Abstract

   Garden-cultivated Ginseng (GG) and mountain-cultivated Ginseng (MG)
   both belong to Panax Ginseng C. A. Meyer. However, the effective
   substances which can be used to distinguish GG from MG remain obscure.
   Therefore, the purpose of this study was to screen for discriminating
   markers that can assist in the correct identification of GG and MG.
   HPLC Q-TOF/MS and various chemometrics methods were used to analyze the
   chemical profiles of 13 batches of Ginseng and to explore the
   characteristic constituents of both GG and MG. The
   hepatocyte-protecting effects of GG and MG were investigated through a
   paclitaxel-induced liver injury model. Through a combination of
   correlation analysis and bioinformatic techniques, markers for
   differentiation between GG and MG were ascertained. A total of 40 and
   41 compounds were identified in GG and MG, respectively, and 15
   characteristic ingredients contributed significantly to the
   discrimination of GG from MG. Correlation analysis and network
   pharmacology were applied and ginsenosides Rg1, Re, Rb1, Rc, Rb2, and
   Rg3 were found to be discriminating markers of GG and MG. Six markers
   for the identification of GG and MG were screened out by a step-wise
   mutually oriented “chemical profiling–pharmaceutical effect”
   correlation strategy, which is of great significance for future quality
   assessment of Ginseng products.

   Keywords: garden-cultivated Ginseng, mountain-cultivated Ginseng,
   chemometric analysis, chemical profiling, hepatoprotective activity,
   discriminating markers

1. Introduction

   Panax Ginseng C. A. Meyer (Ginseng), of genus Panax and family
   Araliaceae, has been used in traditional Chinese medicines (TCMs) for
   thousands of years throughout Asia to reinforce vital energy and tonify
   the spleen [[38]1,[39]2]. Based on extensive studies, Ginseng is rich
   in polysaccharides, ginsenosides, and volatile oils. More recently,
   numerous pharmacological studies have demonstrated that Ginseng
   possesses various biological activities, including antitumor,
   anti-inflammation, antioxidation, and immunoregulatory properties,
   among others [[40]3,[41]4,[42]5]. Due to its diverse pharmacodynamics,
   Ginseng has been utilized in varied health products and dietary
   nutrients to relieve fatigue, deodorize, and ameliorate the toxic side
   effects of chemotherapy [[43]6,[44]7]. Recently, Ginseng has been
   clinically recognized as a hepatic protectant for its significant
   hepatoprotective effects and ability to improve liver function, as
   revealed by multitudinous studies [[45]8,[46]9,[47]10,[48]11].

   Over recent years, inconsistency in the quality and therapeutic
   efficacy of available Ginseng products has become an intricate issue
   due to disparate growth environments and cultivation conditions. In the
   Chinese Pharmacopoeia (2020 edition), Ginseng was classified as either
   garden-cultivated Ginseng (GG, artificially cultivated) or
   mountain-cultivated Ginseng (MG, grown naturally under mountainous
   forest). In general, GG can be harvested after four to six years, while
   MG requires 10–20 years or longer. Both modern pharmacological research
   and traditional clinical experience indicate that the medicinal value
   and health function of MG is significantly better than that of GG
   [[49]12,[50]13]; thus, MG is consistently more expensive than GG. On
   account of the obvious differences in time investment and economic
   benefit, substitution of MG with GG has been increasingly rampant in
   the market, which is a primary cause of the current difficulty in
   maintaining the quality and effectiveness of Ginseng products. Thus,
   recent studies concerning the discrimination of GG from MG have
   attracted increasing attention. Previous research mainly focused on the
   contrastive qualitative-quantitative analysis of their chemical
   components in vitro [[51]14,[52]15,[53]16]; however, it has been
   recognized that pharmaceutical efficacy is essentially the nucleus of
   TCM [[54]17,[55]18,[56]19,[57]20]. Unfortunately, few studies have
   sought to determine differentiating markers of GG and MG on the basis
   of their biological activity. Therefore, it is essential to integrate
   multiple analytical, pharmacological, and statistical approaches to
   establish a systematic strategy to explore the distinctive bioactive
   markers of GG and MG for their identification and differentiation.

   In the present study, pharmacology-based markers for distinguishing GG
   from MG were first suggested by analyzing the chemical components of GG
   and MG using efficiency indexes. In brief, the ingredient database of
   Ginseng was obtained from seven batches of GG and six batches of MG
   analyzed in Liaoning by an established HPLC Q-TOF/MS method. Then, a
   variety of chemometric methods including hierarchical cluster analysis
   (HCA), principal component analysis (PCA), and orthogonal partial least
   squares discriminant analysis (OPLS-DA) were applied to explore the
   characteristic components of GG and MG. A contrastive study on the
   hepatoprotective function of GG and MG utilized a liver injury model
   induced by paclitaxel [[58]21], which indicated that the
   liver-protective efficacy of MG was superior to that of GG. In order to
   further investigate the ingredients contributing to the different
   pharmacodynamics of GG and MG, we carried out a correlation analysis
   between the characteristic constituents and the efficiency indexes.
   Furthermore, in consideration of the “multi components-multi targets”
   characteristic of TCMs, network pharmacology was adopted to
   comprehensively screen for the effective discriminating markers of
   Ginseng. Above all, a step-wise mutually oriented “chemical
   profiling–pharmaceutical effect” strategy was established, and six
   chemical components—including ginsenosides Rg1, Re, Rb1, Rc, Rb2, and
   Rg3—were ascertained to be key markers, underlying the difference in
   hepatoprotective efficacy, for differentiating GG from MG. This study
   is of great significance in setting a more scientific benchmark for the
   analysis of Ginseng, and offers an alternative quality assessment
   system for TCMs.

2. Results and Discussion

2.1. Qualitative and Semi-Quantitative Analysis of Ingredients in Ginseng
with HPLC Q-TOF/MS

   An established HPLC Q-TOF/MS method using digoxin as an internal
   standard (IS) was adopted to conduct the qualitative and
   semiquantitative analysis of the ingredients in Ginseng for seven
   batches of GG and six batches of MG, whose detailed information is
   presented in [59]Table 1. The chemical components were identified by
   calculating possible chemical formulas based on comparisons between the
   determined molecular weight and chromatographic retention behavior and
   reference data [[60]22,[61]23,[62]24]. As a result, a total of 42
   chemical components were identified, including 19 protopanaxadiol
   ginsenosides (PPD-type), 16 protopanaxatriol ginsenosides (PPT-type), 4
   oleanane ginsenosides, and 3 other ginsenosides; their structures are
   presented in [63]Figure 1. A total of 40 and 41 compounds were
   identified in GG and MG, respectively, 40 of which were common to both,
   as shown in [64]Table 2. Moreover, among them, 12 chemical ingredients
   were accurately identified by comparison with the reference substances
   (extraction ion chromatograms of reference substances and Ginseng
   sample solution are presented in [65]Figure S1A–C).

Table 1.

   A summary of the tested samples of Ginseng.
   NO.      Origins       Years
   MG1  Liaoning, Benxi     8
   MG2  Liaoning, Benxi    8–9
   MG3 Liaoning, Huanren   14
   MG4 Liaoning, Huanren   17
   MG5 Liaoning, Shenyang  10
   MG6 Liaoning, Shenyang  15
   GG1 Liaoning, Jinzhou    4
   GG2  Liaoning, Fushun    4
   GG3 Liaoning, Jinzhou    5
   GG4  Liaoning, Fushun    5
   GG5 Liaoning, Jinzhou    6
   GG6 Liaoning, Jinzhou    4
   GG7  Liaoning, Fushun    6
   [66]Open in a new tab

Figure 1.

   [67]Figure 1
   [68]Open in a new tab

   Chemical structures of the 42 components.

Table 2.

   Detected compounds in Ginseng by HPLC Q-TOF/MS.
   No. Identification Formula Adduct Ion m/z Error (ppm) Retention Time
   (min) Fragment Ions (m/z)
   1 Pseudogisnenoside RT2 C[41]H[70]O[14] [COOH]^+ 831.4726 −1.3 14.45
   161.0451, 491.3708, 653.4280, 785.4686
   2 Floralginsenoside A C[42]H[72]O[16] [H]^− 831.4726 −2.7 14.51
   161.0451, 491.3708, 653.4280, 785.4686
   3 Notoginsenoside R3 C[48]H[82]O[19] [COOH]^+ 1007.542 −0.5 15.03
   637.4349, 781.4764, 799.4853, 961.5411
   4 Notoginsenoside R1 C[47]H[80]O[18] [COOH]^+ 977.5311 −0.5 15.34
   131.0352, 475.3777, 637.4310, 799.4866
   5 Ginsenoside Rg1 C[42]H[72]O[14] [COOH]^+ 845.4898 0.5 15.87 161.0455,
   475.3782, 637.4317, 799.4862
   6 Ginsenoside Re C[48]H[82]O[18] [COOH]^+ 991.5478 0.5 16.07 161.0462,
   179.0564, 475.3810, 619.4249
   7 Pseudoginsenoside RT3 C[41]H[70]O[13] [COOH]^+ 815.4792 0.6 17.82
   391.2720, 475.3806, 553.3349, 637.4354
   8 4’-O-acetylpseudoginsenoside F11 C[44]H[74]O[15] [COOH]^+ 887.5006
   0.8 17.92 161.0434, 391.2863, 475.3884, 619.4274
   9 6’-O-acetylginsenoside Rg1 C[44]H[74]O[15] [COOH]^+ 887.5014 1.8
   19.22 475.3800, 619.4013, 637.4266, 781.4766
   10 Pseudoginsenoside F11 * C[42]H[72]O[14] [H]^− 799.485 0.2 20.21
   161.0455, 415.3218, 491.3768, 653.4331
   11 Ginsenoside Rf C[42]H[72]O[14] [COOH]^+ 845.4899 0.6 20.57
   161.0468,415.3225, 653.4201, 799.4903
   12 Notoginsenoside R2 C[41]H[70]O[13] [COOH]^+ 815.4793 0.7 21.49
   475.3778, 619.4207, 637.4388, 769.4760
   13 Ginsenoside F3 C[41]H[70]O[13] [COOH]^+ 815.4793 0.7 21.58 161.0463,
   457.3778, 475.3778, 619.4207
   14 Ginsenoside Ra2 ** C[58]H[98]O[26] [H]^− 1209.6292 1.5 21.83
   149.0419, 323.1010, 621.4420, 783.4917
   15 Ginsenoside Rb1 C[54]H[92]O[23] [COOH]^+ 1153.599 −0.9 22.50
   161.0469, 221.0674, 459.3756, 621.4323
   16 20S Ginsenoside Rg2 C[42]H[72]O[13] [COOH]^+ 829.4951 0.8 22.67
   391.2899, 475.3792, 619.4224, 637.4360
   17 Malonyl ginsenoside Rb1 C[57]H[94]O[26] [H]^− 1193.595 −1.1 22.93
   179.0562, 783.4874, 927.5361, 945.5423
   18 20S Ginsenoside Rh1 C[36]H[62]O[9] [COOH]^+ 683.4369 0.6 22.94
   391.2749, 457.3619, 475.3830, 637.4316
   19 Ginsenoside Ra1 ** C[58]H[98]O[26] [H]^− 1209.6282 0.7 23.08
   621.4382, 783.4733, 945.5652, 1077.5902
   20 Ginsenoside Rc C[53]H[90]O[22] [COOH]^+ 1123.59 0 23.44 149.0461,
   621.4406, 765.4835, 783.4940
   21 Ginsenoside Rb3 C[53]H[90]O[22] [COOH]^+ 1123.589 −0.9 23.85
   621.4402, 783.4904, 945.5373, 1077.5851
   22 Ginsenoside F1 C[36]H[62]O[9] [COOH]^+ 683.4365 −0.1 23.91 161.0456,
   391.2859, 475.3801, 637.4345
   23 Malonyl ginsenoside Rc C[56]H[92]O[25] [H]^− 1163.584 −0.9 23.94
   783.4904, 945.5373, 1059.5737, 1077.5851
   24 Ginsenoside Ro C[48]H[76]O[19] [COOH]^+ 1001.496 0.9 24.4 455.3498,
   523.3764, 569.3854, 613.3544
   25 Ginsenoside Rb2 C[53]H[90]O[22] [COOH]^+ 1123.59 0.2 24.8 621.4420,
   783.4967, 945.5501, 1077.5889
   26 Malonyl ginsenoside Rb2 C[56]H[92]O[25] [H]^− 1163.584 −0.9 24.83
   459.3823, 621.4387, 783.4917, 945.5454
   27 Quinquenoside R1 C[56]H[94]O[24] [COOH]^+ 1195.609 −0.9 25.05
   179.0564, 323.0999, 621.4222, 783.4717
   28 Ginsenoside Rs1 C[55]H[92]O[23] [H]^− 1119.593 −2.1 25.4 1077.5807,
   945.5515, 783.4877, 621.4542
   29 Pseudoginsenoside RT1 C[47]H[74]O[18] [H]^− 925.4797 −0.5 25.48
   523.4065, 569.3840, 613.3785, 701.4410
   30 20R Ginsenoside Rh1 C[36]H[62]O[9] [COOH]^+ 683.437 0.7 25.89
   391.2854, 457.3704, 475.3805, 637.4319
   31 Ginsenoside Rd C[48]H[82]O[18] [COOH]^+ 991.5474 0.2 26.65 161.0465,
   621.4410, 783.4907,945.5418
   32 Malonyl ginsenoside Rd C[51]H[84]O[21] [H]^− 1031.543 −0.6 26.97
   621.4394, 765.4816, 783.4937, 927.5336
   33 Pseudoginsenoside RC1 C[50]H[84]O[19] [COOH]^+ 1033.552 −5.4 27.44
   161.0453, 459.3802, 621.4449, 765.4792
   34 Zingibroside R1 C[42]H[66]O[14] [H]^− 793.4373 −0.8 27.45 455.3508,
   569.3859, 613.3767, 631.3851
   35 Ginsenoside Rg10 C[42]H[70]O[13] [COOH]^+ 827.4781 −0.8 29.98
   161.0415, 221.0695, 619.4169, 781.4760
   36 Ginsenoside Rg6 C[42]H[70]O[12] [COOH]^+ 811.4842 0.5 31.86
   161.0447, 457.3617, 601.4126, 619.4218
   37 Ginsenoside Rg3 C[42]H[72]O[13] [COOH]^+ 829.4949 0.6 32.10
   161.0468, 459.3864, 621.4404,783.4958
   38 Ginsenoside Rk3 C[36]H[60]O[8] [COOH]^+ 665.4262 0.4 32.37 161.0457,
   457.3588, 619.4184, 665.4284
   39 Chikusetsusaponin IVA C[42]H[66]O[14] [H]^− 793.438 0 32.59
   455.3530, 569.3874, 613.3763, 793.4408
   40 Ginsenoside F2 C[42]H[72]O[13] [COOH]^+ 829.4949 0.6 32.73 161.0460,
   459.3853, 621.4396,783.4938
   41 Ginsenoside Rs3 C[44]H[74]O[14] [COOH]^+ 871.5011 −4.4 32.92
   161.0461, 375.2889, 459.3834, 621.4326
   42 Ginsenoside Rk1 C[42]H[70]O[12] [COOH]^+ 811.4845 0.8 34.20
   161.0460, 537.3988, 603.4275, 765.4791
   [69]Open in a new tab

   * indicates ingredients belong to GG; ** indicates ingredients belong
   to MG.

2.2. Exploration of Characteristic Components for Differentiating GG from MG
by Chemometric Analysis

   Characteristic components for the identification of GG and MG were
   explored using multifarious chemometric analysis methods. Internal
   standard normalization (via digoxin) was adopted in this study. The
   peak areas of all 42 compounds were calibrated to the IS and the ratio
   of the peak areas were used for chemometric analysis.

   As seen in the dendrogram [70]Figure 2A, 13 batches of Ginseng from
   Liaoning could be divided into two categories using HCA, of which GG
   was clustered into one group, and MG into another. Moreover, the PCA
   score plot illustrated that there was an efficacious separation between
   groups G1–G7 and M1–M6, as shown as [71]Figure 2B, which was consistent
   with the results of HCA. In order to clarify the characteristic
   components that contributed greatly to this distinction, OPLS-DA was
   applied. The score plot, as shown in [72]Figure 3A, indicated that the
   two groups were highly distinct with R^2Y = 0.999 and Q^2 = 0.973, and
   15 components demonstrated VIP ≥ 1, as shown in [73]Figure 3B. However,
   among them, only six constituents—ginsenosides Rg1, Re, Rb1, Rc, Rb2,
   and Rg3—revealed significant differences between GG and MG (* p < 0.05)
   according to the results of a Student’s t-test, as shown in [74]Figure
   4.

Figure 2.

   [75]Figure 2
   [76]Open in a new tab

   (A) HCA dendrogram of the 13 batches of Ginseng; (B) PCA score plots of
   the 13 batches of Ginseng.

Figure 3.

   [77]Figure 3
   [78]Open in a new tab

   (A) OPLS-DA score plots of the 13 batches of Ginseng; (B) The value of
   VIP of constituents (* p < 0.05).

Figure 4.

   [79]Figure 4
   [80]Open in a new tab

   t-test analysis results of the 15 identified components (* p < 0.05).

   As above, all of the 15 chemometric components were common ingredients
   of GG and MG. Six ginsenosides were established as the qualitative and
   quantitative markers for in vitro identification of GG and MG. It was
   concluded that these six ingredients could be considered as preliminary
   marker candidates to distinguish GG from MG. Although these six
   components had significant differences in content, this seldom
   indicates the quality of TCMs. The selected markers, determined via
   efficacy mining, were considered to be more representative of quality.

2.3. Comparative Study of the Pharmacodynamics of GG and MG in
Paclitaxel-Induced Liver Injury

   To further explore the discriminating markers of GG and MG underlying
   their respective pharmacodynamics, we conducted a comparative study of
   their liver-protective efficacies using a paclitaxel-induced liver
   injury rat model. Batches of GG5 and GG7, and MG1 and MG4, were
   included in the contrastive pharmacodynamic analysis as the contents of
   the 15 chemometric components in these batches were close to their
   average values in GG and MG, respectively.

   As shown in [81]Figure 5, an obvious elevation in white blood cell
   count (WBC) and lymphocyte count (LYM) could be observed in the model
   group compared with the control group (# p < 0.05), indicating abnormal
   regulation of autoimmunity and the development of inflammation.
   Furthermore, expression of the liver function indicators
   glutamic-oxaloacetic transaminase (GOT) and glutamic-pyruvic
   transaminase (GPT) were also increased markedly in the model group
   compared with the control group, as shown as [82]Figure 6 (# p < 0.05),
   suggesting a decrease in liver function and the occurrence of
   parenchymal hepatic disease [[83]25,[84]26,[85]27]. Through the above,
   it could be concluded that the liver injury model was successfully
   established. These escalating alterations were all consistently
   restored after treatment with GG and MG; however, MG exerted stronger
   therapeutic effects than GG based on the results of a t-test (* p <
   0.05; ** p < 0.01). Therefore, our research demonstrates for the first
   time that the hepatoprotective effects of MG are superior to those of
   GG through in vivo pharmacology experiments.

Figure 5.

   [86]Figure 5
   [87]Open in a new tab

   Effect of Ginseng on WBC (A) and LYM (B) in the blood of rats with
   liver injury induced by paclitaxel (compared with control group, # p <
   0.05; compared with model group, * p < 0.05 and ** p < 0.01).

Figure 6.

   [88]Figure 6
   [89]Open in a new tab

   Effect of Ginseng on the activity of GOT (A) and GPT (B) in the serum
   of rats with liver injury induced by paclitaxel (compared with control
   group, # p < 0.05; compared with model group, * p < 0.05 and ** p <
   0.01).

2.4. Discovery of Pharmacodynamic-Based Markers to Distinguish GG from MG

2.4.1. Correlation Analysis between Characteristic Components and
Pharmacodynamic Indicators

   Based on pharmacological comparison, a correlation analysis of the
   relative peak areas of the 15 characteristic components and the levels
   of the four pharmacodynamic indicators was performed with the
   corresponding GG5, GG7, MG1, and MG4 batch sample data. Evaluation of
   the correlation degree by the absolute value of the Pearson correlation
   coefficient: 0.6 ≤ |γ| ≤ 1 indicated correlation. Ginsenosides Re, Rb1,
   Rc, and Rb2 were positively correlated with WBC and LYM, while
   ginsenosides Rg1 and Rg3 were positively correlated with GOT and GPT,
   as shown in [90]Figure 7 and [91]Table S1, indicating that ginsenosides
   Rg1, Re, Rb1, Rc, Rb2, and Rg3 could participate in regulation of the
   milieu interne in morbid state and represent candidate markers for the
   distinction of GG from MG based on their underlying pharmacodynamics.

Figure 7.

   [92]Figure 7
   [93]Open in a new tab

   Bubble chart of Pearson correlation coefficients.

2.4.2. Identification of Effective Representative Substances of GG and MG
Using Bioinformatic Analysis

   Bioinformatic analysis was applied to visualize the large-scale data of
   “components–targets–pathways” and for virtual excavation of the
   bioactive markers capable of differentiating GG from MG, in
   consideration of the “multi components-multi targets” characteristic of
   TCMs.

   Target genes of the 15 characteristic components were predicted by the
   SwissTargetPrediction database. Pathway enrichment analysis was
   performed with Cytoscape. A total of 73 targets and 62 pathways with
   significant differences were obtained. As shown in [94]Figure 8 and
   [95]Figure 9, Ginseng mainly participated in the regulation of
   inflammation and liver function through modulation of the pathways of
   PI3K-Akt, Ras, HIF-1, TNF, PPAR, AMPK, and p53, in addition to various
   pathways related to bile acid and lipid metabolism. These crucial
   targets and pathways were involved in the anti-liver injury effects of
   Ginseng [[96]28,[97]29,[98]30,[99]31,[100]32,[101]33]. Additionally,
   the degree value is a key parameter to assess the interconnectedness of
   components and their corresponding symptoms; the degree values of
   ginsenosides Rg1, Re, Rb1, Rc, Rb2, Rd, and Rg3, as well as malonyl
   ginsenoside Rb1, were higher than the average degree value (mean value
   = 27.4; as shown as [102]Table S2).

Figure 8.

   [103]Figure 8
   [104]Open in a new tab

   “Ingredient–target” network diagram (blue, ingredient; purple, target),
   * indicates ingredients with a higher than the average degree value
   (mean value = 27.4).

Figure 9.

   [105]Figure 9
   [106]Open in a new tab

   Pathway enrichment analysis of 15 characteristic components using
   Cytoscape.

   Integrating all above results, the ginsenosides Rg1, Re, Rb1, Rc, Rb2,
   and Rg3 could be regarded as effective markers for distinguishing GG
   from MG. Moreover, in consideration of previous studies, it has been
   widely reported that ginsenoside are the most important components of
   Ginseng contributing to its beneficial properties. Ginsenosides Rg1,
   Re, and Rg3 act on the PI3K–Akt signaling pathway to reduce oxidative
   stress response [[107]34,[108]35,[109]36]. In addition, ginsenosides
   Rb1, Rc, and Rb2 possess anti-inflammatory activities through the
   inhibition of both MAPK signaling pathways and the expression of TNF
   [[110]37,[111]38,[112]39]. In agreement, our research identified these
   six ginsenoside compounds as pharmacodynamic-based markers to
   distinguish GG from MG.

3. Materials and Methods

3.1. Chemicals and Reagents

   A total of 13 batches of Ginseng collected from Liaoning (including 7
   batches of GG and 6 batches of MG) of different ages were obtained from
   Tong-Ren-Tang TCMs store (Shenyang, China). HPLC-grade acetonitrile,
   methanol, and formic acid were all purchased from Fisher Scientific
   (Fair Lawn, NJ, USA); all other reagents were analytical grade
   (Shandong Yuwang Industrial Co., Ltd., Yucheng, China). The GOT/GPT
   commercial enzyme-linked immunosorbent assay (ELISA) kits were
   purchased from the Nanjing Jiancheng Institute of Biological
   Engineering (Nanjing, China).

3.2. HPLC Q-TOF/MS Analysis

3.2.1. Sample Preparation

   Ginseng powder (1 g, passed through a 5-mesh sieve) was extracted under
   reflux with 10 times the volume of water for 1 h, and the progress was
   repeated twice. A solution with concentration 50 mg/mL (medicinal
   material dosage/volume) was obtained by combining the extracts. Then,
   the solution was centrifuged at 12,000 rpm for 10 min and filtered
   through a 0.22 μm filter membrane for HPLC Q-TOF/MS analysis.
   Appropriate amounts of the reference substance ginsenosides Re, Rf,
   Rb1, Rc, F1, Ro, Rb2, Rd, Rg3, F2, 20R-ginsenoside Rh1, and
   chikusetsusaponin Iva (ShanghaiyuanyeBio-Technology Co., Ltd.,
   Shanghai, China) were weighed and dissolved in methanol to prepare a
   mixed reference solution containing the 12 chemical components. Digoxin
   (25 ng/mL, Chengdu Chroma-Biotechnology Co., Ltd., Sichuan, China) was
   used as an IS for semiquantitative analysis. All samples were kept at 4
   °C during analysis.

3.2.2. HPLC Q-TOF/MS Conditions

   Analyses were performed on a 1260 Infinity HPLC system (Agilent
   Technology, Santa Clara, CA, USA). A reverse-phase column of Phenomenex
   Kinetex XB C18 (100 mm × 4.6 mm, 2.6 μm) was used for the
   chromatographic separations with a flow rate of 0.5 mL/min at 30 °C.
   The mobile phase consisted of 0.1% formic acid-water (A) and 0.1%
   formic acid-acetonitrile (B) with a gradient procedure as follows: 0–13
   min, 5–30% B; 13–27 min, 30–40% B; 27–30 min, 40–70% B; 30–31 min,
   70–90% B; 31.1–31.1 min, 90–5% B; 31.1–35 min, 5–5% B. The injection
   volume of each sample was 5 µL and the temperature of the sample plate
   was maintained at 4 °C.

   A quadrupole time-of-flight mass spectrometry system (Triple TOF 5600,
   AB SCIEX Corporation, Foster City, CA, USA) was used to obtain the
   total ion chromatograms (TIC) in negative electrospray ionization (ESI)
   mode. The optimal instrument parameters were as follows: spray voltage
   of 4500 V, a source heater temperature of 550 °C, the range of m/z set
   to 100–2000 and nitrogen was used as the atomizing gas and other
   auxiliary gases. PeakView software (version 1.2.1, SCIEX) was used for
   data analysis and the precision error threshold was fixed at 5 ppm.

3.3. Animals

   Healthy Sprague-Dawley rats (200–220 g; 8 weeks old;
   NO.SCXK(Liao)2020---0001) were provided by the Experimental Animal
   Center of Shenyang Pharmaceutical University and housed in a
   specialized pathogen free (SPF) standard environment (ambient
   temperature 22 ± 2 °C, with relative humidity of 50 ± 10 % and a
   natural light-dark cycle). All experiments complied with the Animal
   Experiment Code of Shenyang Pharmaceutical University and approval was
   obtained from the animal ethics committee of the institution (Ethical
   Approval number: SYPU-IACUC-G2P-2021-68). After a week of domestication
   of the rats, they were randomly divided into six groups: control,
   model, and both low (300 mg/kg) and high doses (800 mg/kg) of the 2
   batches of GG (GG5/GG7) and 2 batches of MG (MG1/MG4) (n = 6). Food and
   water were available ad libitum.

   Paclitaxel (10 mg/kg, Haikou Pharmaceutical Factory Co., Ltd, Hainan,
   China) was given intraperitoneally to induce liver injury once a day
   for two weeks. In the meantime, the rats in the medication
   administration groups were gavaged with the predetermined dose of
   Ginseng 1 h after the injection of paclitaxel, while rats in the
   control and model groups received the same volume of normal saline. One
   hour after the final administration, two blood samples (each 1.5 mL,
   one sample transferred into the tubes with anticoagulants directly, the
   other transferred into the heparinized tubes to obtain serum samples)
   were collected from the suborbital vein of each rat and stored at −80
   °C until required.

3.4. Detection of Pharmacodynamic Indicators

   In order to detect the levels of blood routine parameters WBC and LYM,
   a 1.5 mL blood sample of each rat was measured with an automatic animal
   blood cell analyzer (BC2800Vet Shenzhen).

   Serum samples were used to measure the expression of liver function
   indicators, including GOT and GPT, according to the manufacturer’s
   instructions.

3.5. Bioinformatics Analysis Process

   Network pharmacology analysis was performed with the characteristic
   ingredients in this study. The PubChem
   ([113]https://pubchem.ncbi.nlm.nih.gov/ accessed on 30 July 2021) and
   SwissTargetPrediction ([114]http://www.swisstargetprediction.ch/
   accessed on 30 July 2021) databases were integrated to obtain the
   corresponding targets of each ingredient. The targets closely
   associated with “liver injury” were retrieved from the Therapeutic
   Target ([115]http://db.idrblab.net/ttd/ accessed on 30 July 2021),
   GeneCards ([116]https://www.genecards.org/ accessed on 30 July 2021),
   and Comparative Toxicogenomics Databases ([117]http://ctdbase.org/
   accessed on 30 July 2021). Finally, Cytoscape (3.8.2, USA) software was
   used to visualize the “component–target–pathway” network.

3.6. Statistical and Data Analysis

   All values are presented as means ± SD. The data were analyzed using
   SIMCA14.0 (UmetricsAB, Umea, Sweden) and SPSS (SPSS Inc., Chicago, IL,
   USA). Data from two groups were analyzed using a two-tailed Student’s
   t-test and p values lower than 0.05 were considered significant.

4. Conclusions

   In the present study, an integrated mutually oriented “chemical
   profiling–pharmaceutical effect” strategy was proposed and successfully
   applied to identify distinct bioactive markers for the identification
   of GG and MG for the first time. First, we established a chemical
   component database of Ginseng using HPLC Q-TOF/MS, which contained 40
   compounds from GG and 41 compounds from MG. Multiple chemometric
   methods were further applied to discover the characteristic components.
   We then conducted a comparative study of the pharmacodynamics of GG and
   MG, which indicated that MG may exert superior therapeutic effects to
   GG. Furthermore, in order to explore the components responsible for
   this difference in efficacy, correlation analysis combined with network
   pharmacology technology was applied. Ginsenosides Rg1, Re, Rb1, Rc,
   Rb2, and Rg3 were identified as markers underlying the difference in
   hepatoprotective efficacy which can be used to distinguish GG from MG.
   These findings may contribute to the development of quality control and
   thus enhance the clinical efficacy of Ginseng.

Acknowledgments