Optimization for decocting later of menthae herba in eungyosan, a herbal formula, using response surface methodology with gas chromatography/mass spectrometry
WooRam Ha, JinHyung Park, JungHoon Kim
Division of Pharmacology, School of Korean Medicine, Pusan National University, Yangsan, Gyeongnam, Korea
Date of Submission  10Mar2017 
Date of Acceptance  25Apr2017 
Date of Web Publication  20Feb2018 
Correspondence Address: JungHoon Kim BusandaehakRo 49, Yangsan, Gyeongnam Korea
Source of Support: None, Conflict of Interest: None  Check 
DOI: 10.4103/pm.pm_97_17
Abstract   
Background: “Decocting later” is important procedure for the extraction of herbal medicines containing volatile compounds. Objective: This study was performed to investigate optimal conditions for “Decocting later” of Menthae herba in Eungyosan (EGS) and correlation between extraction variables and the yields of d/lmenthol, a marker compound of Menthae herba. Materials and Methods: The decocting temperature, total decocting time, and decocting later time were chosen as individual variables, and the yield of d/lmenthol was set as the response value which were calculated by using a BoxBehnken design (BBD). The amount of d/lmenthol was quantified using gas chromatography/mass spectrometry. Results: Response surface methodology (RSM) was used to predict optimal conditions for decocting later of Menthae herba into the formula. Optimal conditions for “Decocting later” from RSM were as follows: 100.63°C of decocting temperature; 82.95 min of total decocting time; 19.11 min of decocting later time. Both decocting temperature and total decocting time showed significant correlation with the yield of d/lmenthol. Conclusions: These results suggest that the decocting temperature and total decocting time were influential factors, and RSM can be applied for optimizing the conditions of “Decocting later” of Menthae herba in EGS. Abbreviations used: KM: Korean medicine; EGS: Eungyosan; GC/MS: Gas chromatography/mass spectrometry; RSM: Response surface methodology; SIM: Selected ion monitoring; LOD: Limits of determination; LOQ: Limits of quantification; RSD: Relative standard deviation; ANOVA: Analysis of variance; BBD: BoxBehnken design.
Keywords: Decocting later, Eungyosan, gas chromatography/mass spectrometry, menthae herba, response surface methodology
How to cite this article: Ha WR, Park JH, Kim JH. Optimization for decocting later of menthae herba in eungyosan, a herbal formula, using response surface methodology with gas chromatography/mass spectrometry. Phcog Mag 2018;14:1721 
How to cite this URL: Ha WR, Park JH, Kim JH. Optimization for decocting later of menthae herba in eungyosan, a herbal formula, using response surface methodology with gas chromatography/mass spectrometry. Phcog Mag [serial online] 2018 [cited 2021 May 12];14:1721. Available from: http://www.phcog.com/text.asp?2018/14/53/17/225673 
Summary
 Gas chromatography/mass spectrometry method developed was applied to quantify the d/lmenthol, a volatile compound in Menthae Herba, in Eungyosan decoction (EGS)
 d/lMenthol was extracted in the chloroform layer of the partition between EGS decoction and chloroform
 A BoxBehnken design produced the predicted response values (yield of d/lmenthol in EGS) from the actual response values with individual variables including decocting temperature, total decocting time, and decocting later time
 Optimal conditions for “Decocting later” of Menthae Herba in EGS obtained from the response surface methodology were 100.63°C of decocting temperature, 82.95 min of total decocting time, and 19.11 min of decocting later time.
Introduction   
Decoction, the extraction method that has been widely used for oral administration, is prepared by boiling plant materials using water. In traditional, Korean Medicine (KM), some herbal medicines containing the compounds which are easily evaporated at higher temperature are added generally 5–10 min before the end of decoction, so called as “Decocting later,” to prevent losing much of volatile characteristics.^{[1],[2],[3],[4],[5]} The extraction efficiency of bioactive compounds from herbal medicines is usually influenced by the extraction factors, such as extraction time, temperature, and material to solvent ratio. In addition, the extraction efficiency of volatile compounds in decoction is affected by decocting later time.^{[6],[7]} Chemical interactions between the components in KM decoction varied the extraction efficiency of herbal medicines, which consequently changed the extraction yields of volatile compounds in KM decoction.^{[8]} Hence, it is necessary to optimize the conditions of decocting later, such as decocting temperature, decocting later time, and total decocting time, to guarantee high efficiency of volatile compounds in herbal medicines.
Eungyosan (EGS) is a KM formula composed of nine herbal medicines: Forsythiae Fructus, Lonicerae Flos, Platycodonis Radix, Menthae Herba, Phyllostachys Folium, Glycyrrhizae Radix et Rhizome, Schizonepetae Spica, Glycine Semen Preparatum, and Arctii Fructus. EGS has been used to treat the symptoms caused by common cold, including headache, fever, sore throat, chills, and dipsia; and those caused by influenza infection or other viral infections, including handfootmouth disease, esophagitis, pneumonia, acute tonsillitis, and mumps.^{[9],[10],[11],[12]} Menthae herba, one of the herbal medicines consisting EGS, contains characteristic volatile compound, d/lmenthol, which can be easily vaporized when decocted together with other herbal medicines at initial decocting time.^{[13]} Therefore, Menthae herba is recommended to be added separately at nearly end of decoction, which can assure high quality of decoction by preventing it from losing bioactive volatile compounds.
In the present study, various types of EGS decoctions were prepared by differing the extraction variables such as decocting temperature (80, 90, and 100°C), total decocting time (90, 120, and 150 min), and decocting later time (5, 10, and 15 min). d/lMenthol, the volatile compound in EGS decoction, was quantified by gas chromatographymass spectrometry (GC/MS), a widely used analytical tool for volatile compounds.^{[14],[15]} Optimum extraction conditions for d/lmenthol in EGS decoction and the correlations between extraction yield of d/lmenthol and extraction variables were investigated using response surface methodology (RSM), which enables researcher to find optimal conditions for the extraction and evaluate the interaction of individual extraction variables.^{[16],[17]}
Materials and Methods   
Reagents and herbal materials
HPLCgrade methanol and chloroform were purchased from J. T. Baker Inc. (Center Valley, PA, USA). D/lmenthol (≥95%) was purchased from ChemFace (Wuhan, Hubei, China). Nine herbal medicines of EGS, Forsythiae Fructus (3.0 g), Lonicerae Flos (3.0 g), Menthae Herba (1.8 g), Arctii Fructus (1.8 g), Platycodonis Radix (1.8 g), Glycine Semen Preparatum (1.5 g), Glycyrrhizae Radix et Rhizome (1.5 g), Phyllostachys Folium (1.2 g), and Schizonepetae Spica (1.2 g), were purchased from a herbal company (Kwangmyungdang Medicinal Herbs; Ulsan, Korea).
Preparation of standard solution
The standard compound was accurately weighed and dissolved in methanol to make stock solution at concentration of 1,000 μg/mL. The stock solution was diluted to produce the working solution for constructing the calibration curve.
Preparation of Eungyosan decoction and samples
Nine herbal medicines of EGS were grinded up and accurately weighed. The mixture of herbal medicines, except for Menthae Herba which was prepared to be added before the end of decoction, were decocted with 300 mL of distilled water and they were boiled using reflux extracting system with temperaturecontrollable heating mantle (MSDM, MTops, Korea). EGS decoction was filtered through a testing sieve (500 μm, Chunggye, Gyeonggido, Korea). The filtered decoction was made up to 250 mL and centrifuged at 3500 rpm in 5 min to separate pellet from the EGS decoction. Supernatant was partitioned with chloroform at 1:1 ratio. The chloroform layer was filtered through 0.2 μm syringe filter (BioFact, Daejeon, Korea) and analyzed by GC/MS to quantify d/lmenthol.
Gas chromatographymass spectrometry conditions
The GC/MS system consisted of Agilent 7890 GC (Agilent Technologies, CA, USA) equipped with 5975C inert MSD tripleaxis detector (Agilent Technologies, CA, USA), autosampler. DB1 capillary column (50 m × 250 μm, 0.25 μm, Agilent Technologies, CA, USA) was used to separate the compounds. Separation was achieved with the temperature program of [Table 1]. N_{2} was used as the carrier gas at a constant flow rate of 0.3 mL/min. Samples were injected at a split ratio of 1:20, and the injection temperature was held at 250°C. The mass spectrometer was operated in scan mode (50–170 m/z) and SIM mode.
Method validation
The stock solution was diluted at six levels to construct calibration curves in which the xaxis was the concentration of marker compound, and the yaxis was the area of the marker compound. The linearity was determined by coefficient of determination (r^{2}). The limits of determination (LOD) and limits of quantification (LOQ) were evaluated as follows: LOD = 3.3 × standard deviation (SD/)S and LOQ = 10 × SD/S, where SD is the SD of the response and S is the slope of the calibration curve. The precisions were measured at three concentrations of the standard compounds (low, medium, and high levels) during a day (intraday) and three successive days (interday), which was represented by the value of the relative SD (RSD). The recovery test was used to evaluate the accuracy of the method. Two known amounts of marker compound (low and high) were added to the samples, and the recovery was evaluated as follows: Recovery (%) = ([detected concentration – initial concentration/spiked concentration] ×100). The repeatability was determined by RSD value of the retention times and the absolute areas of marker compound (n = 6).
Experimental design and statistical analysis
To determine the optimum conditions for extraction of d/lmenthol in ESG, preliminary ranges of the extraction variables, decocting temperature, total decocting time, and decocting later time, were investigated using a singlefactor test. A threelevelthreefactor BBD was employed to determine the optimal conditions for the extraction. The experimental data obtained from the BBD were fitted to a secondorder polynomial model, and the regression coefficients were obtained. The equation is as follows:
Where Y is the estimated response, β_{0}, β_{j}, β_{jj,} and β_{ij} are the regression coefficients for intercept, linearity, square, and interaction terms, respectively. X_{i} and X_{j} are the independent variables, which were coded.
The fitness of the secondorder polynomial model was expressed by the lack of fit and coefficient of determination (r^{2}). The value of ttest and P value resulting from the analysis of variance (ANOVA) were calculated to confirm the significance of the regression coefficients, which was determined at P < 0.05, 0.01, or 0.001. The interaction and influence of the three variables on the yield of d/lmenthol were represented as threedimensional (3D) response surface plots and 2D contour plots, on which the optimum extraction condition was obtained. The opensource software R (version 3.1.1; The R Foundation for Statistical Computing) was used to generate the experimental design, statistical analysis, and regression model.
Results and Discussion   
Method validation
d/lMenthol in chloroform fraction partitioned from EGS decoction was reasonably separated and detected on total ion chromatogram [Figure 1]. Regression equation of calibration curve was y = 117,476.973x− 105,112.000 with concentration range from 3.16 μg/mL to 50.00 μg/mL. The linearity d/lmenthol represented as the correlation coefficient (r^{2}) was 0.9985, and LOD and LOQ were 1.08 and 3.27 μg/mL, respectively. The intra and interday precisions of d/lmenthol which were represented as RSD value were 0.55%–4.37% for intraday precision and 0.86%–6.11% for interday precision [Table 2]. The recoveries of d/lmenthol at low and high concentrations were 84.06% and 80.51% with RSD values 15.04% and 1.81%, respectively [Table 3]. The repeatability was 0.01% for retention time and 4.61% for absolute peak area, respectively.  Figure 1: Total ion chromatograms of d/lmenthol standard solution (a), the chloroform fraction of Eungyosan decoction (b), and mass spectrum of d/lmenthol (c)
Click here to view 
Model fitting
The range of decocting temperature (X_{1}, 80°C–100°C), total decocting time (X_{2}, 90–150 min), and decocting later time (X_{3}, 5–15 min) were determined by preliminary experiments using singlefactor tests (Data not shown). In this study, a threelevelthreefactor BBD composed of 15 experiments was employed, and three replicates (run order 2, 5, and 8) were measured for calculating the pure error sum of squares [Table 4]. The yield of d/lmenthol was set as the response. Three levels of each variable were coded as −1, 0, and 1.  Table 4: BoxBehnken design and the response values for yields of d/lmenthol in Eungyosan
Click here to view 
Multiple regression analysis on the experimental data was expressed by secondorder polynomial equation using coded variables as follows:
Y = 12.987696 + 1.28109X_{1 }+_{ }0.69903X_{2 }−_{ }0.050757X_{3 }−_{ }0.23248X_{1}X_{2 }
−_{ }0.5333488X_{1}X_{3 }+_{ }0.449382X_{1}X_{3 }−_{ }0.17086X_{1}X_{1 }+_{ }
2.704081X_{2}X_{2 }+_{ }0.198502X_{3}X_{3}
Where Y is the yield of d/lmenthol (μg/mL), and the coded variables X_{1}, X_{2}, and X_{3} represent extraction variables, including decocting temperatures, total decocting time, and decocting later time.
[Table 5] shows the regression coefficients of the predicted quadratic polynomial model of d/lmenthol in EGS decoction. The significance of each coefficient was estimated using Student'S ttest and the P value, which larger tvalue and smaller P value represent significant corresponding variables.^{[18]} It was observed that decocting temperature (X_{1}) was significant in linear term (P < 0.05) and total decocting time (X_{2}) was significant in quadratic term (P < 0.01). Coefficients of the other terms (X_{2}, X_{3}, X_{1}:X_{2}, X_{1}:X_{3}, X_{2}:X_{3}, X_{1}:X_{1}, and X_{3}:X_{3}) were not significantly influential on the model (P > 0.05). It was observed that decocting temperature (X_{1}) showed significant influence to the extraction yield of d/lmenthol in linear term (P < 0.05) while total decocting time (X_{2}) showed significant influence in quadratic term (P < 0.01) from the regression coefficients.  Table 5: Regression coefficients of the predicted quadratic polynomial model of d/lmenthol in Eungyosan
Click here to view 
The ANOVA was performed to evaluate the fitness of quadratic polynomial model for the extraction of d/lmenthol [Table 6]. Optimization of the fitted response surface can be obtained from an adequately fitted model which approach to the true system.^{[18]} The coefficient of determination (r^{2}) was 0.8764 with no significant lack of fit at P > 0.05, which indicates that the predicted model could explain 87.64% of the results.  Table 6: Analysis of variance for the fitted quadratic polynomial model for the extraction of d/lmenthol in Eungyosan
Click here to view 
Analysis of response surface
The predicted yield of d/lmenthol calculated from secondorder polynomial equation was visualized through 3D response plots and 2D contour plots by the levels of two independent variables with other variable set at the zero level.
As shown in [Figure 2], the yield of d/lmenthol increased with increasing decocting temperature (X_{1}). However, a significant correlation between decocting temperature (X_{1}) and total decocting time (X_{2}) was not observed. Increase of decocting temperature (X_{1}) also increased the yield of d/lmenthol, however, the interaction between decocting temperature (X_{1}) and decocting later time (X_{3}) was not significant [Figure 3]. The yield of d/lmenthol during total decoction time (X_{2}) decreased from 90 min to about 120 min while increased from 120 min to 150 min. However, it was not observed that significant correlation between total decocting time (X_{2}) and decocting later time (X_{3}) [Figure 4].  Figure 2: Response surface plot and contour plot of decocting temperature (°C, X1) and total decocting time (min, X2)
Click here to view 
 Figure 3: Response surface plot and contour plot of decocting temperature (°C, X1) and decocting later time (min, X3)
Click here to view 
 Figure 4: Response surface plot and contour plot of total decocting time (min, X2) and decocting later time (min, X3)
Click here to view 
These results demonstrate that the extraction of d/lmenthol from EGS decoction was influenced by extraction conditions of decocting later significantly or nonsignificantly as complementary and neutralizing action among the compounds in herbal formula commonly exists during decoction.^{[8]} Furthermore, the correlation between the extraction efficiency of volatile compounds and decocting later time indicates that herbal medicines containing volatile compounds are required to be decocted later.^{[6],[7]}
Optimization and verification of extraction by response surface methodology
Optimized conditions for decocting later of Menthae Herba in EGS decoction were obtained through RSM: 100.63°C of decocting temperature; 82.95 min of total decocting time; and 19.11 min of decocting later time. The optimized extraction yield of d/lmenthol was predicted to be 13.54 μg/mL. The modified conditions were 100°C of decocting temperature, 80 min of total decocting time, and 20 min of decocting later time and were applied. The extraction yield of d/lmenthol from modified conditions was 12.23 ± 0.16 μg/mL, which is slightly different from those calculated from optimized extraction conditions [Table 7].  Table 7: The predicted value of responses at the optimum conditions and the experimental value of responses at the modified conditions
Click here to view 
Conclusions   
In the present study, the optimal conditions of decocting later were investigated: 100.63°C of decocting temperature; 82.95 min of total decocting time; 19.11 min of decocting later time. In addition, the interactions between the extraction variables, such as decocting temperature, total decoction time, and decocting later time, on the yield of menthol in EGS was also observed.
Acknowledgements
This research was supported by the “2015 KIOM Undergraduate Research Program” funded by Korea Institute of Oriental Medicine.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References   
1.  Lee YB, Kook YB, Choi SM. Study on the most effective Yinqiaosan by GC (Gas Chromatography). J Korean Med Classics 2006;19:2026. 
2.  Kim YK, Cho SI. A study on the methods of decocting and taking prescriptions in Sanghanron. Herb Formul Sci 2000;8:1137. 
3.  Kim YK, Kim CS, Cui X. The decocting and taking methods of herbal medicines. Integr Med Res 2004;10:6372. 
4.  Lee NH, Yu YB, Ha HK, Lee HY, Jung DY, Choi JY, et al. A study on the development of medication teaching of herbal formulas based on evidence. J Korean Med Sci 2007;28:144 –55. 
5.  Zhang PY. Discussions about the classification and decoction method of the afterdecoction medicine of traditional Chinese medicine decoction. J Henan Univ Tradit Chin Med 2010;30:9289. 
6.  Thangam R, Suresh V, Kannan S. Optimized extraction of polysaccharides from Cymbopogon citratus and its biological activities. Int J Biol Macromol 2014;65:41523. 
7.  Shu Y, Chen Y, Qin K, Liu X, Cai B. A study on the chemical compositions of the yinqiaosan ( Lonicerae and Forsythiae powder) at different time of laterdecoction by gas chromatography mass spectrometry. Pharmacogn Mag 2016;12:1348. 
8.  Kim HU, Ryu JY, Lee JO, Lee SY. A systems approach to traditional oriental medicine. Nat Biotechnol 2015;33:2648. 
9.  Kim SC, Kim SH, No SH, Park SD, Park SK, Seo BI et al. Herbal Formula. Younglimsa, Seoul, Korea; 1999;835. 
10.  Wang C, Cao B, Liu QQ, Zou ZQ, Liang ZA, Gu L, et al. Oseltamivir compared with the Chinese traditional therapy maxingshiganYinqiaosan in the treatment of H1N1 influenza: A randomized trial. Ann Intern Med 2011;155:21725. 
11.  Tao Z, Yang Y, Shi W, Xue M, Yang W, Song Z, et al. Complementary and alternative medicine is expected to make greater contribution in controlling the prevalence of influenza. Biosci Trends 2013;7:2536. 
12.  Luo FY, Wang SM. Clinical application overview of the Yin Qiao San. Chin J Integr Tradit West Med 2009;18:37813. 
13.  Kim JH, Seo CS, Shin HK. Simultaneous determination of (−)menthone and (−)menthol in Menthae herba by gas chromatography and principal component analysis. Nat Prod Sci 2010;16:1804. 
14.  Lin R, Tian J, Huang G, Li T, Li F. Analysis of menthol in three traditional Chinese medicinal herbs and their compound formulation by GCMS. Biomed Chromatogr 2002;16:22933. 
15.  Gherman C, Culea M, Cozar O. Comparative analysis of some active principles of herb plants by GC/MS. Talanta 2000;53:25362. 
16.  Lee AY, Kim HS, Jo JE, Kang BK, Moon BC, Chun JM, et al. Optimization of extraction condition for major iridoid compompounds in fruit of Corni ( Corus officinalis) by UPLCPDA using response surface methodology. Food Sci Biotechnol 2012;21:10239. 
17.  Sun Y, Liu J, Kennedy JF. Application of response surface methodology for optimization of polysaccharides production parameters from the roots of Codonopsis pilosula by a central composite design. Carbohydr Polym 2010;80:94953. 
18.  Kim SH, Kim HK, Yang ES, Lee KY, Kim SD, Kim YC, et al. Optimization of pressurized liquid extraction for spicatoside A in Liriope platyphylla. Sep Purif Technol 2010;71:16872. 
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]
