Poliumoside (Figure 1, purity > 98%) and caffeic acid (used as internal standard (IS) were purchased from Chengdu Herbpurity Co. Ltd. (Chengdu, China). Analytical grade trichloroacetic acid (TCA) was purchased from Tianjin Chemical Corporation (Tianjin China). Acetonitrile, methanol and formic acid were all of HPLC grade.

Chromatographic separation of the analyte was performed on Acquity UPLC system with BEHC₁₈ column (2.1 × 100 mm, 1.7 μm, Waters, Milford, MA, USA). An aliquot of the sample (5 μL) was injected into the UPLC system for analysis. Gradient elution was performed with formic acid (0.1%) in water (mobile phase A) and acetonitrile (mobile phase B) at a flow rate of 0.4 mL/min. The gradient elution process was used as follows: 5% B at 0 to 1 min, 5 to 95% B at 1 to 3 min, then back to the initial condition within 1 min. Before the next injection, all the samples would undergo a needle wash program to minimize the pollution, the solvent for needle wash containing methanol and water (v/v, 50:50). The column temperature was maintained at 45˚C.

Mass spectrometry was performed on the Xevo G2 Q-TOF MS (Waters) equipped with the electronic spray ion (ESI) [13] under negative ion mode. The extraction cone, sampling cone and capillary voltages were 4, 40 and 3000 V, respectively. Nitrogen used as cone and desolvation gas were set to 30 and 800 L/h, respectively. The temperature of source and desolvation were 100˚C and 250˚C, respectively. Mass spectra were recorded in the range of m/z 50 - 1000. An external reference 2 μg/mL leucine enkephalin (m/z 556.2771) was infused at 5 μL/min to guarantee mass accuracy of the entire assay. The ratio mass-to-charge of poliumoside and IS were 769.2537 and 179.0328 respectively. MS quantitation for pharmacokinetics was performed on the full scan analysis and extracted ion chromatograms using MassLynx version 4.1.

Stock solutions of poliumoside (1 mg/mL) were prepared in saline. Working solutions for control and calibration samples were generated by gradient dilution of the stock solution with methanol:water (60:40). The 10% TCA used as precipitating agent was prepared in purified water. The IS (caffeic acid) was prepared by dilution of the stock solution (100 μg/mL) with the same solvent to a concentration of 10 μg/mL. All of the described solutions were stored at 4˚C before use.

The calibration solutions were obtained by spiking appropriate volume of working solutions into the aliquots of 100 μL blank rat plasma, and the plots were set up at the range of 50 - 10,000 ng/mL (50, 100, 625, 1250, 2500, 5000, 10,000 ng/mL). As the same to calibration standard, quality control (QC) samples were prepared in three different plasma concentration level (200, 1000, 8000 ng/mL). All of the samples were stored at −20˚C.

The rat plasma samples, calibration standard solution and the QC samples were thawed at room temperature before use and put them on the ice for the consideration of stability. Plasma samples were extracted through protein precipitation method by 10% TCA. In brief, an aliquot of 20 μL IS working solution (10 μg/mL) was added into 100 μL plasma samples in a 1.5 mL polypropylenes tube, followed by vortexing for 30 seconds. Subsequently, 80 μL of 10% TCA was added into each tube, vortexing for another 3 min. The mixture was centrifuged under 13,000 g for 15 min at 4˚C. The supernatant was collected and transferred into a new tube. Then, 5 μL of the supernatant was injected into the UPLC/Q-TOF-MS system for the further analysis.

Male Sprague-Dawley (SD) rats weighing 220 - 250 g were obtained from Laboratory Animal center of Jinan University (Guangdong, China). Rats were housed in a humidity-controlled room maintained on 12 h light-dark cycles for 5 days before the experiment. Standard animal food and water were provided ad libitum. Twelve rats were anaesthetized using 10% chloral hydrate and the jugular vein were cannulated for blood collection before 24 h of the experiment. Diet was prohibited for 12 h before the experiment, while water was freely taken. All procedures were approved by the Animal Care and Ethics Committee of Jinan University (Guangzhou, China).

Rats were randomly assigned to two groups (six rats per group). Poliumoside was dissolved in saline and administrated to rats by oral and intravenous injection. Approximately 0.3 mL blood samples were collected from jugular vein into 1.5 mL heparinized centrifuge tubes. Time points were set at 2, 5, 10, 20, 30, 45, 60, 90, 120, 180, 240, 360 min for the intravenous group (10 mg/kg) and 5, 10, 20, 30, 45, 60, 90, 120, 180, 240, 360, 480, 720 min for the oral group (200 mg/kg). Blood samples were immediately centrifuged (5000 g, 4˚C, 8 min), and 100 μL aliquot of supernatant (plasma) was collected and frozen at −80˚C before analysis.

MS parameters were optimized by directly infusing a standard solution into the ESI source. Both positive and negative ion mode were employed to confirm the accurate ions of poliumoside and IS. All analytes showed a better response under the negative mode. Ultimately, the m/z 769.2537 for poliumoside and 179.0328 for IS were used for the further analysis under negative ion mode. Other parameters have been showed in section 2.2.

To separate the interfering compounds from the poliumoside and IS, different columns and mobile phase compositions were compared for chromatographic separation. The Waters BEH C₁₈ column showed better peak shape and appropriate retention time for the analyte than Phenomenex Kinetex C₁₈ column (2.1 × 50 mm, 2.6 μm). Appropriate ionization and chromatographic behavior were based on the proper mobile phase [14] - [16] . The gradient mobile phase with water (containing 0.1% formic acid) and acetonitrile greatly shortened the retention time of the analyte. The gradient was verified to be stable and repeatable according to the following test of method validations. The addition of 0.1% formic acid in water and acetonitrile could improve peak shapes under the optimized conditions. The retention time of IS and poliumoside were 1.91 min and 1.73 min for pharmacokinetic study, respectively.

Appropriate extraction method of the plasma is essential for the quantitative analysis. The technique of liquid-liquid extraction and precipitation of protein was tested for alternative. The method of liquid-liquid extraction showed low recovery (30%) when ethyl acetate was used as extraction solvents, and this method was both complicated and time consuming. Various protein-precipitating reagents, such as methanol, acetonitrile and 10% TCA, were tested. Strong response value, good recovery and clear supernatant for the analytes were observed by using 10% TCA as the protein-precipitating reagent. Comparing with the neutral or alkaline conditions, poliumoside was found more stable in acidic solvents [11] . Analyte was more stable in 10% TCA (as acidic solvent) than in the other two precipitant at room temperature, so 10% TCA was elected to be the most suitable protein- precipitating reagent in the experiment.

The plasma concentration-time profiles of poliumoside were shown in Figure 4. The concentration of poliumoside was calculated using the following equation:

A is the peak area ratio of poliumoside to IS from different samples. The mean pharmacokinetic parameters from non-compartment modeling were listed in Table 4. All the pharmacokinetic parameters were calculated by PK solver software. The validated method was successfully applied to measure the plasma concentration of poliumoside in rats. After oral (200 mg/kg) administration of poliumoside, the T_max was 30 min, suggesting that poliumoside was rapidly absorbed through the gastrointestinal (GI) tract. However, the calculated C_max was only 561 ng/mL and the low absorbing may result from the high polarity of the structure of poliumoside. Plasma concentration experienced rapidly decline after peak concentration. The fast tissue distribution and the sluggish overall elimination may give rise to the result. The AUC_0−t was 146 ± 9.88 μg/mL*min ,the mean residence time (MRT) of the analyte after oral and intravenous administration were 15.41 ± 4.17 h and 0.74 ± 0.29 h, respectively.

Abbreviations: AUC_0−t, area under the concentration-time curve calculated from zero up to the last measured concentration; AUC_0−∞, area under the concentration-time curve extrapolated from zero up to infinity; MRT, mean residence time of the analysis; T_max, time to maximum observed concentartion; C_max, maximum plasma concentration; t_1/2, elimination; AUMC_0−∞, area under the first moment of the concentration-time curve from zero up to infinity; V, volume of distribution; CL, clearance.

The systematic pharmacokinetics of poliumoside in rats was explored for the first time. Furtherly, the bioavailability of poliumoside was firstly reported to be 0.69%, which is similar to the other PhGs [18] - [20] . The metabolism of PhG in GI tract was largely reported, such as poliumoside [21] and echinacoside [22] [23] . The complex microbial ecosystem in GI is generally considered as a separated organ with powerful metabolic capacity. The metabolism in GI may directly decrease absorption of poliumoside and result in the low bioavailability ultimately.