Phenol glycosides have the potential to deglycosylate and this result in its aglycone metabolite. In the present study, salidroside and the potential metabolite p-tyrosol, the aglycone of salidroside, were identified by HPLC-PDA and LC-MS/MS. Comparing HPLC-PDA chromatograms of rat blank plasma with those samples after i.v. and i.g. administrations, two specific peaks at the retention time of 7.68 min and 10.88 min in the samples after i.v. administration were observed, whereas only one peak at 7.68 min was shown in the samples after i.g. administration (Figure 2). UV spectra of the two peaks are similar and match to those of salidroside and p-tyrosol. By comparing retention times with standards of salidroside and p-tyrosol, peak at 7.68 min was identified as salidroside, and the peak at 10.88 min was p-tyrosol.

The LC-MS/MS detection was performed in the negative ion mode, and the conditions for the detection of salidroside and p-tyrosol were optimized with the standards. The daughter ions obtained from the deprotonated molecular ions of salidroside ([M−H]-299.0) and p-tyrosol ([M−H]-137.0) were shown in Figure 3. The deprotonated molecular ion of salidroside produced three ions at m/z 88.7, 118.8 and 179.0 with higher abundance. The presumable structure of the ion at m/z 88.7 was [CHOHC(OH)CH₂OH-H]⁻. The other two daughter ions at m/z 118.8 and 179.0 were [p-tyrosol-OH]⁻ and glucosyl ions, which are typical and characteristic fragments for the salidroside structure. The deprotonated p-tyrosol produced daughter ions at m/z 105.8 (loss of a -CH₂OH fragment) and 118.9 (loss of a -OH fragment). The identification was developed by monitoring the transitions m/z 299.0→118.8 and 299.0→179.0 for salidroside and m/z 137.0→105.8, 137.0→118.9 for p-tyrosol, and the corresponding cone voltage and collision energy were shown in Table 1. The peak area ratio of the two coupled transitions for each analyte should be matched to the standard and also an index for the identification generated, which was 3.17 ± 0.13 for salidroside (the peak area ratio of transitions m/z 299.0→118.8 to m/z 299.0→179.0), and 3.56 ± 0.19 for p-tyrosol (the peak area ratio of transitions m/z 137.0→105.8 to m/z 137.0→118.9). The total run time was 3.5 min. The retention times of salidroside, p-tyrosol and IS were 1.94 min, 2.73 min and 1.80 min respectively. p-Tyrosol was detected from the samples after i.v. administration, but not from those samples obtained after i.g. administration, which further confirmed that salidroside could be metabolized to p-tyrosol after i.v. administration, whereas this metabolism was not detectable after i.g. administration.

The deprotonated molecular ion of paracetamol (IS) ([M−H]⁻ 150.1) only gave one characteristic product ion at m/z 106.9. The transitions of m/z 299.0→118.8 for salidroside, m/z 137.0→118.9 for p-tyrosol, m/z 150.1→106.9 for IS were also used for the quantification of the analytes.

The method was successfully applied to the quantification of salidroside and p-tyrosol in rat plasma samples after i.v. and i.g. administration. The mean concentration-time curves of salidroside and p-tyrosol after 50 mg/kg i.v. administration are shown in Figure 5 and the main pharmacokinetic parameters were calculated and are shown in Table 6. p-Tyrosol concentration reached the maximum within the first 5 min post-administration of the dose, which might be even higher than the concentration we detected, hence the T_max and C_max for p-tyrosol was not calculated. The t_1/2 of elimination phase was prolonged 1.34 fold to 1.64 ± 0.30 h for p-tyrosol, comparing with that of 0.70 ± 0.21 h for salidroside. The AUC_0–4 and AUC_0–∞ values obtained were 7,060.72 ± 1,337.51 h·ng/mL and 7,135.79 ± 1,346.40 h·ng/mL for salidroside, and 122.77 ± 25.90 h·ng/mL and 146.83 ± 32.49 h·ng/mL for p-tyrosol, respectively, which implied that only about 2% salidroside was present as the aglycone metabolite, p-tyrosol, in plasma.

The mean plasma concentration-time curve of salidroside following i.g. administration was presented in Figure 5, and the pharmacokinetic parameters of salidroside were also summarized in Table 6. p-Tyrosol was undetectable post i.g. administration. After i.g. administration, the concentration of salidroside increased within the first 20–30 min then decreased, which meant that the absorption of salidroside reached the maximum within 20–30 min, and then the elimination was dominant. The AUC_0–8 and AUC_0–∞ values obtained were 7,522.82 ± 549.02 h·ng/mL and 7,724.52 ± 446.62 h·ng/mL for salidroside.

According to the data shown in Table 6, T_1/2 of salidroside was observed at 0.70 ± 0.21 h after i.v. administration and 1.32 ± 0.22 h after i.g. administration, which was similar to the reported T_1/2 at 0.5 ± 0.2 h after i.v. administration and 1.1 ± 0.7 h after i.g. administration [13]. Although salidroside underwent an absorption phase after i.g. administration, which resulted in a little longer T_1/2 (~0.6 h in difference) than that after i.v. administration, salidroside was eliminated quickly after both i.v. and i.g. administrations in vivo.

The oral bioavailability in the present study was 51.97%, calculated following the equation given in the literature [26]. It was between previous results of 98% calculated at dosages of 25 mg/kg oral (p.o.) and 5 mg/kg i.v. administration [13] and 32.1% done at dosages of 12 mg/kg i.g. and i.v. administration [19] in S.D. rats, which imply that the intake of salidroside may vary upon different dosages and administrative patterns.

Phenol glycosides are abundant in foods and medicinal herbs; as a result most research on their metabolism was under oral administration conditions. It has proven that during transfer across the jejunum and ileum, phenol glycosides are subject to extensive metabolism by phase I hydrolyzing and oxidizing enzymes, including cytochrome P450 and glucosidase enzymes and by phase II conjugating and detoxifying enzymes. The majority of aglycones, which escape metabolism in the small intestine, are subject to metabolism by phase I and II enzymes including uridine diphosphate (UDP)-glucuronosyl transferases, which lead to the formation of O-glucuronides, sulpho-transferases, which generate O-sulfates and catechol-O-methyl transferases, which produce O-methylated metabolites [27,28,29,30,31], and the aglycones may be negligibly present in the circulation [32]. In the present study, the detection of p-tyrosol after i.v. administration and the oral bioavailability (52.1%) reflect that salidroside may metabolize to p-tyrosol after i.g. administration, whereas it might be further metabolized to its sulfate or glucuronide conjugates, or even the methylate, and resulted in undetectable p-tyrosol in the plasma sample. Thus, more attention will be paid to the further metabolism and metabolites of salidroside and its aglycone in our further studies.

Salidroside was purchased from the National Institutes for Food and Drug Control (Beijing, China). Paracetamol, the IS (purity ≥ 98%) was purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). p-Tyrosol and HPLC-grade acetonitrile was obtained from Sigma (St. Louis, MO, USA). Ultra pure water was produced by a Millipore Milli-Q system (Billerica, MA, USA). All other reagents or solvents used were commercially available and of reagent grade. Blank rat plasma was collected from healthy Male Wistar rats weighting 200 ± 20 g (Laboratory Animal Center of Jilin University, Changchun, Jilin Province, China).

The identification of salidroside and p-tyrosol in plasma samples was conducted by HPLC with a PDA detector and LC-MS/MS. The quantitative analysis of salidroside and p-tyrosol was performed on a LC-MS/MS.

The HPLC system consisted of a 1525 HPLC Pump, a 717 plus autosampler, and a 2996 PDA detector (Waters Co., Milford, MA, USA). The column used for separation was a Thermo Hypersil Gold C18 column (4.6 mm × 250 mm, 5 μm, USA). The mobile phase was a mixture of acetonitrile and water (1/9, v/v).

The LC-MS/MS system was comprised of an Alliance 2695 HPLC and a Quattro-Micro^TM mass spectrometer (Waters Co.). An xTerra C18 mass column (3.5 µm, 50 × 3.0 mm, Waters Co., Milford, MA, USA) was used for separation, which was equilibrated and eluted with an isocratic mixture of acetonitrile-water (1:9, v/v) at a flow rate of 0.3 mL/min. The injection volume was 20 µL. The ESI-MS/MS detection was performed under negative ion mode under the following conditions: Capillary voltage 3.00 kV, extractor 2.00 V, RF Lens 0.1 V, desolvation gas 500 L/Hr, cone gas 50 L/Hr, source temperature 120 °C, desolvation temperature 350 °C, entrance voltage −2 V, exit voltage 1 V. ESI-MS/MS parameters were shown in Table 1. The total run time was 3.5 min. The retention time of salidroside, p-tyrosol and IS were 1.94 min, 2.73 min and 1.80 min respectively. Waters MassLynx 4.0 software was used for system control and data acquisition.

Stock solutions of salidroside, p-tyrosol and IS (paracetamol) were prepared at 200 μg/mL in acetonitrile-water (1:9, v/v), respectively, and further diluted with acetonitrile-water (1:9, v/v) to give a series of working solutions. All solutions were stored at −20 °C until use.

Calibration curves were prepared by spiking 20 μL of appropriate working solution with 100 μL of blank rat plasma. The effective concentrations were 50, 100, 200, 500, 1,000, 1,500 and 2,000 ng/mL for salidroside, and 20, 40, 80, 100, 140, 180, 200 ng/mL for p-tyrosol, respectively. QC samples were prepared in pool as a single batch for each concentration at concentrations of 50, 500 and 2,000 ng/mL for salidroside and of 20, 100, 200 ng/mL for p-tyrosol, and then divided into aliquots and stored in the freezer at −20 °C until use. The IS working solution of 250 ng/mL was diluted from stock solution as needed. The spiked rat plasma (CSs and QCs) were treated following the sample processing procedure as for the unknown samples.

To identify salidroside and the potential metabolite p-tyrosol in plasma sample, an aliquot (500 µL) of plasma sample and methanol (1,500 µL) were mixed by vortex-mixing for 1 min followed by ultrasonic incubation for 10 min. After centrifuging at 45,000 g for 5 min, the clear supernatant was transferred into a new polypropylene tube and evaporated to dryness under a nitrogen stream. The residue was reconstituted in 200 µL of mobile phase. After filtering through a membrane (0.22 µm pore size), each 20 µL aliquot was injected into the HPLC-PDA or LC-MS/MS system.

The above sample preparation method was employed with minor modifications for pharmacokinetic studies. One hundred micro liters of plasma sample (blank plasma, spiked plasma or pharmacokinetics study plasma sample) and IS working solution (20 µL) was pipetted out into a 1.5 mL polypropylene tube, then methanol (280 µL) was added followed by vortex mixing for 1 min and 10 min of ultrasonic incubation. After centrifuging at 45,000 g for 5 min, the clear supernatant was transferred to a new tube and evaporated to dryness under a nitrogen stream. The residue was reconstituted in 200 µL of mobile phase. After centrifuging at 45,000 g for 10 min, an aliquot of 20 µL was injected into the LC-MS/MS system. When the concentration of salidroside in rat plasma was over the range of calibration curve, appropriate dilutions were applied to the plasma sample with blank rat plasma before sample processing.

The method for simultaneous determination of salidroside and p-tyrosol in rat plasma by LC-MS/MS was validated to meet the acceptance criteria as the guidance from Food Drug Administration (FDA) and per guidelines of the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH).

The selectivity was assessed by comparing the chromatograms of six individual blank rat plasma samples with the corresponding spiked plasma samples. Each blank plasma sample was processed through the proposed extraction procedure and tested to ensure no interference of the analyte from the rat plasma.

Lower limit of quantification (LLOQ) was defined as 10 times the signal-to-noise (S/N) ratio, and limit of detection (LOD) was done as three times the S/N ratio. The lowest concentration on the calibration curve was to be accepted as the LLOQ, if the analyte response was at least 10 times more than that of blank plasma sample.

Linearity was assessed by assaying calibration curves ranging from 50 to 2,000 ng/mL for salidroside and 20–200 ng/mL for p-tyrosol in duplicate on six consecutive days. Plasma samples were quantified using the peak area ratio of salidroside or p-tyrosol to that of IS standard curves are in the form of y = A + Bx, where y represents the plasma concentration of analyte and x represents the ratio of analyte peak area to that of IS The acceptance criterion for a calibration curve was a correlation coefficient (r) of 0.99 or better, and each back-calculated standard concentration must be within 100 ± 15% except at LLOQ, for which the maximum acceptable deviation was at 20%.

Accuracy and precision were evaluated at three QC levels of 50, 500 and 2,000 ng/mL for salidroside and of 20, 100, 200 ng/mL for p-tyrosol. The assays of intra- or inter-day accuracy were preformed in six separate runs on the same day or on six consecutive days, and expressed as (observed concentration/spiked concentration) ×100%. Intra- and inter-day precisions were obtained by one-way analysis of variance (ANOVA) test, and were expressed as relative standard deviation (RSD). The accuracy was required to be within 100 ± 15% and the precision should not exceed ±15%.

The extraction recoveries of salidroside and p-tyrosol were determined at three QC levels, respectively. Recoveries were calculated by comparing the analyte/IS peak area ratios of each analyte in spiked plasma samples with those of analytes in the matrices by spiking extracted analyte-free plasma samples prior to chromatography.

The matrix effects from endogenous substances present in extracted rat plasma may cause ion suppression or enhancement of the signal. The matrix effects were investigated by post-extraction spike method in the present study. Peak area (A) of standard analyte in spiked blank plasma was compared with the corresponding peak area (B) obtained by directly injecting the standard analyte in the mobile phase at concentrations of 50, 500 and 2,000 ng/mL for salidroside and 20, 100 and 200 ng/mL for p-tyrosol in triplicates. The peak area ratio of A/B (as a percentage) was used as a quantitative measure of the matrix effects [33].

The stability of standard solutions was tested at room temperature for 2 h and upon refrigeration (4 °C) for 30 days. The stability of analytes was examined by keeping replicates of salidroside and p-tyrosol QC samples in the autosampler tray for 24 h and in a freezer at −20 °C for 30 days; the freeze-thaw stability was obtained over three freeze-thaw cycles, by thawing at room temperature for 2–3 h and then refreezing at −20 °C for 12–24 h. For each concentration and each storage condition, 6 replicates were analyzed in one analytical batch. The concentration of analytes after each storage period was related to the initial concentration, which was determined when the samples were originally prepared and processed.

Male rats (ICR, 200 ± 20 g) were obtained from Laboratory Animal Center of Jilin University (Changchun, Jilin Province, China). Animal handling procedures were according to standard operating procedure approved by the institutional animal care and use committee. All rats were dosed following an overnight fasting (except for water).

For the identification of salidroside and p-tyrosol, six male rats were divided into two groups. Rats in one group underwent an i.v. administration through the vena caudalis with sodium chloride solution of salidroside at the dose of 50 mg of salidroside (12.5 mg/mL in saline) per kg body weight (i.v. 50 mg/kg). Five min after administration, blood samples were collected in heparinized tubes via cardiac puncture. Rats in the other group received an i.g. dose of 100 mg of salidroside (25 mg/mL in saline) per kg body weight (i.g. 100 mg/kg), and blood samples were collected at 0.5 h after administration. Plasma was separated by centrifuging at 4,000 g for 30 min at 4 °C and stored frozen at −20 °C until analysis.

For pharmacokinetic study, 12 male rats underwent jugular vein cannulation [34] and were randomly divided into two groups. Rats in group 1 were i.v. administered with 50 mg/kg. Serial blood samples (about 0.3 mL) were collected in heparinized tubes via the jugular vein before and at the time points of 0, 0.08, 0.17, 0.33, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0 h after administration. In group 2, each rat received the administration of i.g. 100 mg/kg. Blood samples were collected before and at the time points of 0, 0.08, 0.17, 0.25, 0.33, 0.5, 0.75, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0 h. Plasma was separated and stored frozen at −20 °C until analysis.

A noncompartmental pharmacokinetic analysis using Kinetica^TM software package (version 5.0, Thermo Fisher Scientific Inc., Pittsburgh, PA, USA) was performed to determine the key parameters including maximum concentration (C_max), time-to-maximum concentration (T_max), half life time (T_1/2), total body clearance (Cl), mean residence time (MRT), steady state apparent volume of distribution (Vss), area under curve from zero to the last measurable plasma concentration point (AUC_0−t, t = 4.0 h for i.v. administration, t = 8.0 h for i.g. administration), and area under the plasma concentration-time curve from zero to time infinity (AUC_0–∞).

The oral bioavailability (F) is defined as the fraction of unchanged drug reaching the systemic circulation following administration through the i.g. route. The absolute oral bioavailability of a drug is generally measured by comparing the respective AUCs after i.g. and i.v. administration according to the following equation [28]: