Endocrinol Japon 1991, 38 (2), 167-174

Effect of Oral Administration of Glycyrrhizin on the of

MING-FENGCHEN, FUMIESHIMADA, HIROMIKATO, SABUROYANO, ANDMATAO KANAOKA* First Departmentof Internal Medicine,and *ResearchInstitute for WAKAN-YAKU(Oriental Medicines), Toyama Medical and PharmaceuticalUniversity, Toyama 930-01, Japan

Abstract. The pharmacokinetics of total and free prednisolone (PSL) in six healthy men, with or without pretreatment with oral glycyrrhizin (GL), was investigated to confirm whether oral administra- tion of GL influences the metabolism of PSL in man. Each subject received an intravenous administration of 0.096 mg/kg of prednisolone hemisuccinate (PSL-HS) with or without pretreatment with 50 mg of oral GL four times. Blood samples were taken from a peripheral vein at 5, 10, 15, 30, 45 min and 1, 1.5, 2, 3, 4, 6, 8, 10, 12 and 24 h after the start of PSL-HS infusion. The concentrations of total PSL in plasma were analyzed by high-performance liquid chromatography, and the free PSL was measured by an isocolloidosmolar equilibrium dialysis method. The pharmacokinetic parameters of PSL were determined by non-compartment analysis. Oral administration of GL was found to significantly increase the concentrations of total PSL at 6, 8 h, and of free PSL at 4, 6 and 8 h after PSL-HS infusion. Moreover, oral administration of GL was also found to modify the pharmacokinetics of both total and free PSL. After oral administration of GL, the area under the curve (AUC) was significantly increased, the total plasma clearance (CL) was significantly decreased, and the mean residence time (MRT) was significantly prolonged. However, the volume of distribution (Vdss) showed no evident change. This suggests that oral administration of GL increases the plasma PSL concentrations and influences its pharmacokinetics by inhibiting its metabolism, but not by affecting its distribution. (Endocrinol Japon 38: 167-174, 1991)

IN OUR previous paper [I] it was shown that an important to clarify whether oral administration of intravenous infusion of Glycyrrhizin (GL) in- GL also increases the plasma concentration of PSL

creases the plasma concentrations of prednisolone or influences the pharmacokinetics of PSL in

(PSL) and influences the pharmacokinetics of humans. For this purpose, in this present work, we

prednisolone in man. However, the main compo- measured the plasma concentrations of total and nent in plasma is different when GL is adminis- free PSL and investigated the pharmacokinetics of tered by different routes. In intravenous adminis- PSL in six healthy men after intravenous infusion tration, the main component in plasma is GL [1, 2], of a low dose of prednisolone-hemisuccinate (PSL- whereas in oral administration the main compo- HS), with or without an oral dose of GL. nent is Glycyrrhetinic acid (GA), a metabolic derivative of GL [3, 4]. In vitro, the inhibitory

action of GA on the metabolism of Materials and Methods by 5 cr , 5ƒÀ-reductase and 11 ƒÀ-dehydrogenase is

stronger than that of GL [5-7]. Therefore, it is Subjects Six male subjects with an average age of 24.0 Received: October 30, 1990 Accepted: February 21, 1991 (range 20-33) years and with average weight of Correspondence to: Dr. Ming-Feng CHEN, First Department 58.2 (range 48-70) kg were included in this study. of Internal Medicine, Toyama Medical and Pharmaceutical All subjects had the content of the experiments University, 2630 Sugitani, Toyama 930-01, Japan. and the possible side-effects of PSL explained to 168 CHEN et al.

them before the study, and agreed to take part in ml/min. PSL and (an internal the clinical experiments. Physical examination, standard) were detected with a UV detector at 254 routine serum biochemical analysis and hematolo- nm. Under these conditions, the retention times gic analysis were carried out prior to the study. for dexamethasone, and PSL were 11.25, They were found to be free of abnormalities in 12.28 and 16.26 min, respectively. These retention these clinical and laboratory examinations, and times were different from those of endogenous had no history of illness which required long-term and other synthetic corticoids. As shown in Fig.1, medication within the previous 12 months. no interference was found between the peaks of dexamethasone, cortisol, and PSL. The recovery Experimental procedures of PSL was 82.4%. Each subject was injected with 0.096 mg/kg of PSL-HS (equivalent to 0.075 mg/kg of PSL, Equilibrium dialysis for free PSL Shionogi Pharmaceutical Co.) first as a control. On The free PSL percentages (F-P%) in plasma

another occasion, two weeks later, each one was samples were detected by the isocolloidosmolar

given 50 mg of an oral preparation of GL equilibrium dialysis method [8] described pre- (purchased from Minophagen Pharmaceutical viously [1]. In summary, each sample (0.5 ml of Co.) every 8 h, four times, and then injected with plasma) in a visking tube was dialyzed against a 3% the same dose of PSL-HS after the last intake of dextran Krebs Ringer solution, containing 10 the GL preparation. The PSL-HS was adminis- nCi/ml of [2, 4, 6, 7-3H] prednisolone with specific tered by bolus injection into a peripheral vein over activity of 1.68 TBq/mmol (Amersham). After 6 h a 3-minute period between 0830 h and 0900 h in of dialysis, the radioactivities in the outer dialysate the morning. Blood samples (7 ml) were taken and inner plasma were counted by scintillator. The from an indwelling cannula on the opposite arm F-P% and free PSL concentration (F-P concentra- before and at 5, 10, 15, 30, 45 min, and 1, 1.5, 2, 3, tion) were calculated by means of the formula of 4, 6, 8, 10, 12 and 24 h after the start of PSL-HS Matsui et al. [8]: infusion. The cannula was kept patent by flushing dpm in 50ƒÊl of dialysate/ with 0.4 ml of heparized saline (20 units/ml) after F-P%= ×100 dpm in 50ƒÊl of plasma each sample was taken. Plasma was separated by

centrifugation and stored at -20•Ž until analyzed. F-P concentration = Total PSL concentration•~F-P%•~100 High-performance liquid chromatography of PSL

Plasma samples were analyzed for total PSL by Pharmacokinetic analysis and statistical analysis high-performance liquid chromatography The pharmacokinetics of total and free PSL

(HPLC). The extraction method for PSL and were analyzed. The area under the plasma concen- HPLC for PSL was carried out as previously tration vs. time curve (AUC) was determined by

reported [1]. In summary, 0.5 ml of plasma sample means of the linear trapezoidal rule [9]. The was extracted with 6 ml of dichloromethane after clearance rate (CL) was calculated by dividing the

pretreatment with 0.5 ml of sulfuric acid and dose by the AUC. The volume of distribution in injected into an Extrelut precolumn. The organic the steady state (Vdss) was calculated from the eluent was then washed with 1 ml of 0.05 N same data by the noncompartment method [10].

sodium hydroxide, and evaporized with nitrogen The mean residence time (MRT) was determined

gas at 40•Ž. The residue after evaporization was by dividing the Vdss by the CL [11]. resolved with 200 ƒÊl of dichloromethane, 50-100 Statistical comparisons between pharmacokine-

μl of which was provided for analysis by HPLC. tic parameters were performed by means of Chromatography was carried out on a 4.6 mm Student's paired t-test. A statistically signficant

φ ×250mm Zorbax Silica Column (Shimadzu). difference was defined as P<0.05 and results were The mobile solution was an organic solution reported as the mean•}SD. separated from a dichloromethane/ethanol/ methanol/water (1000/50/5/12v/v) solution by placing it overnight in a separating funnel at room temperature. The HPLC had a flow rate of 1 EFFECT OF ORAL GLYCYRRHIZIN 169

A

B

Fig.1. High-performance liquid chromatograpy (HPLC) for prednisolone. A) HPLC for a pool plasma sample, taken from a healthy man at 1600 h. B) The same sample after the addition of 100 ng of cortisol and

prednisolone with 100 ng of dexamethasone as an internal standard. HPLC was done on a 4.6mm ƒÓ•~25 cm Zorbax Silica column with a mobile solution of dichloromethane/ethanol/ methanol / water (1000/ 50/5/12 v/v) at a flow rate of 1 ml/min, under 254 nm at 30•Ž.

Results percentage vs. time. The mean free PSL percen- tages in GL-pretreated group were higher than Plasma concentrations of total PSL those in the control group. A significant difference Figure 2 shows the plasma concentration vs. was found at 3, 4, 6 and 8 h after PSL-HS infusion. time of total PSL. Plasma PSL was detected by HPLC until 8 to 10 h after PSL-HS infusion. The Plasma concentrationsof free PSL mean total PSL concentrations increased to the Figure 4 shows the plasma concentration vs. peak during the first 15 min after PSL-HS infusion time for free PSL. The mean free PSL concentra- and decreased in a two-phase slope. The mean tions in the GL-pretreated group were higher than total PSL concentrations were higher in the GL- those in the control group. A significant difference pretreated group than in the control group. A was found at 4, 6 and 8 h after PSL-HS infusion. significant difference was found at 6 and 8 h after PSL-HS infusion. AUC (Area Under the Curve) Figure 5 shows the AUCs of total and free PSL Free PSL in plasma expressed as a percentage in each subject. After pretreatment with oral GL, Figure 3 shows the free PSL expressed as a the AUCs of both total and free PSL were 170 CHEN et al.

Fig. 3. Mean plasma percentage of free prednisolone (PSL). Mean plasma free PSL expressed as a percentage vs. Fig. 2. Mean plasma concentration of total prednisolone the time curve in all subjects (n=6), after intravenous (PSL). The mean plasma concentration vs. time curve administration of a dose of 0.096 mg/kg of predniso- for total PSL in all subjects (n=6), after intravenous lone hemisuccinate to each subject, with or without administration of a dose of 0.096 mg/kg of predniso- pretreatment with the oral administration of 50 mg of lone hemisuccinate to each subject, with or without glycyrrhizin (GL) four times. *: P<0.05; Statistical pretreatment with the oral administration of 50 mg of analysis was performed by Student's paired t-test. glycyrrhizin (GL) four times. *: P<0.05; Statistical analysis was performed by Student's paired t-test. 1).

significantly increased. The AUC of total PSL was increased from 591•}181 to 882•}202 ƒÊg•Eh/l. The Vdss (Volume of distribution in steady state)

AUC of free PSL was increased from 117•}27 to Figure 8 shows the Vdss of total and free PSL in 182•}50 ƒÊg•Eh/l (Table 1). each subject. No significant difference was found between the Vdss of total and free PSL for the

CL (Total Plasma Clearance) GL-pretreated group and the control group

Figure 6 shows the CLs of both total and free (Table 1). PSL in each subject. After pretreatment with oral

GL, the CLs of both total and free PSL were significantly decreased. The CL of total PSL was Discussion decreased from 0.13•}0.03 to 0.09•}0.02 l/kg•Eh. The CL of free PSL was decreased from We previously reported that an intravenous

0.64•}0.16 to 0.42•}0.10 l/kg•Eh (Table 1). administration of GL results in an increase in the plasma concentrations of PSL and influences the MRT (Mean Residence Time) pharmacokinetics of PSL in healthy men [1]. Figure 7 shows the MRTs of total and free PSL Ojima et al. [7] also reported that an intravenous in each subject. After pretreatment with oral GL, administration of GL prolonged the half-life of

the MRTs of both total and free PSL were PSL in patients during long-term treatment with PSL. However, the effect of oral GL on the prolonged significantly. The MRTs of both total and free PSL were prolonged significantly. The pharmacokinetics of PSL in humans is still un- MRTs of total PSL was prolonged from 4.71•}2.02 known. In this study, the oral administration of

to 7.81•}2.30 h. The MRT of free PSL was GL was found to significantly increase the concen- tration of total PSL at 6, 8 h, and the concentra- prolonged from 3.03•}1.01 to 4.82•}1.27 h (Table EFFECT OF ORAL GLYCYRRHIZIN 171

A B

Fig. 5. The AUC of prednisolone (PSL). The AUCs for total Fig. 4. Mean plasma concentration of free prednisolone (A) and free (B) prednisolone in all subjects, after (PSL). Mean plasma concentration vs. time curve of intravenous administration of a dose of 0.096 mg/kg of free PSL in all subjects (n=6), after intravenous prednisolone hemisuccinate to each subject, with or administration of a dose of 0.096 mg/kg of predniso- without pretreatment with oral administration of 50 lone hemisuccinate to each subject, with or without mg of glycyrrhizin (GL) four times. Statistical analysis pretreatment with the oral administration of 50 mg of was performed by Student's paired t-test. glycyrrhizin (GL) four times. *: P<0.05; Statistical analysis was performed by Student's paired t-test.

Table 1. Pharmacokinetic parameters of total and free pred- nisolone (PSL) with or without oral glycyrrhizin (GL)

A B Fig. 6. The CL of prednisolone (PSL). The CLs for total (A) and free (B) prednisolone in all subjects, after in- AUC: the area under the curve; CL: total plasma clearance travenous administration of a dose of 0.096 mg/kg of rate; Vdss: the volume of distribution in a steady state; MRT: prednisolone hemisuccinate to each subject, with or mean residence time; NS: no significant difference; *: without pretreatment with the oral administration of P<0.05; Values are expressed as the mean•}SD; Statistical 50 mg of glcyrrhizin (GL) four times. Statistical analysis was performed by Student's paired t-test. analysis was performed by Student's paired t-test. 172 CHEN et al.

tions of free PSL at 4, 6 and 8 h after the start of PSL-HS infusion. Pretreatment with oral GL was also found to influence the pharmacokinetics of both total and free PSL. The AUCs were signi- ficantly increased, the CLs were significantly de-

creased, and the MRTs were significantly pro- longed. Nevertheless, the Vdss did not show

evident change. Of total PSL, the mean AUC was increased from 591 to 882 ƒÊg•Eh/l, the mean MRT was increased from 4.71 h to 7.81 h, and the mean

CL was decreased from 0.13 to 0.09l/kg•Eh. Of free PSL, the mean AUC was increased from 117 to

182 ƒÊg•Eh/l, the mean MRT was increased from 3.03 to 4.82 h, and the mean CL was decreased from 0.64 to 0.42l/kg•Eh. These results suggest that

the oral administration of GL influenced the

pharmacokinetics of PSL by inhibiting its clear- ance rate, but not by affecting its distribution. A B There are three possible mechanisms which can

Fig. 7. The MRT of prednisolone (PSL). The MRTs for total decrease the plasma clearance rate of PSL [12]: (1) (A) and free (B) prednisolone in all subjects, after an increase in the binding of PSL to plasma intravenous administration of a dose of 0.096 mg/kg of binding proteins; (2) an inhibition of the metabol- prednisolone hemisuccinate to each subject, with or without pretreatment with the oral administration of ism of PSL by microsomal enzymes, largely in 50 mg of glcyrrhizin (GL) four times. Statistical and; (3) a decrease in the clearance of metabolized analysis was performed by Student's paired t-test or non-metabolized PSL by kidney. The oral administration of contraceptive steroids had been reported to decrease the metabolic clearance rate of PSL, partly due to the inhibition of metabolism

of PSL and a slight increase in the binding of PSL to -binding globulin [13]. A decrease in the clearance rate of PSL was also reported in

patients with liver cirrhosis [14] and renal failure [15]. In this study, the subjects were excluded from liver disease, renal disease, and drug admi-

nistration. Although we did not measure the

plasma concentrations of corticosteroid-binding

globulin (CBG) in subjects, the concentrations of total plasma protein, albumin, and total globulin

were found to be normal. We also measured the creatinine clearance rate (Ccr) for each subject in

the study day. No evident change in the Ccr was found. In a separate in vitro experiment (data not shown), various doses of GL or GA were added

A B into pooled plasma samples containing the fixed concentration of PSL, and the free percentages of Fig. 8. The Vdss of prednisolone (PSL). The Vdss for total (A) and free (B) prednisolone in all subjects, after PSL in pooled plasma samples were detected after intravenous administration of a dose of 0.096 mg/kg of 6 h of isocolloidal dialysis. The addition of GL or prednisolone hemisuccinate to each subject, with or GA did not alter the free percentages of PSL in without pretreatment with the oral administration of 50 mg of glcyrrhizin (GL) four times. NS: no signi- plasma even at high concentrations of GL (60 ficant difference: Statistical analysis was perfbrmed by μg/ml)and GA(10μg/ml).This indicates that Student's paired t-test. neither GL nor GA influences the binding of PSL EFFECT OF ORAL GLYCYRRHIZIN 173 to plasma binding proteins. The above results performed by unpaired Student's t-test and no indicate that oral administration of GL decreased significant difference was found. The dosage and the total plasma clearance of PSL neither by administration regimen still need to be elucidated. inhibiting the urine clearance of PSL by kidney The plasma concentrations of GL and GA were nor by interfering with the binding of PSL to also determined by HPLC and enzyme immuno- assay [19, 20]. Only GA but not GL was detected plasma binding proteins, but by inhibiting the metabolism of PSL by microsomal enzymes. The (data not shown) . The concentrations of GA direct evidence is that GL and GA can inhibit fluctuated between 0.2μg/ml and 0.9μg/ml,and corticosteroid 5ƒ¿- , 5ƒÀ-reductase, and 11 ƒÀ- reached a peak between 45 min and 1.5 h after the dehydrogenase activities in rat liver and kidney in last adminsitration of GL. This agrees with the vivo and in vitro [5-7, 16, 17]. results of our previous studies showing that GL

The free percentages of PSL in the GL- itself cannot be detected in plasma when it is orally administered [3, 4]. It seems that GL has been pretreated group were found to be higher than those in the control group. This is not due to the metabolized to GA before it appears in the influence of GL or GA on the binding of PSL to peripheral blood. The place where oral GL is metabolized to GA is still unknown. Hattori et al. plasma binding proteins, but due to another mechanism. Pickup et al. [18], reported that the [21], reported that GL was found to be metabol- free percentage of PSL increases with an increase ized to GA by human intestinal flora. In one of our in the dose, since the CBG, despite its high affinity other experiments [22], the concentrations of GL for PSL, has low capacity. This indicates that the and GA in the peripheral venous blood, abdomin- increase in plasma total PSL in the GL-pretreated al aortic blood, and portal venous blood were measured after rats were given an oral dose of GL. group might result in an increase in the free percentage of PSL. Another possible mechanism is Only GA but not GL was detected in all blood that PSL must become free before it is metabolized samples. This suggests that oral GL is metabolized by enzymes. Therefore, the influence of GL on the to GA in the intestines, but not in the liver. It is still metabolic clearance of PSL is more evident and unknown whether GL can be metabolized by liver faster in the free PSL than in the total PSL. This enzymes when the oral dose of GL is over the might also induce an increase in the free percen- metabolic ability of intestinal flora. Thus, the main tages of PSL in the GL-pretreated group. active component of GL when given orally is GA. In accordance with our previous study [1], an but not GL. intravenous adminstration of GL also resulted in In clinics, Ojima et al. [6] reported that the ora. an increase in the concentrations of total and free administration of GL potentiated the effects of

PSL. However, it influenced the pharmacokinetics PSL in some patients with rheumatoid arthritis or of PSL significantly in free PSL, but not signi- polyarteritis nodosa. This fact taken together witl- 且cantlyintotalPSLWhetherthepotentiating our results suggests that the oral adminsitration of action of oral GL on the effects of PSL is stronger GL can potentiate the effects of PSL by increasing than that of intravenous GL is still unknown, its plasma concentration through the inhibition of because the subjects in these two experiments were its metabolism. The effects of the combinatior different. Statistical analysis of the pharmacokine- treatment with PSL and GL are worthy of further tic parameters of PSL in these two studies was evaluation.

References

1. Chen MF, Shimada F, Kato H, Yano S, Kanaoka M Minophagen Med Rev Suppl 15: 3-11 (In Japanese). (1990) Effect of glycyrrhizin on the pharmacokine- 3. Nakada T, Kato H, Yano S, Kanaoka M (1986) tics of prednisolone following low dosage of pred- Pharmacokinetics of glycyrrhizin and glycyrrheti- nisolone hemisuccinate. Endocrinol Japon 37 (3): nic acid in healthy humans with long-term treat- 331-341. ment of glycyrrhizin preparations (I). J Med Pharm 2. Kato H, Nakanishi K, Kanaoka M (1984) The Soc WAKAN-YAKU 3: 278-279 (In Japanese). enzyme immunoassay of glycyrrhizin and glycy- 4. Shimada F, Chen MF, Kato H, Yano S, Kanaoka M rrhetinic acid and its' clinical application. (1989) Pharmacokinetics of glycyrrhizin and glycy- 174 CHEN et al.

rrhetinic acid in healthy humans with long-term (1986) Effect of liver function on the metabolism treatment of glycyrrhizin preparations (II). J Med of and prednisolone in humans. Gas- Pharm Soc WAKAN-YAKU 6 (3): 402-403 (In troenterol 90: 819-828.

Japanese). 15. Kawai S, Ichikawa Y, Homma M (1985) Difference 5. Tamura Y, Nishikawa T, Yamada K, Yamamoto in metabolic properties among cortisol, predniso- M, Kumagai A (1979) Effects of glycyrrhetinic acid lone, and dexamethasone in liver and renal dis- and its derivatives on •¢4-5ƒ¿- and 5 ƒÀ-reductase in ease: accelerated metabolism of dexamethasone in rat liver. Arzneim-Forsch, 29: 647-649. renal failure. J Glin Endocrinol and Metabol 60 (5): 6. Ojima M, Itoh N, Satoh K, Fukuchi S (1987) The 848-854. effects of glycyrrhizin preparations on patients 16. Kumagai A, Yano S, Otomo M, Takeuchi K (1957) with difficulty in release from steroids treatment. Study on the corticoid-like action of glycyrrhizin Minophagen Med Rev Suppl 17: 120-125 (In and the mechanism of its action. Endocrinol Japon 4

Japanese). (1): 17-27. 7. Ojima M, Satoh K, Gomibuchi, Itoh N, Kin S, 17. Monder C, Stewart PM, Lakshmi V, Valentino R, Fukuchi S (1990) The inhibitory effects of glycy- Burt D, Edward CRW (1989) Licorice inhibits rrhizin and glycyrrhetinic acid on the metabolism corticosteroid 11-dehydrogenase of rat kidney and of cortisol and prednisolone in in vivo and in vitro liver: in vivo and in vitro studies. Endocrinol 125 (2): studies. Folia Endocrinol 66: 584-589 (In Japanese). 1046-1053. 8. Matsui N, Yamamoto M, Seo H, Ninimi M, 18. Pickup ME, Lowe JR, Leatham PA, Rhind VM, Ariyoshi Y (1979) Determination of serum cortisol Wright V, Dowine WW (1977) Dose dependent fractions by isocolloidosmolar equilibrium dialysis. pharmacokinetics of prednisolone. Europ J Clin Endocrinol Japon 26 (2): 263-273. Pharmacol 12: 213-219. 9. Gibaldi M, Perrier D (1975) Pharmacokinetics. Vol. 19. Kanaoka M, Yano S, Kato H, Nakano N, Kinoshita 1, Mercel Dekker, Inc., New York, pp 293-296. E (1983) Studies on the enzyme immunoassay of 10. Benet LA, Galeazzi RL (1979) Noncompartmental bio-active constituents contained in oriental medi- determination of the steady-state volume of dis- cinal drugs. II. Enzyme immunoassay of glycyrrhi- tribution. J Pharm Sci 68: 1071-1074. zin. Chem Pharm Bull 31 (6): 1866-1873. 11. Oppenheimer JH, Schwartz HL, Surks MI (1975) 20. Kanaoka M, Yano S, Kato H, Nakada T, Kawa- Determination of common parameters of mura K (1988) Studies on the enzyme immunoas- iodothyronine metabolism and distribution in man say of bio-active constituents contained in oriental by noncompartmental analysis. J Glin Endocrinol medical drugs. IV. Enzyme immunoassay of glycy- Metabol 41: 319-331. rrhetinic acid. Chem Pharm Bull 36 (1): 8-14. 12. Frey FJ (1987) Kinetics and Dynamics of Predniso- 21. Hattori M, Sakamoto T, Kobashi K, Namba T lone. Endocrine Rev 8 (4): 453-473. (1983) Metabolism of Glycyrrhizin by human intes- 13. Frey BM, Schaad HJ, Frey FJ (1984) Pharmaco- tinal flora. Plant Medica 48: 38-42. kinetic interaction of contraceptive steroids with 22. Satoh M, Shimada F, Kato H, Yano S, Kanaoka M prednisone and prednisolone. Europ J Clin Pharma- (1990) Metabolic fate of Glycyrrhizin in the rat. J col 26: 505-511. Med Pharm Soc WAKAN-YAKU 7: 382-383 (In 14. Renner E, Horber FF, Jost G, Frey BM, Frey FJ Japanese).