1 Introduction

Eucommia ulmoides cortex, which has been widely used as a traditional Chinese medicine in China, Japan and Korea, is obtained from the bark of 15- to 20-year-old E. ulmoides trees [1, 2]. According to records listed in the Chinese medical classic Shen Nong Ben Cao Jing and Ben Cao Gang Mu, E. ulmoides cortex can “nourish the kidney and reinforce the Yang” and, as such, it is recommended to reinforce the muscles and lungs, lower blood pressure, prevent miscarriages, improve the tone of the liver and kidneys and increase longevity.

Many modern pharmacological studies have focused on the activities of E. ulmoides against hypertension, hyperglycemia, diabetes, obesity, osteoporosis, Alzheimer’s disease, aging and sexual dysfunction [3–10]. Concomitant studies on the chemical components of E. ulmoides have led to the identification of 112 compounds [11], including lignans, phenolics, iridoids, steroids, terpenoids and flavonoids. Of these, the lignans, phenolics and iridoids are the major constituents of E. ulmoides cortex and also important chemotaxonomic markers. In E. ulmoides, the two main components of the lignans, (+)-pinoresinol di-O-β-d-glucopyranoside (PDG) and (+)-pinoresinol 4′-O-β-d-glucopyranoside (PG), have been shown to have anti-hypertensive effects [12], and geniposidic acid (GA), which is the main component of the iridoids, has been reported to exhibit anti-hypertensive and neuroprotective properties [10]. Chlorogenic acid (CA) and protocatechuic acid (PCA) are typical constituents of the phenolics present in the E. ulmoides cortex and show good antimicrobial, antimutagenic, antioxidant and neuroprotective activities [13].

However, most of these studies have focused only on the pharmacological effects, and data on the mechanisms (pharmacokinetics) of these chemical constituents and of the total extract of E. ulmoides is limited. Wang et al. [14] established a sensitive high-performance liquid chromatography/tandem mass spectrometry (HPLC/MS/MS) method for the pharmacokinetic study of pinoresinol diglucoside in rat plasma following the oral administration of E. ulmoides extract. Huang et al. [15] established and applied a new HPLC/MS/MS method for the simultaneous, quantitative determination of six ingredients, namely, aucubin, geniposide, GA, pinoresinol diglucoside, secologanin, and loganin in single and combined extracts of E. ulmoides and Dipsacus asperoides. Zhang et al. [16] developed a specific and sensitive HPLC–MS/MS method for the simultaneous determination of GA and aucubin in rat plasma after oral administration of E. ulmoides tea extract. Although pharmacokinetic studies abound on E. ulmoides, most are based on the HPLC–MS/MS method and not on the ultrahigh performance (UP) LC–MS/MS method. However, compared with traditional HPLC separation, UPLC systems equipped with 1.7-μm particle size columns have a higher efficiency of resolution and greatly reduce the analysis time. Here, we report the development and validation of a novel sensitive and selective UPLC–MS method for the simultaneous determination of GA, PCA, CA, PDG and PG concentrations and the application of this system to investigate the pharmacokinetic characteristics of these compounds following the oral administration of E. ulmoides extract to rats.

2 Methods and Materials

2.1 Chemicals and Standard Substances

2.1.1 Chemicals and Reagents

The reference standards of geniposidic acid (GA) and protocatechutic acid (PCA) (purity of 98 %) were purchased from the Sichuan Weikeqi Biological Technology Co., Ltd. (Chengdu, China). Standard pinoresinol di-O-β-d-glucopyranoside (PDG) was purchased from the Shanghai Shunbo Biological Technology Co., Ltd. (Shanghai, China), standard pinoresinol 4′-O-β-d-glucopyranoside (PG) was provided by Jiangsu Honghui Medical Co., Ltd. (Nanjing, China) and standard chlorogenic acid (CA) was purchased from Beijing Shiji’aoke Biological Technology Co., Ltd. (Beijing, China). The internal standard (IS), puerarin, was obtained from the National Institute for the control of Pharmaceutical and Biological Products (Beijing, China), distilled water was purchased from Watsons Group Co., Ltd (Hong Kong, China) and HPLC-grade acetonitrile was obtained from Merck Company (Darmstadt, Germany). All other reagents were of analytical grade.

2.1.2 Preparation of Standard Solutions and Quality Control (QC) Samples

Primary stock solutions of each of the five standard compounds to be tested (PDG, PG, CA, PCA, GA) and of the IS were prepared in methanol. A mixed stock solution was then obtained by mixing 1.350 mg/mL of GA, 0.974 mg/mL of PCA, 1.004 mg/mL of CA, 1.000 mg/mL of PDG and 1.010 mg/mL of PG. The IS solution was prepared to a final concentration of 200.0 ng/mL in methanol. A series of working standard solutions were prepared by successive dilution of the mixed stock solution with methanol. All solutions were stored at 4 °C in the refrigerator and brought to room temperature before use. Calibration standards and QC samples were also prepared. For both the calibration standards and QC samples, 20 µL of standard working solution was spiked into 100 µL of blank rat plasma. The calibration concentration ranges were 37, 111, 333, 1000, 3000 and 9000 ng/mL for GA; 4.45, 13.4, 40.1, 120, 361 and 1082 ng/mL for PCA; 13.8, 41.3, 124, 372, 1116 and 3347 ng/mL for CA; 54.9, 165, 494, 1481, 4444 and 13,333 ng/mL for PDG; 9.23, 27.7, 83.1, 249, 784 and 2244 ng/mL for PG. In the same manner, the QC samples were prepared at concentrations of 111, 1000 and 9000 ng/mL for GA; 13.4, 120 and 1080 ng/mL for PCA; 13.8, 124 and 1116 ng/mL for CA; 54.9494 and 4444 ng/mL for PDG; 27.7, 249 and 2244 ng/mL PG.

2.2 Animals

Twelve male Sprague–Dawley rats (body weight 220–240 g) were provided by the Animal Supply Center of Guiyang Medical University (Certificate No.: SCXK2013-001, Guiyang, China). All animal maintenance protocols and experimental studies were based on the guidelines of the National Institutes of Health for the Care and Use of Animals and were approved by the Animal Care and Use Committee of Guiyang Medical University. All rats were kept in housing under a controlled temperature regimen (22 ± 1 °C) and relative humidity of 50–70 %. The animals were maintained on a 12-h light/12-h dark cycle and had free access to food and water. They were fasted for 12 h before each experiment, but allowed to drink freely during this time.

2.3 LC–MS/MS Conditions

The LC-MS/MS analyses were carried out using a system which consisted of an ACQUITY™ UPLC system (Waters, Milford, MA) in which the power was coupled to a Waters ACQUITY triple quadrupole mass spectrometer equipped with a Z-spray electrospray ionization (ESI) source. The ESI source was set in positive ion mode for the IS and in negative ion mode for GA, PCA, CA, PDG and PG. All analytes, including the IS, were quantified in SIR mode at m/z transitions of 373.1 for GA, 152.9 for PCA, 353.1 for CA, 681.3 for PDG, 519.3 for PG and 417.2 for the IS (Fig. 1). The parameters of the mass spectrometer were optimized as follows: capillary voltage of 3 kv, source temperature of 120 °C, desolvation temperature of 350 °C, desolvation gas flow of 650 L/h and cone voltage of 35 kv for all analytes tested. The operation of the LC-MS/MS system and data analysis were performed with the MassLynx™ V4.1 workstation (Micromass, Manchester, UK).

Fig. 1
figure 1

Product ion mass spectra of geniposidic acid (GA) and protocatechutic acid (PCA), chlorogenic acid (CA), pinoresinol di-O-β-d-glucopyranoside (PDG), pinoresinol 4′-O-β-d-glucopyranoside (PG) and the internal standard (IS)

Full size image

Liquid chromatography analyses were performed in a gradient elution mode using a Waters C18 BEH column (2.1 mm × 50 mm, 1.7 µm, Waters, Milford, MA, USA) equipped with a guard column at 45 °C. The mobile phase consisted of 0.1 % formic acid in acetonitrile (A) and 0.1 % formic acid water solution (B). A linear gradient at a flow rate of 0.35 mL/min was run at 5–10 % A for 0–0.5 min, 10–20 % A for 0.5–3.5 min, 20–90 % A for 3.5–4 min and 90–5 % A for 4–5 min. The samples were kept at 4 °C in the auto-sampler and a volume of 2 µL was injected into the UPLC system.

2.4 Preparation of E. ulmoides Extract

The cortex of E. ulmoides was crushed and decocted 3 times (2 h each time) with water (1:10 w/v). The decoctions were then combined and condensed to a concentration of 1 g/mL, following which 95 % ethanol was slowly added to the concentrate until a final ethanol concentration of 60 % was reached. The ethanol–E. ulmoides concentrate was then filtered and the filtrate evaporated until there was no smell of ethanol. The concentrated solution was extracted 3 times successively with n-butanol (1:3 v/v) and the n-butanol extracts evaporated until there was no smell of n-butanol. This latter extract was subjected to D101 macroporous resin eluted with 45 % ethanol solution. The 45 % ethanol eluent was concentrated and vacuum-dried to powder. The GA, PCA, CA, PDG and PG contents of the E. ulmoides extracts were determined under the same chromatography conditions as described above and found to be 35.2, 1.7, 1.8, 20.0 and 10.4 mg/g, respectively.

2.5 Sample Preparation

Formic acid is a strong acid which can be used to facilitate the separation of drugs from plasma protein when added as a 1 % formic acid water solution to the plasma. It also makes the sample more conducive to ionization in UPLC-MS analyses and increases the detection sensitivity of the sample. Therefore, we spiked each plasma sample (100 μL) with 50 μL 1 % formic acid distilled water solution, then vortexed the sample for 1 min, following which we added 20 μL methanol and 10 μL IS (200 ng/mL) to the sample and mixed well. Protein was precipitated by adding 270 μL of methanol to the mixture. After being vortexed and sonicated, the samples were centrifuged at 15,000 g for 10 min. The supernatant of each sample was transferred to a new 2-mL centrifuge tube and evaporated to the dryness by N2 at 40 °C. Finally, the residues were dissolved in 200 μL methanol, and 2-μL samples were injected into the UPLC-MS system for analysis.

2.6 Method Validation

The UPLC–MS/MS method was fully validated in terms of selectivity, linearity, the lower limit of quantification (LLOQ), accuracy and precision, extraction recovery and matrix effect and stability according to the U.S. Food and Drug Administration (FDA) guidelines for the validation of bioanalytical method.

Selectivity

The selectivity of the method was assessed by comparing the chromatograms of blank plasma samples collected from six different individual rats, plasma samples spiked with the analytes and IS and plasma samples obtained from a rat at 0.5 h after oral administration of the E. ulmoides extract.

Linearity and the Lower Limit of Quantification

Calibration curves were obtained by plotting the measured peak area ratios of analytes to IS. Standard curves representing peak area ratios versus analyte concentrations were described in the form of y = a + bx (weighing factor 1/x). The LLOQ for each analyte was the lowest concentration with a signal-to-noise of ≥10 which could be quantitatively determined with a precision and accuracy of ≤20 %, based on the analysis of six replicates of each sample.

Accuracy and Precision

Precision and accuracy were evaluated at three concentrations. The intra-day accuracy and precision were estimated by analyzing six QC samples at each concentration level during the same day, while the inter-day accuracy and precision were assessed on 3 consecutive days. The different precisions were defined as the relative standard deviation (RSD%), which should be <15 %. The accuracy was expressed by the relative error (RE%), which was required to be within 15 %.

Extraction Recovery and Matrix Effect

The extraction recoveries of five analytes at three QC levels were evaluated by comparing the peak areas obtained from the extracted QC samples with those obtained from reference standards spiked into post-extracted blank plasma. The matrix effect was assessed by comparing the peak areas of the standard solutions added to post-extraction supernatant with those of the standard solutions dried directly and reconstituted in methanol.

Stability

The stability of the five analytes (GA, PCA, CA, PDG, PG) in rat plasma was evaluated at three concentrations of QC samples, each in six replicates, under different storage and processing conditions, including the stability of the plasma samples kept at room temperature for 6 h, at −20 °C for 14 days and after three freeze–thaw cycles and of the post-preparation sample in the autosampler at 4 °C for 24 h. Stability was evaluated by comparing the mean concentration of the stored QC samples with the mean concentration of freshly prepared QC samples.

2.7 Pharmacokinetic Study

Six male Sprague–Dawley rats (body weight 220–240 g) were used in the pharmacokinetic study. After a 12-h fast, each rat was orally administered E. ulmoides extracts at a dose of 4.8 g/kg (equivalent of 168.96 mg/kg GA, 8.16 mg/kg PCA, 8.64 mg/kg CA, 96 mg/kg PDG and 49.92 mg/kg PG, respectively) dissolved in saline. Blood samples (0.3 mL) were collected from the caudal vein into centrifuge tubes coated with heparin both before the infusion (0 h) and at various time-points after dosing (0.083, 0.25, 0.5, 0.75, 1, 2, 3.5, 5, 7, 9, 11, 13, 24 and 30 h). The plasma was separated by centrifugation at 4000 rpm for 3 min and then stored at −20 °C until analysis. The amount of each analyte in the plasma was estimated by UPLC-MS/MS analysis as described above.

2.8 Pharmacokinetic Data Analysis

Non-compartmental methods using DAS 2.0 software provided by the Mathematical Pharmacology Professional Committee of China were used to analyze plasma concentration versus time profiles and to estimate the following pharmacokinetic parameters: terminal elimination half-life (t 1/2, λ z), area under the plasma concentration versus time curve from zero to last sampling time (AUC0–t), volume of distribution (V d, λ z) and total body clearance (CL). The peak plasma concentration (C max) and time to reach C max (T max) for each dose were read directly from the observed individual plasma concentration–time data.

3 Results and Discussion

3.1 Selection of ESI Mode

The UPLC–MS method described here was developed to facilitate the simultaneous quantification of GA, PCA, CA, PDG and PG in rat plasma. The water: acetronitrile mobile phase of the UPLC–MS system was spiked with 0.1 % formic acid to improve ionization. The observed response of analytes was better in the negative ESI mode than in the positive ESI mode. However, the IS showed a high sensitivity in the positive ESI mode. As it was essential to optimize the ion source to increase the MS intensity for quantifier ions, the ESI ion source was optimized in combination with the drying gas flow and the pressure of nebulizer. The fragmentor was also optimized to enhance the signal response of each standard and to eliminate interference from noise. All analytes, including the IS, were quantified in SIR mode at m/z transitions of 373.1 for GA, 152.9 for PCA, 353.1 for CA, 681.3 for PDG, 519.3 for PG and 417.2 for the IS (Fig. 1). The parameters of the mass spectrometer were optimized as follows: capillary voltage of 3 kv, source temperature of 120 °C, desolvation gas flow of 650 L/h at a temperature 350 °C and a cone voltage of 35 kv. These conditions were found to be optimal for the determination of all analytes tested.

3.2 Method Validation

Selectivity

An Acquity UPLC BEH C18 column (2.1 × 50 mm, internal diameter 1.7 µm) was used for the simultaneous analysis of the five compounds in the plasma. Figure 2 shows the UPLC-MS chromatograms of blank plasma (A), of blank plasma spiked with the solution containing all of the analytes and the IS in LLOQ (B) and of the plasma sample at 30 min after oral administration of E. ulmoides extracts (C). The retention times for GA, PCA,CA, PDG, PG and IS were 0.98, 1.07, 1.46, 2.38, 3.71 and 1.81 min, respectively. A high resolution (R s >1.5) among the peaks was achieved, and no interfering peak in the blank plasma was observed under these experimental conditions.

Fig. 2
figure 2

Chromatograms of the analytes and IS in blank plasma (a), blank plasma spiked with the analytes and IS (b) and plasma sample from a rat orally administered Eucommia ulmoides extracts (c). See caption to Fig. 1 for abbreviations of analytes

Full size image

Linearity and LLOQ

Typical equations for the calibration curves and correlation coefficients (r) were y = 0.0089 x + 0.1823 (r = 0.9998) for GA, y = 0.0306 x + 0.0119 (r = 0.9999) for PCA, y = 0.0068 x + 0.0492 (r = 0.9998) for CA, y = 0.0017 x + 0.0292 (r = 0.9999) for PDG and y = 0.0087 x + 0.0371 (r = 0.9999) for PG. The calibration curves of the five analytes were linear in the ranges of 37–9000 ng/mL for GA, 4.45–1082 ng/mL for PCA, 13.7–3333 ng/mL for CA, 54.9–13,333 ng/mL for PDG and 9.23–2244 ng/mL for PG. The LLOQ samples of six different rat plasma samples independent of the calibration curves were analyzed. The LLOQs were 37 ng/mL for GA, 4.45 ng/mL for PCA, 13.7 ng/mL for CA, 54.9 ng/mL for PDG and 9.23 ng/mL for PG, which was sufficient for pharmacokinetic study of the five analytes following oral administration of E. ulmoides extract to rats.

Precision and Accuracy

The results of the precision and accuracy at three concentration levels are summarized in Table 1. The intra-day and inter-day precisions (RSD%) for all analytes were <10.4 %, and the accuracy (%) ranged from −10.6 to 11 %.

Table 1 Precision and accuracy for all analytes in rat plasma
Full size table

Matrix Effect and Extraction Recovery

The data on extraction recovery and matrix effect are shown in Table 2. The extraction recoveries of GA, PCA, CA, PDG and PG at three different concentrations from rat plasma were found to be 93.9–102.4 %, 84.3–97.9 %, 93.5–102.3 %, 91.0–95.5 %, 85.7–97.9 %, respectively, which indicates a good extraction efficiency. Our results indicated that there was no matrix effect under the given experimental conditions.

Table 2 Matrix effects and extraction recovery of all analytes in plasma samples collected from rats orally administered Eucommia ulmoides extracts (n = 6)
Full size table

Stability

The stability of all analytes was evaluated under various conditions. The results listed in Table 3 indicate that the five analytes were stable in the plasma samples for 4 h at room temperature, 14 days at −20 °C, three freeze–thaw cycles and in the autosampler at 4 °C for 24 h.

Table 3 Stability of all analytes in plasma samples collected from rats orally administered E. ulmoides extracts (n = 5)
Full size table

3.3 Pharmacokinetic Analysis

The validated UPLC–MS method was successfully applied to pharmacokinetic studies of GA, PCA, CA, PDG and PG in plasma samples obtained from Sprague–Dawley rats which had been orally administered E. ulmoides extracts at a dose of 4.8 g/kg. The mean plasma concentration–time profiles of GA, PCA, CA, PDG and PG are shown in Fig. 3, and their estimated pharmacokinetic parameters are given in Table 4. The assay was sensitive for the determination of GA, PCA, CA, PDG and PG in these plasma samples.

Fig. 3
figure 3

Plasma concentration–time curves of the five analytes of the E. ulmoides extract tested following the oral administration of E. ulmoides extracts to Sprague–Dawley rats at a dosage of 4.8 g/kg. The results are given as the mean ± standard deviation (n = 6 replicates)

Full size image
Table 4 Pharmacokinetic parameters of all analytes tested in plasma samples collected from rats orally administered E. ulmoides extracts
Full size table

As seen from Fig. 3 and Table 4, the maximum concentrations (C max) of GA, PCA, CA, PDG and PG in plasma occurred at 8.00 ± 1.67, 0.46 ± 0.33, 0.83 ± 0.63, 0.54 ± 0.25, 0.54 ± 0.10 h, respectively, post-administration of the oral dose to male Sprague–Dawley rats. The half-life (t 1/2) of GA, PCA, CA, PDG and PG was 4.12 ± 0.77, 9.20 ± 4.69, 12.30 ± 11.88, 7.05 ± 9.73, 2.75 ± 0.76 h, respectively. The CLz/F and Vz/F were 80.70 ± 37.33 and 473.80 ± 211.74 μg/L h for GA, 14,034.55 ± 5403.20 and 169,821.02 ± 85,830.96 μg/L h for PCA, 4997.37 ± 2680.00 and 61,671.95 ± 30,085.93 μg/L h for CA, 1398.30 ± 479.80 and 15,023.50 ± 21,536.31 μg/L h for PDG and 9221.89 ± 2829.47 and 36,336.10 ± 12,884.54 μg/L h for PG. The CLz/F and Vz/F of GA markedly decreased relative to these values for the other four compounds analyzed, showing that the visceral distribution of GA was different from that of these other compounds. The concentration–time curves showed that the maximum concentration occurred between 0.46 and 0.83 h for all analytes, with the exception of GA for which the peak concentration occurred at 8 h. This observation indicates that PCA, CA, PDG and PG were rapidly absorbed relative to the absorption of GA, which was the slowest. However, only one peak for the five compounds analyzed was found in the concentration–time curves, which indicated that the absorption of these compounds was scarcely affected by the emptying time of the stomach and that there was no effect of the enterohepatic circulation. The data described here may be helpful for clarifying drug disposition in the body and for use in further studies on the pharmacokinetic evaluation of GA, PCA, CA, PDG and PG.

4 Conclusion

We describe here the development and validation of a rapid and sensitive UPLC-MS method for the simultaneous analysis of GA, PCA, CA, PDG and PG in rat plasma. The method was successfully applied to the pharmacokinetic study of five analytes in rat plasma samples after oral administration of E. ulmoides extracts. Our results also provide useful information for additional clinical applications of formulations of Chinese medicine.