Exploration for the real causative agents of licorice-induced pseudoaldosteronism

Abstract

I investigated the causative agents of licorice-induced pseudoaldosteronism, which is a frequent side effect of Japanese traditional Kampo medicines. Glycyrrhizin (GL), the main ingredient of licorice, is absorbed after being metabolized to glycyrrhetinic acid (GA) by intestinal bacteria, and then metabolized in liver to 3-monoglucuronyl-glycyrrhetinic acid (3MGA). In normal condition, 3MGA is excreted into bile via a multidrug resistance-related protein (Mrp) 2, therefore, 3MGA does not appear in blood circulation. However, under the dysfunction of Mrp2, 3MGA appears in the blood circulation and is excreted into the urine by not glomerular filtration but tubular secretion via organic anion transporter (OAT) 1 and 3. At this time, 3MGA inhibits type 2 11β-hydroxysteroid dehydrogenase (11βHSD2) in tubular cells to cause pseudoaldosteronism. Since GA is not the substrates of these transporters, GA cannot inhibit 11βHSD2 in tubular cells. Therefore, it was considered that 3MGA was the causative agents of licorice-induced pseudoaldosteronism. After that, I isolated and identified three other GL metabolites, 22α-hydroxy-18β-glycyrrhetyl-3-O-sulfate-30-glucuronide (1), 22α-hydroxy-18β-glycyrrhetyl-3-O-sulfate (2), and 18β-glycyrrhetyl-3-O-sulfate (3) from the urine of Mrp2-deficient rats orally treated with GA, and found that their blood and urinary concentrations were much higher than 3MGA and that their pharmacokinetic behaviors were similar to 3MGA. 3MGA was not detected in the blood of patients with pseudoaldosteronism who developed rhabdomyolysis due to licorice, and compound 3 was detected at a high concentration. In addition, a multicenter retrospective study was conducted using the serum and urine of 97 patients who took Kampo medicines containing licorice. Of a total of 97 patients, 67 detected GA in the serum (median 122 nM, 5 nM-1.8 µM) and 68 detected compound 3 (median 239 nM, 2 nM-4.2 µM), and there were no cases of detection of GL, 3MGA, compounds 1, and 2. High blood concentrations of compound 3 were associated with low plasma renin activity, plasma aldosterone levels, and serum potassium levels. It is highly probable that compound 3 is the true causative agent of pseudoaldosteronism.

Keywords: 18β-glycyrrhetyl-3-O-sulfate; Adverse effects; Glycyrrhiza; Glycyrrhizin; Licorice; Pseudoaldosteronism.

Fig. 1

Chemical structures of glycyrrhizin (GL) and its metabolites. GA, 18β-glycyrrhetinic acid; 3MGA, 3-monoglucuronyl-glycyrrhetinic acid; compound 1, 22α-hydroxy-18β-glycyrrhetyl-3-O-sulfate-30-glucuronide (1); compound 2, 22α-hydroxy-18β-glycyrrhetyl-3-O-sulfate (2); compound 3, 18β-glycyrrhetyl-3-O-sulfate (3)

Fig. 2

Eastern blot analysis of urine collected from EHBRs orally treated with GA using anti-3MGA mAb [14]. On the photographs, base lines were added. On TLC plate, the following samples were spotted at the base line; lane 1, double spot of 6 μl of urine collected from EHBRs which were orally administered with GA as drinking water (1 mg/ml) and 2 μl of 3MGA (1 mM); lane 2, 2 μl of 3MGA (1 mM); lane 3, double spot of 6 μl of urine collected from EHBRs and 2 μl of compound 1 (1 mM); lane 4, 2 μl of compound 1 (1 mM); lane 5, 6 μl of urine collected from EHBRs. Then, the spots were spreaded out using H₂O/BuOH/AcOH (2:7:3), and the top line of the solvent was penciled. Photograph of TLC plate detected by UV absorption at 254 nm (left) was taken. On the lanes 2 and 4 used as positive control, the spots appeared were marked by pencil. The spots on the plate were transferred onto PES membrane, fixed, blocked, and strained by anti-3MGA MAb. Photograph of the membrane was taken (right). Rf values of 3MGA and compound 1 were 0.76 and 0.50 respectively

Fig. 3

Pharmacokinetic profiles of GL metabolites in SD rats (a, c, e, f) and EHBRs (b, d) after the administration of GA [16]. GA (200 mg/kg) was administered orally to anesthetized SD rats or EHBRs, and plasma and urine were collected for 12 h (a–d). GA (0.2 mg/kg) was intravenously injected into anesthetized SD rats in which the biliary tract was cannulated. Then, bile samples were collected for 4 h (e). GA (0.2 mg/kg) was intravenously injected into conscious SD rats, and the feces was collected for 24 h. 3MGA and 1 were below detectable levels in the feces (f). Since the determined levels of 3MGA for B and D, 3 for C, GA for D, and all compounds except for 3 for E were relatively low; their magnified versions of their graphs are shown at the top or as an inside of each graph. The concentrations of GA metabolites were measured by LC–MS/MS, and data are plotted as mean ± S.E. (n = 4 for A–D, F; n = 3 for E)

Fig. 4

Inhibitory effects of compounds 1–3 on 11β-HSD2 using rat kidney microsome [14; 16]. [³H] cortisone and each compound were mixed with the rat kidney microsome fraction, and incubated at 37 °C for 30 min. Then, the amount of [³H] cortisol was measured. Data are expressed as mean ± S.E. (n = 4) of the percentage relative to the amount of [³H] cortisol in the mixture without samples. ** p < 0.01 and *** p < 0.001 compared with the groups without the samples by Dunnett’s multiple t test for compounds 1–3, and by Student’s t test for GA

Fig. 5

Dot-blot analysis and ELISA detection range of GL metabolites using anti-3MGA-monoclonal antibody (mAb) [16]. GL metabolites (1 μg each) were spotted on a PES membrane, and colored using an anti-3MGA-mAb (a). Competitive ELISA using anti-3MGA-mAb for GL metabolites was performed. Data are represented from a single analysis (n = 1) (b). Competitive ELISA using an anti-3MGA-mAb for compound 3 and GA was performed. Data are represented as mean ± S.D. (n = 3). Two-way ANOVA indicated a significant difference between GA and compound 3 (F_1,36 = 4441, p < 0.001), concentrations (F_8,36 = 315, p < 0.001), and their interactions (F_8,36 = 281, p < 0.001). The calibrated line for the absorbance-concentration profile of compound 3 was calculated by the linear least-squares method, and the regression formula is shown in graph (c)

Fig. 6

Schematic diagram of the pharmacokinetics of glycyrrhizin (GL) and its metabolites after oral administration of GL in human. GA, glycyrrhetinic acid; Comp3, 18β-glycyrrhetyl-3-O-sulfate (3); MRP, multidrug resistance-related protein; OAT, organic anion transporter; 11βHSD2, type 2 11β-hydroxysteroid dehydrogenase