Separation, Purification And Structure Elucidation Of Active Compounds From Rhodiola Rosea

Rhodiola rosea is a popular plant in traditional medical systems in Eastern Europe and China, with a reputation for eliminating fatigue, improving learning and memory, stimulating the nervous system, enhancing work performance, improving sleep, anti-hypoxia, preventing high altitude sickness and anti-cancer. At present, it is popular as medicine-food plant. In this study, the separation and purification of two major active compounds was investigated in detail. At the same time some monomers were prepared from the extract of R. rosea and their structures were eludicated. On the other hand, a HPLC method was established to determine the six major active compounds of Rhodiola samples from different origins and the established method was used to construct HPLC fingerprints of three species of Rhodiola. The present study can provided help for the deep development of Rhodiola resources and application in pharmaceutical industry.1. Based on the single factor experiments, the effect of three major factors, including temperature,ethanol concentration and solid-liquid ratio, on extraction yield of the two major target compounds from Rhodiola rosea was investigated to optimize the extraction process by central composite design. The regression equation and response surfaee figure from three factor interaction were presented and the obtained optimum extraction conditions were as follows: ethanol concentration 70% (v/v), solid-liquid ratio 1:10 (w/v, g/mL), extraction temperature 75℃, extraction twice 60 minutes for each time. Under the optimum extraction conditions, the yields of rosavin and salidroside can reach 0.80% and 0.37%, respectively. At the same time dynamics and thermodynamics of the extraction were investigated. Thermodynamic parameters, such as standard enthalpy (ΔH0), standard entropy (ΔS0) and standard Gibbs free energy (ΔG0), were evaluated by Van't Hoff equation. The results showed that dynamic data of extraction were well fitted to mathematical model deduced on the basis of the Fick's Second Law and diffusion coefficients of salidroside and rosavin were 0.75×10-9 (m2/s) and 0.41×10-9(m2/s), respectively. The salidroside and rosavin extraction was a endothermic, entropy increase and spontaneous processes according to values ofΔH0,ΔS0 andΔG0.2. In static and dynamic experiments, according to adsorption and desorption properties and selectivity of different resins, HPD-200 was selected to separation and purification of the two major active compounds (salidroside and rosavin) and its adsorption isotherm was well fitted to Langmuir equation. Its adsorption kinetics and thermodynamics for the two compounds were also studied. Sorption data were fitted to pseudo-first-order, pseudo-second-order, intra-particle diffusion models, and found that adsorption kinetics can be described according to the pseudo-second-order model. The intra-particle diffusion model showed a double contribution of the surface and pore diffusivities to the sorption process. The sorption process was a exothermic and spontaneous processes and high temperature was not favorable to adsorption. Dynamic adsorption and desorption tests were carried to optimize the separation process. The optimum separation conditions were as follow: On adsorption, the loading volume was 6.4 mL(sample)/g resin (on dry weight), the concentrations of two comounds in feed solution 1.89mg/mL(salidroside) and 0.24 mg/mL (rosavin), respectively, temperature 25℃,bed height 30 cm, flow rate 1BV/h; on desorption, temperature was 25℃, gradient elution step was firstly 5% and 10%ethanol, respectively, for 2.5 BV, then 30% ethanol for 2.5 BV, and 40% ethanol for 2.5 (BV), finally 60% ethanol for 2BV. After two adsorption and desorption runs, the purity of salidroside and rosavin was increased from 5.57% and 0.69% in Rhodiola rosea crude extract to 91.21% and 15.37% with a overall recovery of 48.82% and 46.52%, respectively. On the base of the conditions optimized above, large-scale preparation of salidroside and rosavin was carried out to allow production of a salidroside-rich fraction with a purity of 90.36% and a overall recovery of 44.75, and a rosavin-rich fraction with a content of 16.02% and a overall recovery of 52.02%.3. Crystallization was performed to purify the salidroside-rich fraction from the large-scale preparation. The optimum crystallization conditions were determined as follows: the salidroside concentration in ethanol solution (w/v) was 20.5mg/mL, temperature 15℃, time 16 h, under which, salidroside crystals with a above 99.18% purity were obtained. The rosavin-rich fraction in large-scale preparation was subsequently subjected to intermediate purification on a silica gel column and a final purification on a ADS-5c resin column . The optimum conditions on the silica gel column were as follows. Silica gel size was 200-300 mesh; packed bed height 25cm; sample amounts 50mg (sample)/g (silica gel); mobile phase chloroform-methanol (4:1, v/v); flow rate 1.5mL/min; elution volume 3BV. After the intermediate purification, the purity of crude rosavin was increased to 50.17 % from 16.02% with a recovery of 82.46%. The optimum conditions on the ADS-5c column were determined as follow. The sample amount was 16.7 mg (sample)/g (dry ADS-5c resin);elution solution 50% ethanol; flow rate 2BV/h. After the final purification, a rosavin monomer with a purity of 98.73% was obtained. In addition, from eluates obtained in the separation process of salidoside and rosavin by macroporous resins, 19 monomers were prepared by silica gel column chromatography and preparative reversed phase high performance liquid phase chromatography after clarification on macroporous resin.4. By UV, HPLC-ESI-MS, 1DNMR (1H-NMR, 13C-NMR, DEPT-135) and 2DNMR (1H-1H-COSY, HMQC and HMBC),15 monomers were elucidated as 4-[4-(β-D- glucopyranosyloxy)phenyl]-2-butanone, [4 - (β-glucopyranosyloxy) phenyl-2-propenoic acid, 2-(4-hydroxyphenyl)ethyl-6-(3,4,5-trihydroxybenzoate)-β-D-glucopyranoside(6'-O-Galloylsalidroside), 7-methoxy-Coumarin, epigallocatechin-3-O-gallate(EGCG),α-phenyl-taloside, 4-hydroxyl-benzoicacid, 3-phenyl-2-propenyl6-O-L-arabinofuranosyl-β-glucopyranoside, 3-phenyl-2-propenyl6-O-L-arabinopyranosyl-β-glucopyranoside, 3-phenyl-2-propenyl 6-O-L- xylopyranosyl-β-glucopyranoside,3-phenyl-2-propenyl-β-glucopyranoside, 2,7-dimethylocta -2,6-diene-1,4-diol (rosiridol), 2,7-dimethylocta-2-6-diene -1,4-diol1-O-β- D-glucopyranoside (rosiridin), 7-methoxycoumarin, p-hydroxyl-phenethylalcohol (tyrosol), of these compounds, 4-[4-(β-D-glucopyranosyloxy)phenyl]-2-butanone,6'-O-galloylsalidroside, 4-[4-(β-glucopyranosyloxy)phenyl]-2-propenoic acid, 7-methoxycoumarin were first identified in Rhodiola rosea.5. By optimizing the extraction, separation and analytical conditions, a sensitive and accurate high performance liquid chromatographic method has been developed for the simultaneous determination of six active compounds in different species of Rhodioa L. The analysis was performed on a Purospher STAR C18 column at 30℃using 20mmol/L aqueous ammonia acetate / methanol gradient system at a flow rate of 1.0mLmin-1 and photodiode array detection (DAD ) at wavelengths 276, 250 and 205 nm, respectively. The method showed good linearity and satisfactory accuracy and recoveries. The newly established HPLC-DAD quantitative method was subsequently utilized to simultaneously determine the quantities of the six active constituents in the samples. The results showed that the contents of the six active compounds ranged from 0.16-11.14 mg/g, 0-2.89 mg/g and 0-1.01 mg/g, 0-6.78 mg/g, 0-1.51 mg/g, 0-7.89 mg/g. The total content of the six active compounds ranged from 1.20-22.88 mg/g and there was a wide variation in content and species of the six active compounds in the samples with different varieties and habitats. Then, the HPLC method was used to conduct fingerprints of three species of Rhodiola. The characteristic analytical fingerprints of them showed 19 common peaks, and molecular weight of 19 peaks were determined by HPLC-ESI-MS method, out of these, according to the characteristic UV spectra, the information of molecular weight and structure provided by ESI–MS, 10 compounds were identified. The resulting fingerprints were examined by similarity analysis, cluster analysis and principal component analysis. The results indicated that there was a good similarity between the same variety of Rhodiola and vice versa, the chromatographic fingerprint can efficiently distinguish Rhodiola from different varieties.