s reconstituted solution (the final solution ofblank plasma after extraction and reconstitution).

Table 1. Mean inter-run back-calculated standard and standard curve results (n ===== 3)

Addedconcentration Found concentration (ng/mL)

(ng/mL) I II III Mean SD RSD (%) RE (%)

Mean inter-run back-calculated results0.1589 0.1299 0.1337 0.1434 0.1357 0.007 5.1 −14.60.3177 0.3148 0.3219 0.3120 0.3162 0.005 1.6 −0.41.059 1.142 1.157 1.068 1.122 0.05 4.2 6.03.177 3.565 3.499 3.499 3.521 0.04 1.1 10.910.59 10.22 9.79 10.48 10.16 0.35 3.4 −4.021.18 22.32 22.48 21.60 22.13 0.47 2.1 4.742.36 40.90 40.94 44.63 42.15 2.1 5.1 −0.663.54 63.61 63.84 60.33 62.59 2.0 3.1 −1.3

Standard curve resultsIntercept 0.005476 0.003629 0.004718 0.004648 0.0009 20.0Slope 0.05948 0.0611 0.05636 0.05893 0.0024 4.1r 0.9993 0.9991 0.9988 0.9991 0.0003 NA

Calibration curves were weighed 1/C. RSD, relative standard deviation; NA, not applicable.


Determination of the active metabolite M-1 of suplatast tosilate 1299ORIGINAL RESEARCH

The stability of M-1 in plasma was investigated under avariety of storage and handling conditions using the low (0.3 ng/mL) and high (50.0 ng/mL) QC samples. The short-termstability was assessed by analyzing the samples that were keptat ambient temperature for 8 h. The freeze–thaw stability(−20°C in plasma) was checked through three cycles. The QCsamples were stored at −20°C for 24 h and thawed unassistedat room temperature. When completely thawed, the sampleswere refrozen for 24 h under the same conditions. Thefreeze–thaw cycles were repeated three times, and thenanalyzed on the third cycle. The long-term stability was per-formed at −20°C in the plasma for 3 weeks.

Pharmacokinetic study. The method was applied to deter-mine M-1 in plasma samples from 10 healthy male Chinesevolunteers who were administered the dose of 100 mgsuplatast tosilate in a clinical study. The mean age of the 10volunteers was 30 years (range 26 –35 years); mean bodyweight was 58.7 kg (range 51.5–67.5 kg). Following an over-night fast, each volunteer received suplatast tosilat granulescontaining 100 mg suplatast tosilat. Standard meals were pro-vided 4 h post-dose. Blood samples were collected pre-doseand at 0.5, 1, 2, 3, 4, 5, 6, 8, 10, 14, 24 and 36 h post-dose. TheM-1 plasma concentrations of these samples were determined,and the pharmacokinetics of M-1 was evaluated. Model-independent pharmacokinetic parameters were calculatedfor M-1. The maximum plasma concentration (Cmax) and thetime to reach it (tmax) were noted directly. The eliminationrate constant (kel) was calculated by linear regression of theterminal points of the semi-log plot of plasma concentra-tion against time. Elimination half-life (t1/2) was calculatedusing the formula t1/2 = 0.693/kel. The area under the plasmaconcentration–time curve, AUC0–36 to the last measurableplasma concentration, was calculated by the linear trapezoidalrule.

RESULTS AND DISCUSSION

LC/MS conditions

The ESI in positive ion mode was adopted for the LC/MS determination of M-1. The LC/ESIMS was per-formed in the selected ion-monitoring (SIM) mode. Inorder to select the target ion for monitoring M-1, theESI mass spectra obtained by scan monitoring at differ-ent fragmentor voltages were investigated. The test re-sults showed that the base peak in the mass spectra ofM-1 obtained at different fragmentor voltages was ofthe same ion at m/z 288.1, which was the ion [M + Na]+

of M-1. The intesity of the [M + Na]+ ion was not stableand reproducible. Therefore, it cannot be selected asthe target ion for M-1, and the assay sensitivity ob-tained by monitoring of the protonated molecule [M +H]+ of M-1 at m/z 266.1 could meet the requirements ofthe pharmacokinetic study. Therefore, the protonatedmolecule of M-1 at m/z 266.1 was selected as the targetion for M-1. Figure 2(A) shows a typical mass spectrumof M-1 at 120 V fragmentor voltage obtained in the

Figure 2. Mass spectra of the positive ion of the metaboliteof suplatast tosilate (A) and the IS (B) at 120 V fragmentorvoltage.

Figure 3. The intensity of M-1 at different fragmentorvoltages.

scan monitoring mode. In order to achieve the highestassay sensitivity for M-1, the optimal fragmentor volt-age of the ESIMS was investigated. The intensities ofthe ion at m/z 266.1 were compared at the fragmentorvoltages of 50, 70, 90, 110, 120 and 150 V. Figure 3



shows the intensity of M-1 at different fragmetorvoltages, which shows that the highest intensity ofthe protonated molecule can be obtained at 120 Vfragmentor voltage. Therefore, the fragmentor voltagewas set at 120 V for the assay. At this fragmentor volt-age, the base peak in the mass spectrum of the IS wasat m/z 180.1, which was the protonated molecule [M +H]+ of the IS, see Fig. 2(B). Therefore, the ion at m/z180.1 was selected as the target ion for the IS.

Phenacetin was chosen as the internal standard be-cause it is structurally similar to M-1. To improve thechromatographic peak shapes of M-1 and the IS, a 10 mM

ammonium acetate buffer solution was adopted in themobile phase. The experimental results showed thatacidification of the mobile phase by adding some for-mic acid could improve the separation and increase theMS sensitivity. Finally, the high sensitivity and goodseparation of M-1 were obtained by using an elutionsystem of acetonitrile–10 mM ammonium acetate solu-tion containing 0.1% formic acid (28:72, v/v) as themobile phase. Under the present chromatographicconditions, the retention time was 3.9 min for M-1 and5.1 min for the IS. The representative chromatogramsare shown in Fig. 4.

Selectivity and calibration

The selectivity was assessed by comparing thechromatograms of six different batches of blank plasmawith the corresponding spiked plasma. Figure 4 showsthe typical chromatograms of blank plasma, spikedplasma sample with the M-1 and the IS, and the plasmasample from a volunteer after oral administration.There was no significant interference from endogenoussubstances at the retention times of the analytes.

Three calibration curve analyses were performed onthree separate runs and the back-calculated values foreach level recorded (see Table 1). The RSD (%) ateach concentration level varied from 1.1 to 5. The RSDof three slopes was 4.1. The calibration curves did notexhibit any nonlinearity within the chosen range. Theback-calculated results showed good accuracy and pre-cision. The LLOQ of M-1 in plasma was 0.15 ng/mL.The data for the LLOQ is shown in Table 2.

Figure 4. Typical chromatograms of blank plasma (A), LLOQfor the metabolite M-1 in plasma (0.15 ng/mL) and the IS(B), plasma spiked with M-1 (40 ng/mL) and the IS (C),plasma obtained from a volunteer at 2 h after oral administra-tion of 100 mg suplatast tosilate (D). This figure is available incolour online at www.interscience.wiley.com/journal/bmc

Precision and accuracy

The intra- and inter-run precision and accuracy of theassay is summarized in Table 3. The precision wascalculated using one-way ANOVA. The results inTable 3 demonstrate that the precision and accuracyvalues are within the acceptable range and the methodis accurate and precise.

Table 2. Accuracy and precision for the assay of LLOQ (n ===== 5)

Added Foundconcentration concentration Mean RSD RE(ng/mL) (ng/mL) (ng/mL) (%) (%)

0.1589 0.1601 0.70.1589 0.1491 −6.20.1589 0.1536 0.1495 6.9 −3.30.1589 0.1325 −16.60.1589 0.1522 −4.2

www.interscience.wiley.com/journal/bmc


Determination of the active metabolite M-1 of suplatast tosilate 1301ORIGINAL RESEARCH

Table 3. Precision and accuracy of the assay for the determination of M-1 inhuman plasma (n ===== 3 runs, five replicates per run)

Added Foundconcentration concentration Intra-assay Inter-assay RE(ng/mL) (ng/mL) RSD% RSD% (%)

0.3177 0.3288 5.6 7.2 1.05.295 5.393 4.8 8.6 1.852.95 54.06 3.7 6.4 3.1

Extraction recovery and matrix effects

When using diethyl ether or ethyl acetate as the extractionsolvent in the plasma sample preparation procedure,higher extraction efficiency of M-1 could be achieved.However, ethyl acetate may extract some endogenousinterference substances of the analytes from the plasma.Therefore, diethyl ether was selected as the extractionsolvent. The extraction recovery of M-1 was evaluatedby analyzing five replicates at 0.3, 5.0 and 50.0 ng/mL ofM-1. The extraction recoveries of the assay were 80.0± 6.4, 85.8 ± 3.9 and 83.4 ± 4.0% (n = 5) for the low,medium and high concentration levels, respectively.

The matrix effect was defined as the direct or in-direct alteration or interference in response due to thepresence of unintended analytes or other interferingsubstances in the samples (Meng et al., 2005). It wasevaluated by comparing the peak area of the analytesdissolved in the blank plasma sample’s reconstitutedsolution (the final solution of blank plasma after ex-traction and reconstitution) with that dissolved in themobile phase. Three different concentration levels ofM-1 (0.3, 5.0 and 50.0 ng/mL) were evaluated byanalyzing five samples at each level. The blank plasmasused in this study were from five different batches. Ifthe peak area ratio is less than 85% or greater than115%, a matrix effect is implied. The ratios of the assaywere 96.7 ± 8.6, 102.0 ± 4.1 and 103.5 ± 5.3% (n = 5)for the low, medium and high concentration levels,respectively. The results showed there was no matrixeffect of the analytes observed in this study.

Stability

The stability results are summarized in Table 4. Thedata showed that no significant degradation of M-1 in

plasma was observed at ambient temperature for 8 hand during the three freeze–thaw cycles. M-1 in plasmaat −20°C was stable for 3 weeks.

Application

The method was successfully applied to determine theplasma concentration of M-1 up to 36 h after oraladministration of 100 mg suplatast tosilate to 10 healthyChinese volunteers. The mean plasma concentration–time curve of M-1 is shown in Fig. 5. The maximumplasma concentration (Cmax) of M-1 was 9.84 ± 2.41 ng/mL, and the time to reach it (tmax) was 3.0 ± 1.3 h. Theelimination half-life (t1/2) of M-1 was 6.5 ± 1.4 h. TheAUC0–36 of M-1 was 77.74 ± 10.65 h ng/mL.

Table 4. Stability data of M-1 in human plasma under various conditions (n ===== 3)

Storage Added concentration Found concentration Inter-run REconditions (ng/mL) (ng/mL) RSD (%) (%)

Room temperature for 8 h 0.3177 0.3054 8.7 −3.952.95 53.43 10.3 1.0

Three freeze–thaw cycles 0.3177 0.2901 2.4 −8.752.95 52.27 9.3 −1.3

Three weeks at −20°C 0.3177 0.3411 3.8 7.452.95 54.02 7.6 2.0

Figure 5. Mean plasma concetration–time profile of M-1 afteran oral administration of 100 mg suplatast tosilate to the 10Chinese volunteers.



CONCLUSION

A sensitive LC/ESIMS method for the quantification ofM-1 in human plasma was developed and successfullyapplied to evaluate the pharmacokinetics of M-1 inhealthy Chinese volunteers. No significant interferencescaused by endogenous compounds were observed. Themethod is sensitive, selective and suitable for routineanalysis of large batches of biological samples.

REFERENCES

CDER. Guidance for Industry, Bioanalytical Method Validation. USDepartment of Health and Human Services, Food and DrugAdministration, Center for Drug Evaluation and Research, May2001.

Keizo K and Dotai Y. Pharmacokinetics studies of suplatast tosilate(TPD-1151T) (I): absorption, distribution and excretion after

administration of 14C-suplatast tosilate(TPD-1151T) to rats. DrugMetabolism and pharmacokinetics 1992; 7: 399.

Meng F, Chen XY, Zeng YL and Zhong DF. Sensitive liquidchromatography–tandem mass spectrometry method for the determi-nation of cefixime in human plasma: application to a pharma-cokinetics study. Journal of Chromatography A 2005; 819: 277.

Satoh T, Sasaki G, Wu MH, Yokozeki H, Katayama I and NishiokaK. Suplatast tosilate inhibits ensinophil production and recruitmentinto the skin in murine contact sensitivity. Clinical Immunology2003; 108: 257.

Taniguchi H, Togawa M, Ohwata K, Kiniwa M, Matsuura N, NagaiH and Koda A. Suplatast tosilate, a new type of antiallergic agent,prevents the expression of airway hyperresponsiveness in guineapigs. Journal of European Pharmacology 1996; 318: 447.

Tei M, Kodama K, Yafune A, Muranushi A, Takayanagi H, TakebeM, Shindoh T, Masuda H, Kuwata K, Matsushima E, Muramoto Kand Umeno Y. Pharmacokinetics of suplatast tosilate (TPD-1151T)in man after oral administration. Clinical Report 1992; 26: 3199.

Ushio T and Yamamoto K. High-performance liquid chromato-ghraphy of enantiomers {2-(4-(3-ethoxy-2-hydroxypropoxy) phenyl-carbamoyl) ethyl} dimethylsulforium p-toluenesulfonate (suplatasttosilate) on a cellulose tris-3,5-dimethyphenylcarbamate column.Journal of Chromatography A 1994; 684: 235.

Liquid chromatography/electrospray ionization mass spectrometry method for the determination of the...

Documents

Transcript of Liquid chromatography/electrospray ionization mass spectrometry method for the determination of the...