Simultaneous Determination and Pharmacokinetic Characterization of Glycyrrhizin, Isoliquiritigenin, Liquiritigenin, and Liquiritin in Rat Plasma Following Oral Administration of Glycyrrhizae Radix Extract

Glycyrrhizae Radix is widely used as herbal medicine and is effective against inflammation, various cancers, and digestive disorders. We aimed to develop a sensitive and simultaneous analytical method for detecting glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, the four marker components of Glycyrrhizae Radix extract (GRE), in rat plasma using liquid chromatography-tandem mass spectrometry and to apply this analytical method to pharmacokinetic studies. Retention times for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin were 7.8 min, 4.1 min, 3.1 min, and 2.0 min, respectively, suggesting that the four analytes were well separated without any interfering peaks around the peak elution time. The lower limit of quantitation was 2 ng/mL for glycyrrhizin and 0.2 ng/mL for isoliquiritigenin, liquiritigenin, and liquiritin; the inter- and intra-day accuracy, precision, and stability were less than 15%. Plasma concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin were quantified for 24 h after a single oral administration of 1 g/kg GRE to four rats. Among the four components, plasma concentration of glycyrrhizin was the highest and exhibited a long half-life (23.1 ± 15.5 h). Interestingly, plasma concentrations of isoliquiritigenin and liquiritigenin were restored to the initial concentration at 4–10 h after the GRE administration, as evidenced by liquiritin biotransformation into isoliquiritigenin and liquiritigenin, catalyzed by fecal lysate and gut wall enzymes. In conclusion, our analytical method developed for detecting glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin could be successfully applied to investigate their pharmacokinetic properties in rats and would be useful for conducting further studies on the efficacy, toxicity, and biopharmaceutics of GREs and their marker components.


Introduction
Glycyrrhizae Radix (licorice root) has been used as a herbal medicine because of a variety of pharmacological activities, including anti-oxidative, anti-cancer, and anti-diabetic activities as well as memory enhancing and inflammation reducing effects [1]. In addition, it has been used as a flavoring agent in food products [2] and also as an adjuvant to increase the therapeutic efficacy of other drugs. For example, Glycyrrhizae Radix lowers the risk of aphthous ulcers caused by aspirin intake and In order to optimize ESI conditions for four components, each compound was injected directly into the mass spectrometer ionization source. Glycyrrhizin showed optimal ionization in positive mode and isoliquiritigenin, liquiritigenin, and liquiritin showed optimal ionization in negative mode. Figure 1 shows the mass spectra and chemical structure of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin. The selection of berberine as an internal standard (IS) was based on its simultaneous determination with glycyrrhizin, which was present at the highest concentration in GRE, in a positive ionization mode [21,22]. In addition, berberine eluted in the middle of glycyrrhizin and the three flavones and it showed a stable extraction recovery with low coefficient of variation (CV). The optimized analytical conditions including mass transition from the precursor to product ion (m/z) for glycyrrhizin, isoliquiritigenin, liquiritigenin, liquiritin, and berberine (IS) are listed in Table 1.  Representative multiple reaction monitoring (MRM) chromatograms of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, and IS ( Figure 2) showed that the four analytes and berberine (IS) peaks were well-separated with no interfering peaks at their respective retention times corresponding to the concentration of the lower limit of quantification (LLOQ). The retention times  Representative multiple reaction monitoring (MRM) chromatograms of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, and IS ( Figure 2) showed that the four analytes and berberine (IS) peaks were well-separated with no interfering peaks at their respective retention times corresponding to the concentration of the lower limit of quantification (LLOQ). The retention times of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, and IS were 7.8 min, 4.1 min, 3.1 min, 2.0 min, and 4.8 min, respectively, and the total run time was 10.0 min. The selectivity of the analytes was confirmed from six different blank rat plasma and plasma samples obtained from rats at 2 h after oral administration of GRE (1 g/kg) ( Figure 3). of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, and IS were 7.8 min, 4.1 min, 3.1 min, 2.0 min, and 4.8 min, respectively, and the total run time was 10.0 min. The selectivity of the analytes was confirmed from six different blank rat plasma and plasma samples obtained from rats at 2 h after oral administration of GRE (1 g/kg) ( Figure 3).   (3), liquiritigenin (2), liquiritin (1), and berberine (4) in rat plasma samples at 2 h following single oral administration of GRE (1 g/kg).

Linearity and LLOQs
To access linearity, the calibration curve consisting of seven different concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin was analyzed, and the calibration curves and equations for each component have been shown in Figure 4 and Table 2. The LLOQs for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin in our analytical system was defined by a signal-to-noise ratio of >5, the precision was ≤15%, and the accuracy was 80-120%; these results have been listed in Table 2. of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin, and IS were 7.8 min, 4.1 min, 3.1 min, 2.0 min, and 4.8 min, respectively, and the total run time was 10.0 min. The selectivity of the analytes was confirmed from six different blank rat plasma and plasma samples obtained from rats at 2 h after oral administration of GRE (1 g/kg) ( Figure 3).

Linearity and LLOQs
To access linearity, the calibration curve consisting of seven different concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin was analyzed, and the calibration curves and equations for each component have been shown in Figure 4 and Table 2. The LLOQs for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin in our analytical system was defined by a signal-to-noise ratio of >5, the precision was ≤15%, and the accuracy was 80-120%; these results have been listed in Table 2.

Linearity and LLOQs
To access linearity, the calibration curve consisting of seven different concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin was analyzed, and the calibration curves and equations for each component have been shown in Figure 4 and Table 2. The LLOQs for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin in our analytical system was defined by a signal-to-noise ratio of >5, the precision was ≤15%, and the accuracy was 80-120%; these results have been listed in Table 2.

Accuracy and Precision
The inter-and intra-day accuracy and precision were assessed using three concentrations of quality control (QC) samples consisting of the mixture of four analytes ( Table 3). The results showed that inter-and intra-day precision for glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin was below 13.6%. The inter-and intra-day accuracy for glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin ranged from 87.4% to 112.2% (Table 3).

Accuracy and Precision
The inter-and intra-day accuracy and precision were assessed using three concentrations of quality control (QC) samples consisting of the mixture of four analytes ( Table 3). The results showed that inter-and intra-day precision for glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin was below 13.6%. The inter-and intra-day accuracy for glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin ranged from 87.4% to 112.2% (Table 3).

Matrix Effect and Recovery
The extraction recoveries of glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin in the low, medium, high QC samples ranged from 70.3% to 99.1% with CV of <14.0%. The matrix effects ranged from 76.2 to 114.2% with CV of <14.8%. These results indicate that no significant interference occurred during the ionization and methanol precipitation process. The extraction recovery and matrix effect of the IS at 0.1 ng/mL were 86.2% and 108.2%, respectively (Table 4). Data represented as mean ± SD from six independent experiments.

Stability
It was found that the precision and accuracy of QC samples consisting of a mixture of the four analytes were within 12.9% for short-term stability, below 6.4% for post-preparative stability, and below 11.1% for three freeze-thaw cycle stability (Table 5). Therefore, glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin in plasma samples were found to exhibit no problems in these three stability tests during the bioanalytical procedure.

Plasma Concentration of Glycyrrhizin, Liquiritin, Isoliquiritigenin, and Liquiritigenin
Next, we investigated the plasma concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin following a single oral GRE administration at a dose of 1 g/kg. Plasma concentrations of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin over time and their PK parameters are shown in Figure 5 and Table 7, respectively. Among the four major components present in rat plasma, glycyrrhizin was found to be maintained at the highest concentration for a period of 24 h. In contrast, the plasma concentration of liquiritin gradually reduced at an elimination half-life (T 1/2 ) of 3.7 ± 2.2 h. Concentrations of isoliquiritigenin and liquiritigenin were similar and increased up to 8 h and then gradually reduced; therefore, T max and MRT values of both these compounds were very similar. The area under the plasma concentration-time curve (AUC) of isoliquiritigenin and liquiritigenin were comparable but higher than that of liquiritin, despite their concentration (0.014% and 0.027%, respectively) in GRE being lower than that of liquiritin (0.38%). Collectively, these results suggested that isoliquiritigenin and liquiritigenin were transformed during the intestinal absorption process. respectively) in GRE being lower than that of liquiritin (0.38%). Collectively, these results suggested that isoliquiritigenin and liquiritigenin were transformed during the intestinal absorption process.  Figure 5. Plasma concentration-time profiles of glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin and after oral administration of GRE to rats. Data were expressed as mean ± SD from four rats per group.

Biotransformation in the Rat Intestine
We investigated whether the biotransformation of isoliquiritigenin and liquiritigenin from liquiritin could occur in the rat intestine. Because the plasma concentration of isoliquiritigenin and liquiritigenin was increased at 4-10 h after oral GRE administration, we measured biotransformation in the rat ileum segment. After a 2 h incubation of a single component of GRE such as isoliquiritigenin, liquiritigenin, and liquiritin with rat ileum segments and intestinal contents, liquiritin was found to transform into isoliquiritigenin and liquiritigenin and the formation rate of both these compounds were similar to each other ( Figure 6A-C). Moreover, although isoliquiritigenin and liquiritigenin were interchangeable, they did not transform into liquiritin ( Figure 6D).

Biotransformation in the Rat Intestine
We investigated whether the biotransformation of isoliquiritigenin and liquiritigenin from liquiritin could occur in the rat intestine. Because the plasma concentration of isoliquiritigenin and liquiritigenin was increased at 4-10 h after oral GRE administration, we measured biotransformation in the rat ileum segment. After a 2 h incubation of a single component of GRE such as isoliquiritigenin, liquiritigenin, and liquiritin with rat ileum segments and intestinal contents, liquiritin was found to transform into isoliquiritigenin and liquiritigenin and the formation rate of both these compounds were similar to each other ( Figure 6A-C). Moreover, although isoliquiritigenin and liquiritigenin were interchangeable, they did not transform into liquiritin ( Figure 6D). LG, and ILG in rat intestine. Data were expressed as mean ± SD from six independent experiments per group.

Discussion
In this study, the newly developed analytical method for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin using an LC-MS/MS system showed relatively higher sensitivity (i.e., LLOQ 2 ng/mL for glycyrrhizin and 0.2 ng/mL for isoliquiritigenin, liquiritigenin, and liquiritin) despite using lower plasma sample volume (50 μL). For example, Wang et al. implemented a proteinprecipitation method and sample preparations via evaporation and reconstitution for detecting 10 active constituents in Shaoyao-Gancao decoction. The LLOQs for glycyrrhizin and three flavone compounds were 5 and 0.5 ng/mL, respectively [17]. Mao et al. applied analytical methods for glycyrrhizin, glycyrrhetinic acid, isoliquiritigenin, liquiritigenin, isoliquiritin, and liquiritin using an LC-MS/MS system with 10 and 0.4 ng/mL of LLOQ for glycyrrhizin and three flavones, respectively [18]. Additionally, previously established methods by Shan et al. applied liquid-liquid extraction which requires acidification with HCl for the extraction of glycyrrhizin and glycyrrhetinic acid and a larger plasma sample volume (100 μL) and the LLOQs for glycyrrhizin and three flavones were 1 and 0.34-0.5 ng/mL, respectively [19]. Herein, we used a protein-precipitation method with methanol containing IS rather than a previously described liquid-liquid extraction method or sample preparation via evaporation and reconstitution method [8,17,19], and then directly injected an aliquot of the supernatant after centrifugation of protein-precipitated plasma samples.
We further validated our simple, sensitive, and simultaneous analytical method by performing a pharmacokinetic study after orally administering rats with 1 g/kg of GRE. We successfully measured the plasma concentrations of glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin for 24 h. However, we should take into account the fact that the pharmacological efficacy was investigated following repeated oral administration for 9 days at a dose range of 0.1-0.5 g/kg. Thus, the pharmacokinetic study involving repeated administration of GRE at a lower dose range of 0.1-0.5 g/kg needs to be performed to understand the pharmacokinetic-pharmacodynamic correlation of GRE.
The pharmacokinetic features of glycyrrhizin, including its long half-life, were consistent with

Discussion
In this study, the newly developed analytical method for glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin using an LC-MS/MS system showed relatively higher sensitivity (i.e., LLOQ 2 ng/mL for glycyrrhizin and 0.2 ng/mL for isoliquiritigenin, liquiritigenin, and liquiritin) despite using lower plasma sample volume (50 µL). For example, Wang et al. implemented a protein-precipitation method and sample preparations via evaporation and reconstitution for detecting 10 active constituents in Shaoyao-Gancao decoction. The LLOQs for glycyrrhizin and three flavone compounds were 5 and 0.5 ng/mL, respectively [17]. Mao et al. applied analytical methods for glycyrrhizin, glycyrrhetinic acid, isoliquiritigenin, liquiritigenin, isoliquiritin, and liquiritin using an LC-MS/MS system with 10 and 0.4 ng/mL of LLOQ for glycyrrhizin and three flavones, respectively [18]. Additionally, previously established methods by Shan et al. applied liquid-liquid extraction which requires acidification with HCl for the extraction of glycyrrhizin and glycyrrhetinic acid and a larger plasma sample volume (100 µL) and the LLOQs for glycyrrhizin and three flavones were 1 and 0.34-0.5 ng/mL, respectively [19]. Herein, we used a protein-precipitation method with methanol containing IS rather than a previously described liquid-liquid extraction method or sample preparation via evaporation and reconstitution method [8,17,19], and then directly injected an aliquot of the supernatant after centrifugation of protein-precipitated plasma samples.
We further validated our simple, sensitive, and simultaneous analytical method by performing a pharmacokinetic study after orally administering rats with 1 g/kg of GRE. We successfully measured the plasma concentrations of glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin for 24 h. However, we should take into account the fact that the pharmacological efficacy was investigated following repeated oral administration for 9 days at a dose range of 0.1-0.5 g/kg. Thus, the pharmacokinetic study involving repeated administration of GRE at a lower dose range of 0.1-0.5 g/kg needs to be performed to understand the pharmacokinetic-pharmacodynamic correlation of GRE.
The pharmacokinetic features of glycyrrhizin, including its long half-life, were consistent with those reported in previous studies [23,24]. The plasma concentrations of isoliquiritigenin and liquiritigenin were similar and showed fast elimination up to 4 h but rebounded to the initial plasma concentration over 10 h. As demonstrated in this study (Figure 6), liquiritin was hydrolyzed to isoliquiritigenin and liquiritigenin, and both isoliquiritigenin and liquiritigenin were interchangeable. The results suggested that liquiritin in GRE could be a precursor for isoliquiritigenin and liquiritigenin and is a more potent pharmacological component [25]; therefore, the hydrolysis of liquiritin could be attributed to an increased plasma concentration of isoliquiritigenin and liquiritigenin during the absorption period (from 4 to 10 h in this study). The results were consistent with the previous report that isoliquiritigenin and liquiritigenin, which were generated from liquiritin, were absorbed from the jejunum to colon with the help of gut wall enzymes and intestinal flora [26]. Mao et al. reported that the metabolism of some active components of Gancao, including glycyrrhizin and liquiritigenin, in rat fecal lysate was changed after co-administration with Jiegeng; consequently, this could be an important factor for alterations in the pharmacokinetic profiles of glycyrrhizin and liquiritigenin [18]. According to their results, not only Jiegeng but also Gancao changed the hydrolysis of liquiritigenin [18], which is similar to this study and could be a factor responsible for higher plasma concentration of liquiritigenin after GRE administration. In addition to this biotransformation, liquiritin permeability was much greater when added as a part of GRE as compared with the addition of an equal amount of liquiritin alone [18]. These results should also be clarified by performing a pharmacokinetic comparison between GRE administration and the administration of an equal amount of a single component as well as a comparison between intestinal absorption and metabolism; these topics are currently being investigated by our research group.
Glycyrrhizae Radix extract (GRE; #KNUNPM GR-2015-001, deposited at the laboratory of Natural Products Medicine in Kyungpook National University) was used [1]. Briefly, dried Glycyrrhizae Radix, imported from China to Republic of Korea in 2015, was extracted with 94% ethanol for 3 h. The extracted solution was filtrated and concentrated by a rotary evaporator to obtain the GRE.

Animals
Male Sprague-Dawley rats (7 weeks of age, 249 ± 6 g), obtained from Samtako Bio Korea (Osan, Kyunggido, Korea) were used for pharmacokinetic experiments. All animal procedures were approved by the Animal Care and Use Committee of Kyungpook National University (approval No. KNU-2017-0126) and conducted in accordance with the National Institutes of Health guidance for the care and the use of laboratory animals. Rats were maintained in an animal facility at the College of Pharmacy, Kyungpook National University at a temperature of 21-27 • C with 13 h light (08:00-21:00) and a relative humidity of 60 ± 5%.

Preparation of Calibration Curve and Quality Control Samples
Calibration curve samples containing glycyrrhizin (2 to 500 ng/mL), liquiritin (0.2 to 100 ng/mL), isoliquiritigenin (0.2 to 50 ng/mL), and liquiritigenin (0.2 to 50 ng/mL) were prepared using an internal standard method. Briefly, aliquots of calibration curve samples (50 µL) was added to 300 µL of methanol containing 0.1 ng/mL berberine, vortex-mixed for 10 min, and centrifuged at 10,000× g for 10 min at 4 • C. An aliquot (10 µL) of the supernatant was injected into the LC-MS/MS system.

Specificity
Six individual blank plasma samples from rats were used and specificity was defined as no interfering signal at the peak region of each analyte (glycyrrhizin, isoliquiritigenin, liquiritigenin, and liquiritin) and internal standard from background and endogenous signal.

Linearity
The linearity of the method was evaluated by analyzing a series of calibration samples consisting glycyrrhizin (from 2 to 500 ng/mL), liquiritin (from 0.2 to 100 ng/mL), isoliquiritigenin and liquiritigenin (from 0.2 to 50 ng/mL). Least square linear regression equation of the peak area ratios of each analyte to IS against the corresponding concentrations was obtained with a weighting factor of 1/x 2 .

Accuracy and Precision
The accuracy and precision of intra-day and inter-day were analyzed by six replicate QC samples on the same day and six consecutive days, respectively.
Accuracy is described in relative percentage of measured concentration compared to the spiked concentration and precision are determined by relative standard deviation compared to the average concentration of QC samples.

Extraction Recovery and Matrix Effect
Extraction recovery was calculated by comparing the peak areas of each analytes in QC samples through the extraction process with those in blank plasma extracts spiked with corresponding concentrations [27]. Matrix effects were monitored by dividing the peak areas in blank plasma extracts spiked with QC concentrations by those in neat solutions of corresponding concentrations [27].

Stability
The stability of four analytes in the rat plasma was tested from QC samples exposed to three different conditions [28]. Short-term stability was calculated by comparing QC samples that were stored for 5 h at 25 • C before sample preparation with the untreated QC samples. The three freeze-thaw cycle stability was analyzed by comparing QC samples that underwent three freeze-thaw cycles (−80 • C to 25 • C and standing for 3 h at 25 • C defined as one cycle) with the untreated QC samples. Post-preparative stability was evaluated by comparing the post-preparative QC samples maintained in the autosampler at 4 • C for 12 h with the untreated QC samples [28].
This method can easily be applied in the bioanalysis and pharmacokinetic studies of GRE, including its administration at multiple therapeutic doses, or for making pharmacokinetic comparisons among individual components in small experimental animals. Moreover, following an appropriate validation, the present method can be extended to determine routine drug monitoring of glycyrrhizin, isoliquiritigenin, liquiritin, and liquiritigenin in plasma samples as well as in other biological samples, and thus, can be applied to in vivo pharmacokinetic-pharmacodynamic correlation studies.