|Home | About | Journals | Submit | Contact Us | Français|
A laboratory-assembled surfactant-assisted pressurized liquid extraction system at room temperature was used for the extraction of glycyrrhizin (GLY) in Radix glycyrrhizae. Environmentally friendly saccharide fatty acid ester such as glucose oleic acid ester is proposed to replace chemical-based surfactants. As the chemical properties of the surfactant obtained were unknown initially, lipase-catalyzed synthesis and liquid chromatography with tandem mass spectrometry were used to ascertain the identity. Surfactant-assisted pressurized liquid extraction (PLE) was carried out dynamically and the extraction efficiencies of the proposed method using different concentration of glucose oleic acid ester were compared with sonication using an organic solvent (ethanol/water, 70:30). The extraction efficiencies of GLY in Radix glycyrrhizae using surfactant-assisted PLE was observed to be higher compared with sonication. The method precision was found to vary from 1.3 to 5.1% (relative standard deviation, RSD, n= 6) on different days. The new method demonstrated the possibility for the extraction to be carried out at room temperature for the production of botanical extracts.
Radix glycyrrhizae is commonly used in traditional medicine (1–6) and is typically consumed as a single plant or as a mixture of medicinal plants (3). Currently, monographs of Radix glycyrrhizae can be found in the United States Pharmacopeia (4), Chinese Pharmacopeia (3), WHO monographs for medicinal plants (1, 2), Japanese Pharmacopeia (5) and Herbal Medicine (Expanded commission E Monographs) (6). From Figure Figure1A,1A, glycyrrhizin (GLY) is the main component found in Radix glycyrrhizae and it should not be <2%, w/w (3). It is responsible for the sweetness of the plant and is commonly used as an expectorant and detoxicant of the spleen (1–3).
The active components from medicinal plants need to be extracted prior to chemical analysis, and selection of the methods of extraction is one of the critical considerations for the production of medicinal plant extracts. Different methods of extraction will significantly affect the chemical profiles obtained. Furthermore, the extraction techniques must be compatible with the analytical instruments used. Traditional methods of extraction include Soxhlet extraction, sonication or heating under reflux. Emerging techniques include microwave-assisted extraction (MAE), supercritical-fluid extraction (SFE), pressurized hot-water extraction (PHWE), pressurized liquid extraction (PLE) and others. Although traditional methods are simpler, they have drawbacks such as long waiting times, high energy consumption, high usage of organic solvents and others (7, 8).
In the move to overcome extraction problems with using organic solvent and heat, systems such as the PLE, PHWE and surfactant-assisted PLE at room temperature were developed (8, 9). The usage of surfactant such as sodium dodecyl sulfate (SDS) and Triton X100 in the extraction media for PLE at room temperature was also discussed for Radix glycyrrhizae and Ephedra sinica (10). Both methods showed results that are comparable with reference methods (9, 10). Currently, the scaling up of PHWE and surfactant-assisted PLE at room temperature for use in the food industry remains to be a challenge. For PHWE, high energy consumption from heating will add to the cost of production. In addition, the use of surfactants may also pose a challenge in the production of botanical extracts for health supplement purposes as additional steps are needed to remove the added surfactants.
Recently, the potential use of saccharide fatty acid esters (such as glucose oleic acid ester, Figure Figure1B)1B) containing hydrophilic groups to replace chemical-based surfactant was reported. This class of compounds is expected to have high solubility in water and strong surface activity. The biodegradability and low toxicity of saccharide fatty acid esters make them ideal for food, cosmetic and pharmaceutical industries (11–14).
Upon obtaining the medicinal plant extracts, various analytical methods can be employed for the determination of the bioactive compounds present. Methods such gas chromatography (GC), gas chromatography mass spectrometry (GC–MS), liquid chromatography (LC), LC with tandem mass spectrometry (LC–MS-MS) and capillary electrophoresis (CE) are widely used for this purpose (1–5). However, the usage of nonvolatile surfactants such as SDS and Triton X100 can be a challenge when characterization of target compounds with LC–MS is needed (8, 10). Currently, the usage of nonvolatile surfactants such as SDS and Triton X100 with PLE had been reported (8, 10). The usage of saccharide fatty acid esters with emerging analytical techniques such as PLE and others had not been noted.
The aim of the current work is to design an environmentally friendly PLE system using saccharide fatty acid ester to extract active components present in Radix glycyrrhizae under ambient conditions. The surfactant provided will be characterized using LC, LC–MS-MS and lipase-catalyzed synthesis. The instrumentation used for PLE will be based on a laboratory-assembled system. The PHWE extracts will be compared with sonication using organic solvent and analyzed using reversed-phase HPLC coupled to a diode array detector.
All reagents were of analytical grade. Sand purified by acid (~50–70 mesh), glycyrrhizic acid (GLY) ammonium salt from glycyrrhiza root (licorice) (purity >70%), 1,1-diphenyl-2-picryl-hydrazyl (DPPH), ascorbic acid, formic acid, acetonitrile, glucose, oleic acid (purity >90%), Novozym 435 (Candida antarctica lipase B immobilized on macroporous resin) and methanol were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ultrapure water (18.2 MΩ cm) was obtained from Thermo Smart2Pure water system (Thermo Scientific, San Jose, CA). Premium grade of saccharide fatty acid ester was obtained from DD Solution (S) Pte Ltd, Singapore.
The reaction mixture for glucose oleic acid ester synthesis starts from glucose and oleic acid. Oleic acid and glucose were fixed at a molar ratio of 1:1 and 0.2 g of lipase was added. This mixture was placed in an open 100-mL beaker covered with watch glass which was agitated on an orbital shaker at 200 rpm and 45°C with 50 mL of acetonitrile. At the indicated time intervals, aliquots were withdrawn for LC–UV and LC–MS-MS analysis (11–14).
Dried roots of Radix glycyrrhizae are purchased from Chinese medical halls in Singapore. To prepare a homogenous sample, the roots were grounded using an IKA A11 basic analytical mill (Staufen, Germany) and passed through a 0.6-mm Retsch sieve (Retsch, Haan, Germany).
Stock solutions of GLY at 1,000 mg L−1 were prepared in methanol. For all analysis, the working solutions of GLY were prepared in the calibration range of 0–100 mg L−1 in water. Linearity of GLY was established between 0 and 400 mg L−1 and correlation coefficient was r2 ≥ 0.999. To quantify the marker compounds in the medicinal plants, a three-point calibration based on the linearity established was used. The system precision (RSD, n = 6) for GLY at different concentrations of the standard prepared was found to be <1.5% on different days.
A Shimadzu LC20 system (Kyoto, Japan) equipped with a quaternary gradient pump, an auto-sampler, a column oven and a UV detector was employed to separate and quantify the amount of successfully extracted markers. The analysis of the extracts was performed with a Waters XTerra hydrophobic and non-polar column (C18, 150 × 3.9 mm, 5 µm) kept at 40°C. Gradient elution was carried out with acidified water (0.1% formic acid) as mobile phase A and acidified acetonitrile (0.1% formic acid) as mobile phase B. The initial condition was set at 20% B and rose to 100% B in 15 min before returning to the initial condition for the next 10 min. A flow rate of 1.0 mL min−1 was used with a detection wavelength at 254 nm. For all experiments, 5 µL of standards and sample extracts were injected.
For the LC–MS assay, a 20-µL aliquot of detergent solution was diluted to 1,000 µL with distilled water. A Shimadzu LC system (Kyoto, Japan) equipped with a binary gradient pump, an auto-sampler, a column oven and a diode array detector was coupled with a Shimadzu LCMS 8040 triple quadrupole mass spectrometer. The gradient elution involved a mobile phase consisting of (A) 0.1% formic acid in water and (B) 0.1% formic acid in acetonitrile. The initial condition was set at 5% of (B), gradient up to 100% in 10 min and returning to the initial condition for 5 min. The oven temperature was set at 40°C and the flow rate was set at 200 µL min−1. For all experiments, 2 µL of samples were injected. The column used for the separation was a reversed-phase Zorbax SB18, 50 × 2.0 mm, 1.8 µm (Agilent Technologies, USA). The ESI–MS was acquired in the positive and negative ion mode. The product ions of m/z ranging from 100 to 800 were collected. The drying gas and nebulizer nitrogen gas flow rates were 10 L min−1 and 1.5 L min−1, respectively. The DL temperature is 250°C and the BH temperature is 400°C.
PLE was performed with a laboratory-assembled system, and the setup was identical to our earlier report (10). It consisted of stainless steel tubes with dimensions 1/16 inch outer diameter (OD) and 0.18 mm inner diameter (ID). The stainless steel extraction cells with the dimensions 10 mm ID × 250 mm were used. The pump used was an isocratic Shimadzu LC10 series pump (Kyoto, Japan). Then, 0.5 g of grounded Radix glycyrrhizae powder was weighed directly in a 20-mL scintillation vial and mixed well with high proportion of sand. The mixture was transferred to an extraction cell.
The premium grade of saccharide fatty acid ester solution obtained was observed to contain 7.59 ± 0.06% w/w of the solute. The extraction efficiency was studied with different concentrations of saccharide fatty acid esters at 0.1, 0.2, 0.4, 1.0 and 2.0% (v/v) prepared from premium grade of saccharide fatty acid ester solution obtained for 40–50 min at a flow rate of 1.5 mL min−1. The extract was collected in a 50-mL volumetric flask and made to the 50 mL mark with water. The solution was filtered with a 0.45-µm membrane before HPLC analysis. During the extraction processes, the pressure in the system was between 8 and 10 bar. The system was washed with fresh extractant for 5 min in between the runs to prevent any carry over products for the new run.
For Radix glycyrrhizae, an accurately weighed 0.5 g of the plant sample was extracted with 3 × 15 mL of ethanol/water (70:30) for 10 min on a sonication bath. The extract was centrifuged at 1,000 rpm, and the supernatant was transferred into a 50 mL volumetric flask. The solution was filtered with a 0.45-µm membrane before HPLC analysis (3).
Stock solutions of DPPH were prepared in methanol, and around 4 mg of DPPH powder was dissolved in 50 mL of methanol. Furthermore, 2.5 mg of ascorbic acid and 6.2 mg of GLY were weighed accurately and dissolved in 50 and 10 mL of water, respectively. Radix glycyrrhizae was extracted using the respective amount of the surfactant and sonication. Different percentage of saccharide fatty acid esters (0.1, 0.2, 0.4, 1 and 2% v/v) was used in PLE at room temperature. The dilution was carried out accordingly using the extractant with the respective percentage of saccharide fatty acid esters added. Two milliliters of each dilution were aliquoted into a new test tube and mixed with 2 mL of DPPH solution. The mixture was left stand for 30 min before taking the absorbance using a UV spectrophotometer. The absorbance was taken at 517 nm, and the readings were tabulated accordingly. The percentage of inhibition was calculated based on the following formula:
Sugar fatty acids esters are nonionic surfactants that are synthesized from sugars and fatty acids based on a lipase-catalyzed reaction in an organic solvent medium and others (11–14). Based on the material safety data sheet and product specifications, the identity and amount of the surfactant provided were unknown. Hence, the characterization of the saccharide fatty acid esters using LC–MS-MS was required. Figure Figure2A2A shows the positive ion ESI–MS spectra of glucose oleic acid ester and a molecular ion at m/z 444.5, and this corresponded to the molecular weight of glucose oleic acid ester. In addition, a series of adducts with glucose oleic acid ester was formed with the observation of m/z at 488.5, 532.5, 576.6 and 620.6. The various adducts were confirmed with the tandem mass spectra obtained for the product ions at m/z 488.5, 532.5, 576.5 and 620.5, respectively. From Supplementary data, Figure S1, it was found that the MSMS spectra observed for the various product ions were consistent with m/z 444.5 (Figure (Figure22B).
The positive ion ESI–MS-MS spectra obtained revealed that the backbones of glucose and oleic acid were present in the compound analyzed. From Figure Figure2B,2B, the m/z of 165.2 indicated that a loss of CH3 from the fragmented glucose side chain had occurred (m/z 165.2: M-RCOO-CH3+H). In addition, a further neutral loss of water from the fragmented glucose side chain generated m/z 147.2 (M-RCOO-CH3–H2O + H). As for the fragmented oleic acid side chain, a neutral loss of CO generated m/z 277.1 (M-C6H10O5-CO + H+Na). Further fragmentation of oleic acid side chain with losses of COO and C2H4 was proposed (M-C6H10O5-COO-C2H4+H + Na) for m/z 233.1. Based on the MS–MS spectra, the chemical structure of the glucose oleic ester present in the solution obtained was confirmed (Figure (Figure11B).
Finally, the target surfactant was synthesized using glucose and oleic acid with lipase as the catalyst. The LC retention time and positive ion ESI–MS-MS of the product synthesized were compared with that obtained from the supplier. Hence, the identity of glucose oleic acid ester was confirmed.
From the chromatograms obtained in Supplementary data, Figure S2, it was clear that the glucose oleic acid ester obtained was essentially hydrophobic. With UV detection at 210 and 254 nm, it was noted that the LC method was highly specific for the various components present in Radix glycyrrhizae with 20% v/v surfactant added. The constituents present in the surfactant used will not co-elute with the various components present in Radix glycyrrhizae. In addition, the presence of GLY in the botanical extracts was confirmed by comparison of the retention time and UV spectra with the standard compound.
For the extraction of bioactive components in medicinal plants, PHWE has been proposed as one of the alternative methods for the elimination and reduction of organic solvents (8, 10). For PHWE, to achieve the extraction efficiency for the target compounds in the medicinal plants, an applied temperature of 80–150°C may be required. It was noted that the higher applied temperature will result in higher energy consumption and possible degradation of the target compounds. Hence, surfactant-assisted PLE at room temperature was proposed as the alternative solution (10). Surfactants are a wide range of compounds that are both synthetic and biological, where all of which have tensioactive properties. However, the use of Triton X100 and SDS as reported will not be ideal for food, cosmetic and pharmaceutical industries (10).
The main parameters that will affect the extraction efficiency of surfactant-assisted PLE at room temperature are flow rate, amount of surfactant added, time for extraction and applied pressure (8, 10). Based on our experiments and other reports, the geometry of the extraction cell and the flow direction of the water had only minor effect on the recoveries of the target analytes in the solid samples (15). In addition, it was noted that a significant proportion of the target compounds will be extracted within the first 30 min using surfactant-assisted PLE (10, 16). To evaluate the effect of time needed for the 1.0% v/v of surfactant solution added in the extractant, the peak area of GLY was monitored at different time points during the course of the extraction. It was noted that a significant portion of GLY was extracted from the medicinal plant within the first 30 min (Supplementary data, Figure S3). Hence, the time for extraction with Radix glycyrrhizae was set at 25–30 min with a flow rate of 1.5 mL min−1. Finally, it was observed that the applied pressure has very little effect on the extraction efficiency (17). As the main objective of the current work was to design a simple system for the extraction of active components from medicinal plants, the effect of applied pressure on the extraction efficiency was not further investigated.
For the extraction of GLY from Radix glycyrrhizae, the effect of surfactant added was evaluated in the range from 0.1 to 2.0% v/v (Figure (Figure3).3). An increasing amount of the surfactant in the extractant from 0.1 to 1.0% v/v was observed to correlate with the higher amount of GLY extracted from Radix glycyrrhizae. A similar trend was observed from GLY extracted from another batch of Radix glycyrrhizae (Supplementary data, Figure S4). However, a higher concentration of surfactant added at 2.0% v/v will result in the degradation of GLY present in Radix glycyrrhizae. As it was reported in our earlier work that the combination of applied pressure with hot water would result in the degradation of certain target compounds in medicinal plants (18), the degradation of GLY was suggested to be due to the effect of applied pressure and higher amounts of surfactant added to the extractant (Figure (Figure44).
Other than monitoring the amount of GLY extracted from Radix glycyrrhizae, an increasing amount of surfactant added into the extractant was found to extract more components and generate a more yellowish solution (Supplementary data, Figure S4A). Hence, a 1.0% (v/v) surfactant solution was selected for comparison with the reference method.
For the validation of new analytical methods with medicinal plants, the extraction efficiencies of the proposed method will need to be compared with reference methods such as sonication, heating under refluxed and others. The bioactive or marker compounds are present naturally, and significant analyte–matrix interaction will be present. It was deduced that the spiking of the target compounds such as GLY into Radix glycyrrhizae will not mimic the real environment (9, 10). Hence, the high recovery obtained in the spiking experiments may not imply that the method has good accuracy.
Based on the optimization experiments performed, the performance of the proposed method was compared with sonication using aqueous ethanol on three different days with three different batches of Radix glycyrrhizae. From Table TableI,I, the extraction efficiencies of surfactant-assisted PLE for different batches of Radix glycyrrhizae were observed to be higher when compared with sonication using ethanol–water (70:30). The addition of glucose oleic acid ester in the extractant for surfactant-assisted PLE further increased the extraction efficiency of GLY in Radix glycyrrhizae as compared with our earlier works on SDS and Triton X100 (11). The method precision was found to vary from 1.3 to 5.1% (RSD, n= 6) on different days (Table (TableII).
The anti-oxidant properties of the botanical extracts obtained were evaluated with the DPPH assay and ascorbic acid was used as the positive control (19). From Supplementary data, Figure S5, it was observed that the presence of glucose oleic acid added will affect the scavenging of DPPH free radicals. Hence, a similar amount of surfactant was added to the various botanical extracts obtained during the dilution process. In addition, it was observed that GLY present in Radix glycyrrhizae has a very low level of percentage inhibition when compared with ascorbic acid (Supplementary data, Figures S5 and S6). From Figure Figure5,5, it was noted that the percentage of inhibition of the various botanical extracts obtained from different amounts of surfactants was rather similar despite of the differences in the color of the extracts obtained and amount of GLY determined.
The extraction of target compounds in medicinal plants will depend on nature of the sample matrix and location of the analytes in the sample core. The mechanism in traditional and emerging methods of extraction is proposed to involve four sequential steps. The first step is the desorption of solutes from the various active sites in the sample matrix in contact with the extractant. Next, it will involve the diffusion of the target compound through the organic part of the matrix to get the matrix–fluid interface. At the same time, the solutes may partition themselves from the sample matrix into the extraction fluid. Finally, the analytes obtained in the extractant can be analyzed with the selected analytical technique (8, 20). It was noted that emerging methods of extraction such as PLE, PHWE and MAE have higher extraction efficiency when compared with traditional methods (7, 9, 21). The enhanced performance of extraction was suggested to be attributed to the higher solubility of target compounds in the extractant. Additionally, the ability of the extractant to be able to penetrate further into the sample core of medicinal plants with the application of microwave energy, higher temperature that is above the boiling point of the extractant used, pressure and others (7, 9, 21).
For surfactant-assisted PLE at room temperature in a dynamic mode, it was noted that the presence of surfactants such as glucose oleic acid ester in the water and the back pressure generated greatly enhanced the desorption and diffusion of GLY from the medicinal plant materials. In our earlier work, it was noted that the extraction efficiency of surfactant-assisted PLE at room temperature with SDS and Triton X100 was comparable to reference methods such as sonication with final analysis by HPLC–UV (3, 10, 16). For the current work, the results obtained showed that glucose oleic acid ester further enhanced the solubility of GLY in the extractant. From Supplementary data, Figure S3, it was deduced that this step took place in the earlier part of the extraction process. For surfactant-assisted PLE at room temperature, the fresh liquid pumped continuously through the sample matrix and sand complete the mass transfer process. Hence, a higher extraction efficiency for surfactant-assisted PLE at room temperature when compared with sonication was observed (Table (TableII).
Most governmental agency regulates the residual solvent in medicinal products because of their unacceptable toxicity or their deleterious environmental effect (22, 23). To our best knowledge, methods for the production of medicinal products vary from manufacturers to manufacturers (24, 25). Hence, standard methods for the production of medicinal products remained to be a challenge. At the same time, manufacturers should also consider alternative methods for the production of medicinal products. The feasibility of extracting GLY present in medicinal plants using surfactant-assisted PLE at room temperature with environmentally friendly surfactant such as glucose oleic acid ester was demonstrated. With its low toxicity as observed in rats (unpublished data), glucose oleic acid ester is a class of surfactants that was found to be highly suitable for the food and pharmaceutical industries. For the current work, the proposed system setup for surfactant-assisted PLE was simpler with no elements of heating and the back pressure was generated by the medicinal plant material and sand used. With the additional step of dispensing the plant samples with sand and accounting of sample nonhomogeneity, good method precision and accuracy were observed. Surfactant-assisted PLE with glucose oleic acid ester added in the extractant proved to be more efficient when compared with sonication with an organic solvent for the extraction of GLY present in Radix glycyrrhizae. The current work showed the feasibility of eliminating organic solvents in analytical methods and extraction processes for production purposes with a biodegradable and environmentally friendly surfactant. In addition, the current work provided the increasingly needed alternative green approach for assuring the quality of botanicals and herbal preparations.
The authors acknowledge the financial support from the SUTD-MIT International Design Centre (IDC). In addition, the technical support from Hwa Chong Institution is acknowledged.