Component Identification of Phenolic Acids in Cell Suspension Cultures of Saussurea involucrata and Its Mechanism of Anti-Hepatoma Revealed by TMT Quantitative Proteomics

Saussurea involucrata (S. involucrata) had been reported to have anti-hepatoma function. However, the mechanism is complex and unclear. To evaluate the anti-hepatoma mechanism of S. involucrata comprehensively and make a theoretical basis for the mechanical verification of later research, we carried out this work. In this study, the total phenolic acids from S. involucrata determined by a cell suspension culture (ESPI) was mainly composed of 4,5-dicaffeoylquinic acid, according to the LC-MS analysis. BALB/c nude female mice were injected with HepG2 cells to establish an animal model of liver tumor before being divided into a control group, a low-dose group, a middle-dose group, a high-dose group, and a DDP group. Subsequently, EPSI was used as the intervention drug for mice. Biochemical indicators and differences in protein expression determined by TMT quantitative proteomics were used to resolve the mechanism after the low- (100 mg/kg), middle- (200 mg/kg), and high-dose (400 mg/kg) interventions for 24 days. The results showed that EPSI can not only limit the growth of HepG2 cells in vitro, but also can inhibit liver tumors significantly with no toxicity at high doses in vivo. Proteomics analysis revealed that the upregulated differentially expressed proteins (DE proteins) in the high-dose group were over three times that in the control group. ESPI affected the pathways significantly associated with the protein metabolic process, metabolic process, catalytic activity, hydrolase activity, proteolysis, endopeptidase activity, serine-type endopeptidase activity, etc. The treatment group showed significant differences in the pathways associated with the renin-angiotensin system, hematopoietic cell lineage, etc. In conclusion, ESPI has a significant anti-hepatoma effect and the potential mechanism was revealed.


Introduction
Saussurea involucrata (S. involucrata) is a rare and slow-growing herb growing in the Tianshan and Altay Mountains of Xinjiang Province, China, at altitudes over 2600 m. S. involucrata has the effect of promoting blood circulation, relaxing tendons, dispersing cold, and removing dampness [1]. It is mainly used for the treatment of coughs, rheumatoid arthritis, high-altitude response, and stomach pain [2].

Extraction of the Phenolic Acids in Cell Suspension Cultures of S. involucrata (EPSI)
Phenolic acids are soluble in organic solvents, so the total phenolic acids were extracted by the organic solvent extraction method [14]. The method is as follows: The cell suspension culture of S. involucrata was mixed 1:8 (m/v) with absolute alcohol, heated at 50 • C, and stirred for 15 min. After the supernatant was collected, 8× absolute alcohol was added to the precipitate and stirred for 2 h; the above steps were repeated twice. The collected supernatant was filtered and subjected to rotary evaporation at 60 • C. After the rotary distillation, the raw materials were collected with water, and the organic phase was extracted with ethyl acetate. The above steps were repeated 3 times. After the materials had been subjected to rotary evaporation at 60 • C, the concentrate (EPSI) was collected and freeze-dried at −80 • C for storage.

Effect of EPSI on the Multiplication of HepG2 Cells In Vitro
Hepatic cellular carcinoma (HCC) cell lines HepG2 were purchased from Bluef Biotechnology Development Co. Ltd. (Shanghai, China) and cultured in Dulbecco's Modified Eagle Medium (DMEM) (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher Scientific, Waltham, MA, USA).
After resuscitation, cells at the logarithmic phase were taken for standby. We adjusted the cell concentration to 10 4 units/mL and added them to a 96-well culture plate at a concentration of 100 µL/well for culturing (37 • C, CO 2 concentration: 5%, 24 h). The old medium was discarded after culturing, and a blank group with 100 µL of 5% medium was added. For the experimental group, EPSI samples of different concentrations were added and then cultured at 37 • C and 5% CO 2 for 72 h.
CCK8 (CCK-8 Cell Proliferation and Cytotoxicity Assay Kit, Solarbio, Beijing, China) and MTT (MTT Cell Proliferation and Cytotoxicity Assay Kit, Solarbio, Beijing, China) were used to evaluate the inhibitory effect of EPSI on HepG2 cells in vitro. All operations were carried out according to the operating instructions.

Acute Toxic Test
A fixed-dose procedure (FDP) was used to verify the acute toxicity of EPSI [18,19]. The dosage was set at 2000 mg/kg. Six-to eight-week-old BALB/c nude female mice (Beijing Vital River Laboratory Animal Technology Co. Ltd., Beijing, China) were used as the test animal. Intraperitoneal injection (I.P.) was used as the mode of administration. It was considered that EPSI had no acute toxicity if the LD50 was >2000 mg/kg.

Establishment of the Animal Model
Six-to eight-week-old BALB/c nude female mice were purchased from Beijing Vital River Laboratory Animal Technology Co. Ltd. (Beijing, China). All mice were housed under controlled conditions in individual cages at 22 ± 3 • C and 60-70% relative humidity with a 12 h dark/light cycle in a specific germ-free environment and were allowed free access to sterile food and water. After 1 week of acclimation, each animal was subcutaneously injected with 10 6 units of HepG2 cells. When there were measurable tumors with an equivalent volume, the experiment animals were randomly divided into five groups. The groups consisted of the control group (no gavage, free feeding, n = 8), the low-dose group (100 mg/kg EPSI gavage for 24 days, free feeding, n = 8), the middle-dose group (200 mg/kg EPSI gavage for 24 days, free feeding, n = 8), the high-dose group (400 mg/kg EPSI gavage for 24 days, free feeding, n = 8) and the DDP (PtCl 2 [NH 3 ] 2 ) group (10% DDP intraperitoneal injection for 10 days, free feeding, n = 8). All operations were performed in a sterile environment.
The weights and the tumor volumes of the experimental animals in each group were measured once every 2 days at night during the intervention. The experimental animals were sacrificed when there were significant differences in tumor volume. The tumors were removed and stored at −80 • C after eyeball blood collection. All animal experiments were approved by the Laboratory Animal Welfare and Ethics Committee of Jilin Agricultural University (No. 20190410005), following National Research Council Guidelines.

Hematoxylin and Eosin (H&E) Staining
The tumor tissues obtained as described in Part 2.4 were separated by about 0.2 mm 3 , without being frozen, and were fixed in a 4% paraformaldehyde solution for 12 h at room temperature and embedded in paraffin. After cutting the paraffin into 5 µm sections, the slides were dewaxed, rehydrated, and stained with 1% H&E at room temperature as described previously.

Tandem Mass Tag (TMT) Quantitative Proteomics
The process of TMT quantitative proteomics was basically consistent with the research methods for animal tissues in other studies [20]. The process is briefly described as follows [21][22][23].

Total Protein Extraction
The sample was ground into powder at a low temperature and transferred to a tube (cooled by liquid nitrogen). A certain amount of a PASP lysis buffer (100 mM NH 4 HCO 3 , 8 M Urea, pH = 8), was added, mixed, and ultrasonicated for 5 min under an ice bath to full splitting. The product was centrifuged at 4 • C and 12,000× g for 15 min. Next, 10 mM DTT (DL-dithiothreitol) was added to the supernatant and reacted at 56 • C for 1 h. Sufficient IAM (iodoacetamide) was then added and reacted at room temperature without light for 1 h. After completing the above process, 4× the volume of acetone (precooled at 20 • C) was added, precipitated at −20 • C for 2 h, and centrifuged at 4 • C and 12,000× g for 15 min, and the precipitate was collected. After that, 1 mL of acetone (precooled at 20 • C) was added to resuspend the precipitate, then the mixture was centrifuged at 4 • C and 12,000× g for 15 min, the precipitate was collected, air-dried, and an appropriate amount of a protein solution dissolution buffer (8 M urea, 100 mM TEAB (triethylammonium bicarbonate); pH 8.5) was added to dissolve the protein precipitate.

Protein Quality Test
The Bradford protein quantitative kit was used to prepare a BSA (bovine serum albumin) standard protein solution according to the instructions, and the concentration gradient range was from 0 to 0.5 g/L. The BSA standard protein solution with different concentration gradients and the sample solution were taken to be tested at different dilution ratios and were added into a 96-well plate, the volume was made up to 20 µL, and each gradient was repeated 3 times. After that, 180 µL of a G250 staining solution was added immediately, and the mixture was placed at room temperature for 5 min, and the absorbance at 595 nm was measured. The standard curve was drawn with the absorbance of the standard protein solution, and we calculated the protein concentration of the sample to be tested. For gel electrophoresis, 20 µg of the protein sample was loaded to 12% SDS- PAGE (Sodium dodecyl sulfate-Polyacrylamide gel electrophoresis). The electrophoresis conditions were 80 V for 20 min in concentrated gel, then 120 V for 90 min in separation gel. After electrophoresis, Coomassie brilliant blue R-250 staining was performed to decolorize the sample until the bands were clear.

TMT Labeling of Peptides
A protein sample was taken and a DB dissolution buffer (8 M urea, 100 mM TEAB; pH 8.5) was added to make up the volume to 100 µL. Trypsin and the 100 mM TEAB buffer were added, mixed well, and digested at 37 • C for 4 h, then trypsin and CaCl 2 were added overnight. Formic acid was added to adjust the pH to be less than 3, then the sample was mixed well and centrifuged at room temperature at 12,000× g for 5 min. The supernatant was passed through the C18 demineralizer column slowly, then washed 3 times with a washing buffer (0.1% formic acid, 3% acetonitrile). After that, an appropriate amount of an elution buffer (0.1% formic acid, 70% acetonitrile) was added to collect the filtrate and freeze-dried. The lyophilized sample was reconstituted by adding 100 µL of 0.1 M TEAB buffer and 41 µL of TMT labeling reagent dissolved in acetonitrile. The reaction was reversed and mixed at room temperature for 2 h. After that, 8% ammonia was used to stop the reaction, and the samples marked with equal volume were mixed, desalted, and lyophilized [24].

Separation of Fractions
The mobile phase composition was as follows: Mobile Phase A: 2% acetonitrile, with the pH adjusted to 10.0 using ammonium hydroxide; B: 98% acetonitrile. The lyophilized powder was dissolved with Solution A and centrifuged at 12,000× g at room temperature for 10 min. The L-3000 HPLC (high-performance liquid chromatography) system with a C18 column (Waters BEH C18, 4.6 × 250 mm, 5 µm; column temperature: 45 • C) was used to test the sample. The specific elution gradient is shown in Table 1. One tube was collected every minute and divided into 10 fractions. After freeze-drying, 0.1% formic acid was added to dissolve each fraction. The details of the elution gradient are shown in Table 2. One tube was collected per minute and divided into 10 fractions. After freezedrying, 0.1% formic acid (FA) was added to dissolve each fraction. The mobile phase composition was as follows: Mobile Phase A: 0.1% formic acid; Mobile Phase B: 80% acetonitrile and 0.1% formic acid. Next, 1 µg of the supernatant of each fraction was injected and tested. The EASY-nLC 1200 UHPLC system (Thermo Fisher) was coupled with the C18 Nano-Trap column (4.5 cm × 75 µm, 3 µm) as a homemade analytical column and the C18 Nano-Trap column (15 cm × 150 µm, 1.9 µm) as a homemade analytical column. The linear elution gradient is shown in Table 3. A Q Exactive HF-X mass spectrometer (Thermo Fisher) was used with the ion source of Nanospray Flex (ESI) to analyze the samples. The conditions were as follows: spray voltage: 2.1 kV; ion transport capillary temperature: 320 • C, full scan range: m/z 350 to 1500; primary MS resolution: 60,000 (at m/z 200); automatic gain control (AGC) target value: 3 × 10 6 ; maximum ion injection time: 20 ms. The top 40 precursors with the highest abundance in the full scan were selected and fragmented by higher-energy collisional dissociation (HCD) and analyzed by MS/MS. The conditions were as follows: resolution: 30,000 (at m/z 200); AGC target value: 5 × 10 4 ; maximum ion injection time: 54 ms; normalized collision energy: 32%; intensity threshold: 1.2 × 10 5 ; dynamic exclusion parameter: 20 s. The resulting spectra from each run were searched separately against the homo_sapiens _uniprot_2021_3_9 (194,557 sequences) database by the search engine Proteome Discoverer 2.4 (PD 2.4, Thermo). The search parameters were set as follows: mass tolerance for the precursor ion was 10 ppm and the mass tolerance for production was 0.02 Da. Carbamidomethyl was specified as a fixed modification; oxidation of methionine (M), and TMT plex were specified as dynamic modifications. Acetylation, TMT plex, Met loss, and Met-loss + Acetyl were specified as N-terminal modifications in PD 2.4. A maximum of 2 missed cleavage sites was allowed.
The software package PD 2.4 was used to improve the quality of the results for further filtering. Peptide spectrum matches (PSMs) with a reliability of more than 99% were identified as PSMs, which had to contain 1 unique peptide (5 unique peptides) or more. The identified PSMs and proteins with a FDR no more than 1.0% were retained and tested. The protein quantitation results were statistically analyzed by T-tests. The proteins whose quantitation was significantly different between the high-dose group and the control group (p < 0.05 and |log 2 FC| > 0.25 (FC > 1.2 or FC < 0.83) (fold change, FC)) were defined as differentially expressed proteins (DEP).

Functional Analysis of Protein and DEP
Gene Ontology (GO) and InterPro (IPR) functional analyses were conducted using the interproscan program against the non-redundant protein database (including Pfam, PRINTS, ProDom, SMART, ProSite, and PANTHER) [21], and the COG (Clusters of Orthologous Groups) and KEGG (Kyoto Encyclopedia of Genes and Genomes) databases were used to analyze the protein families and pathways. DEPs were used for volcanic map analysis, cluster heat map analysis, and GO, IPR, and KEGG enrichment analysis [22]. The probable protein-protein interactions were predicted using the STRING-db server (http://string.embl.de/, accessed on 23 August 2021) [23].

Statistical Analysis
All the data are presented as means ± standard deviation (SD). Statistical analysis was carried out by GraphPad Prism version 6 (GraphPad Software, Inc., San Diego, CA, USA). Difference comparisons were carried out using the T-test or one-way analysis of variance (ANOVA) using IBM SPSS 25.0 (SPSS Inc., Chicago, CA, USA); p < 0.05 was considered statistically significant.

Statistical Analysis
All the data are presented as means ± standard deviation (SD). Statistical analysis was carried out by GraphPad Prism version 6 (GraphPad Software, Inc., San Diego, CA, USA). Difference comparisons were carried out using the T-test or one-way analysis of variance (ANOVA) using IBM SPSS 25.0 (SPSS Inc., Chicago, CA, USA); p < 0.05 was considered statistically significant.
Thus, after the identification of the components of EPSI, a variety of components with pharmacological effects were identified. These proved that EPSI may have potential anti-hepatoma ability.

Growth Inhibition of EPSI in HepG2 Cells
The CCK8 and MTT methods were used to determine EPSI's inhibition of HepG2 cell proliferation. As shown in Figure 2A (CCK8), the inhibitory ability of EPSI on the proliferation of HepG2 was at a low level (< 20%) when the concentration was 100-200 µg/mL. When the dose concentration was increased to 300 µg/mL, the proliferation inhibition of EPSI in HepG2 cells increased to about 50%, which was significantly different from that at a low dose (p < 0.05). When the dosage was over 400 µg/mL, the inhibitory ability of EPSI in HepG2 increased slightly, and the maximum rate was about 58%. This result showed that the IC50 value of EPSI was between 100-200 µg/mL for the CCK8 method. When the dosage was between 100-300 µg/mL, EPSI showed an obvious inhibition ability in HepG2 cells.
Thus, after the identification of the components of EPSI, a variety of components with pharmacological effects were identified. These proved that EPSI may have potential antihepatoma ability.

Growth Inhibition of EPSI in HepG2 Cells
The CCK8 and MTT methods were used to determine EPSI's inhibition of HepG2 cell proliferation. As shown in Figure 2A (CCK8), the inhibitory ability of EPSI on the proliferation of HepG2 was at a low level (< 20%) when the concentration was 100-200 μg/mL. When the dose concentration was increased to 300 μg/mL, the proliferation inhibition of EPSI in HepG2 cells increased to about 50%, which was significantly different from that at a low dose (p < 0.05). When the dosage was over 400 μg/mL, the inhibitory ability of EPSI in HepG2 increased slightly, and the maximum rate was about 58%. This result showed that the IC50 value of EPSI was between 100-200 μg/mL for the CCK8 method. When the dosage was between 100-300 μg/mL, EPSI showed an obvious inhibition ability in HepG2 cells.
In Figure 2B, we measured the growth inhibition of EPSI in HepG2 cells by the MTT method. When the dose concentration was 25-100 μg/mL, the proliferation of HepG2 cells was still at a high level and the maximum inhibition was less than 25%. When the dosage was 200 μg/mL, EPSI increased the inhibition of HepG2 cells significantly (55%, p < 0.05). When the dosage was 400 μg/mL or higher, the inhibition ability of HepG2 improved slightly. The maximum inhibition ability of EPSI in HepG2 cells was about 67%. These results showed that the IC50 value of EPSI was 100-200 μg/mL for the MTT method. When the dosage was 100-300 μg/mL, EPSI showed good inhibition in vitro.  In Figure 2B, we measured the growth inhibition of EPSI in HepG2 cells by the MTT method. When the dose concentration was 25-100 µg/mL, the proliferation of HepG2 cells was still at a high level and the maximum inhibition was less than 25%. When the dosage was 200 µg/mL, EPSI increased the inhibition of HepG2 cells significantly (55%, p < 0.05). When the dosage was 400 µg/mL or higher, the inhibition ability of HepG2 improved slightly. The maximum inhibition ability of EPSI in HepG2 cells was about 67%. These results showed that the IC50 value of EPSI was 100-200 µg/mL for the MTT method. When the dosage was 100-300 µg/mL, EPSI showed good inhibition in vitro.

Acute Toxicity Test
The fixed-dose procedure (FDP), a classic way to determine the acute toxicity, was used to verify the acute toxicity of EPSI. The acute toxicity test of EPSI in BALB/c mice showed a LD50 of >2 g/kg, indicating that the product was safe for the experiment of the next stage. This toxicity was lower than that of conventional drugs with anticancer activity [37][38][39].

Effect of EPSI on Tumors in BALB/c Nude Mice
The bodyweight changes of all experimental animals are shown in Figure 3A. Compared with the control group, the weight of the animals in the high-dose group showed a Foods 2021, 10, 2466 9 of 20 significant reduction (p < 0.05), which happened from Day 15. During the use of anticancer drugs, the situation of weight loss occurs [40,41]. The intersection of liver cancer, anticancer drugs, and weight change may be the thyroid gland [42,43], which is closely related to the proliferation of liver cancer cells and changes in bodyweight. Whether EPSI affects the thyroid needs further study. However, for cancer patients, keeping correct weight is necessary to maintain efficacy [44,45].
The fixed-dose procedure (FDP), a classic way to determine the acute toxicity, was used to verify the acute toxicity of EPSI. The acute toxicity test of EPSI in BALB/c mice showed a LD50 of >2 g/kg, indicating that the product was safe for the experiment of the next stage. This toxicity was lower than that of conventional drugs with anticancer activity [37][38][39].

Effect of EPSI on Tumors in BALB/c Nude Mice
The bodyweight changes of all experimental animals are shown in Figure 3A. Compared with the control group, the weight of the animals in the high-dose group showed a significant reduction (p < 0.05), which happened from Day 15. During the use of anticancer drugs, the situation of weight loss occurs [40,41]. The intersection of liver cancer, anticancer drugs, and weight change may be the thyroid gland [42,43], which is closely related to the proliferation of liver cancer cells and changes in bodyweight. Whether EPSI affects the thyroid needs further study. However, for cancer patients, keeping correct weight is necessary to maintain efficacy [44,45].  The changes in tumor volume are shown in Figure 3B. The state of the tumor at the sacrifice of experimental animals is shown in Figure 4. Because of the individual differences of experimental animals, the tumor volume was not very consistent. However, EPSI can certainly inhibit the growth of tumors, because the tumor volume of the treatment group (low/middle/high-dose groups) was lower than that of the control group ( Figure 3B). A significant difference was shown between the high-dose group and the control group (p < 0.05). Meanwhile, the tumor volume of the medium-dose group and the high-dose group was significantly smaller than that of the DDP group (p < 0.05), showing good tumor inhibitory activity ( Figure 3B). Observation of the tumor state after animal sacrificed also revealed the same situation ( Figure 4): the tumor state of the treatment group was significantly different from that of other groups.
The results of H&E staining showed the state of the tumor tissue directly, and the results of the control group and the high-dose group are shown in Figure 4. Compared with the control group, the tumors in the high-dose group showed the significant reduction and destruction of tumor tissue. The tumor tissue in the control group was dense and uniform, while that of the high-dose group was disordered and less density, with even a lack of tissue. This result showed that EPSI had a significant inhibitory effect on hepatoma tumors, especially at high doses. Because EPSI passed the acute toxicity test, the high-dose group and the control group were selected for quantitative proteomic analysis.
The changes in tumor volume are shown in Figure 3B. The state of the tumor at the sacrifice of experimental animals is shown in Figure 4. Because of the individual differences of experimental animals, the tumor volume was not very consistent. However, EPSI can certainly inhibit the growth of tumors, because the tumor volume of the treatment group (low/middle/high-dose groups) was lower than that of the control group ( Figure 3B). A significant difference was shown between the high-dose group and the control group (p < 0.05). Meanwhile, the tumor volume of the medium-dose group and the high-dose group was significantly smaller than that of the DDP group (p < 0.05), showing good tumor inhibitory activity ( Figure 3B). Observation of the tumor state after animal sacrificed also revealed the same situation ( Figure 4): the tumor state of the treatment group was significantly different from that of other groups.
The results of H&E staining showed the state of the tumor tissue directly, and the results of the control group and the high-dose group are shown in Figure 4. Compared with the control group, the tumors in the high-dose group showed the significant reduction and destruction of tumor tissue. The tumor tissue in the control group was dense and uniform, while that of the high-dose group was disordered and less density, with even a lack of tissue. This result showed that EPSI had a significant inhibitory effect on hepatoma tumors, especially at high doses. Because EPSI passed the acute toxicity test, the high-dose group and the control group were selected for quantitative proteomic analysis. The effects of EPSI intake on TNF-α are shown in Figure 5. As a tumor necrosis factor, TNF-α is an important serological indicator used to measure the severity of a tumor. As shown in Figure 5, the serum TNF-α level in the control group was the lowest (116.11 The effects of EPSI intake on TNF-α are shown in Figure 5. As a tumor necrosis factor, TNF-α is an important serological indicator used to measure the severity of a tumor. As shown in Figure 5, the serum TNF-α level in the control group was the lowest (116.11 pg/mL), which was basically the same as that in the DDP group (120.42 pg/mL). The level of TNF-α in the treatment group showed an obvious upward trend with an increase in the dosage. Although the low-dose group (124.82 pg/mL) increased slightly, no significant differences (p > 0.05) were seen compared with the control group. TNF-α in the middle-dose group (149.89 pg/mL) and the high-dose group (169.29 pg/mL) were significantly increased (p < 0.05) compared with the control group.
BALB/c mice are innately immune-deficient animals. The test animals have no thymus, resulting in T-cell defects and dysfunction. At this time, we can judge that the higher the TNF-α content, the stronger the immune response of the tested animals and the more obvious the improvement of their immunity. Therefore, these results mean that with the increase in the EPSI dosage, the limited immune system of the test animals was activated and produced a stronger immune response in the case of a lack of immune function. pg/mL), which was basically the same as that in the DDP group (120.42 pg/mL). The level of TNF-α in the treatment group showed an obvious upward trend with an increase in the dosage. Although the low-dose group (124.82 pg/mL) increased slightly, no significant differences (p > 0.05) were seen compared with the control group. TNF-α in the middledose group (149.89 pg/mL) and the high-dose group (169.29 pg/mL) were significantly increased (p < 0.05) compared with the control group.
BALB/c mice are innately immune-deficient animals. The test animals have no thymus, resulting in T-cell defects and dysfunction. At this time, we can judge that the higher the TNF-α content, the stronger the immune response of the tested animals and the more obvious the improvement of their immunity. Therefore, these results mean that with the increase in the EPSI dosage, the limited immune system of the test animals was activated and produced a stronger immune response in the case of a lack of immune function.

Protein Expression Differences Induced by EPSI
The proteomic data of the tumors from the high-dose group of BALB/c mice are shown in Figure 6A. In total, 116 differentially expressed proteins (DE proteins) and 6787 proteins were identified in this experiment. On the whole, the number of upregulated DE proteins (99) was far more than that of downregulated DE proteins (17) between the control group and high-dose group after the intervention, and the upregulated DE protein amounts in the high-dose group were over 3 times higher than that in the control group when the fold change was 1.2×. The volcano map of the DE proteins with a 1.2× fold change is shown in Figure 6B. According to Figure 6B, the number of upregulated and downregulated proteins in the treatment group had unique proteins.

Protein Expression Differences Induced by EPSI
The proteomic data of the tumors from the high-dose group of BALB/c mice are shown in Figure 6A. In total, 116 differentially expressed proteins (DE proteins) and 6787 proteins were identified in this experiment. On the whole, the number of upregulated DE proteins (99) was far more than that of downregulated DE proteins (17) between the control group and high-dose group after the intervention, and the upregulated DE protein amounts in the high-dose group were over 3 times higher than that in the control group when the fold change was 1.2×. The volcano map of the DE proteins with a 1.2× fold change is shown in Figure 6B. According to Figure 6B, the number of upregulated and downregulated proteins in the treatment group had unique proteins.

GO Enrichment of Differentially Quantified Proteins
Following the GO classifications for biological process (BP), cellular component (CC), and molecular function (MF), the enrichment results were examined to compare the functional correlations of DE proteins; the results are shown in Figure 7. In the upregulated proteins ( Figure 7A), the proteins had functions in biological process and molecular function, but not cell component. These proteins are mainly concentrated (>

GO Enrichment of Differentially Quantified Proteins
Following the GO classifications for biological process (BP), cellular component (CC), and molecular function (MF), the enrichment results were examined to compare the functional correlations of DE proteins; the results are shown in Figure 7. In the upregulated proteins ( Figure 7A), the proteins had functions in biological process and molecular function, but not cell component. These proteins are mainly concentrated (>20%) in protein metabolic processes (BP), metabolic processes (BP), catalytic activity (MF), and hydrolase activity (MF). In the downregulated proteins ( Figure 7B), by comparison, all three classifications of proteins existed, and their functions showed a high concentration. Proteolysis (BP), endopeptidase activity (MF), and serine-type endopeptidase activity (MF) were the main roles of these proteins. These results suggest that the intake of EPSI associated with the anti-hepatoma effect may be achieved through specific mechanisms.

GO Enrichment of Differentially Quantified Proteins
Following the GO classifications for biological process (BP), cellular component (CC), and molecular function (MF), the enrichment results were examined to compare the functional correlations of DE proteins; the results are shown in Figure 7. In the upregulated proteins ( Figure 7A), the proteins had functions in biological process and molecular function, but not cell component. These proteins are mainly concentrated (> 20%) in protein metabolic processes (BP), metabolic processes (BP), catalytic activity (MF), and hydrolase activity (MF). In the downregulated proteins ( Figure 7B), by comparison, all three classifications of proteins existed, and their functions showed a high concentration. Proteolysis (BP), endopeptidase activity (MF), and serine-type endopeptidase activity (MF) were the main roles of these proteins. These results suggest that the intake of EPSI associated with the anti-hepatoma effect may be achieved through specific mechanisms.

Specific Regulation Pathways for Inhibiting Liver Tumor Proliferation by EPSI
To further explore the related pathway of EPSI for anti-hepatoma effects and inhibition of liver tumors, KEGG pathway enrichment analysis was carried out (Figure 8). In the upregulated proteins ( Figure 8A), the highest degree of protein enrichment was in caffeine metabolism, while most DE proteins with higher reliability appeared in the reninangiotensin system. In the downregulated proteins ( Figure 8B), two kinds appeared in neuroactive ligand-receptor interaction and hematopoietic cell lineage. The results showed that the inhibition of liver tumors by EPSI was closely related to the above pathways.
In the detailed KEGG pathway enrichment results (Table 5), we found 12 typical enrichment pathways in the process of inhibiting liver tumors by EPSI (p < 0.05). The names of the enriched KEGG pathways and the enriched proteins with significant changes are also listed in Table 5, while a description of the proteins is shown in Table 6.
Structural domain enrichment can identify the domain entries that are statistically significantly enriched. This function or positioning may be the cause of the differences. From the statistics of the enrichment results, a bubble chart of the structural domain is shown in Figure 9. As for the previous expression, peptidase S1/S6, chymotrypsin/Hap, peptidase cysteine/serine, and trypsin-like should be considered as the typical structural domains.    Christmas factor (Fragment) Prot ID: the enriched protein list; description: the function of the protein described in the database.

Interaction Analysis of Differentially Expressed Proteins
According to the KEGG enrichment results (Table 5), we found that a few DE proteins appeared in multiple pathways, such as Q08426, E7ESP4, etc. These results showed that there should be a close relationship between the DE proteins, and between the pathways. The protein interaction network is shown in Figure 10. There are three main associated relationships in the network: the first is centered on P17301, the downregulation of which caused upregulation of other three proteins (E1B4S8, P12830, Q8TDR6) and downregulation of the other four (Q19UG4, A0A0F7G8J1, Q597H1, and Q99542). The second line established the relationships among Q6B051, B2R941, and P23946. In addition, there was also an association between Q08426 and P48163. The above results established the relationship between hematopoietic cell lineage, vitamin digestion and absorption, ARVC, complement and coagulation cascades, and fat digestion and absorption.
However, we were also aware that the situation that the metabolic pathways and related DE proteins summarized above may not be all related to EPSI. Because EPSI is not soluble in water, a certain amount of ethanol must be used as the solvent of EPSI during the feeding of experimental animals. Therefore, the above differences might have the effect of ethanol intervention in the body. We should probably pay more attention to what other pathways the related DE proteins appeared in. Therefore, we summarized the relevant information (Table 7).

Interaction Analysis of Differentially Expressed Proteins
According to the KEGG enrichment results (Table 5), we found that a few DE proteins appeared in multiple pathways, such as Q08426, E7ESP4, etc. These results showed tha there should be a close relationship between the DE proteins, and between the pathways The protein interaction network is shown in Figure 10. There are three main associated relationships in the network: the first is centered on P17301, the downregulation of which caused upregulation of other three proteins (E1B4S8, P12830, Q8TDR6) and downregulation of the other four (Q19UG4, A0A0F7G8J1, Q597H1, and Q99542). The second line established the relationships among Q6B051, B2R941, and P23946. In addition there was also an association between Q08426 and P48163. The above results established the relationship between hematopoietic cell lineage, vitamin digestion and absorption ARVC, complement and coagulation cascades, and fat digestion and absorption. In this result, nine DE proteins and 54 related KEGG pathways were shown; meanwhile, the pathways that have been reported to be directly associated with cancer were marked . Among these, some are directly related to liver cancer or tumors, such as platelet activation [75], the PI3K-Akt signaling pathway [76], and the PPAR signaling pathway [77], etc. We found that, besides liver cancer, these signaling pathways appeared in studies on cancers of the breast, colon, and thyroid frequently. There are reasons to believe that the inhibitory effect of EPSI on liver tumors may be the same as the inhibitory mechanism of these cancers. We will carry out a demonstration of the metabolic pathways based on the results of this study in follow-up studies.  Figure 10. Protein interaction network. Each node in the interaction network represents a protein, and the node size represents the number of interacting proteins. The larger the node is, the more proteins interact with it. The color of the node indicates the expression level of the protein in the comparison pair: red represents significantly high expression of the protein and blue represents significantly low expression of the protein.
However, we were also aware that the situation that the metabolic pathways and related DE proteins summarized above may not be all related to EPSI. Because EPSI is not soluble in water, a certain amount of ethanol must be used as the solvent of EPSI during the feeding of experimental animals. Therefore, the above differences might have the effect of ethanol intervention in the body. We should probably pay more attention to what other pathways the related DE proteins appeared in. Therefore, we summarized the relevant information (Table 7). Table 7. KEGG pathways associated with DE proteins which had relationships in the network.

P17301
Hematopoietic cell lineage * Platelet activation * Arrhythmogenic right ventricular cardiomyopathy (ARCV) ECM-receptor interaction * Hypertrophic cardiomyopathy (HCM) Dilated cardiomyopathy (DCM) Small-cell lung cancer * Proteoglycans in cancer * Focal adhesion * Phagosome * Regulation of actin cytoskeleton * PI3K-Akt signaling pathway * Human papillomavirus infection * Pathways in cancer * Figure 10. Protein interaction network. Each node in the interaction network represents a protein, and the node size represents the number of interacting proteins. The larger the node is, the more proteins interact with it. The color of the node indicates the expression level of the protein in the comparison pair: red represents significantly high expression of the protein and blue represents significantly low expression of the protein.

Conclusions
In summary, we identified the main components of EPSI by LC-MS and determined that it contained a variety of anti-inflammatory and anti-cancer components. On this foundation, we found that the EPSI showed good growth inhibition of HepG2 cells in vitro, while the high dose of EPSI had an obvious anti-hepatoma effect in vivo, and this effect was better than that of DDP without toxicity. To explore its anti-hepatoma mechanism, a TMT quantitative proteomic approach was used to examine the inhibition of liver tumors by EPSI. DE proteins and their relevance, and the related metabolic pathways were counted and displayed. The upregulated pathways, such as the renin-angiotensin system, and the downregulated pathways, such as hematopoietic cell lineage, and the changes in the PI3K-Akt signaling pathway and the PPAR signaling pathway may be the significant keys to the anti-hepatoma effects. Moreover, the key DE protein intervening in the pathways is probably a new target for the treatment of liver tumors. Moreover, our work shines a new light on the crucial inhibitive function of S. involucrata in the progression of liver disease.