A rapid and sensitive UFLC-MS/MS method for thesimultaneous determination of gentiopicroside andswertiamarin in rat plasma and its application inpharmacokineticsBo Fenga,b, Heyun Zhub, Jiao Guanb, Longshan Zhaoc, Jingkai Gua, Lei Yina,e, J. Paul Fawcettd andWenshu Liue

aResearch Center for Drug Metabolism, College of Life Science, Jilin University, bSchool of Pharmacy, Jilin Medical College, eDepartment ofStomatology, First Hospital of Jilin University, Changchun, Jilin, cSchool of Pharmacy, Shenyang Pharmaceutical University, Shenyang, Liaoning,China and dSchool of Pharmacy, University of Otago, Dunedin, New Zealand

Keywordsgentiopicroside; pharmacokinetics; RadixGentianae; swertiamarin; UFLC-MS/MS

CorrespondenceWenshu Liu, First Hospital of Jilin University,71 Xinmin Street, Changchun 130021, China.E-mail: liuwenshu256@sina.com

Received January 09, 2014Accepted March 16, 2014

doi: 10.1111/jphp.12266

Abstract

Objectives Radix Gentianae is a traditional Chinese medicine derived frommedicinal plants of the family Gentianaceae. Its pharmacological effects have beenprimarily attributed to the presence of a number of secoiridoid glycosides, in par-ticular gentiopicroside and swertiamarin. In this study, a rapid and sensitivemethod based on ultrafast liquid chromatography-tandem mass spectrometry hasbeen developed for the simultaneous determination of gentiopicroside andswertiamarin in rat plasma using paeoniflorin as internal standard (IS).Methods After liquid-liquid extraction with ethyl acetate-isopropanol (95 : 5,v/v), separation was achieved on a Shim-pack XR-ODS C18 column(75 mm × 3.0 mm, 2.2 μm) with a mobile phase consisting of methanol : 0.1%formic acid (30 : 70, v/v) at a flow rate of 0.4 ml/min. Detection on an API 3200QTRAP mass spectrometer equipped with an electrospray ionization source oper-ated in the negative ionization mode was performed by multiple reactionmonitoring of the precursor-to-product ion transitions of gentiopicroside,swertiamarin and IS at m/z 401.0 → 179.0, 419.0 → 179.1 and 525.1 → 121.0respectively. The calibration curves were linear over the concentration range of20–10 000 and 2–1000 ng/ml for gentiopicroside and swertiamarin with corre-sponding lower limits of quantification of 20 and 2 ng/ml. The limits of detectionwere 4 and 0.5 ng/ml for gentiopicroside and swertiamarin, respectively. Theintraday and interday precisions were below 11.9% for gentiopicroside and below9.5% for swertiamarin in terms of relative standard deviation, and the accuracywas within ±8.3% for gentiopicroside and within ±10.2% for swertiamarin interms of relative error. Extraction recovery, matrix effect and stability weresatisfactory in rat plasma. The method was fully validated and applied to apharmacokinetic study involving oral administration of a Radix Gentianae extractto groups of male and female rats.Key findings Results showed that in female rats, both compounds were absorbedto a greater extent and eliminated more slowly than in male rats, although the rateof absorption was similar in the two groups.Conclusions There were remarkable differences in pharmacokinetic properties ofgentiopicroside and swertiamarin between male and female rats. The results willprovide helpful information for the development of suitable dosage forms andclinical references on rational administration.

bs_bs_banner

And PharmacologyJournal of Pharmacy

Research Paper

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–•• 1

Introduction

Radix Gentianae consists of the root and rhizome of anumber of important medicinal plants of the familyGentianaceae including Gentiana scabra Bge., G. manshuricaKitag., G. triflora Pall. and G. rigescens Franch. It has beencommonly recommended for treatment of poor appetite,infectious hepatitis and jaundice in China and other Asiancountries.[1] Pharmacological studies have demonstratedthat secoiridoid glycosides are the major bioactive compo-nents in Radix Gentianae[2] of which gentiopicroside(Fig 1a) and swertiamarin (Fig 1b) are two of most abun-dant. These compounds have been shown to display many

pharmacological effects including hepatoprotection,[3–5]

gastroprotection,[6] antibacterial,[7] anti-inflammatory[8] andhypoglycaemic[9] activity in either in vitro or in vivopharmacodynamic studies. Taken together, these studiessuggest that secoiridoid glycosides are promising candidatesfor new drug investigation.

To further investigate and explain the pharmacodynamicprofile of Radix Gentiana, a rapid, sensitive and robustmethod for the high-throughput determination of its activeingredients in biological fluids is required. However, despiteextensive research into its bioactivity, the simultaneousdetermination of gentiopicroside and swertiamarin inplasma and application to a pharmacokinetic study has notbeen reported.

To date, several methods have been reported for thedetermination of gentiopicroside or swertiamarin in plantextracts based on capillary electrophoresis[10,11] HPLC[12–14]

and liquid chromatography tandem mass spectrometry.[15,16]

However, these methods suffer from long run time, theneed for large volumes of biological samples and compli-cated sample preparation. In addition, the sensitivity ofthese methods (lower limits of quantitation of 30 and5 ng/ml for gentiopicroside and swertiamarin, respectively)is inadequate to meet the requirements of pharmaco-kinetic and metabolism studies of gentiopicroside andswertiamarin.

In this paper, a sensitive and rapid ultrafast liquid chro-matography tandem mass spectrometric (UFLC-MS/MS)method for the simultaneous determination of gentiopi-croside and swertiamarin in rat plasma is reported for thefirst time. The method was fully validated and applied to apharmacokinetic study involving oral administration of aRadix Gentianae extract to groups of male and female rats.It is expected that the results of this study will provideuseful information to facilitate the clinical development ofRadix Gentianae extract.

Materials and Methods

Materials and animals

Radix Gentianae was purchased from the TongrentangDrug Company (Changchun, China) and authenticatedby Associate Professor Jinghua Li (School of Pharmacy, JilinMedical College, China). Gentiopicroside, swertiamarinand paeoniflorin for use as internal standard (IS) werepurchased from the National Institute for Food andDrug Control (Beijing, China). Chemical structures ofanalytes and IS are shown in Figure 1. HPLC grademethanol was purchased from Fisher Scientific (Fair Lawn,NJ, USA). All other reagents were analytical grade andusedas received. Deionized water purified using a Milli-Q

(a)

(b)

(c)

4.0 e4

3.6 e4

3.2 e4

2.8 e4

2.4 e4

2.0 e4

1.6 e4

1.2 e4

3.6 e5

3.2 e5

2.8 e5

2.4 e5

2.0 e5

1.6 e5

1.2 e5

8.0 e4

5.0 e44.6 e44.2 e43.8 e43.4 e43.0 e42.6 e42.2 e41.8 e41.4 e41.0 e46000.02000.0

4.0 e4

0.0

8000.0

4000.0

0.0

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

100 140 180 220 260 300 340 380 420m/z, Da

100 140 180 220 260 300 340 380 420m/z, Da

100 140 180 220 260 300 340 380 420 460 500 540m/z, Da

113.0

118.7

149.2

143.2 160.8

192.8

225.0

355.1401.0

179.0

179.0

179.1

118.8

113.0

O O

O

O

O

HO

OHOH

OH

HCOO–

HCOO–

143.1161.0

179.1

355.2 373.1

419.0

O O

O

O

O

HO

OH

OHOH

OH

164.7

121.0

121.0

327.1

449.1

525.1

OH

O

O

HOHO

HO

HOO

O

C O

OH3C

Figure 1 Full-scan product ion spectra of (a) gentiopicroside, (b)swertiamarin and (c) internal standard in the negative ionization mode.

Bo Feng et al.Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–••2

system (Millipore, Milford, MA, USA) was used throughoutthe study.

Twelve Sprague-Dawley rats (6 male, 6 female, weight200 ± 20 g) were obtained from the Experimental AnimalCenter of Jilin University (Changchun, China). Animalwelfare and experimental procedures were carried out inaccordance with the guidance for the care and use oflaboratory animals of the National Research Council ofUSA, 1996 and related ethical regulations of Jilin University.

All experimental procedures were performed in accord-ance with the Guidelines of the Experimental Animal Careand Use Committee of Jilin University (Changchun, China).

Instrumental and analytical conditions

The assay was performed on a Prominence™ UFLC system(Shimadzu, Kyoto, Japan) coupled to an API 3200 QTRAP™mass spectrometer (AB Sciex, Foster, CA, USA) operated inthe negative ionization mode. Chromatography was per-formed on a Shim-pack XR-ODS C18 column (75 mm ×3.0 mm, 2.2 μm) maintained at 40°C preceded by a highpressure column prefilter (2 μm). The mobile phase con-sisted of methanol: 0.1% formic acid (30 : 70, v/v) deliveredat a flow rate of 0.4 ml/min. The injection volume was10 μl.

Quantitation was performed using multiple reactionmonitoring (MRM) of the transitions at m/z 401.0 → 179.0for gentiopicroside, m/z 419.0 → 179.1 for swertiamarinand m/z 525.1 → 121.0 for IS. To optimize electrosprayionization conditions, gentiopicroside, swertiamarin and ISwere dissolved in mobile phase and directly injected intothe mass spectrometer. The optimized ionspray voltage wasset at −4500 V and source temperature at 500°C. High-purity nitrogen was used as nebulizing (50 psi), auxiliary(50 psi) and curtain gas (30 psi). Values of declusteringpotential, entrance potential, collision energy (CE) and col-lision cell exit potential were −40 V, −4 V, −20 eV and −4 Vfor gentiopicroside, −40 V, −4 V, −22 eV and −4 V forswertiamarin, and −30 V, −5 V, −40 eV and −2 V for IS.Analyst 1.5.1 software (AB Sciex) was used to control theequipment and for data acquisition and analysis.

Preparation of standard and qualitycontrol samples

Stock solutions (1 mg/ml) of gentiopicroside, swertiamarinand IS were prepared in methanol. Stock solutions ofanalytes were then mixed and serially diluted with methanolto obtain standard solutions containing 0.2–100 μg/mlgentiopicroside and 0.02–10 μg/ml swertiamarin. A10 μg/ml IS working solution was also prepared by dilutionof the stock solution with methanol. All solutions werestored at 4°C and brought to room temperature before use.

Mixed calibration standards containing, respectively,gentiopicroside 20, 50, 200, 500, 2000, 5000 and 10 000 ng/ml, and swertiamarin 2, 5, 20, 50, 200, 500 and 1000 ng/mlwere prepared by spiking 100-μl aliquots of blank plasmawith 10 μl of the appropriate standard solutions. Low,medium and high mixed quality control (QC) samples con-taining, respectively, gentiopicroside 40, 800 and 8000 ng/ml, and swertiamarin 4, 80 and 800 ng/ml were preparedindependently in the same way.

Sample preparation

Plasma samples (calibration standards, QC samples and testsamples) (100 μl) were spiked with 10 μl IS (10 μg/ml),10 μl methanol and 20 μl 0.1% formic acid, vortex-mixedfor 0.5 min and extracted with 2 ml ethyl acetate-isopropanol (95 : 5, v/v) by vortex-mixing for 5 min.After centrifugation at 12 000 rpm for 5 min, the organicphase was transferred to another vial and evaporated todryness at 40°C under a gentle stream of nitrogen. Theresidue was then reconstituted in 100 μl mobile phaseby vortexing for 1 min after which the centrifugationprocedure was repeated. Finally, the supernatant wasanalysed by UFLC-MS/MS.

Preparation of Radix Gentianae extract

Preparation of Radix Gentianae extract and determinationof gentiopicroside and swertiamarin were performedaccording to our previous researches.[17,18] Radix Gentianae(40 g) was extracted with 600 ml 70% ethanol (1 : 15, w/v)by refluxing for 2 h. The extracted solution was filteredthrough five-layer gauzes, and the procedure was repeatedagain. The combined extract was evaporated to remove allethanol, and the residue was diluted with water to a totalvolume of 200 ml. This gives 0.2 g/ml Radix Gentianaeextract containing gentiopicroside 3.3 mg/ml and swertia-marin 0.19 mg/ml. For oral administration, the concentra-tions of gentiopicroside and swertiamarin were determinedby reverse phase-HPLC method.

Assay validation

The method was fully validated in accordance with US FDAguidelines[19,20] with respect to selectivity, linearity, lowerlimits of quantification (LLOQ), accuracy and precision,recovery, matrix effects, and stability.

Selectivity was investigated by comparing chromato-grams of blank plasma from six different rats with andwithout spiking with gentiopicroside, swertiamarin at theLLOQ and IS, and plasma samples obtained after oraladministration of the Radix Gentianae extract.

Linearity was determined by weighted least-squareslinear regression using 1/x2 as weighting factor of

Bo Feng et al. Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–•• 3

calibration curves based on peak area ratios. The LLOQ wasdefined as the lowest drug concentration that could bedetermined with accuracy (as relative error, RE) and preci-sion (as relative standard deviation, RSD) of ±20% based onanalysis of six replicate QC samples.

Accuracy and precision were evaluated by six replicateassays of QC samples on the same day and on 3 consecutivedays. Intraday and interday precision (RSD) should notexceed 15%, and the accuracy (RE) should be ±15%.

Recovery was determined at three QC levels andcalculated by comparing the analyte standard peak areasobtained from extracted samples with post-extractedsamples spiked with the analytes. Matrix effects wereassessed by comparing analyte standard peak areas dis-solved with blank matrix extracts against those dissolvedwith mobile phase at the same concentration.

Stability was assessed by analysing three replicates of QCsamples under the following storage conditions, and all thesamples were analysed with the calibration curves that werefreshly prepared: 4 h at room temperature, after threefreeze-thaw cycles, and after storage at −80°C for 2 weeks.Stability in processed samples in autosampler vials at roomtemperature for 24 h was also assessed.

Pharmacokinetic study

After oral administration of Radix Gentianae extract at adose of 2 g/kg (dried extract/body weight equivalent to33.0 mg/kg gentiopicroside and 1.9 mg/kg swertiamarin),blood samples (approximately 0.3 ml) were collected fromthe fosse orbital vein using heparinized 1.5 ml polythenetubes before the dose and at 0, 0.25, 0.5, 1, 2, 3, 4, 5, 6, 7, 8,9, 10 and 12 h after the dose. Blood samples were immedi-ately centrifuged at 12 000 rpm for 5 min and plasma trans-ferred into clean tubes and stored at −80°C until analysis.

Non-compartmental pharmacokinetic parameters werecalculated by DAS 2.1 pharmacokinetic program (ChinesePharmacological Society, Shanghai, China). Significant dif-ference between two groups was determined using SPSS16.0 (Statistical Package for the Social Science, IBM,Chicago, IL, USA) by independent sample t-tests afternatural logarithmic transformation for area under curve(AUC)0-t, AUC0-∞, Cmax and by the non-parametric Mann–Whitney test for t1/2 and Tmax. A P value less than 0.05 wasconsidered statistically significant.

Results and Discussion

Optimization of chromatographic andmass conditions

As regards the mobile phase, methanol gave better peakshapes and lower background noise than acetonitrile andthe presence of 0.1% formic acid rather than acetic acid

significantly improved the sensitivity, repeatability and peaksymmetry by promoting the ionization of analytes and IS.Finally methanol: 0.1% formic acid (30 : 70, v/v) wasadopted as the mobile phase.

In terms of ESI, higher sensitivity was observed in thenegative ionization mode. The most abundant quasi-molecular ions or adduct ions were (M + HCOO)− forgentiopicroside (m/z 401) and swertiamarin (m/z 419) and(M-H)− for IS (m/z 525). Using low CEs (10–25 eV), bothanalytes produced a major fragment ion at m/z 179. Usinghigher CEs, the intensity of the [M + HCOO]− ion wasreduced as more fragment ions were formed, resulting in aweaker MS response. Thus, in MRM, a CE of −20 eV wasused for the gentiopicroside transition and −22 eV for theswertiamarin transition. For the MRM transition of IS, a CEof −40 eV was found to be optimum. The full-scan production mass spectra of gentiopicroside, swertiamarin and ISare shown in Figure 1.

Sample preparation

Protein precipitation with methanol or acetonitrile gave riseto severe matrix effects for the two analytes as found in pre-vious studies.[15,16] Therefore, liquid-liquid extraction wasevaluated using various organic solvents including ethylacetate, chloroform, and different ratios of ethyl acetate andisopropanol (5–95%, v/v). Ethyl acetate-isopropanol (95 : 5,v/v) showed optimum recovery for the two analytes withno matrix effects. Moreover, formic acid added to sampleswas found to increase recovery. Finally, the addition of 20 μl0.1% formic acid to 100 ml plasma before extraction using2 ml ethyl acetate-isopropanol (95 : 5, v/v) gave highrecovery for both analytes.

Assay validation

Specificity

No endogenous interference was observed at the retentiontimes of gentiopicroside (3.2 min), swertiamarin (2.7 min)and IS (4.0 min). Typical chromatograms of blank plasma,blank plasma spiked with gentiopicroside and swertia-marin, and a rat plasma sample after oral administration ofRadix Gentianae extract are shown in Figure 2.

Calibration curve and lower limitsof quantification

The assay was linearity over the concentration range20–10 000 ng/ml for gentiopicroside (r = 0.9958) and2–1000 ng/ml (r = 0.9964) for swertiamarin. Typical cali-bration curves for gentiopicroside and swertiamarin werey = 6.1 × 10−4x + 1.43 × 10−2 and y = 1.6 × 10−3x + 5.1 × 10−3.The LLOQ of gentiopicroside was 20 ng/ml with RE of3.4% and RSD of 6.5%, and the LLOQ of swertiamarin was2.0 ng/ml with RE of −4.9% and RSD of 9.2%. Based on the

Bo Feng et al.Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–••4

S/N of 3, the limits of detection were 2 and 0.5 ng/ml forgentiopicroside and swertiamarin.

Precision and accuracy

Intraday and interday precision and accuracy data are sum-marized in Table 1. The intraday and interday precisionswere ≤11.9% for gentiopicroside and ≤9.5% for

swertiamarin in terms of RSD, and the accuracy waswithin ±8.3% for gentiopicroside and within ±10.2% forswertiamarin in terms of RE.

Recovery and matrix effects

The extraction recoveries for gentiopicroside and swertia-marin were in the ranges 84.1–86.7% and 68.9–71.4%,

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

Inte

nsi

ty, c

ps

70

60

50

40

30

20

10

0

(1a)

(1b)

(1c) (2c) (3c)

•XIC of -MRM (3 pairs): 401.000/179.000 Da

•XIC of -MRM (3 pairs): 401.000/179.000 Da

•XIC of -MRM (3 pairs): 401.000/179.000 Da

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5Time, min

Time, min

Inte

nsi

ty, c

ps

70

60

50

40

30

20

10

0

(2a)

(2b) (3b)

•XIC of -MRM (3 pairs): 419.000/179.100 Da

•XIC of -MRM (3 pairs): 419.000/179.100 Da

•XIC of -MRM (3 pairs): 419.000/179.100 Da

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5Time, min

Inte

nsi

ty, c

ps

24

20

16

12

8

4

0

(3a)

•XIC of -MRM (3 pairs): 525.100/121.000 Da

•XIC of -MRM (3 pairs): 525.100/121.000 Da

•XIC of -MRM (3 pairs): 525.100/121.000 Da

0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min

400

2400

2000

1600

1200

800

400

0

100

80

60

40

280

240

200

160

120

80

40

0

20

0

320

240

160

80

00.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

Time, min0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5

3.2

3.2

2.7 6000

8000

7000

6000

5000

4000

3000

2000

1000

0

5000

4000

3000

2000

1000

0

4.0

2.7 4.0

Figure 2 Typical chromatograms of (a) blank plasma, (b) blank plasma sample spiked with gentiopicroside, swertiamarin at the lower limits ofquantification and internal standard and (c) a plasma sample collected 1.0 h after oral administration of a Radix Gentianae extract. (1)gentiopicroside, (2) swertiamarin, (3) paeoniflorin (internal standard).

Table 1 Precision, accuracy, recovery and matrix effects for analysis of gentiopicroside and swertiamarin in rat plasma (n = 6)

AnalytesConcentration(ng/ml)

IntradayRSD (%)

InterdayRSD (%)

Accuracy(RE%)

Recovery(%, mean ± SD)

Matrix effect(%, mean ± SD)

Gentiopicroside 40 4.7 6.8 5.6 84.6 ± 5.6 91.2 ± 7.8800 5.2 6.1 8.3 86.7 ± 3.7 89.7 ± 6.2

8000 11.9 9.3 −4.9 84.1 ± 6.8 90.2 ± 3.6Swertiamarin 4 9.5 5.5 10.2 68.9 ± 4.2 96.4 ± 8.1

80 3.6 9.3 8.1 70.6 ± 7.0 102.5 ± 5.5800 8.4 4.2 −3.8 71.4 ± 5.7 94.4 ± 8.8

RE, relative error; RSD, relative standard deviation; SD, standard deviation.

Bo Feng et al. Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–•• 5

respectively (Table 1), which indicated that recoveries wereconsistent and reproducible at different concentrations. Themean extraction recovery of IS was 84.7%.

Matrix effects of gentiopicroside and swertiamarin werein the range 89.7–102.5% at the three QC levels (Table 1).Matrix effects of IS was 98.2%. The results indicated that nosignificant matrix effects were observed for gentiopicroside,swertiamarin and IS.

Stability

Results of stability studies under different conditions aresummarized in Table 2. The measured concentrations forgentiopicroside and swertiamarin at each QC level werewithin ±10.2% of nominal concentrations indicating thatgentiopicroside and swertiamarin are stable in rat plasmaunder the various conditions tested and in processedsamples in autosampler vials for 24 h.

Pharmacokinetic study

The validated analytical method was applied to apharmacokinetic study of gentiopicroside and swertia-marin in rats after oral administration of Radix Gentianaeextract. The plasma concentration versus time curves forgentiopicroside and swertiamarin are presented in Figure 3with corresponding pharmacokinetic parameters shown inTable 3.

As can be seen from Table 3, the value of Tmax and t1/2

were approximately 2–3 and 0.8–2 h for both gentiopi-croside and swertiamarin, indicating that gentiopicrosideand swertiamarin were slowly absorbed but rapidly elimi-nated; the results were in good agreement with previousreports.[12,15] Results demonstrate that, compared with malerats, Cmax, AUC0-t, AUC0-∞ and t1/2 of gentiopicroside andswertiamarin in female rats were significantly greater

(P < 0.05), while no remarkable distinction in Tmax wasobserved. Both gentiopicroside and swertiamarin presentedgreater extent of absorption and slower elimination infemale rats, whereas the rate of absorption was similar inmales and females.

To summarize, pharmacokinetics of gentiopicroside andswertiamarin in rats are subject to a gender effect, which

Table 2 Stability of gentiopicroside and swertiamarin in rat plasma (n = 3)

Conditions

Gentiopicroside Swertiamarin

Concentration(ng/ml)

RE(%)

RSD(%)

Concentration(ng/ml)

RE(%)

RSD(%)

Frozen for 14 days 40 8.5 8.2 4 −7.5 11.2800 −6.5 8.7 80 −5.0 8.3

8000 −3.7 6.3 800 −4.8 9.23 freeze/thaw cycles 40 10.2 10.4 4 −2.5 8.3

800 −4.7 7.5 80 6.3 4.58000 −4.4 5.8 800 −5.1 11.6

Room temperature for 4 h 40 −3.0 11.2 4 7.5 7.8800 3.9 5.9 80 −8.8 8.2

8000 −2.4 7.1 800 −3.4 6.1Processed samples in

autosampler for 24 h40 9.0 9.5 4 5.0 7.4

800 −6.4 3.8 80 −3.8 8.68000 −2.9 8.4 800 −2.3 4.6

RE, relative error; RSD, relative standard deviation.

Co

nce

ntr

atio

n (

ng

/ml)

Co

nce

ntr

atio

n (

ng

/ml)

3000

2000

1000

0

150

100

50

0

0 2 4 6 8 10 12

0 2 4 6 8 10

Female

Male

Female

Male

(a)

(b)

Time (h)

Time (h)

Figure 3 Mean plasma concentration-time profiles for (a)gentiopicroside and (b) swertiamarin in male and female rats after oraladministration of Radix Gentianae extract (data are mean + standarddeviation, n = 6).

Bo Feng et al.Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–••6

could give rise to individual differences in drug efficacy andtoxicity if it translates to human. It has been reported thatthe activity of most CYP enzymes are significantly lower infemale rats than in male rats,[21,22] which may explain thegender difference in the pharmacokinetics of gentiopi-croside and swertiamarin observed in this study. However,further studies are required to explore the mechanism ofthis gender effect to better understand the pharmacokineticprofiles of gentiopicroside and swertiamarin.

Conclusion

A rapid, sensitive and reliable UFLC-MS/MS method hasbeen established and applied to a pharmacokinetic study ofgentiopicroside and swertiamarin after oral administrationto rats. The study is the first to develop a UFLC-MS/MSmethod to simultaneously determine gentiopicroside andswertiamarin in plasma and investigate the influence ofgender on the pharmacokinetics of iridoid glycosides afteroral administration of Radix Gentianae extract. There wereremarkable differences in pharmacokinetic properties ofgentiopicroside and swertiamarin between male and female

rats. The results will provide helpful information for thedevelopment of suitable dosage forms and clinical refer-ences on rational administration.

Declarations

Conflict of interest

The Author(s) declare(s) that they have no conflicts ofinterest to disclose.

Acknowledgements

This research was supported by the National NaturalScience Foundation of China (Grant no. 30973587 and no.81102383), the Science and Technology Major SpecializedProjects for ‘significant new drugs creation’ of the 12th5-year plan (2012ZX09303-015), the National Key Technol-ogy R&D Program of the Ministry of Science andTechnology (2012BAI30B00), and the department ofScience and Technology of Jilin Province, China (No.200905113).

References

1. Pharmacopoeia Commission of theMinistry of Health of the People’sRepublic of China. Pharmacopoeia ofthe People’s Republic of China, Beijing,China: China Medical Science Press,2010.

2. Shen T et al. Research progress inthe chemical constituents of RadixGentianae. J Anhui Agr Sci 2010; 38:16868–16870.

3. Kondo Y et al. Suppression of chemi-cally and immunologically inducedhepatic injuries by gentiopicroside inmice. Planta Med 1994; 60: 414–416.

4. Jaishree V, Badami S. Antioxidantand hepatoprotective effect of

swertiamarin from Enicostemmaaxillare against D-galactosamineinduced acute liver damage in rats.J Ethnopharmacol 2010; 130: 103–106.

5. Lian LH et al. Anti-apoptotic activityof gentiopicroside in d-galactosamine/lipopolysaccharide-induced murinefulminant hepatic failure. Chem BiolInteract 2010; 188: 127–133.

6. Niiho Y et al. Gastroprotective effectsof bitter principles isolated fromGentian root and Swertia herb onexperimentally-induced gastric lesionsin rats. J Nat Med 2006; 60: 82–88.

7. Kumarasamy Y et al. Bioactivity ofgentiopicroside from the aerial partsof Centaurium erythraea. Fitoterapia2003; 74: 151–154.

8. Chen L et al. Down-regulation ofNR2B receptors partially contributesto analgesic effects of Gentiopicrosidein persistent inflammatory pain.Neuropharmacol 2008; 54: 1175–1181.

9. Shin MH et al. Gentiana scabraextracts stimulate glucagon-likepeptide-1 secretion via G protein-coupled receptor pathway. BioChip J2012; 6: 114–119.

10. Zhao SG et al. Separation and deter-mination of gentiopicroside andswertiamarin in Tibetan medicines bymicellar electrokinetic electrophoresis.Biomed Chromatogr 2004; 18: 10–15.

11. Takei H et al. Analysis of swertiamarinin swertia herb and preparations

Table 3 Pharmacokinetic parameters of gentiopicroside and swertiamarin in groups of male and female rats after oral administration of RadixGentianae extract (n = 6)

Parameters Unit

Gentiopicroside Swertiamarin

Male Female Male Female

AUC0-t ng/h/ml 4880 ± 1020 8734 ± 2292* 242 ± 24 363 ± 34*AUC0-∞ ng/h/ml 4950 ± 1010 8867 ± 2330* 246 ± 24 367 ± 34*Cmax ng/ml 1134 ± 233 2431 ± 278* 62.7 ± 13.0 116 ± 10*t1/2 h 1.41 ± 0.41 2.42 ± 0.24* 0.89 ± 0.07 1.27 ± 0.09*

Tmax h 2.17 ± 0.41 2.83 ± 0.75 2.17 ± 0.41 2.50 ± 0.55CL (l/h/kg) 6.96 ± 1.72 4.08 ± 0.98* 9.25 ± 2.35 6.75 ± 0.91*

AUC, area under curve; CL, clearance. *P < 0.05 compared with male group.

Bo Feng et al. Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–•• 7

containing this crude drug by capil-lary electrophoresis. Anal Sci 2001; 17:885–888.

12. Wang CH et al. Pharmacokineticbehavior of gentiopicroside fromdecoction of Radix Gentianae,Gentiana macrophylla after oraladministration in rats: a pharma-cokinetic comparison with gentiopi-croside after oral and intravenousadministration alone. Arch Pharm Res2007; 30: 1149–1154.

13. Wang CH et al. Pharmacokinetics andbioavailability of gentiopicroside fromdecoctions of Gentianae and LongdanXiegan Tang after oral administrationin rats-comparison with gentiopi-croside alone. J Pharm Biomed Anal2007; 44: 1113–1117.

14. Chang J et al. Simultaneous determina-tion of swertiamarin and its metabo-lites (5Z)-5-ethylidene-8-hydroxy-3,4,5,6,7,8-hexahydro-1H-pyrano[3,4-c]

pyridin-1-one and erythrocentaurin inbroth of Aspergillus niger by HPLC.Biomed Chromatogr 2008; 22: 191–195.

15. Li HL et al. Development and valida-tion of a LC-ESI–MS/MS method forthe determination of swertiamarin inrat plasma and its application inpharmacokinetics. J Chromatogr B2011; 879: 1653–1658.

16. Feng Y et al. Urine pharmacokineticsof gentiopicroside in healthy volun-teers. Chin J Clin Pharmacol 2009; 25:22–25.

17. Guo XC et al. Optimization on extrac-tion technology of total iridoid glyco-side from Gentiana scabra Bge. byorthogonal design. J Jilin Med Coll2011; 32: 283–286.

18. Zhu HY et al. RP-HPLC simultaneousdetermination of three active compo-nents in Radix Gentianae. Chin JPharm Anal 2011; 31: 2298–2301.

19. Food and Drug AdministrationGuidance for Industry. BioanalyticalMethod Validation, 2001. http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM070107.pdf. (Acces-sed May 2001).

20. Shah VP et al. Bioanalytical methodvalidation-a revisit with a decade ofprogress. Pharm Res 2000; 17: 1551–1557.

21. Asaoka Y et al. Changes in the geneexpression and enzyme activity ofhepatic cytochrome p450 in juvenilesprague-dawley rats. Toxcol 2010; 72:471–479.

22. Iwano S et al. Focused DNA micro-array analysis for sex-dependentgene expression of drug metabo-lizing enzymes, transporters andnuclear receptors in rat livers andkidneys. J Toxcol Sci 2012; 37: 863–869.

Bo Feng et al.Radix Gentianae: pharmacokinetics

© 2014 Royal Pharmaceutical Society, Journal of Pharmacy and Pharmacology, ••, pp. ••–••8