The conditions of experimental design methodology (Table 4) was used to obtain the yields of PLE described at Table 1. The results showed the yield of dry extract from each part of the ginseng plant, and the percentage of β-ecdysone in the dry extract (%EMSβ). The %EMSβ was obtained by multiplying PLE Extraction (w/w) (Table 2) by the Yield PLE (%).Table 1 Experimental and calculated for PLE differents conditions, (×1) temperature (×2) flow rate.Table 2 Significance test, standard-error and the respective confidence interval of the %EMSβ. The %EMSβ values will be the response variable for test planning. The response surface methodology (RSM) was applied to transform variables into factors and optimize the experiment12,13,14. The planning results show the importance of optimizing the experiments and reaffirm their statistical importance. The RSM was applied to determine the levels of factors that affect %EMSβ. Table 2 shows the p-value coefficients, which have a significance effect of 0.05 in our predictive model.The experimental responses encoded by factors × 1 and × 2 can also be represented by general mathematical solutions that locate stationary points, with a p-value of 0.05, such as Eqs. (1–3).$$\% {\text{EMS}}\upbeta {\text{-Leaves}} = \, 0.{499 } + \, 0.0{\text{9 x}}_{{1}} + 0.{1}00{\text{x}}_{{2}}$$
(1)
$$\% {\text{EMS}}\upbeta {\text{-Stem}} = \, 0.{764 } + \, 0.{\text{181 x}}_{{1}} + 0.0{\text{49x}}_{{2}}$$
(2)
$$^{{\% {\text{EMS}}\upbeta {\text{-Roots}} = \, 0.{65}0 \, + \, 0.0{\text{89 x}}_{{1}} + 0.{\text{185x}}_{{2}} }}$$
(3)
Equations (1–3) show the extractions of the parts of the plant, offering a unique analysis of each equation associated with its respective graph. Equation (1) shows that the midpoint of %EMSβ—Leaves is 0.499%. The two factors interfere with the extraction efficiency of the β-ecdysone compound. In Eq. (2), the %EMSβ midpoint was the one with the highest value, 0.764, its efficiency can be increased by increasing the temperature and solvent flow. Equation (3) also presents significance between the factors, and its efficiency is also proportional to the increase in flow and temperature.The estimated p-value of %EMSβ shows that both factors are significant for planning (Table 2). Therefore, it is necessary to evaluate the optimal condition of the analyzed factors. The extraction yield of β-ecdysone can be shown by observing the RSM graph, as shown in Figs. 1 and 2.Figure 1(a) Response surface obtained by the Statistica software, for leaf samples; (b) stem samples; (c) root samples.Figure 2Pareto chart: estimation of the linear effects of the variables. (a) Leaf samples, (b) stem samples, (c) root samples.Figure 1 shows that the closer to the stationary dark red region, the higher the %EMSβ. This means that the best region to obtain %EMSβ is between 2.1 and 2.0 solvent flow and 352 (K) to 354 (K). In Fig. 1a–c, it is observed that the increase in yield is not proportional to the increase in solvent flow and temperature. In Fig. 1b, this relationship has a more proportional tendency. In general, the yields obtained at the two temperatures studied were discrepant, and the energy spent on extractions at 2.0 mL min−1 353 K leads to more extractions. As in the study by Chen15, which obtained the best process condition using 70% ethanol solvent, retention pressure time of 5 min, extraction temperature of 353 K.Figure 2 proves that the two linear factors tested in this study are significant. Finally, the data presented demonstrates that the statistical analysis was valid for the extraction PLE for root, stem, and leaves of ginseng and the results obtained by the software were consistent with the extraction performed. The best part of the plant for extracting β-ecdysone is the stem, as it presents the highest yield compared to the other parts of the plant.Chromatographic analysis-HPLCThe quantification of 20-Hydroxyecdysterone in optimized samples of PLE extracts (root—T25, stem—T11, and leaf—T1) were carried out by by high-performance liquid chromatography (HPLC). The calibration plot was linearly related to β-ecdysone concentration over the range of 30 mg L−1 to 250 mg L−1 and the analytical equation is expressed as: Y = − 479x – 5157 (R = 0.9995)3. The limit of detection (LOD) and the limit of quantification (LOQ) values were calculated using as LOD = 3 s/S and LOQ = 10 s/S, where “s” is the standard deviation of y-intercepts (n = 3) and “S” is the slope of the calibration curve16. Thus, the LOD and LOQ values are 2 mg L−1 and 8 mg L−1, respectively3.Figure 3 shows the leaf, stem, and root extracts chromatograms, as well as the beta-ecdysone standard chromatogram.Figure 3Chromatogram of the leaf extract, stem extract, root extract and β-ecdysone standard.The results for the quantification of β-ecdysone in 50 mg of the extract for the different samples are showed in Table 3.Table 3 Quantification of β-ecdysone in PLE extracts Ginseng.The highest yields (%) found were Stem (3.92), Leaf (1.77), and Root (1.53), respectively. In the root samples, it is observed that there was a lower percentage yield of β-ecdysone than that obtained by Soxhlet extraction (1.63%) reported by17 using a 90:10 (v/v) ethanol: water mixture. However, the result was higher than that reported by Vigo, Narita, and Marques, 200418, where the authors used methanol in the Soxhlet extraction and found 1.07% of β-ecdysone. Compared to the study reported by18, the result obtained in this study showed 0.69% more β-ecdysone. Vardanega et al.19, analyzing the extract of Pfaffia glomerata roots obtained by subcritical water extraction (SWE), obtained a yield of 0.70%, and for the aerial parts of the plant, 0.30% of β-ecdysone. This indicates that the PLE extraction can be more efficient to obtain 20E than the classical method.About the aerial parts of the plant, a study conducted by20 used maceration as the extraction method for flower, root, leaves, and stem samples. The HPLC results for the extracts were 0.82%, 0.66%, 0.60%, and 0.24%, respectively. In the case of stem and leaf samples in this study, the yield was higher than 1.17% for the leaf and 3.06% for the stem. In the latter case, the 3.3% yield can be explained due to the initial accumulation of secondary metabolites in the stem, which are later transferred to the roots during their development. According to21 reserves are relocated to the roots near their maturation stage. Another contributing factor may have been the conditions applied to the PLE extraction, where the use of a 70/30 (v/v) ethanol: water solution, combined with high pressure, facilitated the rupture of stem cells and subsequent release of metabolites such as β-ecdysone.Regarding the leaf samples, the different methodologies showed little percentage variation, with PLE extraction being 0.48% higher. This result is significant when compared to the study conducted by20 which obtained a yield of 0.60% for their samples.The presence of β-ecdysone has been reported by different researchers, and the levels vary from accession to accession, where environmental and anthropogenic factors directly interfere with the accumulation of the compound in different parts of the plant.Analysis FT-Raman spectroscopyFT-Raman spectroscopy was carried out to identify molecular structure in the PLE extracts of P. glomerata obtained from leaves, roots and stems.The Fig. 4A–C show the spectra obtained from PLE extraction of leaf, stem and root, in different extraction conditions, which exhibit similar Raman spectral features. A large band of hydroxyl group are present at the region between 3700 and 3000 cm−122. Well defined bands of CH3 stretching of aliphatic chains from lipids can be observed in the region 3000–2790 cm−1. The presence of the peak at 1730 cm−1 was attributed to C=O stretching of esters, while ~ 1600 cm−1 to amide I23. Between 1500 to 1200 cm−1 indicate deformations of HCH and CH2OH bonds. Stretching vibrations of C–O bonds, with contributions from C–C bonds of aliphatic chains were associated with spectral region 1130–1015 cm−1. Nearby 1050 cm−1 is characteristic band of polysaccharides, indicating the existence of amylose and amylopectin23,24.Figure 4FT-Raman spectra from PLE extracts of leaf, stem, root and 20E samples. Extraction conditions: (A) leaf: T1—flow rate 2.0 mL min−1 and 80 °C; T2: flow rate 1.5 mL min−1 and 80 °C; T4—flow rate 2.0 mL min−1 and 60 °C; T5—flow rate 1.5 mL min−1 and 60 °C. (B) Stem: T10—flow rate 1.5 mL min−1 and 80 °C; T11—flow rate 2.0 mL min−1 and 80 °C; T12—flow rate 2.0 mL min−1 and 60 °C; T13—flow rate 1.5 mL min−1 and 60 °C. (C) Root: T24—flow rate 1.5 mL min−1 and 80 °C; T25—flow rate 2.0 mL min−1 and 80 °C; T26: flow rate 1.5 mL min−1 and 60 °C; T27—flow rate 2.0 mL min−1 and 60 °C. (D) β-ecdysone (20E) standard main peaks.The stem presents a peak at 820 cm−1 characteristic of lipids and sulfolipids23. In the root spectra the Raman band at 1460 cm−1 was observed as characteristic of compounds containing the trace element selenium (Se), such as selenophenes, and it is also characteristic of carbohydrates and lipids23. The peak at 930 cm−1 coud be associated to esters, carboxylic acids, salts, and complexes of diselenocarbamic acid, selenates, selenites, seleninyl halogenates, amino acids, and disaccharides23.Figure 4D shows most intense Raman bands of 20E. Comparing on PLE extracts of leaf, stem and root spectra with 20E spectrum is possible to note the presence of 20E bands in all spectra. This finding is consistent with the results obtained from HPLC analyses (Table 3).
Pressurized liquid extraction followed by high-performance liquid chromatography for determination of beta-ecdysone extracted from Pfaffia glomerata (Spreng.) Pedersen
