Metabolite Profiling, Pharmacokinetics, and In Vitro Glucuronidation of Icaritin in Rats by Ultra-Performance Liquid Chromatography Coupled with Mass Spectrometry

Abstract

Icaritin is a naturally bioactive flavonoid with several significant effects. This study aimed to clarify the metabolite profiling, pharmacokinetics, and glucuronidation of icaritin in rats. An ultra-performance liquid chromatography coupled with mass spectrometry (UPLC-MS) assay was developed and validated for qualitative and quantitative analysis of icaritin. Glucuronidation rates were determined by incubating icaritin with uridine diphosphate glucuronic acid- (UDPGA-) supplemented microsomes. Kinetic parameters were derived by appropriate model fitting. A total of 30 metabolites were identified or tentatively characterized in rat biosamples based on retention times and characteristic fragmentations, following proposed metabolic pathway which was summarized. Additionally, the pharmacokinetics parameters were investigated after oral administration of icaritin. Moreover, icaritin glucuronidation in rat liver microsomes was efficient with C[subscript]Lint[/subscript] (the intrinsic clearance) values of 1.12 and 1.56 mL/min/mg for icaritin-3-O-glucuronide and icaritin-7-O-glucuronide, respectively. Similarly, the C[subscript]Lint[/subscript] values of icaritin-3-O-glucuronide and icaritin-7-O-glucuronide in rat intestine microsomes (RIM) were 1.45 and 0.86 mL/min/mg, respectively. Taken altogether, dehydrogenation at isopentenyl group and glycosylation and glucuronidation at the aglycone were main biotransformation process in vivo. The general tendency was that icaritin was transformed to glucuronide conjugates to be excreted from rat organism. In conclusion, these results would improve our understanding of metabolic fate of icaritin in vivo.

Beibei Zhang 1 and Xiaoli Chen 1 and Rui Zhang 1 and Fangfang Zheng 1 and Shuzhang Du 1 and Xiaojian Zhang 1

Academic Editor:Filomena Conforti

Department of Pharmacy, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, Henan 450052, China

Received 5 April 2017; Accepted 23 May 2017; 10 July 2017

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

1. Introduction

Herba Epimedii, the dried aerial parts of Epimedium L. (Berberidaceae), are a widely used Chinese medicine for impotence, bone loss, and cardiovascular diseases [1-3]. Prenylflavonoids are reported to be a group of major active constituents present in Epimedium for the antioxidative stress, anti-inflammatory, antitumor, and antiosteoporosis activities [4-8]. Icaritin is the common aglycone with many biological effects, especially antiosteoporosis activities [5, 7]. Besides, icaritin could induce cell death in activated hepatic stellate cells through mitochondrial activated apoptosis and ameliorate the development of liver fibrosis in rats [9]. Meanwhile, icaritin is able to target androgen receptor and androgen receptor COOH-terminal truncated splice variants, to inhibit androgen receptor signaling and tumor growth with no apparent toxicity [10]. Additionally, icaritin has neuroprotective effects against MPP⁺ -induced toxicity in MES23.5 cells. IGF-I receptor mediated activation of PI3K/Akt and MEK/ERK1/2 pathways are involved in the neuroprotective effects of icaritin against MPP⁺ -induced neuronal damage [11]. Recently, icaritin had been shown as a potential agent for the treatment of systemic lupus erythematosus [12].

These biological activities above had stimulated increasing interests in the in vivo metabolism of icaritin or its related prenylflavonoids. Poor bioavailability of prenylated flavonoids results from their poor intrinsic permeation and transporter-mediated efflux by the human intestinal Caco-2 model and the perfused rat intestinal model [13]. Meanwhile, it is shown that Epimedium flavonoids could be hydrolyzed into secondary glycosides or aglycone by intestinal flora or enzymes, thereby enhancing their absorption and antiosteoporosis activity [14]. So far, numerous researches of total prenylflavonoids or individual flavonoid had been conducted in the fields of in vivo metabolites profiling, biliary excretion, and pharmacokinetics [15-19]. Generally, the in vivo metabolism of Herba Epimedii extracts or its prenylflavonoids could easily be metabolized in gastrointestinal tract following deglycosylation reaction. Additionally, icaritin was easily metabolized into glucuronidation conjugates to be preferentially eliminated and excreted from rat organism [16, 18, 20]. Though the data on metabolic researches of icaritin abounds, its metabolic profile is not so clear. It is essential to systematically characterize the in vivo metabolites in order to better understand its mechanism of action. Hence, the present study aimed to conduct the metabolites screening, quantitative determination, and in vitro glucuronidation of icaritin.

Recently, liquid chromatography coupled with mass spectrometry (LC-MS) had been widely introduced to rapidly screen trace components in biological samples [21, 22]. In this study, icaritin-related metabolites were analyzed based on characteristic fragmentation by UPLC-MS after oral administration. Meanwhile, possible disposing pathway of icaritin was proposed. Furthermore, a UPLC-MS method was developed and applied to perform the pharmacokinetics of icaritin. Moreover, glucuronidation rates were determined by incubating icaritin with uridine diphosphate glucuronic acid- (UDPGA-) supplemented rat liver microsomes (RLM) and rat intestine microsomes (RIM). Kinetic parameters were derived by appropriate model fitting. Icaritin was subjected to significant hepatic and gastrointestinal glucuronidation.

2. Materials and Methods

2.1. Materials

Icaritin, epimedin C, icariside I, icariside II, and desmethylicaritin (purity > 98%) were purchased from Nanjing Jingzhu Medical Technology Co., Ltd. Uridine diphosphate glucuronic acid (UDPGA), magnesium chloride (MgCl₂ ), alamethicin, D-saccharic-1, and 4-lactone were provided from Sigma-Aldrich (St. Louis, MO). Rat liver microsomes (RLM) and rat intestine microsomes (RIM) were prepared in our laboratory based on the protocol [21]. HPLC grade methanol and acetonitrile were purchased from Dikma Scientific and Technology Co., Ltd. All other chemicals were of analytical grade.

2.2. Animals

Male Sprague-Dawley rats (180~220) g were provided by Guangdong Medical Laboratory Animal Center. The rats were kept in an animal room at constant temperature (24±2)°C and humidity (60±5)% with 12 h of light/dark per day and free access to water and food. The animal protocols were approved and conducted in accordance with the guidelines of Laboratory Animal Ethics Committee of Zhengzhou University.

2.3. Samples Collection and Preparation for Qualitative Analysis

After the rats were fasted for 12 h with free access to water before experiments, icaritin dissolved in 0.3% sodium carboxymethyl cellulose solution was orally administrated to rats at a dose of 100 mg/kg. Blood samples were collected from external jugular vein into heparinized tubes and were separated by centrifuging at 13800g for 10 min at 4°C, respectively. Bile samples were collected and recorded during 0-24 h period after an abdominal incision anesthetized with 10% aqueous chloral hydrate. The urine and feces samples were collected separately during 0-24 h period after oral administration. Small intestinal samples were obtained after oral administration for 24 h. All blank samples were obtained in the same way.

Before experiments, all biosamples were stored at -20°C. In this work, solid phase extraction method was applied to pretreat all samples. Before use, C18 columns (3 cm³ , 60 mg) were first preconditioned and equilibrated with 3 mL of methanol and 3 mL of water, respectively. Urine samples were evaporated and concentrated at 40°C under reduced pressure. Feces samples and small intestinal samples were dried in air and stirred into powder. And then they were treated with an ultrasonic bath for 30 min. The filtrate was combined and evaporated to dryness at 40°C in vacuum. The residue was reconstituted with water. Plasma, urine, bile, feces, and small intestinal samples were loaded on pretreated columns. The residue was reconstituted in 200 μ L of 60% methanol and filtered through a 0.22 μ m membrane until injection.

2.4. Samples Preparation for Quantitative Analysis

Plasma sample (200 μ L) was treated with methanol (1.2 mL), after which the mixture was vortex-mixed for 30 s and centrifuged at 13800g for 10 min at 4°C. The supernatant was then transferred and evaporated to dryness using N₂ at room temperature. The residue was dissolved in 200 μ L of 60% methanol and was then injected into the UPLC-MS system.

2.5. Preparation of Standard Solutions

Blank rat plasma was spiked with standard working solutions to achieve final concentration of icaritin of 2.0, 4.0, 16.0, 64.0, 128.0, 256.0, and 512.0 ng/mL. All reference standard solutions were stored at 4°C until use.

2.6. Glucuronidation Assay

Icaritin was incubated with RLM and RIM to determine the rates of glucuronidation as published references previously [23]. Briefly, the incubation mixture mainly contained 50 mM Tris-hydrochloric acid buffer (pH = 7.4), 0.88 mM MgCl₂ , 22 μ g/mL alamethicin, 4.4 mM saccharolactone, and 3.5 mM UDPGA. The reaction was terminated by adding ice-cold acetonitrile. The samples were vortexed and centrifuged at 13800g for 10 min. The supernatant was subjected to UPLC-MS analysis. All experiments were performed in triplicate.

2.7. UPLC-MS Conditions

UPLC was performed using an ACQUITY(TM) UPLC system (Waters, Milford, MA, USA). Separation was achieved on a Waters BEH C18 column (1.7 μ m, 2.1 × 50 mm) maintained at 35°C. The mobile phase consisted of water (A) and acetonitrile (B) (both containing 0.1% formic acid), and the flow rate was 0.5 mL/min. The gradient elution program was as follows: 0 min, 15% B; 3 min 35% B; 7 min 60% B; 8 min 100% B. An aliquot of 4 μ L sample was then injected into the UPLC-MS system.

The UPLC system was coupled to a Waters Xevo TQD (Waters, Milford, MA, USA) with electrospray ionization. The operating parameters were as follows: capillary voltage, 2.5 kV (ESI+); sample cone voltage, 30.0 V; extraction cone voltage, 4.0 V; source temperature, 100°C; desolvation temperature, 300°C; and desolvation gas flow, 800 L/h. The method employed lock spray with leucine enkephalin (m/z 556.2771 in positive ion mode and m/z 554.2615 in negative ion mode) to ensure mass accuracy.

2.8. Pharmacokinetic Application

After fasting with free access to water for 12 h, icaritin was given to rats as a dosage of 100 mg/kg. Plasma samples were then obtained at 0.083, 0.25, 0.5, 1, 2, 3, 4, 6, 8, 12, 24, 36, and 48 h after administration. For pharmacokinetic application, DAS 2.0 was used to calculate the pivotal pharmacokinetic parameters.

2.9. Enzymes Kinetic Evaluation

Serial concentrations of icaritin (0.4~20 μ M) were incubated with RLM and RIM to determine icaritin glucuronidation rates. The kinetic models Michaelis-Menten equation and substrate inhibition equation were fitted to the data of metabolic rates versus substrate concentrations and displayed in (1) and (2), respectively. Appropriate models were selected by visual inspection of the Eadie-Hofstee plot [24]. Model fitting and parameter estimation were performed by Graphpad Prism V5 software (San Diego, CA).

The parameters were as follows. V is the formation rate of product. _Vmax is the maximal velocity. _Km is the Michaelis constant and [S] is the substrate concentration. _Ksi is the substrate inhibition constant. The intrinsic clearance (C_Lint ) was derived by _Vmax /_Km for Michaelis-Menten and substrate inhibition models. [figure omitted; refer to PDF]

3. Results

3.1. Fragment Pattern of Icaritin

As had already been reported in the previous study [16], besides the typical adduct ion [M+Na]⁺ at m/z 391.1151 (C₂₁ H₂₀ O₆ Na) and [M+H]⁺ at m/z 369.1336 (C₂₁ H₂₁ O₆ , -0.5 ppm), the ion at m/z 313.0714 (C₁₇ H₁₃ O₆ ) in positive ion mode was considered as the characteristic fragment ion (see Figure S1a in the Supplementary Material available online at https://doi.org/10.1155/2017/1073607).

3.2. Screening of Metabolites

On the basis of MS/MS fragmentation pattern, the metabolites were deduced, clarifying the general metabolism in vivo. The extracted ion chromatograms (EICs) of prototype (M0) and metabolites (M1~M30) were shown in Figure 1, while the individual EICs of M1~M30 were exhibited in Figure S2. The UV, MS, and MS/MS data of M0~M30 were all exhibited in Table 1.

Table 1: UPLC-MS analysis of icaritin and its observed metabolites in biosamples.

Number	tR	UV	[M + H]+	Error	Formula	(+) ESI-MS/MS	(-) ESI-MS/MS	Sources	Characterization
min	nm	ppm
M0#	7.20	272	369.1336	-0.5	C21 H20 O6	391.1152 369.1336 313.0714 215.0748	367.1180 352.0941 309.0417 297.0399 281.0443 253.0485	PUBFI	Icaritin
M1	2.48	270	721.1978	-0.3	C33 H36 O18	743.1882 721.1978 545.1666 369.1345 313.0717	719.1825 543.1521 367.1194	PUBI	Icaritin-di-O -gluA
M2	2.62	269	563.1768	0.5	C27 H30 O13	563.1768 387.1442 369.1332 313.0716	561.1608 385.1288	UBI	Hydrated icaritin-O -gluA
M3	2.89	345	531.1511	1.5	C26 H26 O12	531.1511 355.1166 299.0576	529.1348 353.1027 119.0504	UBI	Desmethylicaritin-O -gluA
M4	3.15	341	561.1617	1.6	C27 H28 O13	561.1617 385.1301 367.1186	559.1456 383.1175 367.1135 175.0263	UBI	Hydroxylated icaritin-O -gluA
M5	3.22	341	561.1613	0.9	C27 H28 O13	561.1613 385.1296	559.1458 383.1172 175.0262	UBI	Hydroxylated icaritin-O -gluA
M6	3.36	345	531.1503	0.0	C26 H26 O12	531.1503 355.1175 329.0741 314.0455	529.1345 353.1025 119.0506	UBI	Desmethylicaritin-O -gluA
M7	3.41	341	561.1605	-0.5	C27 H28 O13	561.1605 385.1295 313.0715	559.1460 383.1171	BI	Hydroxylated icaritin-O -gluA
M8	3.46	270	721.1982	0.3	C33 H36 O18	721.1982 545.1662 369.1340 313.0717	719.1823 543.1519 367.1196	UBI	Icaritin-di-O -gluA
M9	3.52	341	561.1608	0	C27 H28 O13	561.1608 385.1283	559.1454 383.1170 175.0260	BI	Hydroxylated icaritin-O -gluA
M10	3.58	269	563.1761	-0.7	C27 H30 O13	563.1761 387.1452 369.1350 313.0735	561.1605 385.1289	UBI	Hydrated icaritin-O -gluA
M11	3.74	341	561.1615	1.2	C27 H28 O13	561.1615 385.1292 367.1186	559.1462 383.1175	BI	Hydroxylated icaritin-O -gluA
M12	4.29	n.a.	543.1501	-0.4	C27 H26 O12	543.1501 367.1189	541.1343 365.1034 351.0765 175.0221 113.0253	I	Dehydrogenated icaritin-O -gluA
M13	4.34	271	545.1663	0.7	C27 H28 O12	567.1486 545.1663 369.1335 313.0716	543.1501 367.1188 352.0948 309.0408 297.0398 281.0498	PUBFI	Icaritin-O -gluA
M14#	4.40	270	823.3024	-0.1	C39 H50 O19	823.3024 369.1324 313.0721	821.2863 659.2331 367.1168 351.0867	I	Epimedin C
M15	4.45	273	387.1440	-1.0	C21 H22 O7	387.1440 369.1338 313.0722	385.1288	I	Hydrated icaritin
M16	4.49	270	823.3021	-0.5	C39 H50 O19	823.3021 369.1336 313.0725	821.2860 659.2333 367.1165 351.0869	I	Epimedin C isomer
M17#	4.53	270	531.1866	1.5	C27 H30 O11	531.1866 369.1350 313.0718	529.1705 367.1182 352.0948 297.0395 253.0496	FI	Icariside I
M18	4.57	271	545.1661	0.4	C27 H28 O12	545.1661 369.1346 313.0723	543.1505 367.1183 352.0945 297.0394	PUBFI	Icaritin-O -gluA
M19	4.59	271	647.2349	1.4	C32 H38 O14	669.2215 647.2349 515.1930 369.1346 313.0724	645.2168 367.1162 351.0863 323.0945 295.0595	I	Icaritin-rha-xyl
M20	4.61	270	661.2503	1.1	C33 H40 O14	683.2326 661.2503 369.1339 313.0719	659.2343 367.1165 351.0865 295.0602 217.0499	FI	Icaritin-rha-rha
M21#	4.88	269	515.1927	1.9	C27 H30 O10	537.1757 515.1927 369.1340 313.0726	513.1766 367.1157 352.0900 323.0910 295.0604 217.0498	FI	Icariside II
M22	4.93	300	543.1500	-0.6	C27 H26 O12	565.1453 543.1500 367.1172	541.1345 365.1033 175.0222 113.0254	BI	Dehydrogenated icaritin-O -gluA
M23	4.97	273	387.1445	0.3	C21 H22 O7	387.1445 369.1343 313.0715	385.1286	UFI	Hydrated icaritin
M24	5.02	300	543.1505	0.4	C27 H26 O12	543.1505 367.1134 313.0720	541.1346 365.1032 175.0223 113.0255	BI	Dehydrogenated icaritin-O -gluA
M25#	5.08	270	355.1180	-0.6	C20 H18 O6	355.1180 299.0582	353.1036 309.0409 297.0373 281.0481	UFI	Desmethylicaritin
M26	5.13	271	545.1664	0.9	C27 H28 O12	567.1540 545.1664 369.1343 313.0720 299.0585	543.1504 367.1180 352.0942 297.0391	BI	Icaritin-O -gluA
M27	5.37	268	385.1288	0.3	C21 H20 O7	407.1136 385.1288 367.1186 313.0715	383.1175 365.1035 175.0226	UFI	Hydroxylated icaritin
M28	5.54	268	385.1285	-0.5	C21 H20 O7	385.1285 367.1181 313.0719	383.1170 365.1046	UFI	Hydroxylated icaritin
M29	5.66	n.a.	385.1290	0.9	C21 H20 O7	385.1290 367.1176 313.0712	383.1172 367.1043 175.0262	UFI	Hydroxylated icaritin
M30	6.36	272	367.1180	-0.5	C21 H18 O6	367.1180 352.0948 313.0721 297.0395 253.0496	365.1047 351.0765 175.0223 113.0250	FI	Dehydrogenated icaritin

M0, parent drug; M1~M34, metabolites; n.a., not available; P, plasma; U, urine; B, bile; F, feces; I, intestine. gluA, glucuronide conjugates; glc, glucose; rha, rhamnose; xyl, xylose.

# means that the metabolites are exactly identified with reference standards.

Figure 1: EICs of all metabolites in rat biosamples after oral administration of icaritin. (a) Plasma; (b) urine; (c) bile; (d) feces; (e) intestine.

(a) [figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

(c) [figure omitted; refer to PDF]

(d) [figure omitted; refer to PDF]

(e) [figure omitted; refer to PDF]

3.3. Structure Elucidation of Metabolites

M0 (Parent Drug). M0 (7.20 min, C₂₁ H₂₀ O₆ , -0.5 ppm) in biological samples was unambiguously identified by comparing with references.

M15 and M23 (Hydration of Isopentene Group). Based on the [M+H]⁺ ion at m/z 387.1445 (C₂₁ H₂₃ O₇ , 0.3 ppm) and [M-H]^- ion at m/z 385.1286 (C₂₁ H₂₁ O₇ ), the molecular formula of M15 (4.34 min) and M23 (4.97 min) was determined as C₂₁ H₂₂ O₇ , with one H₂ O more than M0. The MS/MS spectrum of [M+H]⁺ ion (C₂₁ H₂₃ O₇ ) showed predominant [M+H-H₂ O]⁺ ion at m/z 369.1343 and 313.0715 (Figure S1b), which indicated that M15 and M23 were the hydration products at isopentene group of icaritin and agreed with previous study of icariin [18].

M25 (Demethylation of Flavonoid Aglycone) . According to the [M+H]⁺ ion at m/z 355.1180 (C₂₀ H₁₉ O₆ , -0.6 ppm) and [M-H]^- ion at m/z 353.1036 (C₂₀ H₁₇ O₆ ), the formula of M25 (5.08 min) was supposed as C₂₀ H₁₈ O₆ , with a methyl group less than M0. The MS/MS experiments (Figure S1c) showed a significant loss of neutral loss of C₄ H₈ (56.0626 Da) from the ion at m/z 355.1180 to 299.0582 in positive ion mode or from the ion at m/z 353.1036 to 297.0373 in negative ion mode. Meanwhile, the demethylation position was purposed at 4[variant prime] position of B ring of flavonoid aglycone. Moreover, M25 was identified as desmethylicaritin by comparison of reference standard.

M30 (Dehydrogenation of Isopentene Group) . The formula of M30 (6.36 min) was C₂₁ H₁₈ O₆ , with two hydrogens fewer than M0, based on the [M+H]⁺ ion at m/z 367.1180 (C₂₁ H₁₉ O₆ , -0.5 ppm) and [M-H]^- ion at m/z 365.1047 (C₂₁ H₁₇ O₆ ). In MS/MS spectrum (Figure S1d), the ions at m/z 352.0948 and 313.0721 were attributed to obvious loss of CH₃ (15.0235 Da) and C₄ H₆ (54.0470 Da) group, respectively, which indicated that the dehydrogenation position was at isopentene group [18].

M27~M29 (Hydroxylation of Isopentene Group) . From the [M+H]⁺ ion at m/z 385.1288 (C₂₁ H₂₁ O₇ , 0.3 ppm) and [M-H]^- ion at m/z 383.1175 (C₂₁ H₁₉ O₇ ), the formulae of M27 (5.37 min, _λmax 268 nm), M28 (5.54 min), and M29 (5.66 min) were speculated as C₂₁ H₂₀ O₇ , which was one oxygen more than M0. In (+) ESI-MS/MS spectrum (Figure S1e), the ion at m/z 385.1288 could lose a H₂ O and C₄ H₆ group to produce the daughter ions at 367.1186 ([M+H-H₂ O]⁺ ) and 313.0715 ([M+H-C₄ H₆ ]⁺ ), respectively. This illustrated that M27~M29 were tentatively characterized as the hydroxylated products of M0 at the isopentene group.

M14, M16, M17, M19, M20, and M21 (Glycosylation of Flavonoid Aglycone) . M21 (4.88 min) was given a [M+H]⁺ ion at m/z 515.1927 (C₂₇ H₃₁ O₁₀ , 1.9 ppm) and [M-H]^- ion at m/z 513.1766 (C₂₇ H₂₉ O₁₀ ) in full scan mass spectrum. The ion at m/z 515.1927 could easily yield the characteristic fragment ions at m/z 369.1340 and 313.0726 by subsequent loss of C₆ H₁₀ O₄ and C₄ H₈ (Figure S1f). So M21 could be the glycosylation product of M0 by adduct of rhamnose (C₆ H₁₀ O₄ , 146.0579 Da). Similarly, M17 (4.53 min, C₂₇ H₃₀ O₁₁ , 1.5 ppm) was the glucose conjugate of M0, while M19 (4.59 min, C₃₂ H₃₉ O₁₄ , 1.4 ppm) and M20 (4.61 min, C₃₃ H₄₁ O₁₄ , 1.1 ppm) were the xylose and rhamnose glycosylation derivates of M21, respectively. M14, M17, and M21 were identified as epimedin C, icariside I, and icariside II, respectively.

M14 (4.40 min, _λmax 270 nm, C₃₉ H₅₁ O₁₉ , -0.1 ppm) and M16 (4.49 min, _λmax 270 nm, C₃₉ H₅₁ O₁₉ , -0.5 ppm) both with the formula of C₃₉ H₅₀ O₁₉ (Figure S1g) were tentatively characterized as the glucose glycosylation conjugate of M20. These glycosylation reactions were the same as the metabolism of epimedin C in rats reported in reference (Liu et al., 2011). By comparing with references, M14, M17, and M21 were identified as epimedin C, icariside I, and icariside II, respectively. And the MS/MS spectra of M17, M19, and M20 were shown in Figures S1h-S1j, respectively.

M1, M8, M13, M18, and M26 (Glucuronidation of Flavonoid Aglycone) . In full scan mass spectrum, M13 (4.34 min), M18 (4.57 min), and M26 (5.10 min) all exhibited the [M+H]⁺ ion at m/z 545.1663 (C₂₇ H₂₉ O₁₂ , 0.7 ppm) and [M-H]^- ion at m/z 543.1501 (C₂₇ H₂₇ O₁₂ ) with a formula of C₂₇ H₂₈ O₁₂ of 176.0325 Da larger than M0. The MS/MS spectrum (Figure S1k) displayed an obvious loss of C₆ H₈ O₆ group from parent ion at m/z 545.1663 to the daughter ion at m/z 369.1338, which suggested an existing glucuronic acid of these three metabolites. Just like reported studies [16], monoglucuronide conjugate and diglucuronide conjugate were widely distributed in biological samples after oral administration of Epimedium-related total flavonoids or individual flavonoid. Therefore, M13, M18, and M26 were tentatively identified as monoglucuronidation conjugate of M0, while M1 (2.48 min, C₃₃ H₃₆ O₁₈ , -0.3 ppm) and M8 (3.45 min, C₃₃ H₃₆ O₁₈ , 0.3 ppm) (Figure S1l) were characterized as diglucuronidation derivates based on two molecules of C₆ H₈ O₆ fragment larger than M0.

Similarly, M2 (2.62 min, _λmax 269 nm, C₂₇ H₃₀ O₁₃ , 0.5 ppm) and M10 (3.57 min, _λmax 269 nm, C₂₇ H₃₀ O₁₃ , -0.7 ppm) with the MS/MS spectrum shown in Figure S1 m were tentatively considered as the monoglucuronidation products of M15 and M23. M3 (2.89 min, _λmax 345 nm, C₂₆ H₂₆ O₁₂ , 1.5 ppm) and M6 (3.36 min, _λmax 345 nm, C₂₆ H₂₆ O₁₂ , 0 ppm) were characterized as monoglucuronide conjugate of M25. The MS/MS spectrum of M3 and M6 was exhibited in Figure S1n. Meanwhile, M4 (3.16 min, _λmax 341 nm, C₂₇ H₂₈ O₁₃ , 1.6 ppm), M5 (3.21 min, _λmax 341 nm, C₂₇ H₂₈ O₁₃ , 0.9 ppm), M7 (3.40 min, _λmax 341 nm, C₂₇ H₂₈ O₁₃ , -0.5 ppm), M9 (3.50 min, _λmax 341 nm, C₂₇ H₂₈ O₁₃ , 0 ppm), and M11 (3.74 min, _λmax 341 nm, C₂₇ H₂₈ O₁₃ , 1.2 ppm) were regarded as monoglucuronidation derivates of M27~M29 and their MS/MS spectrum was displayed in Figure S1o. M12 (4.29 min, not available _λmax , C₂₇ H₂₆ O₁₂ , -0.4 ppm), M22 (4.94 min, _λmax 300 nm, C₂₇ H₂₆ O₁₂ , -0.6 ppm), and M24 (5.02 min, _λmax 300 nm, C₂₇ H₂₆ O₁₂ , 0.4 ppm) were tentatively identified as glucuronidation conjugates of M30. And their MS/MS spectrum was shown in Figure S1p.

3.4. Method Validation

The method was validated for specificity, linearity, extraction recovery, matrix effects, precision, accuracy, and stability according to the US Food Drug Administration guidelines for bioanalytical method validation [25].

Specificity was determined by comparing the chromatograms obtained for six blank plasma samples, blank plasma samples spiked with standard solutions at LLOQ concentrations, and drug plasma samples obtained 4 h after oral administration. As shown in Figure S3, no interference peaks were detected at the retention times of icaritin.

The LOD and LOQ were calculated as 3-fold and 10-fold of the ratio of signal-to-noise, respectively. The LLOQ was defined as the lowest concentration in the calibration curve with accuracy of 80~120% and precision of 20%. Calibration curves were acquired by plotting peak area (y) versus respective plasma concentrations (x) using a 1/^x2 weighting factor and linear least-squares regression analysis. A series of standard solutions were used to generate calibration curve. The correlation coefficients (^r2 ) of calibration curves were greater than 0.9926 within 2.0~512.0 ng/mL and LLOQ was 2.0 ng/mL. The regression equations, correlation coefficients, and LLOQ were shown in Table S1.

The experiments to evaluate matrix effect and recovery were conducted by the protocol [26]. According to the protocol, the peak areas from QC samples at three concentrations were defined as A1; those from extracted control plasma reconstituted with standard solutions at 4.0, 64.0, and 256.0 ng/mL were A2. The responses of icaritin found by direct injection of the corresponding pure reference standards at three QC levels were A3. The matrix effect and recovery were calculated as follows: matrix effect (%) = A2/A3×100%. Recovery (%) = A1/A2 × 100%. The results (as shown in Table S2) illustrated that matrix effect was between 89.1% and 113.5%, and the recovery was from 96.3% to 102.7%.

The accuracy and inter/intraday precision of the method were evaluated by determining six replicates of QC samples on three consecutive days. The measured concentrations of QC samples were determined with a calibration curve obtained on the same day. Relative error and relative standard deviation were used to describe accuracy and inter/intraday precision, respectively. They both should not exceed 15%. As exhibited in Table S3, the intraday and interday precision were less than 13.2% and 10.2%, respectively, while the intraday and interday precision of LLOQ were no more than 17.4% and 15.6%, respectively.

Stability of icaritin in rat plasma was assessed under different conditions at three concentration levels, including extracted samples for 12 h at room temperature, kept at -20°C for 60 h, three cycles of freezing at -20°C and thawing at 25°C, and plasma sample at room temperature for 8 h. Each was compared by three QC replicates of the same concentration with a calibration curve in the same day. The RE was within 13.8% and RSD was less than 11.3%. Stability results (Table S4) indicated that icaritin were stable under different storage conditions.

3.5. Pharmacokinetics Application

The mean concentration-time profiles of these bioactive components were shown in Figure 2. The main pharmacokinetic parameters were illustrated in Table 2. In this study, _Cmax was (294.5±22.7) ng/mL when _Tmax was (5.3±1.1) h after oral administration. The area under the concentration-time curve (_AUC0-∞ ) and mean residence time (MRT_0-∞ ) were (3145.0±302.3) ng·h/mL and (10.9±1.3) h, respectively. The results illustrated that icaritin had a poor absorption after oral administration. The reason may be that icaritin stepped into small intestine to undergo mass phase I and phase II metabolism by intestinal flora, especially the glycosylation and glucuronidation conjugates.

Table 2: Pharmacokinetic parameters of icaritin in rat plasma after oral administration.

Parameters	100 mg/kg
T m a x (h)	5.3 ± 1.1
C m a x (ng/mL)	294.5 ± 22.7
t 1 / 2 (h)	8.3 ± 1.0
A U C 0 - t (ng[sm middot]h/mL)	3048.5 ± 289.0
A U C 0 - ∞ (ng[sm middot]h/mL)	3145.0 ± 302.3
M R T 0 - t (h)	9.6 ± 1.1
M R T 0 - ∞ (h)	10.9 ± 1.3

Figure 2: Concentration-time curve of icaritin in rat plasma after oral administration.

[figure omitted; refer to PDF]

3.6. Glucuronidation of Icaritin in RLM and RIM

Due to lack of reference standard, quantification of icaritin glucuronide was based on the standard curve of the parent compound (icaritin) according to the assumption that parent compound and its glucuronide have closely similar UV absorbance maxima [27-29]. The detection wavelength of icaritin and icaritin glucuronides was 270 nm. The linear range of icaritin was 0.02~20 μ M, with LOD (S/N = 3~5) and LOQ (S/N = 8~10) of 0.01 and 0.02 μ M, respectively. And the acceptable linear correlation (Y=12149X) was confirmed by correlation coefficients (^r2 ) of 0.9994. The accuracy and precision of the intraday and interday error were both less than 3.4%. There were no matrix effects observed and no other sample preparation performed except those mentioned in the manuscript.

Kinetic profiling revealed that formation of icaritin-3-O -glucuronide (M13) and icaritin-7-O -glucuronide (M18) in RLM was well modeled by the substrate inhibition equation (Figure 3(a)), whereas they followed the classical Michaelis-Menten kinetics in RIM (Figure 3(b)). In contrast, the glucuronide formation of M13 (4.06 nmol/min/mg) and M18 (2.39 nmol/min/mg) in RLM was similar as well as M13 (11.88 nmol/min/mg) and M18 (8.23 nmol/min/mg) in RIM. Icaritin glucuronidation in RLM was efficient (C_Lint = 1.12 and 1.56 mL/min/mg for M13 and M18, resp.), following the substrate inhibition kinetics with _Km values of 3.62 and 1.53 μ M, respectively. Similarly, the C_Lint values of M13 and M18 in RIM were 1.446 and 0.861 mL/min/mg, respectively, whereas the _Km values of M13 and M18 in RIM in Michaelis-Menten model were 8.22 and 9.56 μ M, respectively. In addition, _Ki values of M13 and M18 in RLM were 11.31 and 17.07 μ M, respectively. The detailed parameters of M13 and M18 were listed in Table 3.

Table 3: Kinetic parameters of icaritin glucuronidation by RLM and RIM (mean ± SD).

Protein source	Metabolite	V m a x (nmol/min/mg)	K m (μ M)	K i (μ M)	C L i n t (mL/min/mg)	Model
RLM	M13	4.06 ± 0.70	3.62 ± 0.99	11.31 ± 3.51	1.12 ± 0.36	SI
M18	2.39 ± 0.26	1.53 ± 0.34	17.07 ± 4.38	1.56 ± 0.38	SI
RIM	M13	11.88 ± 0.60	8.22 ± 0.92	N.A.	1.45 ± 0.18	MM
M18	8.23 ± 0.63	9.56 ± 1.51	N.A.	0.86 ± 0.15	MM

Note. SI, substrate inhibition model; MM, Michaelis-Menten model; N.A., not available.

Figure 3: Kinetic profiles for glucuronidation of icaritin by various types of microsomes. (a) Pooled rat liver microsomes (RLM); (b) pooled rat intestine microsomes (RIM). In each panel, the insert figure showed the corresponding Eadie-Hofstee plot.

(a) [figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

4. Discussion

Normally, only the prototypes or metabolites in blood with a high enough exposure in target organs for a finite period of time are considered as potential effective components for therapeutic benefits [30]. In this study, M0 , M1 , and M13 were the main xenobiotics in plasma (Figure 1(a)), which may be the potential in vivo effective components directly. After circulation, M2 , M5 , M13 , M23 , and M28 were passed out with the urine (Figure 1(b)).

Due to poor oral bioavailability, several components were limited to be absorbed in blood. But they could influence intestinal dysfunction to exert efficacy by their prototypes, secondary metabolites, or finally the aglycone in intestinal tract [31]. Massive metabolites containing M6, M8, M13, M17, M25, M28, and M30 were detected in rat feces and small intestinal samples (Figure 1(d)). Moreover, icaritin underwent phase II metabolism by main conjugating enzymes including UDP-glucuronosyltransferases (UGTs) to produce extensive mono- or diglucuronic acid conjugates. In rat bile, M3, M6, M13, M18, and M24 mainly were biotransformed in rat liver and excreted into bile (Figure 1(c)).

Characterization of icaritin glucuronidation assumed a great role in the understanding of its pharmacokinetics and bioavailability. Oral bioavailability is a major factor in determining the biological actions of icaritin in vivo following oral administration of the compound [32]. This study suggested that the oral bioavailability of icaritin would be influenced by first-pass glucuronidation in the liver. The glucuronidation activity was obtained by kinetic profiling and modeling. Kinetic profiling required the determination of the rates of icaritin glucuronidation at a series of icaritin concentrations. The relative activities of RLM and RIM toward icaritin glucuronidation were evaluated by the derived C_Lint values (Table 3). Use of C_Lint (=_Vmax /_Km ) as an indicator of enzymes activity was advantageous, because (1) C_Lint represents the catalytic efficiency of the enzyme and is independent of the substrate concentration; (2) compared with other kinetic parameters such as _Km and _Vmax , C_Lint is more relevant in an attempt to predict hepatic clearance in vivo [33]. Therefore, C_Lint values were used to determine icaritin glucuronidation activity in this study.

Based on the metabolite profiles, the metabolic pathways of icaritin were proposed and shown in Figure 4(a), and the metabolic sites were shown in Figure 4(b). In summary, icaritin was hard to be absorbed into the rat blood. In small intestine, icaritin could form flavonoid glycoside by the sequential glycosylation metabolism. Meanwhile, icaritin could easily conjugate with a glucuronic acid to form phase II metabolites in liver, which indicated that the biliary clearance was one of the major routes of excretion. Phase I metabolism of icaritin mainly included demethylation, dehydrogenation, and hydration. The general tendency was that the saponins were metabolized and transformed into the high polar metabolites to be eliminated and excreted from the rat organism.

Figure 4: The proposed metabolic pathway (a) and metabolic sites (b) of icaritin in rats.

(a) [figure omitted; refer to PDF]

(b) [figure omitted; refer to PDF]

5. Conclusion

As a result, a total of 30 metabolites were identified or tentatively characterized based on the retention time behaviors and fragmentation patterns. Dehydrogenation at isopentenyl group and glycosylation and glucuronidation at the flavonoid aglycone were the main biotransformation process of icaritin in vivo. Meanwhile, a validated method was successfully applied to a pharmacokinetic study. Moreover, icaritin glucuronidation in RLM was efficient with C_Lint values of 1.12 and 1.56 mL/min/mg for M13 and M18, respectively. Similarly, the C_Lint values of M13 and M18 in RIM were 1.45 and 0.86 mL/min/mg, respectively. Taken altogether, this study could provide an experimental basis to understand the metabolic fate of icaritin in rat.

References

[1] Editorial Committee of Pharmacopoeia of Ministry of Health PR China, The Pharmacopeoia of People^' s Republic of China (Part 1) , Chemical Industry Press, Beijing, China, 2015.

[2] H. Ma, X. He, Y. Yang, M. Li, D. Hao, Z. Jia, "The genus Epimedium: an ethnopharmacological and phytochemical review,", Journal of Ethnopharmacology , vol. 134, no. 3, pp. 519-541, 2011.

[3] F. Xu, Y. Ding, Y. Guo, B. Liu, Z. Kou, W. Xiao, J. Zhu, "Anti-osteoporosis effect of Epimedium via an estrogen-like mechanism based on a system-level approach,", Journal of Ethnopharmacology , vol. 177, pp. 148-160, 2016.

[4] J. Zhou, J. Wu, X. Chen, "Icariin and its derivative, ICT, exert anti-inflammatory, anti-tumor effects, and modulate myeloid derived suppressive cells (MDSCs) functions,", International Immunopharmacology , vol. 11, no. 7, pp. 890-898, 2011.

[5] S.-H. Chen, M. Lei, X.-H. Xie, L.-Z. Zheng, D. Yao, X.-L. Wang, W. Li, Z. Zhao, A. Kong, D.-M. Xiao, D.-P. Wang, X.-H. Pan, Y.-X. Wang, L. Qin, "PLGA/TCP composite scaffold incorporating bioactive phytomolecule icaritin for enhancement of bone defect repair in rabbits,", Acta Biomaterialia , vol. 9, no. 5, pp. 6711-6722, 2013.

[6] X. Lai, Y. Ye, C. Sun, X. Huang, X. Tang, X. Zeng, P. Yin, Y. Zeng, "Icaritin exhibits anti-inflammatory effects in the mouse peritoneal macrophages and peritonitis model,", International Immunopharmacology , vol. 16, no. 1, pp. 41-49, 2013.

[7] S. Peng, G. Zhang, B.-T. Zhang, B. Guo, Y. He, A. J. Bakker, X. Pan, W. Zhen, L. Hung, L. Qin, W.-N. Leung, "The beneficial effect of Icaritin on osteoporotic bone is dependent on the treatment initiation timing in adult ovariectomized rats,", Bone , vol. 55, no. 1, pp. 230-240, 2013.

[8] S. W. Lei, G. Cui, G. P. H. Leung, S. C. W. Luk, M. P. M. Hoi, L. Wang, G. B. Mahady, S. M. Y. Lee, "Icaritin protects against oxidative stress-induced injury in cardiac H9c2 cells via Akt/Nrf2/HO-1 and calcium signalling pathways,", Journal of Functional Foods , vol. 18, pp. 213-223, 2015.

[9] J. Li, P. Liu, R. Zhang, L. Cao, H. Qian, J. Liao, W. Xu, M. Wu, Z. Yin, "Icaritin induces cell death in activated hepatic stellate cells through mitochondrial activated apoptosis and ameliorates the development of liver fibrosis in rats,", Journal of Ethnopharmacology , vol. 137, no. 1, pp. 714-723, 2011.

[10] F. Sun, I. R. Indran, Z. W. Zhang, M. H. E. Tan, Y. Li, Z. L. R. Lim, R. Hua, C. Yang, F.-F. Soon, J. Li, H. E. Xu, E. Cheung, E.-L. Yong, "A novel prostate cancer therapeutic strategy using icaritin-activated arylhydrocarbon-receptor to co-target androgen receptor and its splice variants,", Carcinogenesis , vol. 36, no. 7, pp. 757-768, 2015.

[11] M.-C. Jiang, X.-H. Chen, X. Zhao, X.-J. Zhang, W.-F. Chen, "Involvement of IGF-1 receptor signaling pathway in the neuroprotective effects of Icaritin against MPP+-induced toxicity in MES23.5 cells,", European Journal of Pharmacology , vol. 786, pp. 53-59, 2016.

[12] J. Liao, Y. Liu, H. Wu, M. Zhao, Y. Tan, D. Li, H. Long, Y. Dai, S. Yung, T.-M. Chan, Q. Lu, "The role of icaritin in regulating Foxp3/IL17a balance in systemic lupus erythematosus and its effects on the treatment of MRL/lpr mice,", Clinical Immunology , vol. 162, pp. 74-83, 2016.

[13] Y. Chen, Y. H. Zhao, X. B. Jia, M. Hu, "Intestinal absorption mechanisms of prenylated flavonoids present in the heat-processed Epimedium koreanum Nakai (Yin Yanghuo),", Pharmaceutical Research , vol. 25, no. 9, pp. 2190-2199, 2008.

[14] J. Zhou, Y. H. Ma, Z. Zhou, Y. Chen, Y. Wang, X. Gao, "Special section on drug metabolism and the microbiome intestinal absorption and metabolism of epimedium flavonoids in osteoporosis rats,", Drug Metabolism and Disposition , vol. 43, no. 10, pp. 1590-1600, 2015.

[15] Y.-T. Wu, C.-W. Lin, L.-C. Lin, A. W. Chiu, K.-K. Chen, T.-H. Tsai, "Analysis of biliary excretion of icariin in rats,", Journal of Agricultural and Food Chemistry , vol. 58, no. 18, pp. 9905-9911, 2010.

[16] H. Zhao, M. Fan, L. Fan, J. Sun, D. Guo, "Liquid chromatography-tandem mass spectrometry analysis of metabolites in rats after administration of prenylflavonoids from Epimediums,", Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences , vol. 878, no. 15-16, pp. 1113-1124, 2010.

[17] Q. Chang, G.-N. Wang, Y. Li, L. Zhang, C. You, Y. Zheng, "Oral absorption and excretion of icaritin, an aglycone and also active metabolite of prenylflavonoids from the Chinese medicine Herba Epimedii in rats,", Phytomedicine , vol. 19, no. 11, pp. 1024-1028, 2012.

[18] Q. Qian, S.-L. Li, E. Sun, K.-R. Zhang, X.-B. Tan, Y.-J. Wei, H.-W. Fan, L. Cui, X.-B. Jia, "Metabolite profiles of icariin in rat plasma by ultra-fast liquid chromatography coupled to triple-quadrupole/time-of-flight mass spectrometry,", Journal of Pharmaceutical and Biomedical Analysis , vol. 66, pp. 392-398, 2012.

[19] Y. Jin, C.-S. Wu, J.-L. Zhang, Y.-F. Li, "A new strategy for the discovery of epimedium metabolites using high-performance liquid chromatography with high resolution mass spectrometry,", Analytica Chimica Acta , vol. 768, no. 1, pp. 111-117, 2013.

[20] L. Cui, F. Xu, J. Jiang, E. Sun, Z. Zhang, X. Cheng, J. Wang, X. Jia, "Identification of metabolites of epimedin A in rats using UPLC/Q-TOF-MS,", Chromatographia , vol. 77, no. 17-18, pp. 1223-1234, 2014.

[21] H. Liu, H. Sun, D. Lu, Y. Zhang, X. Zhang, Z. Ma, B. Wu, "Identification of glucuronidation and biliary excretion as the main mechanisms for gossypol clearance: in vivo and in vitro evidence,", Xenobiotica , vol. 44, no. 8, pp. 696-707, 2014.

[22] F. Li, Y. Tan, H. Chen, Y. Yan, K. Zhai, D. Li, J. Kou, B. Yu, "Identification of schisandrin as a vascular endothelium protective component in YiQiFuMai powder injection using HUVECs binding and HPLC-DAD-Q-TOF-MS/MS analysis,", Journal of Pharmacological Sciences , vol. 129, no. 1, pp. 1-8, 2015.

[23] D. Lu, Z. Ma, T. Zhang, X. Zhang, B. Wu, "Metabolism of the anthelmintic drug niclosamide by cytochrome P450 enzymes and UDP-glucuronosyltransferases: metabolite elucidation and main contributions from CYP1A2 and UGT1A1,", Xenobiotica , vol. 46, no. 1, pp. 1-13, 2016.

[24] J. M. Hutzler, T. S. Tracy, "Atypical kinetic profiles in drug metabolism reactions,", Drug Metabolism and Disposition , vol. 30, no. 4, pp. 355-362, 2002.

[25] US FDA Guidance for Industry: Bioanalytical Method Validation, US Department of Health and Human Services, Center for Drug Evaluation and Research (CDER), http://www.fda.gov/cder/guidance, 2001

[26] B. K. Matuszewski, M. L. Constanzer, C. M. Chavez-Eng, "Matrix effect in quantitative LCMSMS analyses of biological fluids: a method for determination of finasteride in human plasma at picogram per milliliter concentrations,", Analytical Chemistry , vol. 70, no. 5, pp. 882-889, 1998.

[27] J. Troberg, E. Järvinen, G. Ge, L. Yang, M. Finel, "UGT1A10 Is a high activity and important extrahepatic enzyme: why has its role in intestinal glucuronidation been frequently underestimated?,", Molecular Pharmaceutics, Ahead of Print , 2016.

[28] D. Lu, H. Liu, W. Ye, Y. Wang, B. Wu, "Structure- and isoform-specific glucuronidation of six curcumin analogs,", Xenobiotica , pp. 1-10, 2016.

[29] H. Sun, H. Wang, H. Liu, X. Zhang, B. Wu, "Glucuronidation of capsaicin by liver microsomes and expressed UGT enzymes: reaction kinetics, contribution of individual enzymes and marked species differences,", Expert Opinion on Drug Metabolism and Toxicology , vol. 10, no. 10, pp. 1325-1336, 2014.

[30] X. Wang, "Studies on serum pharmacochemistry of traditional Chiese medicine,", World Science And Technology/Modernization of Traditional Chinese Medicine , vol. 4, no. 2, pp. 1-4, 2002.

[31] R. Zhang, S. Gilbert, X. Yao, "Natural compound methyl protodioscin protects against intestinal inflammation through modulation of intestinal immune responses,", Pharmacology Research Perspectives , vol. 3, article e00118, no. 2, 2015.

[32] B. Wu, K. Kulkarni, S. Basu, S. Zhang, M. Hu, "First-pass metabolism via UDP-glucuronosyltransferase: a barrier to oral bioavailability of phenolics,", Journal of Pharmaceutical Sciences , vol. 100, no. 9, pp. 3655-3681, 2011.

[33] B. Wu, D. Dong, M. Hu, S. Zhang, "Quantitative prediction of glucuronidation in humans using the in vitro-in vivo extrapolation approach,", Current Topics in Medicinal Chemistry , vol. 13, no. 11, pp. 1343-1352, 2013.