Pharmacokinetics and metabolism of icaritin in rats by UPLC‐MS/MS

Abstract Icaritin (ICT) has distinct bioactivities, especially known for its beneficial effects on bone‐related degenerative disorders; however, its pharmacokinetic properties remain unknown. A novel developed UPLC‐MS/MS method for the determination of ICT and its main metabolite glucuronidated icaritin (GICT) was firstly applied to pharmacokinetic and metabolism studies of ICT in female rats, which were intraperitoneally given 40 mg/kg ICT. Following the protein precipitation of plasma samples with acetonitrile, ICT and GICT were separated on a C18 column using gradient elution mode and quantified in the multiple reaction monitoring mode. The linearities were acceptable for ICT (r = 0.9960) and GICT (r = 0.9968), and the lower limit of quantification values was 0.5 and 5 ng/ml, respectively. The accuracy fell in the range of 92.0%–103.1% and precisions were within 9.5%. Good linearity, accuracy, precision, and recovery were achieved for the UPLC‐MS/MS method. ICT was predominantly and rapidly biotransformed to GICT which was slowly eliminated in vivo with a terminal half‐life value of 4.51 hr. Pharmacokinetics of pure ICT eliminated biotransformation interference of Epimedium extract and disclosed genuine pharmacokinetic manner of ICT, as well as firstly elucidated low concentration and bioavailability of ICT in rat plasma.

from the crude Epimedium extract, in which, besides ICT, also contained epimedin, icariside, and icariin. More than those, icariin was observed to be transformed to ICT, icariside I, and icariside II in rats , which resulted in increasing plasma ICT concentration after the administration of Epimedium brevicornu extract to rats Wong et al., 2009); the other component, epimedin, was also metabolized to ICT in vivo that elevated plasma ICT level (Han, Shan, Zhou, Zhang, & Hou, 2013). Only one investigation reported the pharmacokinetics of pure ICT in rats using an ultraviolet detection. Unfortunately, glucuronidase and sulphatase were added to plasma to result in the conversion of phase II metabolites to ICT, and faulty pharmacokinetic parameters and bioavailability were obtained by total ICT (free ICT and released ICT) (Chang et al., 2012). Phase II enzymes were also employed to determine ICT concentration after Epimedium brevicornu extract was treated to rats ). In our preliminary study, a phase II metabolite glucuronidated icaritin (GICT, Figure 1b) was found to be the major metabolite of ICT following the administration of pure ICT to rats.
Based on the results by Wong et al. (2009) and Chang et al. (2012), it was inferred that the exposure of phase II metabolites was several hundred times than that of ICT itself, and it was imperative to study the major metabolite of ICT for the safety assessment according to the guideline "safety testing of drug metabolites: guidance for industry (FDA, 2016)." In the present study, a reliable method based on UPLC-MS/MS was firstly established for the simultaneous measurement of plasma ICT and GICT using protein precipitation and was used to study pharmacokinetics and metabolism of pure ICT in rats for exploring the fate of ICT in vivo and further supporting safety evaluation of ICT.

| Sample preparation
Each plasma sample taken from rats was successively spiked with 10 μl of IS working solution (1 μg/ml) and 200 µl of ACN, and subsequently was stirred by a vortex for 5 min and centrifuged at 10,000 g for 10 min at room temperature. The supernatant was collected with a 200-µl polypropylene sample vial for GICT (b), and coumestrol (c) the UPLC-MS/MS analysis. Plasma samples with high ICT/GICT above the ULOQ were diluted by control plasma collected from rats without ICT injection to the concentration range of calibration curve and were measured repeatedly.

| Method validation
According to the " Bioanalytical Method Validation-Guidance for Industry" of the USA Food and Drug Administration (USFDA, 2018), the bioanalytical method of ICT and GICT in plasma was fully validated.

| Pharmacokinetics
Six adult female Sprague Dawley rats, weighing 302 ± 21 g, were already catheterized in the right jugular vein and were intraperitoneally given at a dose of 40 mg/kg of ICT, which was prepared in a

| Selectivity
Representative chromatograms of control plasma and spiked plasma at LLOQ concentration of ICT and GICT are illustrated in Figure 2.
The retention times of ICT, GICT, and the IS were 3.13, 2.72, and 2.66 min, respectively, and interferences for three analytes were not observed, indicating that this method was selective.

| Linearity and LLOQ
Calibration curves of the ratios of peak areas of ICT/GICT to coume-

| Accuracy and precision
The accuracy of both ICT and GICT ranged from 92.0% to 103.1%, and precisions were within 9.5%, as shown in Table 1, indicating that the UPLC-MS/MS method was reproducible.

| Recovery and matrix effect
As illustrated in Table 1, the recoveries for ICT at the levels of LQC, MQC, and HQC were 93.1%, 94.6%, and 96.2%, while those for GICT were 94.8%, 97.6%, and 95.8%, respectively, with the RSD of less than 9.5% for ICT and 8.9% for GICT. In the case of matrix effect, RSD values at three levels were 5.3%-10.1% for ICT and 7.4%-11.2% for GICT. The recovery and matrix effect values of the IS were 81.3 ± 3.6% and 92.2 ± 3.3%, respectively.

| Carryover and dilution integrity
Carryover was not found in blank plasma samples injected immediately following the analysis of the ULOQ samples in three independent runs.

| Stability
The stability results are presented in Table 2. When plasma ICT/GICT was subject to three freeze-thaw cycles, to postextraction storage for 6 hr at ambient temperature, to 4 hr in unprocessed sample at ambient temperature, and storage at −80°C for 7 days, the accuracy of ICT fell in the range of 94.2%-98.3% and that of GICT shifted within 89.9%-104.3%, as well as the precisions were within 10.2% and 6.9% for ICT and GICT, respectively, indicating that ICT and GICT were stable during sample preparation and 7-day storage at −80°C.  Table 3.

| Pharmacokinetics
Poor aqueous solubility and high lipophilicity of ICT resulted in its poor bioavailability (Zhang & Zhang, 2017); therfore, surfactant Cremophor EL and PEG 400, as well as organic solvent ethanol, were employed to solubilize ICT in water. C max and T max were 541.1 ng/ml and 4 hr for ICT, and 4,236.7 ng/ml and 2 hr for GICT, respectively. C max value of 1617 ng/ml appeared at 2 hr when multiple doses of ICT were intraperitoneally injected to rats daily for 7 consecutive days at a dose level of 40 mg kg -1 day -1 (Zhang, 2014). AUC 0-24h ratio of GICT to ICT was 5.92 and concentration ratio range of GICT to ICT at each corresponding time point was 1.0-18.8, implying that ICT was predominantly and rapidly biotransformed to GICT although ICT absorption was slow, and it was necessary to evaluate the safety of GICT in vivo. GCIT was slowly eliminated in vivo with a terminal half-life (t 1/2, λz ) value of 4.51 hr, and t 1/2, λz value of ICT was 3.14 hr. For intraperitoneal administration, t 1/2, λz values of ICT and GICT in plasma were much shorter than those in liver, spleen, kidney, lung, muscle, adipose, and brain in the range of 8.80-50.56 hr (Zhang, 2016). t 1/2, λz value of intraperitoneal route fell between those of intravenous route o(1.72 hr) and oral route (7.37 hr) (Zhang & Zhang, 2017). Similarly, systemic clearance (Cls) and volume of distribution (V) of intraperitoneal administration were between those of intravenous administration and oral administration (Zhang, 2016).
In others' studies (Chang et al., 2012;Wong et al., 2009), conjugated ICT, a phase II metabolite of ICT, was misused as a part of total ICT TA B L E 1 Accuracy, precision, matrix effect, and recovery of the method (Mean ± SD, n = 6) to calculate pharmacokinetic parameters. The LLOQ value (20 ng/ ml) was too high to detect the plasma ICT concentration failed after 2 hr postdosing (Chang et al., 2012). In our pharmacokinetic study, a parent compound ICT (not conjugated ICT) was quantified at a very low concentration based on a small amount of plasma using the present method. The genuine pharmacokinetic properties of ICT were achieved for the evaluation of the compound in animals.

| CON CLUS ION
A reliable UPLC-MS/MS method was firstly established for the quantification of ICT and its dominant metabolite GICT in plasma. The linearities were acceptable for ICT (r = 0.9960) and GICT (r = 0.9968), and LLOQ values were 0.5 and 5 ng/ml, respectively. The accuracy fell in the range of 92.0%-103.1%, and precision was within 9.5%.
The UPLC-MS/MS method performed well in linearity, accuracy, precision, and recovery and was firstly utilized for pharmacokinetic and metabolism studies of ICT in female rats following a single intraperitoneal injection of ICT. ICT was predominantly and rapidly biotransformed to GICT which was slowly eliminated in vivo with a terminal half-life value of 4.51 hr.

AUTH O R CO NTR I B UTI O N S
Shuang-Qing Zhang designed the experiments; Zhen-Wu Huang, Yue-Xin Yang, and Ling-He Huang performed the experiments; Zhen-Wu Huang wrote the manuscript. All authors revised and approved the manuscript.

E TH I C A L A PPROVA L
The animal experiments were approved by our Institutional Animal