Metabolomic Approach for Discrimination of Four- and Six-Year-Old Red Ginseng (Panax ginseng) Using UPLC-QToF-MS

Abstract

Panax ginseng C.A. MEYER is one of the most popular medicinal herbs in Asia and the chemical constituents are changed by processing methods such as steaming or sun drying. Metabolomic analysis was performed to distinguish age discrimination of four- and six-year-old red ginseng using ultra-performance liquid chromatography quadruple time of flight mass spectrometry (UPLC-QToF-MS) with multivariate statistical analysis. Principal component analysis (PCA) showed clear discrimination between extracts of red ginseng of different ages and suggest totally six discrimination markers (two for four-year-old and four for six-year-old red ginseng). Among these, one marker was isolated and the structure determined by NMR spectroscopic analysis was 13-cis-docosenamide (marker 6-1) from six-year-old red ginseng. This is the first report of a metabolomic study regarding the age differentiation of red ginseng using UPLC-QToF-MS and determination of the structure of the marker. These results will contribute to the quality control and standardization as well as provide a scientific basis for pharmacological research on red ginseng.

Korean ginseng (Panax ginseng C.A. MEYER) is one of the most widely used and acclaimed herbs in the world.^1–3) Traditionally, the root of P. ginseng, the most used and valuable part, is physically subdivided into three groups; the main root, lateral root, and root hairs.¹⁾ It has already been reported that the chemical constituents and efficacy of each part of the ginseng root are quite different.¹⁾ It has been processed to make white ginseng (by air-drying the roots after peeling or not peeling) and red ginseng (by steaming the roots at 98–100°C without peeling) to enhance its preservation and efficacy. Red ginseng is more common as an herbal medicine, because steaming induces changes in the chemical constituents and enhances the biological activities of ginseng.⁴⁾

P. ginseng is generally cultivated for four or six years in the field before harvest. Thus, four and six year cultivated P. ginseng is a common item in the Korean ginseng market. However, six year cultivated P. ginseng and products thereof (including white ginseng and red ginseng) are produced and consumed much more than others. Because the cultivation age and harvest season have a significant effect on the quality and efficacy of ginseng, products of six year cultivated P. ginseng are also more expensive than products of four year cultivated P. ginseng. Additionally, a low survival rate and higher cultivation costs contribute to the expensive price of six year cultivated P. ginseng. The accurate determination of the cultivation age of ginseng is a very important problem in the market, although the cultivation age of ginseng can hardly be determined by the physical appearance alone, such as by the number of stem vestiges in rhizome.⁵⁾

In recent years, attempts have been made to overcome this difficulty in discriminating between the two categories of products using metabolomics by various instrumental methods such as NMR,^6–8) GC-MS^9,10) and LC-MS.^11,12) However, the studies regarding red ginseng were unable to produce reliable results. In this study, we focused on reliable methods to discriminate the cultivation ages of red ginseng for quality control and prevention of adulteration of the market.

Herein, we report metabolomics by ultra performance liquid chromatography quadruple time-of-flight mass spectrometry (UPLC-QToF-MS) for discriminating between four- and six-year-old red ginseng. We proposed one potential marker substance was isolated from six-year-old red ginseng and identified by NMR spectroscopic analyses. As previously mentioned, age discrimination of ginseng samples has been studied previously,^6–12) but only considered fresh ginseng; this is the first report of marker substances for the discrimination of four- and six-year-old red ginseng. These results are also helpful for quality control and standardization and to provide a scientific basis for pharmacological research of red ginseng.

Experimental

Plant Materials

The four- and six-year-old ginseng (P. ginseng C.A. MEYER) used in this study were cultivated in the same area (I-cheon, Gyeonggi-do, Korea; 37°16′19.18″N, 127°26′5.36″E) by the same farmer. It was prepared with only main roots to an exact choice of age differentiation between two groups because of that ginseng root consisted of only main root until two-year cultivation. Fresh ginseng roots were prepared by steaming and drying to make red ginseng in the red ginseng manufacturing factory of Korea Ginseng Coporation (Buyeo, Chung-nam, Korea). The processing method was as follows. Washed fresh ginseng roots were steamed at 100°C for 8 h and then dried at 95°C for 24 h. A secondary drying process was performed in a closed chamber at 60°C for 5 d. When we performed UPLC-QToF-MS for determining the experimental conditions, total roots extraction did not show a clear principal component analysis (PCA) plot (see Supplementary materials, S-13–S-16).

Chemicals and Reagents

Leucine-enkephalin was purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). Phosphoric acid was purchased from Junsei Chemical Co., Ltd. (Tokyo, Japan). HPLC-grade acetonitrile and methanol were purchased from Merck (Darmstadt, Germany). All distilled water used in this experiment was purified by the Milli-Q gradient system (Millipore, Bedford, MA, U.S.A.) and the resistance value was measured as 18 MΩ prior to use.

Sample Preparation for Metabolomics

For the UPLC-QToF-MS analysis, the powdered red ginseng samples were extracted in a similar manner to our previous studies using ultrasonic extraction.¹³⁾ Three hundreds milligram of each powered red ginseng was weighed in a centrifugal tube (15 mL, PP-single use; BioLogix Group, Jinan, Shandong, China) and shaken vigorously after the addition of 6 mL of methanol. The extraction was then placed in an ultrasonic cleaner (60 Hz; Wiseclean, Seoul, Korea) for 30 min. The solution was centrifuged (Legand Mach 1.6R; Thermo, Frankfurt, Germany) at a speed of 3000 rpm for 10 min and an aliquot of the supernatant solution was filtered (0.2 µm; Acrodisk, Gelman Sciences, Ann Arbor, MI, U.S.A.) and injected into the UPLC system (Waters Co., Milford, MA, U.S.A.).

UPLC-QToF Analysis

The instrumental analysis was performed by UPLC using an ACQUITY BEH C18 column (100×2.1 mm, 1.7 µm; Waters Co.) on a Waters ACQUITY UPLC system with a binary solvent manager. The column temperature was 40°C. The binary gradient elution system consisted of 0.01% formic acid in water (A) and 0.01% formic acid in acetonitrile (B). Separation was achieved using the following protocol: 0–0.5 min (15% B), 14.5 min (30% B), 15.5 min (32% B), 18.5 min (38% B), 24.0 min (43% B), 27.0 min (55% B), 27.0–31.0 min (55% B), 35.0 min (70% B), 38.0 min (90% B), 38.1 min (15% B), 38.1–43.0 min (15% B). The flow rate was 0.4 mL/min and the sample injection volume was 2.0 µL.

Metabolite profiling of red ginseng was performed by coupling a Waters ACQUITY UPLC system to a Waters Xevo QToF mass spectrometer (Waters MS Technologies, Manchester, U.K.) with positive mode of electrospray ionization (ESI⁺) interface. The source and desolvation gas temperatures were kept at 400 and 120°C, respectively. N₂ was used as the nebulizer and desolvation gas. The flow rates of the nebulizer gas and cone gas were set at 800 and 50 L/h, respectively. The capillary and cone voltages were adjusted to 2150 and 40 V, separately. The mass accuracy and reproducibility were maintained by infusing lockmass (leucine-enkephalin, 200 pg/L) through Lockspray™ at a flow rate of 7 µL/min. Centroided data were collected for each sample from 150 to 1500 Da and the m/z values of all acquired spectra were automatically adjusted during acquisition based on lockmass and dynamic range enhancement. The accurate mass and molecular formula assignments were obtained with MassLynx™ 4.1 software (Waters MS Technologies).

For finding applied condition, we already performed negative and positive mode with two type extractions of each red ginseng (50% aq. MeOH and MeOH). However, only the above condition resulted in PCA analysis (positive mode, MeOH ext of main roots; see Supplementary materials, S-13–S-16).

Multivariate Analysis

To evaluate the potential characteristic components of four- and six-year red ginseng, the ESI⁺ raw data of all samples were calculated with MassLynx™ application manager version 4.1 (Waters MS Technologies). The method parameters were as follows: retention time range, 2–37 min; mass range, 150–1500 Da; mass tolerance, 0.07 Da. The parameters of peak widths at 5% height and peak-to-peak baseline noise were automatically calculated for peak integration. Additionally, the noise elimination level was set to 0.10, and the retention time tolerance was set to 0.2 min. No specific mass or adduct ions were excluded, but the isotopic peaks were removed in the multivariate analysis. For data analysis, a list of the intensities of the detected peaks was generated using the pair of retention time (t_R) and mass data (m/z) as the identifier of each peak. A temporary ID was assigned to each of these t_R-m/z pairs for data adjustment that was based on their chromatographic elution order of UPLC. Upon completion, the correct peak intensity data for each t_R-m/z pair for all samples was sorted in a table. Ions from different samples were considered to be the same when they showed the identical t_R and m/z values. MarkerLynx™ (Waters MS Technologies) was used for the normalization of each detected peak against the sum of the peak intensities within that sample. The resulting data consisted of a peak number (t_R-m/z pair), sample name and ion intensity. Then, the resultant data sets were analyzed by principal component analysis (PCA) and orthogonal partial least squared discriminant analysis (OPLS-DA) using the MarkerLynx™.

Isolation of Selected Marker Substances

The six-year-old red ginseng samples (P. ginseng C.A. MEYER; dry weight, 5 kg) were macerated, and extracted repeatedly with MeOH (20 L×2) at 40°C. The combined crude extract (800 g) was partitioned between dichloromethane and H₂O, and then the dichloromethane layer (35 g) was repartitioned between 15% aqueous MeOH (11 g) and n-hexane (24 g). The n-hexane layer was subjected to silica vacuum flash chromatography using sequential mixtures of MeOH and H₂O as eluents (elution order: 100% n-hexane, 10, 20, 30, 40, 50% ethyl acetate in n-hexane, 100% ethyl acetate) and finally using MeOH. Guided by the results of LC/MS and UPLC-QToF-MS analyses, fraction 7 containing secondary metabolites were combined (1.5 g), purified by silica semi-preparative HPLC (YMC silica column, 1×25 cm, 100% EtOAc), and then purified by silica semi-preparative HPLC (YMC-silica column, 5% EtOAc in n-hexane) to yield 0.5 mg of compound 1 (marker 6-1) as a white solid.

General Experimental Procedures for Compound 1

Optical rotation was measured on a JASCO P-1020 polarimeter (Jasco, Tokyo, Japan) using a 1 cm cell. The UV spectrum was recorded on a Hitachi U-3010 spectrophotometer (Hitachi High-Technologies, Tokyo, Japan), and the IR spectrum was recorded on a JASCO 4200 FT-IR spectrometer (Jasco) using a ZnSe cell. NMR spectra were recorded in CDCl₃ containing Me₄Si as an internal standard on Bruker Avance 600 spectrometers (Bruker, MA, U.S.A.). Proton and carbon NMR spectra were measured at 600 and 150 MHz, respectively. High-resolution (HR)-ESI-QTOF-MS/MS mass spectrometric data was obtained on an Agilent Technologies 6530 Accurate-Mass Q-TOF LC/MS spectrometer (Santa Clara, CA, U.S.A.) with an Agilent Technologies 1260 series HPLC. Low-resolution (LR)-ESI-MS data were recorded on an Agilent Technologies 6130 quadrupole mass spectrometer with an Agilent Technologies 1200 series HPLC. HPLC was performed on a Shimadzu LC-6AD equipped with a Shimadzu RID-10A refractive index detector. All solvents were spectroscopic grade or distilled in a glass prior to use.

Compound 1

White amorphous solid; [α]_D²⁵ −3.9 (c=0.5, MeOH); IR (NaCl) ν_max 3360, 2930, 2840, 1660, 1460 cm¹; UV (MeOH) λ_max (log ε) 203 nm (2.28); ¹H-NMR (CDCl₃) δ: 5.35 (2H, dt, J=9.6, 4.7 Hz, H-13, 14), 2.22 (2H, t, J=7.9 Hz, H-2), 2.01 (2H, dtd, J=10.0, 9.5, 6.5 Hz, H-12), 1.64 (2H, tt, J=7.0, 7.3 Hz, H-3), 1.27 (30H, m, H-4–11, H-15–21), 0.88 (3H, t, J=7.3 Hz, H-22); ¹³C-NMR (CDCl₃) δ: 175.6 (C-1), 130.1 (C-14), 130.0 (C-13), 36.1 (C-2), 32.1 (C-20), 29.9 (C-15), 29.7–29.4 (C-4–11, C-16–19), 27.4 (C-12), 25.7 (C-3), 22.8 (C-21), 14.3 (C-22); HR-FAB-MS m/z 338.3425 [M+H]⁺ (Calcd for C₂₂H₄₄ON, m/z 338.3417).

Results and Discussion

Non-targeted Component Analysis

Fifty roots of both four- and six-year-old red ginseng, which were cultivated on the same farm, were analyzed for potential marker substances using UPLC-QToF-MS. Additionally, the UPLC-QToF-MS data were used in a non-targeted component analysis to obtain potential marker substances for discrimination of four- and six-year-old red ginseng. Red ginseng samples were analyzed using a 43 min gradient method, as in our previous research,^13,14) in order to maximize chromatographic performance with respect to simultaneous data acquisition and appropriate retention times and integration values. The resulting chromatographic data were extracted for multivariate analysis. Figure 1 shows the total ion chromatograms (TIC) of four- and six-year-old red ginseng under the given UPLC conditions. Accurate mass of the sample components was determined by the simultaneous and independent analysis of reference ions of leucine-enkephalin (m/z 556.2771) via the LockSpray™ interface. This system offers several advantages for non-targeted component analysis, including the minimization of ion suppression according to the reference ions, and the prevention of fluctuations in the reference ionization efficiency as a result of the gradient elution. Using this system, highly improved mass accuracy data were acquired in the range of 0.1 to 20 ppm and the acquired exact mass significantly reduces the uncertainty regarding the possible structures of metabolites and isotopes.

Fig. 1. Total Ion Count Chromatogram of Four- (A) and Six-Year-Old (B) Red Ginseng Samples

In order to find novel marker substances for four- and six-year-old red ginseng, an unsupervised PCA and supervised OPLS-DA were performed using the UPLC-QToF-MS data. After creating a process for mean-centering and pareto scaled data sets, the data were displayed as PCA score plots (Fig. 2). As shown in Fig. 2, most four- and six-year-old red ginseng samples were very clearly clustered into two groups, i.e., 4Y and 6Y groups. This means that the holistic qualities of four- and six-year-old red ginseng differed in the levels of their metabolites.

Fig. 2. PCA Score Plot of Four- and Six-Year-Old Red Ginseng Samples Using Pareto Scaling with Mean Centering; Measuring in Positive Mode with MeOH ext of Main Root

To explore the potential markers that contributed most to the differences between the two groups, UPLC-QToF-MS data from these samples were processed by supervised OPLS-DA. As shown in Fig. 3, the first seven ions, a (t_R 37.20 min, m/z 637.3127), b (t_R 18.89 min, m/z 425.1416), c (t_R 18.90 min, m/z 409.1687), d (t_R 37.20 min, m/z 659.3025), e (t_R 37.19 min, m/z 654.3506), f (t_R 18.89 min, m/z 387.1870) and g (t_R 18.89 min, m/z 105.0883) at the lower left of the “S-curve” were the ions from four-year-old red ginseng that contributed most to the differences between the two red ginseng groups. Further, the ion intensity trend plots of these ions in all the tested samples were relatively high in all four-year-old red ginseng samples, but were undetectable or present in very little amounts in the six-year-old red ginseng (see Supplementary materials). Similarly, fourteen ions, h (t_R 36.79 min, m/z 336.3295), i (t_R 34.10 min, m/z 336.3294), j (t_R 38.52 min, m/z 336.3298), k (t_R 35.51 min, m/z 352.3225), l (t_R 36.90 min, m/z 376.3217), m (t_R 12.76 min, m/z 226.1837), n (t_R 39.36 min, m/z 321.3168), o (t_R 36.79 min, m/z 354.3377), p (t_R 40.60 min, m/z 366.3739), q (t_R 38.29 min, m/z 310.3092), r (t_R 40.61 min, m/z 340.3551), s (t_R 35.19 min, m/z 352.3493), t (t_R 39.35 min, m/z 338.3437) and u (t_R 39.49 min, m/z 338.3331) at the top right corner of the “S-curve” were the ions from six-year-old red ginseng that contributed most to the differences between the two groups. The intensities of these ions (ion h–u) were relatively high in all six-year-old red ginseng samples, but they were undetectable in the four-year-old red ginseng samples (see Supplementary materials). These ion intensity trends suggest that components related to ions h–u could be used as potential chemical markers of six-year-old red ginseng to distinguish it from four-year-old red ginseng.

Fig. 3. OPLS-DA S-Plot and Selected Ion Intensity Trend Plots of Four- and Six-Year-Old Red Ginseng Samples Using Pareto Scaling with Mean Centering

To identify these potential marker ions, we attempted to exclude false-positive ions and fragment ions from the same potential markers. Thus, the individual potential marker ions were compared to each other using their extracted chromatogram and mass spectrum; chromatograms and mass spectra of selected ions are shown in the Supplementary materials. The six ions (ion h, i, j, k, l, o, s) were determined to be false-positives because the observed signal did not result from a single compound. Additionally, some of the potential marker ions were fragment ions of the same molecule. The remaining promising results are summarized in Table 1, which shows that six potential marker substances were extracted from four- and six-year-old red ginseng. We attempted to identify these marker substances through the spectroscopic data and separation techniques.

Table 1. Selected Potential Marker Substances

Markers for four-year-old red ginseng	Markers for six-year-old red ginseng
Marker 4-1: ion b, ion c, ion e, ion f	Marker 6-1: ion h, ion t, ion u
Marker 4-2: ion a, ion d, ion g	Marker 6-2: ion p, ion r
	Marker 6-3: ion q
	Marker 6-4: ion m

Additionally, when we performed PCA with total root samples of four- and six-year-old red ginseng, the results did not show clear discrimination, which differed from the result of main root (see S-13–S-16).

Identification of the Selected Marker Substances

Six-year-old red ginseng was provided by the Korea Ginseng Corporation (KGC), Buyeo, Korea. The specimens were extracted twice with MeOH at 40°C. Crude extracts were combined and fractionated by solvent partitioning. LC/MS and UPLC-QToF-MS trace-guided separation of non-polar fractions was then accomplished by silica phase vacuum flash chromatography followed by silica HPLC to afford compound 1 (Fig. 4). The structure of compound 1 was identified, based on combined spectroscopic analyses and comparison to published spectral data, to be 13-cis-docosenamide.^15,16)

Fig. 4. Chemical Structure of Compound 1

Compound 1 was isolated as a white powder with molecular formula C₂₂H₄₃ON deduced from HR-FAB-MS. The observation at δ_C 175.6 in the ¹³C-NMR spectrum and an absorption band at 1660 cm⁻¹ in the IR spectrum were indicative of a carbonyl carbon. The fatty acid amide chain of 1 possessed a double bond (δ_H 5.35) that was placed at C-13 by conspicuous ion clusters, including those at m/z 212 and 128, attributed to the characteristic γ-cleavage in FAB collision-include dissociation (FAB-CID) MS/MS data. The small vicinal coupling constant (J_13,14=9.6 Hz) suggested a cis-orientation for the double bond. The Z configuration of the C-13 double bond was assigned by the upfield shifts of allylic methylene carbons at δ_C 27.4 (C-12) and 29.9 (C-15).

According to the results of MS/MS analyses by LC/MS and UPLC-QToF-MS within the isolation fraction from six-year-old red ginseng extract, we expected that three other compounds (marker 6-2, -3, -4) had a similar fatty acid amide structure to 1. However, due to the limited amounts of material in the extract of red ginseng, the isolation of these was not possible. Efforts to identify markers from four-year-old red ginseng, a structurally different compound, are currently underway.

In 1995,¹⁷⁾ new brain lipids (9-unsaturated fatty amides) were reported as a sleep inducing material from the cerebrospinal fluid of cats. The chain length of these compounds is different from compound 1. Additionally, fatty acid amides found in root exudates of aquatic duckweed, including 13-docosenamide, have been reported to stimulate nitrogen removal in the rhizosphere of aquatic duckweed. However, the mechanism of action remains to be fully defined.¹⁶⁾

The identification of 1 and its investigation as a marker for six-year-old red ginseng suggests a significant utility in the quality control and standardization of red ginseng. A more thorough examination of the mechanism of action in six-year-old red ginseng is necessary.

Conclusion

Six new differentiation markers distinguished between four- and six-year-old red ginseng have been found for the quality control and standardization of red ginseng using UPLC-QToF-MS. Various analysis parameters, such as selection of ginseng, preparation of red ginseng, and measurement conditions of UPLC-QToF-MS were successfully obtained. These methods were successfully applied to measure clear discrimination between extracts of red ginseng of different ages and suggest a total of six discrimination markers.

In the process of developing this analytical method, a 13-cis-docosenamide (1) was isolated, and its structure was elucidated by a combined spectroscopic analysis with the aid of reference data. These results demonstrate the potential of using the markers for the quality control and standardization of four- and six-year-old red ginseng. More importantly, the developed method is widely useful as a tool for monitoring manufactured foods containing six-year-old red ginseng.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

References

• 1) Soldati F., “Biologically active natural products; Panax ginseng, Standardization and Biological Activity,” ed. by Cutler S. J., Cutler H. G., CRC Press, New York, 2000, pp. 209–304.
• 2) Park J. D., Rhee D. K., Lee Y. H., Phytochem. Rev., 4, 159–175 (2005).
• 3) Baeg I.-H., So S.-H., J. Ginseng Res., 37, 1–7 (2013).
• 4) Tung N. H., Song G. Y., Minh C. V., Kiem P. V., Jin L. G., Boo H.-J., Kang H.-K., Kim Y. H., Chem. Pharm. Bull., 58, 1111–1115 (2010).
• 5) Lee S.-S., Lee J.-H., Ahn I.-O., Kim Y.-C., Bang K.-H., Hyun D.-Y., J. Ginseng Res., 31, 142–146 (2007).
• 6) Shin Y.-S., Bang K.-H., In D.-S., Kim O.-T., Hyun D.-Y., Ahn I.-O., Ku B. C., Kim S.-W., Seong N.-S., Cha S.-W., Lee D., Choi H.-K., Arch. Pharm. Res., 30, 1625–1628 (2007).
• 7) Lin W.-N., Lu H.-Y., Lee M.-S., Yang S.-Y., Chen H.-J., Chang Y.-S., Chang W.-T., Am. J. Chin. Med., 38, 205–218 (2010).
• 8) Yang S.-O., Shin Y.-S., Hyun S.-H., Cho S., Bang K.-H., Lee D., Choi S.-P., Choi H.-K., J. Pharmaceut. Biomed., 58, 19–26 (2012).
• 9) Qiu Y., Lu X., Pang T., Ma C., Li X., Xu G., J. Sep. Sci., 31, 3451–3457 (2008).
• 10) Park H. E., Lee S.-Y., Hyun S.-H., Kim D. Y., Marriott P. J., Choi H.-K., J. AOAC Int., 96, 1266–1272 (2013).
• 11) Kim N., Kim K., Choi B. Y., Lee D., Shin Y.-S., Bang K.-H., Cha S.-W., Lee J. W., Choi H.-K., Jang D. S., Lee D., J. Agric. Food Chem., 59, 10435–10441 (2011).
• 12) Kim N., Kim K., Lee D., Shin Y.-S., Bang K.-H., Cha S.-W., Lee J. W., Choi H.-K., Hwang B. Y., Lee D., J. Nat. Prod., 75, 1777–1784 (2012).
• 13) Park H.-W., In G., Han S.-T., Lee M.-W., Kim S.-Y., Kim K.-T., Cho B.-G., Han G.-H., Chang I.-M., J. Ginseng Res., 37, 457–467 (2013).
• 14) Park H.-W., In G., Kim J.-H., Cho B.-G., Han G.-H., Chang I.-M., J. Ginseng Res., 38, 59–65 (2014).
• 15) Cravatt B. F., Prospero-Garcia O., Siuzdak G., Gilula N. B., Henriksen S. J., Boger D. L., Lerner R. A., Science, 268, 1506–1509 (1995).
• 16) Lu Y., Zhou Y., Nakai S., Hosomi M., Zhang H., Kronzucker H. J., Shi W., Planta, 239, 591–603 (2014).
• 17) Kolter T., Sandhoff K., Angew. Chem. Int. Ed. Engl., 34, 2363–2364 (1995).