Metabolite identification and pharmacokinetic study of Lamiophlomis rotata in rats

Abstract

An ultra-high performance liquid chromatography coupled with time-of-flight mass spectrometry technique and a subsequent LC-MS/MS method were developed for metabolite profile study of Lamiophlomis rotata extract after its oral administration. Metabolite profile analysis for plasma, bile, urine and feces were performed on a sub-2-μm analytical column, and showed that 39 prototypes and 47 metabolites were identified or tentatively characterized. More than half the metabolites underwent phase II reaction, including sulfate, glucuronide, taurine, glycine, and glutathione conjugation. The targeted and valid LC-MS/MS method, using one-step protein precipitation for rat plasma, was applied for the pharmacokinetic study of shanzhiside methyl ester, 8-O-acetylshanzhiside methyl ester, lamalbid, verbascoside, forsythoside B, and luteolin-7-O-β-D-glucopyranoside. Pharmacokinetics results showed that six bioactive components did not vary significantly in rats after administration of the aerial parts and the whole plant, which was consistent with the metabolite profile comparison between two groups. This article contributes to the understanding of the metabolism of L. rotata, and provides detailed performance analysis and scientific data to support the medicine parts alteration in the Chinese Pharmacopoeia.

---

Introduction

Lamiophlomis rotata (Benth.) Kudo (Duyiwei in Chinese), a famous Traditional Chinese Medicine (TCM) recorded in the Chinese Pharmacopoeia,^1,2 is commonly used in clinical practice for its variety of pharmaceutical effects, including pain relief, anti-inflammation, hemostasis, blood circulation promotion, and osteoporosis alleviation.³ The major bioactive compounds of L. rotata could be classified into three groups: iridoids (such as shanzhiside methyl ester and 8-O-acetylshanzhiside methyl ester), flavonoids (such as luteolin and luteolin-7-O-β-D-glucopyranoside), and phenylethanoid glycosides (such as forsythoside B and verbascoside).^4–9

In our previous report, ultra-high performance liquid chromatography coupled with time-of-flight mass spectrometry (LC-TOF/MS) method has been established to study the chemical constituents of L. rotata, revealing a total of 48 compounds,¹⁰ which delineated the chemical profile in vitro. As grown awareness that TCM metabolism and pharmacokinetic studies played as determinants of the TCM action in vivo, many modern research on TCMs and TCM preparations were engaged in these studies, to discover the phytochemical constituents necessary for therapeutic efficacy.¹¹ However, there was sparse data of its in vivo metabolism study, except one report which only investigated pharmacokinetics of four iridoids after rabbits following intravenous administration of iridoids extraction of L. rotata.¹² Thus, to clarify components “which were absorbed” (chemical constituents of the original TCM) and “which were produced” (bio-transformed metabolites) became the basis for metabolism and pharmacokinetics studies of L. rotata, which was the core of the plasma pharmacochemistry.^13–16 Moreover, both qualitative and quantitative analysis for L. rotata in vitro showed that the chemical profile for the aerial parts and the whole plant from different areas did not vary significantly,¹⁰ which approved the reasonable medicine parts alteration,^1,2,17 but whether in vivo metabolism evidence consistency for the two medicine parts was the same as the above conclusion was still unclear.

In this study, a LC-TOF/MS method was established to analyze L. rotata metabolism from different biological matrix including rat plasma, urine, feces and bile after oral administration. Besides, the pharmacokinetics of shanzhiside methyl ester, 8-O-acetylshanzhiside methyl ester, lamalbid, verbascoside, forsythoside B, and luteolin-7-O-β-D-glucopyranoside (Fig. 1) were studied by a fully validated LC-MS/MS method. Then, the qualitative and quantitative in vivo metabolism data were compared in rats after administration of the aerial parts and the whole plant.

Fig. 1 Chemical structures of the six bioactive marker compounds: shanzhiside methyl ester (1), 8-O-acetylshanzhiside methyl ester (2), lamalbid (3), verbascoside (4), forsythoside B (5), luteolin-7-O-β-D-glucopyranoside (6) and tinidazole (IS).

Materials and methods

Chemicals, reagents, drug preparation and animals

31 standard compounds (ESI, Table S1†) were isolated in our laboratory from L. rotata. All structures were confirmed by UV, MS, ¹H and ¹³C NMR spectral data with reported literature values (≥98%).

Acetonitrile (Merck, Darmstadt, Germany) and formic acid (Tedia, Fairfield, OH, USA) were of HPLC grade. Deionized water was prepared by a Milli-Q50SP Reagent Water System (Bedford, MA, USA). Other reagents were of analytical grade.

The aerial parts (0351A) and the whole plant (0351W) of L. rotata (Yushu, Qinghai) were extracted by 50% ethanol and the extract was prepared according to previous methods, and the amounts of the investigated six biomarkers were analyzed by the established HPLC-UV method to assess the quality homogeneity.¹⁰

Experiments on animals were approved by the Animal Ethics Committee of Second Military Medical University. Sprague-Dawley rats (200 ± 20 g), obtained from Shanghai SLAC laboratory animal Co. Ltd. (Shanghai, China), were housed for 5 days acclimation, with temperature of 20 ± 1 °C, relative humidity of 50 ± 10% and 12 h light–dark cycle, and fed with water and normal food ad libitum.

Biological samples collection and pretreatment for qualitative analysis

For metabolites elucidation, 15 male Sprague-Dawley rats (200 ± 20 g) were randomly divided into three groups: A1-aerial parts group (n = 6), B1-whole plant group (n = 6) and C1-control group (n = 3). For groups A1 and B1, a single dose of 0.3 g kg⁻¹ L. rotata extract which was dissolved in 0.5% CMC-Na solution (4 mL, equivalent to 3.0 g kg⁻¹ of crude drugs), was orally administrated to rats twice a day for two consecutive days. Rats in group C1 received normal saline (4 mL).

Blank blood, bile, urine and feces samples were collected from the control group. All samples were stored at −80 °C until analysis. The retro-orbital blood samples were withdrawn via the cannular into the heparinized tubes at 30 min post-dosing of L. rotata extract. Plasma was obtained by centrifuging for 10 min at 4000 × g at 4 °C. Urine and feces samples were collected from the metabolism cages during the 24–36 h after the first oral administration. Two rats in groups A1 and B1 were fixed on a wooden plate and anesthetized by intraperitoneal injection of 10% aqueous chloral hydrate after the last dose. An abdominal incision was made and common bile duct was cannulated with polyethylene tubes for consecutive collection of bile samples, which were collected for 15 min once they were secreted. Then the incision was closed by suture.

Each sample of plasma, bile and urine was mixed and vortexed with methanol for 30 s and centrifuged at 20 000 × g for 10 min at 4 °C. Plasma or bile (100 μL) was mixed with methanol (200 μL) with a ratio of 1 : 2 in volume, while urine (1.5 mL) was 1 : 1. The supernatant was evaporated to dryness by N₂ (gentle stream, 37 °C) and then the obtained residue was dissolved by 6% acetonitrile (100 μL) and centrifuged at 20 000 × g for 10 min. The subsequent supernatants were applied to the LC-TOF/MS analysis.

Feces (1 g) were extracted exhaustively with 20 mL methanol by ultrasonication, and 2 mL of the supernatant was evaporated to dryness by N₂. The dried residue was dissolved by 1 mL of 6% acetonitrile and centrifuged at 20 000 × g for 10 min. The obtained supernatants were applied to injection.

Plasma collection and pretreatment for quantitative analysis

For pharmacokinetic study, 15 rats were grouped into three groups. Rats in drug groups were orally administrated with L. rotata extract at a single dose of 0.8 g kg⁻¹: A2-aerial parts group (n = 6) and B2-whole plant group (n = 6). Rats in group C2 received normal saline (4 mL).

Blood samples (about 250 μL) were collected in 1.5 mL heparinized polythene tubes at 0.083, 0.25, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12 and 24 h after administration. Plasma was obtained by centrifuging for 10 min at 4000 × g at 4 °C. Methanol adding 0.1% formic acid and IS (tinidazole 200 ng mL⁻¹, 300 μL) was added to plasma samples (100 μL). The mixture was centrifuged at 13 000 rpm for 10 min after vortex-mixing for 3 min, and then the obtained supernatant (200 μL) was transferred into another tube and evaporated to dryness by N₂. The dried residue was reconstituted with 8% acetonitrile (400 μL) and the subsequent supernatants (10 μL) were applied to the LC-MS/MS analysis. Pharmacokinetic parameters were calculated using Drug and Statistics (DAS) 3.2.6 (2.0, Mathematical Pharmacology Professional Committee of China, Shanghai, China) using noncompartmental pharmacokinetic model. Significance of pharmacokinetic parameters between two groups was assessed by t-tests and P value less than 0.05 was considered as significant (SPSS 17.0 statistical software).

Standard solutions preparation for quantitative analysis

The standard stock solutions were prepared by dissolving six analytes in methanol and then diluted by 10% methanol at appropriate amount to obtain a series of standard working solutions. A 200 ng mL⁻¹ IS working solution was also prepared through diluting the stock solution by methanol with 0.1% formic acid. Calibration solutions were prepared by mixing the standard working solutions with blank rat plasma at appropriate amount: 1.00, 2.50, 7.50, 12.50, 25.00, 37.50, and 62.50 ng mL⁻¹ for lamalbid, 4.00, 10.00, 30.00, 50.00, 100.00, 150.00, and 250.00 ng mL⁻¹ for shanzhiside methyl ester, 8-O-acetylshanzhiside methyl ester and verbascoside, 3.50, 7.00, 25.00, 50.00, 100.00, 150.00, and 250.00 ng mL⁻¹ for forsythoside B, and 0.40, 1.00, 3.00, 5.00, 10.00, 15.00, and 25.00 ng mL⁻¹ for luteolin-7-O-β-D-glucopyranoside. Quality control (QC) samples were prepared in the same way: 2.25, 4.50, and 22.50 ng mL⁻¹ for lamalbid, 9.00, 18.00, and 90.00 ng mL⁻¹ for shanzhiside methyl ester and 8-O-acetylshanzhiside methyl ester, 8.00, 16.00, and 90.00 ng mL⁻¹ for verbascoside, 4.50, 9.00, and 45.00 ng mL⁻¹ for forsythoside B, and 1.00, 2.00, and 10.00 ng mL⁻¹ for luteolin-7-O-β-D-glucopyranoside. All samples were stored at −20 °C before analysis.

LC-MS method for qualitative analysis of L. rotata in rat biological samples

LC-TOF/MS analysis was carried out on an Agilent 1290 series UHPLC coupled with a 6538 TOF mass spectrometer equipped with an electrospray ionization interface (Agilent Technologies, Santa Clara, CA, USA). Waters ACQUITY TM HSS T3 C18 column (100 mm × 2.1 mm, 1.8 μm) was applied for separation: flow rate, 0.3 mL min⁻¹; column temperature, 35 °C; injection volume, 2 μL; post time, 5 min. Mobile phases were composed of acetonitrile (phase A) and 0.1% aqueous formic acid solution (phase B), with a gradient elution program: 6–13% A (0–14 min), 13–20% A (14–24 min), 20–31% A (24–28 min) and 31–95% A (28–30 min). Mass-spectrometric conditions were optimized as follows: drying gas flow 11.0 L min⁻¹, capillary voltage 4 kV, nebulizer pressure 40 psi, drying gas temperature 350 °C, sheath gas temperature 400 °C. Mass spectra were recorded in the range of m/z 100–1000 in both negative and positive ion modes.

LC-MS method for quantitative analysis of L. rotata in rat plasma

Assays were performed on an Agilent 1200 series HPLC and interfaced to a 6410 triple-quadrupole ESI-MS (Agilent Technologies, Santa Clara, CA, USA). Agilent Zorbax SB-C18 column (2.1 mm × 100 mm, 3.5 μm) was applied for separation: flow rate, 0.4 mL min⁻¹; column temperature, 35 °C; injection volume, 10 μL; post time, 5 min. Mobile phases were composed of acetonitrile with 0.1% formic acid (phase C) and 0.1% aqueous formic acid solution (phase D), with a gradient elution program: 8–40% C (0–1 min), 40–95% C (1–4 min) 95% C (4–5 min). Multiple reaction monitoring (MRM) in the negative mode with tinidazole as internal standard (IS) was applied for compounds analysis (Table 1). Parameters for detection were optimized as follows: drying gas flow 10 L min⁻¹, capillary voltage 4.0 kV, nebulizer pressure 40 psi, gas temperature 350 °C, and collision-induced dissociation (CID) gas pressure 0.1 MPa. All data were acquired and analyzed using Agilent 6410 Quantitative Analysis version B.01.02 analyst data processing software (Agilent Corporation, MA, USA).

Table 1 Optimized MRM parameters for detected compounds

Compound	Mass	Prec ion (m/z)	Prod ion (m/z)	Frag (V)	CE (eV)
Shanzhiside methyl ester	406.1475	451.3	243.2	125	15
8-O-Acetylshanzhiside methyl ester	448.1581	493.2	225.1	120	18
Lamalbid	422.1424	467.2	259.1	140	14
Verbascoside	624.2054	623.2	161.1	150	38
Forsythoside B	756.2477	755.2	161.1	140	44
Luteolin-7-O-β-D-glucopyranoside	448.1006	447.1	285.2	190	30
Tinidazole (IS)	246.0674	246.0	125.9	80	4

---

The developed method was validated for selectivity, linearity, sensitivity, precision, accuracy, matrix effects, recovery and stability according to the US Food and Drug Administration (FDA) bioanalytical method validation guidance.¹⁸ Selectivity was tested by comparison of six individual blank rat plasma samples with corresponding spiked plasma samples with the analytes and IS. Calibration curves were presented as y = ax + b, where y was the peak area ratios (analyte/IS) and x was the analytes concentration prepared (seven different levels) using weighed (1/x²) least squares regression analysis. The lowest concentration of the calibration curve was determined as the lower limit quantification (LLOQ), with a maximum relative standard deviation (RSD, viz. precision) and relative error (RE, viz. accuracy) less than 20%. Accuracy and precision were evaluated in six replicates analyses for analytes and IS at three QC levels on three consecutive days. Matrix effect was assessed by comparing the analytes peak areas in the standard solutions dissolved in the pretreated bland rat plasma with those in the 10% methanol at the same concentration. Extraction recovery was calculated by comparing the analytes peak areas in the pre- with post-extracted samples. Stability was investigated under four storage conditions, respectively: 6 h at autosampler at room temperature; after three frozen (−20 °C)–thaw (room temperature, spontaneously) cycles; after storage at −20 °C for one month. Data was obtained from comparing the detected concentrations with the original analyte concentrations of QC samples.

Result and discussion

Qualitative analysis

Optimization of LC-TOF/MS conditions. In order to get comprehensive metabolism information of L. rotata extract from different biological samples, it is necessary to develop effective and reliable analytical methods for detection. Iridoids, phenylethanoid glycosides, and flavonoids were considered as the main constituents of L. rotata, thus, 31 standard compounds from it (ESI, Table S1†) were used for optimizing the HPLC-MS conditions. To separate and determine these compounds in a single run, a gradient elution method within a relative long period was developed. The ACN–water system showed more powerful separation ability, less interference in MS, and elution power for compounds investigation both in TCMs and its metabolites after administration, when compared with methanol–water system.^19,20 Moreover, 0.1% acetic acid was added to the aqueous phase, which usually has the function of enhancing the intensity of compounds with hydroxyl groups. As a result, the acetonitrile 0.1% aqueous formic acid system with optimized gradient elution was applied in both positive and negative ion scan modes. UPLC was developed largely by using sub-2-μm analytical column, which shortened the analysis time and increased the peak resolution, capacity and sensitivity, owing to the containing particles with a diameter of sub-2 μm.^20–23 Therefore, waters ACQUITY UPLC HSS T3 column (1.8 μm, 2.1 mm × 100 mm) was chosen for qualitative analysis.

Identification of multiple constituents in rat biological samples. For qualitative analysis, L. rotata was given to the rat consecutively during two days, in a relative low dose, to identify more absorbed and produced compounds. As 48 compounds in six classes were identified or tentatively characterized in L. rotata in the previous work,¹⁰ herein 39 of them were identified from the rat biological samples (ESI, Fig. S1†). In these absorbable constituents, 17 were iridoids (peaks 2–7, 10, 13, 16–21, 24, 34 and 36), eight were flavonoids (peaks 25, 26, 29, 30, 32, 33, 35 and 37), six were phenylethanoid glycosides (peaks 8, 14, 22, 27, 28 and 31), one was nortriterpenoid (peak 39), two were phenolic compounds (peaks 9 and 15), and five were glucide compounds (peaks 1, 11, 12, 23 and 38). 32 constituents were confirmed by comparison of the standard samples, and the remaining constituents by the fragment ions in MS2 spectra upon CID at different energies (10, 25 and 40 V) and the comparison of literature data (ESI, Table S1†). Among 39 prototype components, most (35 compounds) could be detected in plasma, 30 were from urine, 31 were from feces, and only 11 were from bile. Those compounds which were in relative high concentrations in L. rotata extract were found in bile, including iridoids (peaks 3, 4, 6, 7, 17 and 24), phenylethanoid glycosides (peaks 27 and 31), and flavonoid (peak 29).

The other peaks (M1–M47), which were found to exist only in the drug-containing rat biological samples but not in the extract, would be the possible metabolites of the L. rotata constituents. According to the origins of parent compounds and structural types, all these metabolites were categorized as iridoids-, phenylethanoid glycosides-, flavonoids-, norisoprenoid-, phenolic-, and glucide-related (Table 2). 47 metabolites were detected mainly including sulfate, glucuronide and taurine conjugates. TOF/MS analysis gave high-accuracy [M − H]⁻ ions for most of these metabolites. Their structures were identified as followed, in accordance with the drug metabolism rules.²⁴

Table 2 Identification of metabolites from L. rotata in biological samples by UPLC/Q-TOF-MS^a

Peak no.	RT (min)	Formula	Precursor ions	Selected mode	Metabolite name	MS/MS fragment ions	Class	Source
								P	U	F	B
a I – iridoids, F – flavonoids, PG – phenylethanoid glycosides, N – nortriterpenoid, P – phenolic compounds and G – glycoside compounds; P-plasma, U-urine, F-feces and B-bile; M-metabolite.
M1	1.580	C17H29NO9	392.1909	[M + H]^+	Glutathione conjugation	262.6784[M + H–GSH]	PG-M	✓	✓	✓
M2	4.046	C27H40O18	651.2101	[M − H]^−	Glucuronide conjugation	476.1887[M − H–GlcUA]	PG-M		✓	✓
M3	5.372	C34H44O22S	835.1976	[M − H]^−	Sulfation conjugation	755.2475[M − H–SO3]	PG-M		✓	✓
					Hydroxylation	593.2120[M − H–Glc]
M4	5.686	C29H36O18S	703.1546	[M − H]^−	Sulfation conjugation	623.1983[M − H–SO3]	PG-M		✓	✓
					Hydroxylation	461.1672[M − H–Glc]
M5	8.775	C40H52O25	931.2727	[M − H]^−	Glucuronide conjugation	755.2388[M − H–GlcUA]	PG-M		✓	✓
					Hydroxylation	593.2100[M − H–Glc]
M6	9.813	C35H44O21	799.2306	[M − H]^−	Glucuronide conjugation	623.1914[M − H–GlcUA]	PG-M	✓	✓
M7	14.337	C31H41NO17S	730.2045	[M − H]^−	Taurine conjugation	622.9417[M − H–Tau]	PG-M		✓
M8	1.314	C17H26O15S	501.094	[M − H]^−	Sulfation conjugation	421.1834[M − H–SO3]	I-M		✓
M9	1.330	C23H32O17	581.1733	[M + H]^+	Glucuronide conjugation	405.0991[M + H–GlcUA]	I-M	✓	✓	✓
M10	2.184	C17H24O14S	483.0828	[M − H]^−	Sulfation conjugation	403.0824[M − H–SO3]	I-M	✓	✓	✓
M11	2.386	C17H20O13S	467.0837	[M − H]^−	Sulfation conjugation	388.0905[M − H–SO3]	I-M		✓
M12	2.386	C17H20O13S	467.0841	[M − H]^−	Sulfation conjugation	388.0905[M − H–SO3]	I-M		✓
M13	2.841	C17H24O14S	483.0823	[M − H]^−	Sulfation conjugation	403.2533[M − H–SO3]	I-M		✓	✓
M14	2.880	C20H30O15S	543.1384	[M + H]^+	Sulfation conjugation	463.0844[M + H–SO3]	I-M		✓	✓
M15	3.020	C26H38O18	639.2118	[M + H]^+	Glucuronide conjugation	463.0845[M + H–GlcUA]	I-M	✓	✓	✓
						287.0491[M + H–GlcUA-GlcUA]
M16	3.134	C17H24O15S	499.0773	[M − H]^−	Sulfation conjugation	419.9153[M − H–SO3]	I-M			✓
M17	3.138	C23H34O16	565.1771	[M − H]^−	Glucuronide conjugation	389.0320[M − H–GlcUA]	I-M	✓	✓		✓
M18	3.525	C19H29NO12S	540.1364	[M + HCOO]^−	—	349.0853, 317.2356, 293.2886	I-M		✓
M19	3.630	C23H34O17	583.1872	[M + H]^+	Glucuronide conjugation	406.5946[M + H–GlcUA]	I-M		✓	✓	✓
M20	3.868	C23H32O17	579.1359	[M − H]^−	Glucuronide conjugation	403.0996[M − H–GlcUA]	I-M		✓
M21	4.743	C17H25ClO14S	519.0602	[M − H]^−	Sulfation conjugation	438.6176[M − H–SO3]	I-M		✓
M22	5.623	C23H32O18	595.1502	[M − H]^−	Glucuronide conjugation	421.1166[M − H–GlcUA]	I-M		✓	✓
M23	5.854	C22H31NO15	548.1656	[M − H]^−	Glutathione conjugation	420.2198[M − H–GSH]	I-M		✓
M24	6.970	C23H34O18	597.1666	[M − H]^−	Glucuronide conjugation	421.1990[M − H–GlcUA]	I-M		✓
M25	7.002	C23H32O16	563.1640	[M − H]^−	Glucuronide conjugation	387.0956[M − H–GlcUA]	I-M			✓
						211.0351[M − H–GlcUA–GlcUA]
M26	7.510	C25H36O18	623.1257	[M − H]^−	Glucuronide conjugation	487.2290[M − H–GlcUA]	I-M		✓
M27	9.225	C22H31NO12S	532.1684	[M − H]^−	Hydroxylation	442.9396[M − H + H2O–Tau]	I-M		✓
M28	1.512	C12H23NO7	292.1414	[M − H]^−	Glycine conjugation	236.1141[M − H–Gly]	G-M	✓	✓
M29	2.390	C16H25NO10S	424.1996	[M + H]^+	Methylation	409.1213[M + H–CH3]	G-M		✓
M30	3.600	C10H20O9S	315.0734	[M + H]^+	Sulfation conjugation	237.1216[M + H–SO3]	G-M	✓	✓
M31	3.920	C16H28O10S	411.1343	[M − H]^−	Sulfation conjugation	331.3911[M − H–SO3]	G-M		✓
					Glucuronide conjugation	235.1557[M − H–GlcUA]
M32	5.847	C22H36O13	507.209	[M − H]^−	Glucuronide conjugation	331.1453[M − H–GlcUA]	G-M		✓
M33	3.860	C21H18O11	447.0949	[M + H]^+	Glucuronide conjugation	270.8024[M + H–GlcUA]	F-M	✓	✓
M34	4.760	C28H29NO15	620.1631	[M + H]^+	Glutathione conjugation	489.1439[M + H–GSH]	F-M			✓
M35	5.878	C22H30O13	501.094	[M − H]^−	Decarbonation	457.1026[M − H–CO2]	F-M	✓	✓
					Glucuronide conjugation	325.1548[M − H–GlcUA]
M36	6.566	C26H27NO13	560.1423	[M − H]^−	Glutathione hydroxylation	432.6875[M − H–GSH]	F-M			✓
M37	7.579	C26H28O18S	659.0925	[M − H]^−	Sulfation conjugation	579.2313[M − H–SO3]	F-M	✓	✓
M38	9.320	C32H36O21	755.1686	[M − H]^−	Glucuronide conjugation	579.2633[M − H–GlcUA]	F-M		✓
M39	9.350	C27H28O17	623.1257	[M − H]^−	Hydroxylation	462.1888[M − H–Glc]	F-M	✓	✓		✓
M40	10.510	C21H20O14S	527.0503	[M − H]^−	Sulfation conjugation	447.2047[M − H–SO3]	F-M	✓	✓		✓
M41	12.220	C28H36O19S	707.1523	[M − H]^−	Sulfation conjugation	627.2069[M − H–SO3]	F-M		✓
M42	13.044	C23H25NO12S	584.107	[M + HCOO]^−	Taurine conjugation	431.4099[M + HCOO–Tau]	F-M	✓	✓	✓
M43	14.970	C21H18O12	463.0874	[M + H]^+	Glucuronide conjugation	287.1385[M + H–GlcUA]	F-M			✓	✓
M44	15.423	C15H10O8S	349.0033	[M − H]^−	Sulfation conjugation	269.0487[M − H–SO3]	F-M		✓
M45	9.248	C29H46O7S	537.2894	[M − H]^−	Sulfation conjugation	457.3324[M − H–SO3]	N-M	✓	✓
M46	26.355	C35H54O10	633.3654	[M − H]^−	Glucuronide conjugation	456.4401[M − H–GlcUA]	N-M		✓		✓
M47	28.635	C31H51NO6S	564.3319	[M − H]^−	Taurine conjugation	504.3043[M − H–C2H4N–O]	N-M	✓

---

A total of seven peaks detected in rat urine were identified or tentatively characterized as phenylethanoid glycosides-related metabolites (M1–M7). Among them, two and five were detected in plasma and feces, respectively. M5 and M6 were assigned as glucuronide conjugates owing to the presence of characteristic neutral loss of 176 Da and corresponding [M − H]⁻ ions in MS/MS spectra. Based on the similar rule, M3 and M4 were identified as sulfate conjugates (adding 80 Da), M1 and M7 were identified as glutathione (adding 129 Da) and taurine (adding 107 Da) conjugates, respectively. Details were presented in Table 3.

Table 3 Matrix effects, recoveries, intra- and inter-day precisions and stability of six analytes

Compound	Spiked (ng mL^−1)	Matrix effect (%)		Extraction recovery (%)		Intra-day (n = 6) (%)		Inter-day (n = 6) (%)		Stability (RE^a, %)
		Mean	RSD	Mean	RSD	RE	precision	RE	precision	Three freeze–thaw	Short-term	Long-term
a RE is expressed as (measured concentration/spiked concentration − 1) × 100%.
Acetylshanzhiside methyl ester	9.00	92.8	7.6	95.9	8.6	7.0	7.9	4.5	10.3	1.9	9.7	9.0
	18.00	90.1	6.3	80.5	3.0	−5.4	3.8	−1.1	7.2	−2.4	7.8	4.9
	90.0	85.9	1.2	67.0	1.8	3.0	3.8	9.4	6.0	2.8	5.5	2.4
8-O-Acetylshanzhiside methyl ester	9.00	90.3	17.9	90.9	16.5	21.2	14.7	13.6	16.2	−18.8	0.7	−18.0
	18.00	94.1	9.1	84.2	7.3	0.0	10.4	−1.1	10.0	9.5	−0.3	8.7
	90.00	91.3	6.8	68.8	3.4	2.1	5.3	7.2	6.8	14.1	4.1	0.7
Lamalbid	2.25	89.0	9.3	97.4	7.5	7.4	8.4	1.6	13.7	−4.1	−1.7	−8.0
	4.50	96.7	5.7	77.1	8.8	−1.0	10.7	1.3	11.0	5.3	−4.4	−1.4
	22.50	86.9	2.9	69.0	1.5	0.4	2.9	8.2	6.9	6.5	4.6	8.2
Verbascoside	4.50	84.3	9.1	58.2	12.3	−5.5	9.9	−10.9	14.7	−11.3	8.1	7.3
	9.00	92.1	1.8	81.3	6.2	−2.2	6.0	−17.8	14.8	−5.0	−10.5	−11.4
	45.00	84.3	1.4	60.8	1.5	−1.5	4.2	10.5	9.3	3.7	1.8	−2.5
Forsythoside B	8.00	82.4	8.0	62.5	11.8	0.8	9.6	−4.5	15.5	−8.8	5.7	1.8
	16.00	94.2	3.4	73.4	3.8	0.3	3.4	−13.1	12.4	0.9	−4.1	−9.4
	90.00	80.9	3.1	59.8	3.0	−4.0	4.8	9.5	10.1	1.6	2.0	−3.7
Luteolin-7-O-β-D-glucopyranoside	1.00	84.8	7.9	85.2	5.8	7.8	5.6	−3.8	13.3	−5.3	6.3	−2.7
	2.00	97.0	5.4	81.8	9.2	−6.4	9.4	−7.9	8.5	9.3	−1.0	1.6
	10.00	91.0	2.0	68.3	3.1	−3.8	3.5	−0.7	3.7	8.9	4.2	4.4

---

Among 20 iridoids metabolites (M8–M27), most of them (18 compounds) could be were detected in urine, four were detected in plasma, nine were detected in feces and two were detected in bile. Peaks M9, M15, M17, M19, M20, M22 and M24–M26 were assigned as glucuronide conjugates due to the diagnostic neutral loss of 176 Da. Peaks M8, M10–M14, M16 and M21 were sulfate conjugates as their [M − H]⁻ ions could further eliminate SO₃ unit (80 Da). Peak M23 was found as glutathione conjugation.

Although amino acid conjugation reactions were less frequently for drug detoxification, products glycine (adding 57 Da) conjugation was found as M28 for glucide metabolites. The other four glucide metabolites in rat urine were tentatively characterized as sulfate (M30 and M31) and glucuronide (M32) conjugates, and methylation product (M29). Similarly, 12 flavonoids metabolites were detected in rat plasma (six peaks), urine (nine peaks), feces (four peaks) and bile (three peak), including four glucuronide conjugates (M33, M35, M38 and M43), four sulfate conjugates (M37, M40, M41 and M44), two glutathione conjugates (M34 and M36), and one taurine conjugate (M42). It was in good agreement with the previous report that phase II metabolism was considered as the major detoxification pathway and glucuronidation and sulfation were the major metabolites of flavonoid glycosides.^25–27 Notohamosin B, the only nortriterpenoid found in the L. rotata, was found metabolized into three varieties compounds: sulfate (M45), glucuronide (M46), and taurine (M47) conjugates.

In general, metabolic profiling showed that iridoids, phenylethanoid glycosides, and flavonoids from L. rotata were detected in larger numbers of metabolites than in prototype in rat biological samples after its oral administration. Most of them underwent phase II metabolism to produce conjugates with endogenous compounds by conjugation enzymes, and more than one detoxified reactions. It mainly attributed to the multi-hydroxyl groups in the chemical structures of these compounds, which made the above endogenous water-soluble molecules easy to be attached. Moreover, for one prototype compound, the sequence of metabolite polarities was proposed as: sulfate > glucuronide > taurine conjugates. Although products of liver detoxification often lead the xenobiotics to be secreted into the intestines in bile, but sometime the metabolites could be transported into the blood, processed by kidney and excreted in urine. Therefore, among 47 metabolites, most (41 compounds) could be detected in urine, 16 were from plasma, and 20 were from feces (ESI, Fig. S1†). There is little trace for drug in the TIC of bile (only six), might owing to the limited collection time (15 min) in our experiment comparing with that in other researches (about 12 h).^28–33 Furthermore, a 12 (object) × 59 (variable) data matrix containing the absolute peak areas was submitted to partial least squares (PLS) and the principal components (PCs) were calculated. Object 12 represented 12 rats administrated in A1 and B1 groups, and variable 59 meant that 59 common peaks were found as the “absorbed and produced” compounds in rat plasma for each rat. Then, peak areas and retention times of 59 variables for 12 rats were used as input data, and the PLS displayed scattered spots from A1 (red inverted triangle) and B1 (black triangle) groups (ESI, Fig. S2†). Results did not show clear classification for the drug-containing plasma between two groups. Furthermore, the in vivo metabolites profiles from two groups were also compared by the same method, and the PLS result showed that in vivo metabolites spots also scattered for two groups (ESI, Fig. S3†).

Quantitative analysis

Optimization of LC-MS/MS condition and sample preparation. In order to get a fast and sensitive analytical method, a number of commercially available reversed phase HPLC columns and various mobile phases were evaluated. Among Agilent Zorbax SB-C18 column (2.1 mm × 100 mm, 3.5 μm), Agilent Poroshell 120 SB-AQ column (2.1 mm × 100 mm, 3.5 μm), Agilent Zorbax SB-C8 column (2.1 mm × 100 mm, 3.5 μm), and Atlantis T3-C18 column (2.1 mm × 100 mm, 3 μm), the first column provided sufficient retention and suitable separation for the six bioactive compounds quantification, including lamalbid, 8-O-acetylshanzhiside methyl ester, forsythoside B, luteolin-7-O-β-D-glucopyranoside and verbascoside. Similar to TOF/MS analysis, the ACN–water system showed more powerful separation ability and elution power for compounds than methanol–water system. Moreover, 0.1% formic acid was added into both the organic and aqueous phases, with a gradient elution mode after optimization. MRM ion pairs and other MS parameters were also optimized (Table 1). For sample preparation, liquid–liquid extraction (LLE) was evaluated first, using various organic solvents, including methyl acetate, acetone, dichloromethane and n-butyl alcohol or methyl acetate and dichloromethane combination in different ratios, but extraction efficiencies for all compounds were not satisfied under one condition. Then protein precipitation (PPT) with methanol or ACN, with or without addictive (a proper concentration of formic acid) were tested for sample preparation. Finally, methanol with 0.1% formic acid was tried and showed optimum recoveries for all analytes and IS.

Method validation. No obvious endogenous interference was observed when analytes and IS were detected in the optimized MRM modes (Fig. 2). All calibration curves exhibited good linearity (r ≥ 0.995) and the LLOQs were shown in ESI, Table S2.† Matrix effects for all analytes and IS were found to be within the acceptable range (84.3–97.0%), and the RSD values were below 17.9%. Extraction recoveries of the analytes were in the range 58.2–102.1%, with RSD less than 16.5%. RSD of intra- and inter-day precisions were in the range of 2.9–14.7% and 6.8–16.2%, with RE from −13.1 to 13.6% (except two REs −17.8 and 21.1%), respectively. The detected concentrations for all analytes were in the range of −18.8 to 14.1% of original values, indicating that they were stable in plasma samples after three freeze–thaw cycles, at room temperature and autosampler for 24 h, at 20 °C for 1 month and postpreparative for 24 h (Table 3).

Fig. 2 Representative MRM chromatograms of six analytes in different conditions. (A) Blank plasma; (B) blank plasma spiked with the analytes at LOQ; (C) 0.5 h rat plasma after oral administration of whole plant of L. rotata.

Pharmacokinetic study. For quantitative analysis, L. rotata was given to the rat at one time in a relative higher dose rather than that in the qualitative analysis. The validated method was applied to a comparative pharmacokinetic study of the six analytes after oral administration of the aerial parts (group A2) and the whole plant (group B2) of L. rotata extract in male rats. The mean plasma concentration time profiles of the six analytes were shown in Fig. 3 and revealed no significant difference in concentration–time curve for them between two groups. The main pharmacokinetic parameters were listed in Table 4. Unfortunately, as LLOQ of verbascoside and forsythoside B were above the 1/10 of their average C_max, which results to their pharmacokinetic parameters could not exactly describe the change of the two compounds. For other compounds, only lamalbid displayed higher T_max in group A2 than that in group B2 (P < 0.05). It seemed that some numerical difference in these parameters between two groups, such as t_1/2 for shanzhiside methyl ester, and C_max for shanzhiside methyl ester and 8-O-acetylshanzhiside methyl ester. However, there was no significant difference in these parameters between two groups. It might be the variances of the component contents in the blood samples of one rat in one group (data not shown). Compounds in the form of a series of analogues with the same skeleton but different functionalities underwent similar concentration–time curves, such as shanzhiside methyl ester, 8-O-acetylshanzhiside methyl ester and lamalbid, forsythoside B and verbascoside. Although 8-O-acetylshanzhiside methyl ester was in high concentration than shanzhiside methyl ester in L. rotata extract, four important pharmacokinetic parameters, including t_1/2 (h), AUC_0–t (ng h mL⁻¹), AUC_0–∞ (ng h mL⁻¹) and MRT_0–t (h), for 8-O-acetylshanzhiside methyl ester, were lower than those of shanzhiside methyl ester after administration of L. rotata extract. This could be caused by the transformation of 8-O-acetylshanzhiside methyl ester to shanzhiside methyl ester by hydrolyzation in vivo. Of note, all analytes were quickly absorbed with short T_max, which was in accordance with the previous report.¹² Fast distribution from blood into tissues and organs was considered the main cause; and the plasma concentrations for all analytes were under 5 ng mL⁻¹ after 2 or 3 h-post administration. Besides, forsythoside B, luteolin-7-O-β-D-glucopyranoside and verbascoside showed a multi-peaks phenomenon in both groups, which were found in relative high abundance in bile in TOF/MS analysis. Pharmacokinetic study results showed that metabolism behaviors were similar for six bioactive compounds in groups A2 and B2, which made sense that the medicine parts was changed from the whole plant to the aerial parts of L. rotata extract in current Chinese Pharmacopoeia (2015 edition).

Fig. 3 Mean plasma concentration–time profile of six analytes in six rats after oral administration of the aerial parts and whole plant of L. rotata.

Table 4 Mean pharmacokinetic parameters of six analytes in rat plasma after single oral administration of the aerial parts and the whole plant of L. rotata extracts (means ± SD)

Compound	Group	Parameters (n = 6)
		t1/2 (h)	Tmax (h)	Cmax (ng mL^−1)	AUC0–t (ng h mL^−1)	AUC0–∝ (ng h mL^−1)	MRT0–t (h)
a Means P < 0.05.
Shanzhiside methyl ester	Aerial parts	18.00 ± 13.43	0.67 ± 0.04	94.59 ± 25.02	416.62 ± 126.09	980.02 ± 616.18	4.99 ± 0.71
	Whole plant	37.24 ± 22.50	0.50 ± 0.00	68.44 ± 19.40	363.70 ± 100.95	1837.29 ± 1407.71	4.98 ± 0.81
8-O-Acetylshanzhiside methyl ester	Aerial parts	8.20 ± 8.02	0.50 ± 0.00	101.39 ± 27.81	172.94 ± 30.48	226.08 ± 87.69	2.87 ± 0.44
	Whole plant	8.42 ± 8.32	0.45 ± 0.10	75.32 ± 20.76	151.75 ± 22.20	195.02 ± 47.84	3.09 ± 0.28
Lamalbid	Aerial parts	22.32 ± 17.38	0.50 ± 0.00^a	22.05 ± 6.86	63.07 ± 8.30	157.38 ± 124.51	4.47 ± 0.63
	Whole plant	21.22 ± 12.92	0.33 ± 0.13^a	15.45 ± 4.04	56.70 ± 7.73	127.92 ± 62.12	4.71 ± 0.49
Verbascoside	Aerial parts	34.96 ± 28.68	0.39 ± 0.18	15.72 ± 4.38	31.33 ± 4.59	104.36 ± 62.06	4.07 ± 0.54
	Whole plant	126.14 ± 267.78	0.38 ± 0.14	11.68 ± 4.82	29.14 ± 3.33	289.91 ± 551.88	4.35 ± 0.33
Forsythoside B	Aerial parts	24.24 ± 33.97	0.46 ± 0.10	21.03 ± 5.67	43.25 ± 6.51	148.79 ± 167.14	2.91 ± 0.37
	Whole plant	18.04 ± 15.71	0.38 ± 0.14	15.51 ± 6.42	38.49 ± 3.75	115.16 ± 68.42	3.06 ± 0.25
Luteolin-7-O-β-D-glucopyranoside	Aerial parts	5.91 ± 1.90	0.15 ± 0.17	4.72 ± 3.66	3.86 ± 0.72	5.20 ± 1.45	2.63 ± 0.51
	Whole plant	4.36 ± 1.34	0.21 ± 0.17	3.83 ± 3.74	3.70 ± 0.73	4.79 ± 1.13	2.59 ± 0.53

---

Conclusions

This paper developed qualitative and quantitative analyses of bioactive compounds of L. rotata in vivo. Results clearly indicated that 39 of 48 prototype components in five classes in the extract could be metabolized to generate 47 metabolites, including sulfate, glucuronide, taurine, glycine, and glutathione conjugates. Iridoids, flavonoids, phenylethanoid glycosides were major absorbed chemical components of L. rotata. Additionally, the urine samples contained the most abundant metabolites, feces secondly, plasma thirdly and bile finally. Besides, there was no significant difference in metabolites profile between drug-containing rat plasma samples after oral administration of the aerial parts and the whole plant of L. rotata. The further comparative pharmacokinetic study between two groups also verified the metabolites profiles similarity. This investigation provided helpful information to confirm the L. rotata medicine parts alteration in Chinese Pharmacopoeia.

Acknowledgements

This work were supported by the National Natural Science Foundation of China (81202866 and 81325024) and National Significant Projects of New Drugs Creation (2011ZX09302). The authors greatly appreciate the kind help of Chang Shu and Zhipeng Wang, from Department of Pharmacy, Changzheng Hospital, Second Military Medical University.

Notes and references

1. Chinese Pharmacopoeia Commission, Pharmacopoeia of the People's Republic of China, The First Division of 2010 Edition, Chemical and Technologic Press, Beijing, 2010, pp. 245–246.
2. Chinese Pharmacopoeia Commission, Pharmacopoeia of the People's Republic of China, The First Division of 2015 Edition, Chemical and Technologic Press, Beijing, 2015, p. 262.
3. State Administration of Traditional Chinese Medicine, China Herb, Science and Technology Press, Shanghai, 2000, pp. 6074–6075.
4. F. Zhang, Z. J. Wu, L. N. Sun, J. Wang, X. Tao and W. S. Chen, Bioorg. Med. Chem. Lett., 2012, 2, 4447–4452.
5. F. Zhang, L. N. Sun and W. S. Chen, Chem. Nat. Compd., 2009, 45, 360–362.
6. M. X. Lia, X. F. Shang, R. X. Zhang, Z. P. Jia, P. C. Fan, Q. Ying and L. L. Wei, Fitoterapia, 2010, 81, 167–172.
7. W. L. Jiang, F. H. Fu, S. G. Zheng, D. L. Zhang, H. B. Zhu and J. Hou, Eur. J. Pharmacol., 2010, 629, 20–24.
8. W. L. Jiang, F. H. Fu, B. M. Xu, J. W. Tian, H. B. Zhu and J. Hou, Phytomedicine, 2010, 17, 635–639.
9. B. Zhu, N. Gong, H. Fan, C. S. Peng, X. J. Ding, Y. Jiang and Y. X. Wang, Anesthesiology, 2014, 121, 835–851.
10. M. P. La, F. Zhang, S. H. Gao, X. W. Liu, Z. J. Wu, L. N. Sun, X. Tao and W. S. Chen, J. Pharm. Biomed. Anal., 2015, 102, 366–376.
11. S. Bent, J. Gen. Intern. Med., 2008, 23, 854–859.
12. P. C. Fan, M. X. Li, H. P. Ma, L. L. Jing, J. G. Qiu and Z. P. Jia, Biomed. Chromatogr., 2012, 26, 1543–1551.
13. M. Homma, K. Oka, C. Taniguchi, T. Niitsuma and T. Hayashi, Biomed. Chromatogr., 1997, 11, 125–131.
14. C. Xiang, X. Qiao, Q. Wang, R. Li, W. J. Miao, D. A. Guo and M. Ye, Drug Metab. Dispos., 2011, 39, 1597–1608.
15. T. Yi, L. Zhu, Y. N. Tang, J. Y. Zhang, Z. T. Liang, J. Xu, Z. Z. Zhao, Z. L. Yu, Z. X. Bian, Z. J. Yang and H. B. Chen, J. Ethnopharmacol., 2014, 153, 701–713.
16. P. Jiang, Q. Wang, Y. Q. Jia, R. Shi, Y. M. Ma, P. Liu, C. H. Liu, F. Y. Ye and N. N. Cheng, RSC Adv., 2015, 5, 21786–21796.
17. Chinese Pharmacopoeia Commission, Pharmacopoeia of the People's Republic of China, The first division of 2005 edition, Chemical and Technologic Press, Beijing, 2005, pp. 184–185.
18. Food and Drug administration, Guidance for Industry: Bioanalytical Method Validation, 2013.
19. S. M. Ni, D. W. Qian, J. A. Duan, J. M. Guo, E. X. Shang, Y. Shu and C. F. Xue, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2010, 878, 2741–2750.
20. F. Zhang, Q. Zhan, S. H. Gao, X. Dong, B. Jiang, L. N. Sun, X. Tao and W. S. Chen, J. Ethnopharmacol., 2014, 152, 497–507.
21. C. Y. Li, L. W. Qi, P. Li, X. D. Wen, Y. F. Zhu, E. H. Liu, Z. Gong, X. L. Yang, M. T. Ren, Y. J. Li and X. X. Ge, Rapid Commun. Mass Spectrom., 2009, 23, 1977–1988.
22. X. W. Liu, F. Zhang, S. H. Gao, B. Jiang and W. S. Chen, Chin. J. Nat. Med., 2015, 13, 145–160.
23. X. M. Cai, L. J. Zou, J. Dong, L. L. Zhao, Y. Y. Wang, Q. Xu, X. Y. Xue, X. L. Zhang and X. M. Liang, Anal. Chim. Acta, 2009, 650, 10–15.
24. M. Holcapek, L. Kolárová and M. Nobilis, Anal. Bioanal. Chem., 2008, 391, 59–78.
25. V. Crespy, C. Morand, C. Besson, C. Manach, C. Demigne and C. Remesy, J. Agric. Food Chem., 2002, 50, 618–621.
26. W. Mullen, A. Boitier, A. J. Stewart and A. Crozier, J. Chromatogr. A, 2004, 1058, 163–168.
27. Z. X. Yan, Y. Chen, T. X. Li, J. Zhang and X. H. Yang, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2012, 885–886, 73–82.
28. L. Y. Du, D. W. Qian, E. X. Shang, P. Liu, S. Jiang, J. M. Guo, S. L. Su, J. A. Duan, J. Xu and M. Zhao, J. Ethnopharmacol., 2015, 169, 156–162.
29. Z. W. Chen, L. Tong, S. M. Li, D. X. Li, Y. Zhang, S. P. Zhou, Y. H. Zhu and H. Sun, J. Pharm. Anal., 2014, 4, 14–25.
30. X. Qiao, J. Han, M. Xu, Z. C. Zhang, M. Ye, G. Ye and D. A. Guo, J. Pharm. Biomed. Anal., 2011, 54, 789–798.
31. S. P. Li, W. Liu, L. Teng, X. M. Cheng, Z. T. Wang and C. H. Wang, J. Pharm. Biomed. Anal., 2014, 92, 53–62.
32. J. L. Geng, Y. Dai, Z. H. Yao, Z. F. Qin, X. L. Wang, L. Qin and X. S. Yao, J. Pharm. Biomed. Anal., 2014, 96, 90–103.
33. L. Y. Du, D. W. Qian, E. X. Shang, P. Liu, S. Jiang, J. M. Guo, S. L. Su, J. A. Duan, J. Xu and M. Zhao, J. Ethnopharmacol., 2015, 169, 156–162.

---

Footnotes

† Electronic supplementary information (ESI) available: Fig. S1 and S2 and Tables S1 and S2. See DOI: 10.1039/c5ra25264d

‡ M. P. La contributed equally to this study.

---