Optimization of Quality Control Method and Ethanol Extraction Process of Psoralen and Bergapten in Ficus pandurata

Weiwen PENG, Chang CHEN, Weibo DAI, Yangcun LU, Yingjing WANG

Zhongshan Hospital, Guangzhou University of Traditional Chinese Medicine, Zhongshan 528400, China

Abstract [Objectives] To establish a qualitative identification and content determination method for psoralen and bergapten in the roots and leaves of Ficus pandurata, and the multi-indicator scoring method to optimize the ethanol extraction process for the effective parts of F. pandurata. [Methods] Qualitative identification was carried out by thin-layer chromatography (TLC) and content determination was made by high-performance liquid chromatography (HPLC); the content of psoralen and bergapten and extract yield were used as indicators, to investigate the effects of ethanol volume fraction, solid-to-liquid ratio, extraction time and extraction times on the ethanol extraction process of F. pandurata; multi-indicator scoring method was adopted, orthogonal test method was designed to optimize the extraction process, and verification test was performed. [Results] TLC chart of F. pandurata shows clear spots and good separation; the detected concentrations of psoralen and bergapten in the roots are in the range of 1.02-32.64 μg/mL and shows a good linear relationship with the peak area (r=0.999 7); these components have not been detected in the leaves, and the precision, RSD of stability and repeatability test is less than 2%; the average recovery rate was 99.8%-100.2%, and the RSD value is 1.12%-1.13% (n=6); the optimized extraction process is to use 50% ethanol as the extraction solvent, the solid-to-liquid ratio of 1∶10, reflux extraction of 3 times, and 1.5 h each time; the results of the three batches of verification tests show that the content of the two indicator components obtained is high, and the average value of the total amount is 0.34 mg/g (RSD=0.30%, n=3). [Conclusions] The established quality control method for F. pandurata is simple, easy, accurate and reproducible; the preferred extraction process is stable and feasible, suitable for the extraction and purification of coumarin effective parts in F. pandurata roots, so it is expected to provide references for further development of F. pandurata.

Key words Ficus pandurata Hance, Quality control, Extraction process, Orthogonal experiment, Multi-indicator scoring method

1 Introduction

FicuspandurataHance belongs to genusFicus, family Moraceae. Its medicinal part is root, has a warm nature, sweet slightly spicy taste, and has the effects of removing wind and dampness, detoxification and subsiding swelling, promoting qi flow, blood circulation and regulating menstruation, relaxing sinews and activating collaterals, treating malaria, and good for increasing breast milk supply. In the folk areas, people often treat trauma injury, irregular menstruation, backache, jaundice, breast carbuncle, malaria, mastitis, stomach pain, whooping cough, snake bite, dampness and heat resistance, nausea and no appetite in the form of single medicine or compound medicine ofF.pandurata[2]. The main chemical components ofF.pandurataethanol extract include triterpenes, coumarins, flavonoids, phenolic acids,etc. Among them, only Lu Huiqingetal. separated and identified three coumarin compounds: psoralen, 7-hydroxycoumarin, and bergapten; pharmacological studies have shown thatF.pandurataethanol extract has antioxidant and anti-inflammatory effects, the effect of removing DPPH and hydroxyl free radicals is significantly concentration-dependent,IC50is 118.4 and 192.9 μg/mL[3], and has excellent inhibitory repair effect for liver injury[4-5]. Except this, there has been no relevant report on medicinal material quality control methods and ethanol extraction process research. In view of this, this experiment initially adopted thin-layer chromatography (TLC) and high-performance liquid chromatography (HPLC) to make qualitative identification and content determination of psoralen and bergapten in medicinal parts ofF.pandurata[6-8]. In addition, a single factor experiment and orthogonal experiment were performed to comprehensively investigate and select the optimal extraction process of coumarin effective parts in roots ofF.pandurata, in order to provide scientific basis for the rational development and utilization ofF.panduratamedicinal resources.

2 Instruments and medicinal materials

2.1 InstrumentsAgilent 1100 high performance liquid chromatography (HPLC) system; BS224S electronic precision balance (Sartorius Sartorius Scientific Instruments (Beijing) Co., Ltd., d=0.000 1 g); KQ3200E ultrasonic cleaner (Kunshan Ultrasonic Instruments Co., Ltd.); HH-S4 digital constant temperature water bath (Jintan Medical Instrument Factory); Dongfang-A type direct heating electric constant temperature drying oven (Guangzhou Dongfang Electric Heating Drying Equipment Factory).

2.2 Medicinal materialsThe medicinal material was collected in Jiangmen City, Guangdong Province, identified by Professor Huang Haibo from Guangzhou University of Traditional Chinese Medicine as dry roots ofF.pandurataHance. psoralen reference substance (China National Institute for the Control of Pharmaceutical and Biological Products, batch No.110878-200102); silica gel G plate (Qingdao Haiyang Chemical Co., Ltd.); methanol is chromatographically pure; the remaining reagents are analytically pure, and water is Watsons distilled water.

3 Qualitative identification by thin layer chromatography (TLC)

Separately took 0.5 g of roots and leaves ofF.pandurata, added 20 mL of ethyl acetate, conducted ultrasonic treatment for 15 min, filtered, evaporated the filtrate, and added 1 mL of ethyl acetate to dissolve the residue as the test solution. Separately took psoralen and bergapten reference substance, added ethyl acetate to prepare 2 mg/mL solution, as the reference substance solution. With reference to TLC experiment method in General Rule 0502 of 2015ChinesePharmacopoeia(Volume IV), separately pipetted 4 μL of the above six solutions, spotted on the same silica gel G thin layer chromatography prefabricated plate, developed with n-hexane-ethyl acetate (2∶1) as the developing agent, took out, naturally dried, sprayed with 10% potassium hydroxide methanol solution, and placed the silica gel plate under the UV light (365 nm) for observation. The results showed that at the position corresponding to the chromatogram of the reference substance, onlyF.pandurataroot extract showed the same fluorescent spots, and the separation effect was good, as shown in Fig.1.

Note: 1. psoralen reference substance, 2. F. pandurata root, 3. F. pandurata leaf, 4. bergapten reference substance, 5. F. pandurata root, 6. F. pandurata leaf.

Fig.1 Thin layer chromatograph for Ficus pandurata

4 Determination of content

4.1 Chromatographic conditions and system suitability test

Chromatographic column adopted Agilent ZORBAX Eclipse Plus C18(250 mm × 4.6 mm, 5 μm); mobile phase was methanol-water (55∶45); volume flow rate was 1.0 mL/ min; detection wavelength was 246 nm; column temperature was 30 ℃; injection volume was 10 μL, made records of chromatographs. The number of theoretical plate (calculated as per psoralen) should be not less than 3 000. The chromatograms of reference substance and sample were illustrated in Fig.2.

4.2 Preparation of reference solutionPrecisely weighed 10.2 mg of the psoralen reference substance, placed it in a 25-mL volumetric flask, dissolved with methanol and fixed the volume to the desired value. Precisely weighed 1 mL and placed it in a 10-mL volumetric flask, diluted with methanol to the mark, shook well, and used as psoralen reference stock solution, using the same method to prepare bergapten reference stock solution.

4.3 Preparation of test sample solutionFirstly, pulverizedF.pandurataroot into the coarsest powder, took about 0.5 g crude powder of medicinal materials, precisely weighed, put it in a 100-mL conical flask, precisely added 20 mL of methanol, weighed, ultrasonically extracted for 30 min, cooled down, then weighed again, made up the lost weight with methanol, shook and filtered, took the filtrate and filtered through 0.22 μm microporous filter membrane, to obtain the test sample solution.

4.4 Linear relationship testAccurately absorbed appropriate amount of each reference stock solution prepared under Section4.2, and added methanol to make a reference standard solution containing psoralen and bergapten at a concentration of 1.02, 2.04, 4.08, 8.16, 16.32, 32.64 μg/mL, and shook well. Separately took 10 μL of the above-mentioned reference solutions of different concentrations and injected into the high performance liquid chromatography system, and made records of the chromatographic peak area. Taking the injection concentration (μg/mL) as the abscissa (X) and the peak area as the ordinate (Y), the regression analysis was carried out, and the regression equation of psoralen wasY=76.48X-29.35 (r=0.999 7); the regression equation of bergapten isY=46.61X-18.33 (r=0.999 7). The results showed that the concentration of psoralen and bergapten in the range of 1.02-32.64 μg/mL showed a good linear relationship with the peak area.

Note: 1. psoralen, 2. bergapten (A. reference solution, B. test solution).

Fig.2 HPLC chromatograph for Ficus pandurata

4.5 Precision testPrecisely absorbed the solution of the reference substance of the same concentration, continuously injected and measured 6 times under the condition of Section4.1, and made a record of the peak area of the chromatogram. Results showed that the relative standard deviationRSDof the integrated peak area was less than 1.00%.

4.6 Stability testPrecisely pipettedF.panduratasolution in Section4.3, injected at 0, 1, 2, 4, 6, 8, and 12 h, made a record of the peak area,RSDwas less 1.20%, indicating that the test sample solution was stable within 12 h.

4.7 Reproducibility testApproximately took 0.5 g ofF.pandurataroot coarse powder, 6 pieces, precisely weighed, processed according to the preparation method of the test sample solution, and then injected according to the chromatographic conditions under Section4.1to determine and make a record of the peak area. Results showed thatRSDwas less than 0.90%, indicating that this method has high reproducibility.

4.8 Sample recovery rate testPrecisely weighed 6 pieces ofF.pandurataroot coarse powder (0.170 mg/mL) with the measured content, about 0.5 g for each piece, and added different amounts of psoralen reference substance (0.128, 0.256, 0.513 mg, two pieces for each), prepared the test solution according to the method in Section4.3, determined the content of psoralen according to the above chromatographic conditions, and calculated the recovery rate, the average recovery rate was 99.8%-100.2%.RSDwas 1.12%-1.13% (n=6).

4.9 Determination method of samplesTook 0.5 g ofF.panduratacoarse powder, prepared the test solution according to the method in Section4.2, and determined the content of each indicator according to the chromatographic conditions in Section4.1.

5 Study on ethanol extraction process of F. pandurata root

5.1 Extraction method testPrecisely weighed 5 g ofF.pandurataroot coarse powder, put it in a 100-mL conical flask, added at the solid-to-liquid ratio of 1∶8, extracted with ethanol (200 W, 40 KHz) once, 2 h/time, and measured the content of the indicator components. Precisely weighed 5 g ofF.pandurataroot coarse powder, put it in a 100-mL conical flask, added at the solid-to-liquid ratio of 1∶8, extracted with ethanol reflux once, 2 h/time, and measured the content of the indicator components.

The results showed that the extraction rate of ethanol reflux extraction is superior to ultrasonic extraction, so the ethanol reflux extraction method is preferred for single factor and orthogonal experiment, as indicated in Table 1.

5.2 Extraction solvent testPrecisely weighed 5 g ofF.pandurataroot coarse powder, added at the solid-to-liquid ratio of 1∶8, extracted one time, 2 h/time, to explore the effects of different ethanol volume fractions (40%, 50%, 60%, 70%, 80%, and 90%) on the extraction of psoralen and bergapten content inF.pandurataroots. The results showed that the content of the two components reaches a peak when the ethanol volume fraction is 80%, as shown in Table 2.

5.3 Solvent amount testPrecisely weighed 5 g ofF.pandurataroot coarse powder, 80% ethanol as the extraction solvent, extracted one time, 2 h/time, to explore the effects of different solid-to-liquid ratios (1∶6, 1∶8, 1∶10, 1∶12, and 1∶15) on the extraction of psoralen and bergapten content inF.pandurataroots. The results showed that when the solid-to-liquid ratio is 1∶12, the total amount reaches the peak, as shown in Table 3.

Table 2 Effects of extraction solvent on psoralen, bergapten and their total amount in roots of Ficus pandurata

Table 3 Effects of solid-to-liquid ratio on psoralen, bergapten and their total amount in roots of Ficus pandurata mg/g

5.4 Extraction time testPrecisely weighed 5 g ofF.pandurataroot coarse powder, 80% ethanol as the extraction solvent, added at the solid-to-liquid ratio of 1∶8, extracted one time, 2 h/time, to explore the effects of different extraction time (1, 1.5, 2, 2.5, and 3 h) on the extraction of psoralen and bergapten content inF.pandurataroots. The results showed that when the extraction time is 3.0 h, the total amount reaches the peak, as shown in Table 4.

Table 4 Effects of extraction time on psoralen, bergapten and their total amount in roots of Ficus pandurata

5.5 Extraction times testPrecisely weighed 5 g ofF.pandurataroot coarse powder, 80% ethanol as the extraction solvent, added at the solid-to-liquid ratio of 1∶8, extracted one time, 2 h/time, to explore the effects of different extraction times (1, 2, 3, and 4) on the extraction of psoralen and bergapten content inF.pandurataroots. The results showed that when the extraction times is four, the total amount reaches the peak, as indicated in Table 5.

5.6 Orthogonal experimentPrecisely weighed 5 g ofF.pandurataroot coarse powder, a total of 9 pieces, and conducted experiment according to the L9(34) orthogonal experiment table. Based on the single factor experiment results, the ethanol concentration (A), the amount of solvent (B), the extraction time (C), and the extraction times (D) were selected as the investigation factors, and three levels were selected to optimize the optimal extraction process.

Table 5 Effects of extraction times on psoralen, bergapten and their total amount in roots of Ficus pandurata mg/g

In this experiment, the total content of psoralen and bergapten inF.pandurataroot and extract yield were used as scoring indicators, and the data were processed using a comprehensive weighted scoring method[10]. The full score is set to 100 points, where the weight coefficient of the total content of the ingredients is 0.9 and the weight coefficient of the extraction yield is 0.1. Comprehensive score=Total amount/Maximum total amount×90+Yield rate/Maximum yield rate×10. Factor levels and results were listed in Table 6, and variance analysis was shown in Table 7.

Orthogonal results showed that the analysis of variance is carried out with the smallest C factor as the error term, the primary and secondary effects of each factor are D>A>B>C, and the main factors affecting the extraction of indicator components are the extraction times (P<0.05) and ethanol volume fraction (P<0.10). The extreme difference analysis results showed that the extraction time has the smallest effect on the extraction rate, and there is no significant difference between the levels. Combining theKvalue of each factor, the single factor result and considering the production cost, the optimal extraction process for the composition of psoralen and bergapten fromF.panduratais A1B2C1D3, that is, the volume fraction of ethanol is 50%, and the solid-to-liquid ratio is 1 ∶10, the extraction time is 1.5 h, and the extraction times is 3.

Table 6 L9(34) orthogonal design and intuitive analysis

Table 7 Variance analysis

Note:F0.10(2,2)=9.00,F0.05(2,2)=19.00,F0.01(2,2)=99.00.

5.7 Verification testPrecisely weighed 5 g ofF.pandurataroot coarse powder, a total of 3 pieces, added 50% ethanol at the solid-to-liquid ratio, extracted three times with reflux, 1.5 h/time, and prepared 3 test solutions in parallel. The content of the components of psoralen and bergapten was determined by the chromatographic conditions under Section4.1. The results are shown in Table 8. The total value of the components obtained in the three experiments is about 0.34 mg/g, and theRSDvalue is 0.30%, indicating that this optimal process is stable and reliable.

Table 8 Content determination results of verification test mg/g

6 Discussions

This study initially made a qualitatively identification and content determination of psoralen and bergapten in the roots and medicinal parts ofF.pandurata, to provide reference for the establishment of quality standards. In addition, a previous full-scale scan found that the content of psoralen and bergapten in the roots ofF.panduratawas relatively high, while the remaining components that had been reported were small. According to findings of Cai Qingyuanetal.[11], both psoralen and bergapten have the effect of reducing ALT. Among them, psoralen has a significant anticocaine liver injury effect, specifically reducing ALT and AST in mice serum and increasing LDH content, showing the effect of enzyme lowering and anti-lipid peroxidation.

In summary, this experiment took psoralen and bergapten as indicator components for the follow-up test. The single-factor inspection was used to explore the effects of extraction method ofF.pandurataroot and the factor level of the orthogonal experiment, the extraction time and the extraction times on the extraction process ofF.pandurataroot, so as to select the optimal extraction process. The results showed that the extraction rate of ethanol reflux extraction is the highest, so the ethanol reflux extraction method is preferred for single factor and orthogonal experiment. However, the content of psoralen and bergapten inF.pandurataroot is not regularly increased or decreased with the change of extraction solvent. Therefore, it is more reasonable to select the multi-indicator scoring method to evaluate the quality of the process in this experiment. Using the optimal extraction process, the extraction rate of the indicator components inF.panduratais high and stable, so it is favorable for the further development and utilization ofF.panduratamedicinal materials.