Mangiferin and 4-Nitrophenol, which was used as an internal standard (I.S.), were purchased from Sigma (Sigma-Aldrich, Tokyo, Japan). The chromatographic solvents and reagents were obtained from the National Institute for the Control of Pharmaceutical and Biologic Products (Beijing, China). All substances were of chromatographic grade. De-ionized water was prepared using a Milli-Q water purifying system from Millipore Corp. (Bedford, MA).

Adult male Sprague–Dawley rats (280~320 g) were supplied by the Animal Research Center at Harbin Medical University (Harbin, Heilongjiang Province, China). The rats were specifically pathogen-free and housed in a limited access rodent facility for at least 5 days. The room controls were set to maintain the temperature at 22 °C±2 °C, the relative humidity at 55%±15%, the illumination intensity at 150~300 lx, an air ventilation frequency of 15~20 times/h, and a 12 h illumination period (07:00~19:00) each day. The animals drank sterilized drinking water, and standard chow diet was supplied ad libitum to each cage. The animal experiments were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and were approved by the Animal Ethics Committee of Harbin Medical University.

The standard stock solution was prepared by dissolving 10 mg of mangiferin in a 10 ml methanol to obtain a nominal concentration of 1,000 μg/ml. The internal standard (I.S.) stock solution was prepared by dissolving 0.20 mg of 4-Nitrophenol in 100 ml of methanol to obtain a nominal concentration of 2 μg/ml. All the stock solutions were maintained at 4 °C until use.

Mangiferin standard samples of plasma (0.50, 1.00, 5.00, 10.00, 20.00, 50.00, 100.00, and 150.00 μg/ml) were prepared by spiking control rat plasma with appropriate amounts of the standard stock solution prepared above. Quality control (QC) samples were used to determine accuracy and precision of the method, and were independently prepared at low (1.00 μg/ml), medium (10.00 μg/ml) and high (100.00 μg/ml) concentrations in the same manner. The I.S. was added to each standard sample immediately before sample processing. All the samples were stored at −20 °C until analysis.

Mangiferin standard samples of eyes (0.10, 0.50, 1.00, 2.00, 4.00, 6.00, 8.00, and 10.00 μg/ml) were prepared by spiking control rat eye samples with appropriate amounts of the standard stock solution prepared above. QC samples were used to determine accuracy and precision of the method, and were independently prepared at low (0.50 μg/ml), medium (2.00 μg/ml) and high (8.00 μg/ml) concentrations in the same manner. The I.S. was added to each standard sample immediately before sample processing. All the samples were stored at −20 °C until analysis.

Each rat (n=6) was fasted for 24 h with free water during the experiment. Then, they were intravenously administrated of mangiferin, which was suspended in 2 ml of dimethyl sulphoxide (10, 25, and 50 mg/kg).

Blood samples (200 μl) collected from the cervical artery at 0.5, 1.0, 1.5, 2, 2.5, 3.5, 4.5, 5.5, 7.5, 9.5, 11.5, and 24 h post-dose were seated for approximately 30 min and centrifuged at 4,000xg for 20 min to separate the plasma supernatant. To 100 μl of plasma, 100 μl of I.S. solution (2 μg/ml) and 400 μl of acetonitrile-glacial acetic acida (9:1,v:v) were added. Each tube was mixed thoroughly by vortexing for 90 s. After centrifugation for 10 min at 3,000 xg, the supernatant was transferred into labeled, clean test tubes and evaporated to dryness in a water bath at 50 °C under a stream of nitrogen. The residues were reconstituted in 100 μl of mobile phase with vortexing for 90 s, and the centrifugation procedure was repeated. Standard and QC samples were prepared following the method described above. A 20 μl aliquot was then injected onto the chromatographic column.

As to eye samples, samples from both eyes were collected at 0.5, 1.0, 1.5, 2, 2.5, 3.5, 4.5, 5.5, and 7.5 h post-dose and placed in a normal saline solution to remove blood and the conjunctiva. The eyes were then cut from the equator, and the lens and vitreous were removed. The retina was carefully dissected from the choroid and optic nerve, and the homogenate was prepared. Briefly, retina samples were weighed and homogenized in acetonitrile at 0 °C using an FJ-2000 homogenizer (Biaoben Model Factory, Shanghai, China). The homogenates were then ultrasonically extracted using a Branson Sonifier 150D (Branson, Danbury, CT) and centrifuged at 1,000× g for 10 min. The supernatants were taken out and transferred into labeled, clean test tubes. To 100 μl of supernatant, 100 μl of I.S. solution (2 μg/ml) and 400 μl of acetonitrile-glacial acetic acida (9:1, v:v) were added and mixed thoroughly by vortexing for 90 s. After centrifugation, the supernatant was transferred into labeled, clean test tubes evaporated to dryness in a water bath at 50 °C under a stream of nitrogen. The residues were reconstituted in 100 μl of mobile phase with vortexing for 90 s, and the centrifugation procedure was repeated. Standard and QC samples were prepared following the method described above. A 20 μl aliquot was then injected onto the chromatographic column.

HPLC separation was performed using the Agilent 1200 Series Rapid Resolution system. A COSMOSIL 5C₁₈—MS—IIanalytical column (4.6 mm×250 mm, 5 μm, Nacalai Tesque, Inc., Kyoto Japan) was used and operated at 25 °C. The mobile phase consisted of methanol −2% glacial acetic acid (40:60,v:v) and was delivered by a G1322A pump at a flow-rate of 1.0 ml/min. The G1315BDIO diode array detector was set at 318 nm and all the measurements were performed at 35 °C. A HP1090 chromatographic work station was used for data acquisition. The sensitivity was 0.2 AUFS.

Method validation and documentation were performed according to guidelines set by the United States Food and Drug Administration (FDA) for bioanalytical method validation [2].

For the evaluation of the linearity of the standard calibration curve, the analyses of mangiferin in plasma and eye samples were performed on three independent days using fresh preparations. The calibration curves were prepared over a linear range of 0.50~150.00 μg/ml for plasma and 0.10~10.00 μg/ml for eye samples. Each calibration curve consisted of a double blank sample (without internal standard), a blank sample (with internal standard) and eight calibrator concentrations.
Each calibration curve was constructed by plotting the analyte to internal standard peak area ratio (y) against analyte concentrations (x). The calibration curves were fitted using a least square linear regression model y=ax + b, weighted by 1/x² (the reciprocal of the square of the mangiferin concentration) using the Analyst® software. The resulting a, b parameters were used to determine back-calculated concentrations, which were then statistically evaluated. All calibration curves of mangiferin were constructed before the experiments with correlation values of at least 0.995.

The specificity was defined as: non-interference when mangiferin was being retained from the endogenous plasma components and eye samples, and no cross-interference between mangiferin and internal standard using the proposed extraction procedure and HPLC conditions. Six different lots of blank (mangiferin-free plasma) were evaluated both with and without internal standard to assess the specificity of the method.

The intra- and inter-assay accuracies were expressed as the percentage difference between the measured concentration and the nominal concentration. Intra-assay precision and accuracy were calculated using replicate (n=6) determinations for each concentration of the spiked plasma sample during a single analytical run. Inter-assay precision and accuracy were calculated using replicate (n=6) determinations of each concentration made on three separate days. Accuracy (% Bias)=[(C_obs -C_nom)/C_nom] × 100. The precision (relative standard deviation; RSD) was calculated from the observed concentrations as follows: RSD=[standard deviation (SD)/C_obs] ×100. Accuracy (Bias) and precision (RSD) values within ±15%, covering the range of actual experimental concentrations, were considered acceptable [2].

The extraction efficiency of mangiferin was determined by analyzing replicate sets (n=6) of QC samples: 1.00, 10.00, and 100.00 μg/ml for rat plasma and 0.50, 2.00, and 8.00 μg/ml for rat eyes, representing low, medium and high QCs, respectively. Recovery was calculated by comparing the peak areas of mangiferin added into blank samples and extracted using the protein precipitation procedure, with those obtained from mangiferin spiked directly into post-protein precipitation solvent at three QC concentration levels.

The stability of mangiferin in rat plasma and rat eyes was assessed by analyzing replicates (n=6) of QC samples at concentrations of 1.00, 10.00, and 100.00 μg/ml for rat plasma and 0.50, 2.00, and 8.00 μg/ml for rat eyes. The investigation covered expected conditions during all of the sample storage and process periods, which included the stability data from freeze/thaw, bench-top, autosampler and long-term stability tests. For all stability studies, freshly prepared and stability testing QC samples were evaluated by using freshly prepared standard curves for the measurement. The concentrations obtained from all stability studies were compared with the freshly prepared QC samples, and the percentage concentration deviation was calculated. The analytes were considered stable in mouse plasma and eye when the concentration difference between the freshly prepared samples and the stability testing samples was less than 15%.

To evaluate the suitability of the assay for pharmacokinetic studies, 10, 25 or 50 mg/kg of mangiferin were intravenously administered to rats. Six animals were used in each dosage. Pharmacokinetic calculations were performed using the observed data. All data were subsequently processed using the DAS 2.0 program (Drug and Statistics 2.0). All values obtained were expressed in mean±standard deviation.

The sampling technique and liquid chromatographic detection were then applied to the pharmacokinetic characterization using the DAS 2.0 program (Drug and Statistics 2.0, Hefei, China). After a single intravenous administration of 10, 25, or 50 mg/kg of mangiferin, the concentration-time data conformed to a two-compartment model and the major parameters correlated positively with the dosage given. The dosages for Cmax were 24.75±0.85, 56.77±21.13, and 200.77±32.12 μg/ml, respectively, for AUC(0-t), they were 121.48±12.92, 205.28±104.30, and 577.71±63.30l μg/ml/h, respectively. The pharmacokinetic parameters are listed in Table 4. The mean plasma concentration–time profile is illustrated in Figure 4. Results of the study are consistent with the works by Lai [17] and with other previously published studies [12–16]. Detectable mangiferin was found in the eye, as shown in Table 5.The mangiferin concentration reached 5.69±1.48 μg/ml 0.5 h after intravenous administration (50 mg/kg) and then dropped gradually to 0.30±0.02 μg/ml 5.0h later. The pharmacokinetic parameters of the retina are listed in Table 6.

Because of the blood-retina barrier, a study of whether a drug can reach the target tissue in the eye will play an important role in illustrating the barrier’s mechanism and investigating the impact of the drug. The results of the present study are useful for the administration of mangiferin in eye diseases and understanding mangiferin’s pharmacokinetic behavior.