Saikosaponin A Alleviates Hepatocellular Carcinoma Growth Induced by Chronic Stress via Regulating Tumor Immune Microenvironment Authors: Juanjuan Yang, Wei Jiang Published: 2026 ABSTRACT BackgroundThe tumor immune microenvironment is one of the main factors contributing to tumor metastasis, recurrence and drug resistance. Chronic stress is an important factor mediating the occurrence and progression of tumors by regulating tumor immune microenvironment. Saikosaponins have anti-stress and immunomodulatory effects. However, the therapeutic effects and mechanisms of Saikosaponins in chronic stress-induced tumor progression remain unclear.AimThis study aimed to investigate the therapeutic effects of Saikosaponins on stress-induced tumor growth and further elucidate the underlying mechanisms.MethodsA mouse model was established using chronic restraint stress. The open field test was used to assess mouse behaviors. Stress hormone levels were measured by ELISA. The levels of cytokines and chemokines were measured by using Luminex. The proportions of T lymphocytes, monocytic-like myeloid-derived suppressor cells (mMDSC), granulocyte-like MDSC (gMDSC), and macrophages were measured by flow cytometry. A CCK8 assay was used to assess tumor cell proliferation. The expressions of adrenergic receptor and Ki67 were measured by immunohistochemistry. Propranolol and isoprenaline were used to investigate the underlying mechanisms.ResultsFirst, we found that Saikosaponin A (SSA) reversed stress-induced depressive behavior. Second, SSA significantly inhibited hepatocellular carcinoma growth and reversed the stress-induced upregulation of Ki67 in tumor tissues. Furthermore, SSA reversed the changes of CD3+CD4+ T lymphocytes, CD3+CD8+ T lymphocytes, M2 macrophages, gMDSC and mMDSC induced by chronic restraint stress. Mechanistically, SSA decreased norepinephrine (NE) levels and reduced the expression of β2 adrenergic receptor in chronically stressed mice. Finally, we found that SSA inhibited hepatocellular carcinoma growth and reversed tumor immune microenvironment alterations in stressed mice by blocking β-adrenergic signaling.ConclusionOur results suggested that SSA inhibits hepatocellular carcinoma growth by blocking β-adrenergic signaling, thereby enhancing the anti-tumor immune response. Our study provides new insights into the importance of the psychological neuroendocrine immune intervention in the treatment of hepatocellular carcinoma. Introduction Primary liver cancer has a high incidence and poor prognosis, encompassing hepatocellular carcinoma (HCC), cholangiocarcinoma, and mixed cell-type liver cancer. Primary liver cancer ranks sixth in incidence and third in mortality among all malignant tumors.1 Among these, HCC is the most prevalent primary liver cancer. The main treatments for HCC include surgical resection, chemotherapy, and targeted therapy; however, treatment efficacy remains unsatisfactory, and the prognosis for patients with HCC is extremely poor.2 HCC remains a significant health issue in China. Therefore, elucidating the mechanisms underlying HCC progression and exploring novel therapeutic approaches are of critical importance. The tumor immune microenvironment comprises inflammatory factors, stromal cells, and various immune cell populations, including tumor-associated neutrophils (TAN), myeloid-derived suppressor cells (MDSC), tumor-associated macrophages (TAM), and lymphocytes. The tumor immune microenvironment is one of the main contributing factors to tumor metastasis, recurrence, and drug resistance.3 Accumulating evidence indicates that chronic stress is an important factor mediating tumor occurrence and progression through excessive activation of neuroendocrine pathways, such as the hypothalamic-pituitary-adrenal axis and the sympathetic nervous system, leading to the release of norepinephrine (NE), epinephrine (Epi), and corticosterone (Cort).4,5 Our previous studies have shown that chronic stress promotes HCC growth by mobilizing splenic MDSC and regulating the M1-M2 polarization balance of macrophage via activating β-adrenergic receptor.6,7 Pharmacologically reducing NE and Epi levels, interfering with their signaling pathways, or modulating the tumor immune microenvironment has become a priority strategy in the search for antitumor drugs for cancer patients experiencing psychological stress. However, non-selective β-adrenergic receptor blockers and immunotherapeutic agents are associated with side effects, including cardiac disorders, rash, fatigue, thyroid dysfunction, and immune-related pneumonia. Therefore, it is necessary to seek more effective and safer treatments or drugs. Traditional Chinese medicine (TCM) is an important source of pharmaceutical products and has made significant contributions to disease treatment. Chaihu (Radix Bupleuri) is a key component of the classic antidepressant formulation Xiaoyao San in China.8 Saikosaponins are isolated from Radix Bupleuri. Modern medical research has demonstrated that Saikosaponin A (SSA) and Saikosaponin D (SSD) possess antidepressant, antitumor, and immunomodulatory properties.9–12 However, the therapeutic effects of SSA and SSD on chronic stress-induced tumor progression remain unclear. Their effects on the tumor immune microenvironment are also not well understood. In this study, we aimed to investigate the therapeutic effects of SSA and SSD on stress-induced HCC growth and to further elucidate the underlying mechanisms. We found that SSA inhibited the HCC growth and reversed the changes in tumor immune microenvironment in stressed mice by blocking β-adrenergic signaling. Materials and Methods Cell Lines and Mice HCC cell line Hepa1-6 and H22 were obtained from the Cell Bank of the Chinese Academy of Sciences (No. TCM39, Shanghai, China) and China Center for Type Culture Collection (No. GDC901, Wuhan, China), respectively. H22 and Hepa1-6 cells were cultured in RPMI-1640 (HyClone, USA) supplemented with 10% fetal bovine serum (FBS, Excell, China) and 1% penicillin/streptomycin. Male C57BL/6 mice were provided by the Experimental Animal Center of Xi’an Jiaotong University. Mice were acclimated to the environment at 24°C for 14 days prior to the experiment, with 4–5 mice per cage. The in vivo animal study was conducted in accordance with the guidelines of the National Institutes of Health of the United States and approved by the Ethics Committee of Xi’an Jiaotong University College of Medicine. Reagents The following reagents were purchased from BioLegend (CA, USA): Purified anti-CD16/32 (# 101325), APC anti-CD11b (# 101212), Pacific BlueTM anti-CD45 (#103126), FITC anti-F4/80 (# 123107), PE-Cy7 anti-Ly6C (# 128017), PerCP-Cy5.5 anti-Ly6G (# 127615), PerCP-Cy5.5 anti-CD206 (# 141715), FITC anti-CD3 (# 100203), PE anti-CD4 (# 100407), and APC anti-CD8 (# 100712). PE anti-iNOS (# 12–5920-82) was purchased from eBioscience (MA, USA). The following reagents were purchased from Abcam (England, UK): propranolol hydrochloride (# 120757), isoprenaline hydrochloride (# 120710), and anti-β2 adrenergic receptor (# 182136). Anti-Ki67 mouse mAb (# GB121141) was purchased from Servicebio (Wuhan, China). SSA (# 20736–09-8, purity ≥ 98%) and SSD (# 20874–52-6, purity ≥ 98%) were purchased from Psaitong (Beijing, China). Animal Models The restraint stress model and H22 HCC tumor model were established as previously described.13 Briefly, the mice were placed in ventilated 50-mL conical tubes for 2 hours each day between 17:00–21:00 PM Beijing time. For the tumor model, 5×105 H22 cells (in 100 µL) were subcutaneously injected into the right flank after 5 days of stress exposure. Mice in the no-stress group were inoculated with tumor cells at the same time point, and mice in the stress group were subjected to an additional 21 days of stress. For pharmacological treatments, SSA (5 mg/kg), propranolol hydrochloride (Pro, 15 mg/kg) and isoprenaline hydrochloride (ISO, 10 mg/kg) were injected intraperitoneally (IP) 30 min before stress.7,14,15 SSD (0.5 mg/kg) was administered by gavage.10 Mice were anesthetized by intraperitoneal injection of tribromoethanol (0.33mL/20g) before sacrifice. After anesthesia, the mice were sacrificed by cervical dislocation, and blood and tumors were collected. Mice were randomly assigned to each group using a random sequence generated by Excel. Each group consists of 4–5 mice. The study scheme is presented in Figure 1A. Figure 1Effect of SSA on depression-like behaviors in mice. The study scheme (A). The effect of SSA and SSD on body weight (B) and gain weight (C) of chronic stress mice. Representative images of mouse trajectory in open field (D). The effect of SSA and SSD on total distance traveled (E) latency time first enter the center (F) and entries in the center (G) of chronic stress mice in the open field. Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01.A scientific figure with panels A to G. Panel A outlines a study with four groups: Vehicle, Vehicle+stress, Stress+SSA and Stress+SSD. Each group involves a 30-minute pre-stress treatment, followed by a 5-day restraint, H22 cell inoculation and a 21-day restraint, leading to various analyses. Panel B shows body weight (22-24g) across groups. Panel C displays weight gain (3-4g). Panel D presents open-field trajectories for each group. Panel E shows distance traveled: Vehicle (2100 cm), Vehicle+stress (1500 cm), Stress+SSA (2000 cm), Stress+SSD (1900 cm), with significant differences marked. Panel F illustrates latency to center: Vehicle (25s), Vehicle+stress (65s), Stress+SSA (25s), Stress+SSD (35s), with significant differences. Panel G shows center entries: Vehicle (18), Vehicle+stress (8), Stress+SSA (16), Stress+SSD (15), with significant differences. Data are mean ± SEM (n=5), with significance levels <0.05 and *<0.01.A multi-panel scientific figure on SSA and SSD effects on stress mouse open-field behavior and body weight. Effect of SSA on depression-like behaviors in mice. The study scheme (A). The effect of SSA and SSD on body weight (B) and gain weight (C) of chronic stress mice. Representative images of mouse trajectory in open field (D). The effect of SSA and SSD on total distance traveled (E) latency time first enter the center (F) and entries in the center (G) of chronic stress mice in the open field. Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01. Open Field Test The open field test was used to assess mouse behavior as previously described.16 Briefly, mice were placed into a corner of the open field behavioral observation box for 5 minutes. A computerized video tracking system was used to record their movements. The total distance traveled, latency to enter the center, and the number of entries into the center were analyzed. Open field test was conducted by a blinded approach. Flow Cytometric Analysis Single-cell suspensions were prepared as previously described.7 Flow cytometric was performed as previously described.7 Cells were detected by flow cytometer (BECKMAN CytoFLEX, Brea, California, USA) and data were analyzed using Flow Jo (Tree Star, Ashland, OR, USA). Stress Hormones Measurements Blood samples were allowed to coagulate to obtain serum. NE levels were measured using the norepinephrine ELISA kit (Elabscience, # E-CL-0047, Wuhan, China) and Cort levels were measured using the corticosterone ELISA kit (Elabscience, # E-OSEL-M0001, Wuhan, China), following the manufacturer’s instructions. Luminex Serum samples of stress group mice and stress + SSA group mice were collected as we described above. Cytokines and chemokines were measured using the mouse Luminex kit (Univ, # LXR-MultiDTM-31, Shanghai, China) according to the instructions provided by the merchant. CCK-8 Assay H22 cells and Hepa1-6 cells (5*103/well) were plated in 96-well plates with 100 μL 1640 medium. Different concentrations of SSA and SSD (5 μM, 10 μM, 20 μM, 40 μM) were added to stimulate tumor cells for 48 hours. Then we added 10 μL of CCK-8 solution (NCM Biotech, # C6005, Suzhou, China) to each well, and incubated for 2 hours. We used microplate reader to measure the optical density (OD) at 450 nm of the medium. Hematoxylin and Eosin (HE) Staining Tumor tissues were sectioned to a thickness of 5 μm, deparaffinized in xylene and graded ethanol solutions, and rehydrated in water. Sections were stained with hematoxylin for 5 minutes, washed, differentiated, and rinsed with tap water, followed by staining with eosin for 1 minute, dehydration, and mounting in neutral resin. HE analysis was conducted using a blinded approach. Immunohistochemistry (IHC) Tumor tissues were sectioned to 5 μm thickness, deparaffinized in xylene and graded ethanol solutions, rehydrated in water, subjected to antigen retrieval, and treated to inactivate endogenous peroxidase. Sections were incubated with primary antibodies against β2 adrenergic receptor and Ki67 (1:50 dilution) overnight at 4°C, followed by incubation with appropriate secondary antibodies for 30 minutes at room temperature. Signals were detected using a DAB kit (Servicebio, Wuhan, China), and sections were counterstained with hematoxylin. Images were analyzed using ImageJ software. IHC analysis was conducted by a blinded approach. Statistical Analysis Statistical analysis was performed using SPSS 24.0 (IBM Corp., Armonk, NY, USA). Data were presented as mean ± standard error of the mean (SEM). The Kolmogorov–Smirnov test was used to assess normality. Student’s t-test, Mann–Whitney U-test, and one-way ANOVA were used for group comparisons. Pearson’s correlation and linear regression analysis were used to evaluate correlations. P < 0.05 was considered statistically significant. Results The Effect of SSA on Depression-Like Behaviors in Mice We investigated the effects of SSA and SSD on depression-like behaviors in chronically stressed mice. SSA (5 mg/kg) and SSD (0.5 mg/kg) were administered the mice 30 min before stress (Figure 1A). Neither SSA nor SSD affected body weight (Figure 1B, P > 0.05) or weight gain (Figure 1C, P > 0.05) in stressed mice. The behaviors of mice were examined using open field test. Representative trajectories of mice are shown in Figure 1D. Restraint stress reduced the total distance traveled in the open field (Figure 1E, P < 0.05). Notably, SSA blocked this reduction (Figure 1E, P < 0.05). Chronic stress increased the latency to first enter the center of the open field (Figure 1F, P < 0.05). Similarly, SSA reversed this effect (Figure 1F, P < 0.05). The number of center entries was lower in the stress group than in the control group (Figure 1G, P < 0.01). However, both SSA and SSD increased the number of center entries compared with the stress group (Figure 1G, all P < 0.01), indicating that both compounds reversed the stress-induced decrease in center entries. The Effect of SSA on HCC Growth in Stressed Mice We next investigated the effects of SSA and SSD on HCC growth in stressed mice. Consistent with our previous findings, chronic restraint stress significantly promoted HCC growth (Figure 2A and B, P < 0.01). SSA reversed the HCC growth induced by chronic stress (Figure 2A and B, P < 0.01). SSD also partially reversed chronic stress-induced HCC growth, although the difference did not reach statistical significance (Figure 2A and B, P > 0.05). The HE staining of tumors were shown in Figure 2C. Similarly, SSA blocked the stress-induced upregulation of Ki67 in tumor tissues (Figure 2D). To determine whether SSA and SSD directly inhibit HCC growth, we examined their effects on the proliferation of Hepa1-6 and H22 cells in vitro. Low concentration of SSA (5μM) and SSD (5μM) did not inhibit H22 and Hepa1-6 cell proliferation (all P > 0.05), whereas higher concentrations (10 μM, 20 μM, 40 μM) significantly reduced tumor cell proliferation (Figure 2E and F). Figure 2Effect of SSA on HCC growth in stressed mice. Photos of tumor blocks collected from mice in each group (A). The effect of SSA and SSD on the HCC growth of chronic stress mice (B). Representative images of HE staining (C), and Ki67 (D) The effect of SSA and SSD on the proliferation of H22 (E) and Hepa1-6 (F) cells in vitro. Data were presented as mean ± SEM (n = 5–8). *p < 0.05, **p < 0.01.Image A displays tumor blocks in rows: Vehicle, Stress, Stress + SSA, Stress + SSD. Image B is a bar chart of tumor volume (mm³) with categories: Vehicle, Vehicle + Stress, Stress + SSA, Stress + SSD. Bar heights: Vehicle ~150; Vehicle + Stress ~550-600; Stress + SSA ~250; Stress + SSD ~420-450. Significance indicated by asterisks. Image C shows HE images labeled Vehicle, Stress, Stress + SSA, Stress + SSD, with 100 μm scale bars. Image D shows Ki67 images with similar labels and scale bars. Image E is a bar chart titled H22, showing cell proliferation (% of control) across concentrations 5μM, 10μM, 20μM, 40μM, grouped by Control, SSA, SSD. Image F is a bar chart titled Hepa1-6, also showing cell proliferation (% of control) across the same concentrations, grouped similarly. Double asterisks mark several dose groups.A multi-panel infographic on SSA and SSD effects on HCC tumors and cell proliferation under stress. Effect of SSA on HCC growth in stressed mice. Photos of tumor blocks collected from mice in each group (A). The effect of SSA and SSD on the HCC growth of chronic stress mice (B). Representative images of HE staining (C), and Ki67 (D) The effect of SSA and SSD on the proliferation of H22 (E) and Hepa1-6 (F) cells in vitro. Data were presented as mean ± SEM (n = 5–8). *p < 0.05, **p < 0.01. The Effect of SSA on Immune Cell in Stressed Mice Our previous studies have shown that chronic stress promotes HCC growth by mobilizing splenic MDSC and regulating the M1-M2 macrophage balance. We therefore investigated the effects of SSA and SSD on immune cells in stressed mice. Chronic stress reduced the frequencies of CD3+CD4+ T lymphocytes (Figure 3A and B, P < 0.01) and CD3+CD8+ T lymphocytes (Figure 3C and D, P < 0.01) in the blood. SSA, but not SSD, partly reversed the stress-induced decrease in blood CD3+CD4+ T lymphocytes (Figure 3A and B) and CD3+CD8+ T lymphocytes (Figure 3C and D). Similarly, SSA, but not SSD, reversed the stress-induced increase in blood gMDSC (Figure 3E) and mMDSC (Figure 3F) frequencies. No significant differences in the proportions of CD3+CD4+ or CD3+CD8+ T lymphocytes in tumor tissues were observed among the chronic restraint stress, SSA, and SSD groups (Figure 3G and H; all P > 0.05). SSA, but not SSD, reversed the stress-induced increase in tumor gMDSC frequencies (Figure 3I). No significant differences in tumor mMDSC or CD11b+F4/80+iNOS+ M1 macrophage frequencies were observed among the groups (Figure 3J and K; all P > 0.05). Finally, SSA significantly reversed the stress-induced increase in tumor CD11b+F4/80+CD206+ M2 macrophage frequencies (Figure 3L and M). SSD also appeared to reverse this effect, although the difference was not statistically significant (Figure 3L and M, P > 0.05). Figure 3Effect of SSA on immune cell of stressed mice. Representative dot plot flow figures of blood CD3+CD4+ T lymphocytes (A). The effects of SSA and SSD on the frequencies of blood CD3+CD4+ T lymphocytes of chronic stress mice (B). Representative dot plot flow figures of blood CD3+CD8+ T lymphocytes (C). The effects of SSA and SSD on the frequencies of blood CD3+CD8+ T lymphocytes (D) blood gMDSC (E), blood mMDSC (F), tumor CD3+CD4+ T lymphocytes (G) tumor CD3+CD8+ T lymphocytes (H), tumor gMDSC (I), tumor mMDSC (J), tumor M1 macrophages (K) and tumor M2 macrophages (L) of chronic stress mice. Representative dot plot flow figures of tumor M2 macrophages (M). Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01.This study examines the effects of SSA and SSD on immune cells in stressed mice using flow cytometry and bar charts. Stress reduces CD3+CD4+ and CD3+CD8+ T lymphocyte frequencies, with SSA partially restoring these levels, while SSD does not. Stress increases granulocytic and monocytic myeloid-derived suppressor cells, but SSA reverses this increase. In tumors, no significant differences are observed in CD3+CD4+ and CD3+CD8+ T lymphocytes. However, stress increases tumor granulocytic myeloid-derived suppressor cells, with SSA reversing this effect. Tumor monocytic myeloid-derived suppressor cells show no significant differences. M1 macrophages remain unchanged, while M2 macrophages increase with stress, but SSA significantly reverses this. Flow cytometry confirms stress-induced M2 macrophage increase and SSA-mediated reversal. Overall, SSA shows potential in mitigating stress-induced immune changes, unlike SSD.Flow cytometry plots show immune cell changes in Vehicle, Vehicle+stress, Stress+SSA and Stress+SSD groups. Effect of SSA on immune cell of stressed mice. Representative dot plot flow figures of blood CD3+CD4+ T lymphocytes (A). The effects of SSA and SSD on the frequencies of blood CD3+CD4+ T lymphocytes of chronic stress mice (B). Representative dot plot flow figures of blood CD3+CD8+ T lymphocytes (C). The effects of SSA and SSD on the frequencies of blood CD3+CD8+ T lymphocytes (D) blood gMDSC (E), blood mMDSC (F), tumor CD3+CD4+ T lymphocytes (G) tumor CD3+CD8+ T lymphocytes (H), tumor gMDSC (I), tumor mMDSC (J), tumor M1 macrophages (K) and tumor M2 macrophages (L) of chronic stress mice. Representative dot plot flow figures of tumor M2 macrophages (M). Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01. Correlation analyses between tumor volume and immune cell populations were also performed. The percentages of blood CD3+CD4+ and CD3+CD8+ T lymphocytes were negatively correlated with tumor volume, with Pearson’s r values of −0.705 (P = 0.001; Figure 4A) and −0.745 (P < 0.001; Figure 4B), respectively. In contrast, blood gMDSC and mMDSC percentages were positively correlated with tumor volume, with Pearson’s r values of 0.791 (P < 0.001; Figure 4C) and 0.740 (P < 0.001; Figure 4D), respectively. Similarly, tumor gMDSC and M2 macrophage percentages were positively correlated with tumor volume, with Pearson’s r values of 0.839 (P < 0.001; Figure 4E) and 0.626 (P = 0.003; Figure 4F), respectively. Figure 4Correlation between tumor size and immune cells. The correlation between tumor size and blood CD3+CD4+ T lymphocytes (A). The correlation between tumor size and blood CD3+CD8+ T lymphocytes (B). The correlation between tumor size and blood gMDSC (C). The correlation between tumor size and blood mMDSC (D). The correlation between tumor size and tumor gMDSC (E). The correlation between tumor size and tumor CD11b+F4/80+CD206+ M2 macrophages (F).Image A: Scatter plot with Tumor (mm³) vs. Blood CD3+CD4+ T cells (%). Range: (50, 12) to (900, 4). r = -0.705, P = 0.001. Downward trend. Image B: Tumor (mm³) vs. Blood CD3+CD8+ T cells (%). Range: (50, 9) to (850, 3). r = -0.745, P < 0.001. Downward trend. Image C: Tumor (mm³) vs. Blood CD11b+Ly6C+Ly6G+ cells (%). Range: (80, 20) to (900, 65). r = 0.791, P < 0.001. Upward trend. Image D: Tumor (mm³) vs. Blood CD11b+Ly6C+Ly6G- cells (%). Range: (80, 5) to (850, 18). r = 0.740, P < 0.001. Upward trend. Image E: Tumor (mm³) vs. Tumor CD11b+Ly6C+Ly6G+ cells (%). Range: (80, 3) to (850, 9). r = 0.839, P < 0.001. Upward trend. Image F: Tumor (mm³) vs. Tumor CD11b+F4/80+CD206+ cells (%). Range: (80, 0.5) to (900, 1.7). r = 0.626, P = 0.003. Upward trend.A set of six scatter plots showing tumor volume correlations with immune cell percentages. Correlation between tumor size and immune cells. The correlation between tumor size and blood CD3+CD4+ T lymphocytes (A). The correlation between tumor size and blood CD3+CD8+ T lymphocytes (B). The correlation between tumor size and blood gMDSC (C). The correlation between tumor size and blood mMDSC (D). The correlation between tumor size and tumor gMDSC (E). The correlation between tumor size and tumor CD11b+F4/80+CD206+ M2 macrophages (F). The Effects of SSA on Cytokines and Chemokines in Stressed Mice SSA reversed the changes in CD3+CD4+ T lymphocytes, CD3+CD8+ T lymphocytes, gMDSC, mMDSC, and M2 macrophages induced by chronic restraint stress. We next investigated the effects of SSA on cytokines and chemokines levels in stressed mice. Serum samples were collected from mice in the stress group and the stress + SSA group. The levels of cytokines and chemokines were measured by Luminex (Figure 5A). SSA decreased interleukin-6 (IL-6) levels in stressed mice (Figure 5B, P < 0.05). Similarly, SSA decreased CX3CL1 (Figure 5C, P < 0.05) and CXCL16 (Figure 5D, P < 0.01) levels. Conversely, SSA increased CCL20 levels (Figure 5E, P < 0.05). Figure 5Effect of SSA on cytokines and chemokines of stressed mice. The heatmap of cytokine and chemokine detection results using Luminex (A). The effects of SSA on the levels of IL-6 (B) CX3CL1 (C), CXCL16 (D) and CCL20 (E) of stressed mice. Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01.Image A displays a heatmap with rows S1-S4, A1-A5 and columns CCL20, IL-1beta, CXCL5, TNFalpha, GM-CSF, CX3CL1, CCL1, CXCL13, IL-16, IFN-gamma, CXCL10, IL-2, CXCL11, IL-6, IL-4, CCL7, CCL12, CCL2, CCL22, CCL5, IL-10, CXCL1, CCL27, CCL17, CCL19, CXCL16, CXCL12, CCL24, CCL11, CCL4, CCL3. Color scale: 1 to 50000. Image B shows a bar graph with x-axis: Stress, Stress plus SSA; y-axis: IL-6 (pg/ml) 0-400. Bar heights: Stress ~270; Stress plus SSA ~210, with an asterisk. Image C shows a bar graph with x-axis: Stress, Stress plus SSA; y-axis: CX3CL1 (pg/ml) 0-1500. Bar heights: Stress ~1200; Stress plus SSA ~1000, with an asterisk. Image D shows a bar graph with x-axis: Stress, Stress plus SSA; y-axis: CXCL16 (pg/ml) 0-1500. Bar heights: Stress ~1100; Stress plus SSA ~850, with two asterisks. Image E shows a bar graph with x-axis: Stress, Stress plus SSA; y-axis: CCL20 (pg/ml) 0-300. Bar heights: Stress ~170; Stress plus SSA ~230, with an asterisk.A heatmap and four bar graphs showing cytokine and chemokine levels for Stress and Stress plus SSA groups. Effect of SSA on cytokines and chemokines of stressed mice. The heatmap of cytokine and chemokine detection results using Luminex (A). The effects of SSA on the levels of IL-6 (B) CX3CL1 (C), CXCL16 (D) and CCL20 (E) of stressed mice. Data were presented as mean ± SEM (n = 5). *p < 0.05, **p < 0.01. SSA Inhibited HCC Growth in Stressed Mice by Blocking β-Adrenergic Signaling Our previous studies have demonstrated that chronic stress promotes HCC progression through activation of β-adrenergic receptors. We therefore examined the effects of SSA and SSD on the stress hormones. SSA, but not SSD, significantly reversed the stress-induced increase in NE levels (Figure 6A). Both SSA and SSD appeared to reduce the stress-induced increase in Cort levels, although the differences were not statistically significant (Figure 6B). Notably, chronic stress upregulated β2 adrenergic receptor expression in tumor tissues, whereas SSA reduced its expression in stressed mice (Figure 6C). We next investigated the role of β-adrenergic signaling in SSA-mediated inhibition of HCC growth in stressed mice. Stressed mice were treated with SSA, Pro, SSA + Pro, or SSA + ISO, respectively (Figure 6D). Tumor volume in the SSA + ISO group was significantly larger than that in the SSA, Pro, and SSA + Pro groups (Figure 6E and F, all P < 0.01). No significant differences in tumor volume were observed among the SSA, Pro, and SSA + Pro groups (Figure 6E and F, all P > 0.05). These results indicate that SSA inhibits HCC growth in stressed mice by blocking β-adrenergic signaling. Body weight (Figure 6G) and weight gain (Figure 6H) were significantly lower in the SSA + ISO group than in the SSA + Pro group (all P < 0.05). Ki67 expression in tumor tissues was significantly higher in the SSA + ISO group than in the SSA, Pro, and SSA + Pro groups (Figure 6I). Figure 6SSA inhibited HCC growth in stressed mice by blocking β-adrenergic signaling. The effects of SSA and SSD on the NE (A) and Cort (B). Representative images of β2 adrenergic receptor in tumor tissues (C). The study scheme (D). Photos of tumor blocks collected from mice in each group (E). Comparison of tumor volume (F) body weight (G) and gain weight (H) among SSA, Pro, SSA + Pro, SSA + ISO groups. Representative images of Ki67 in tumor tissues (I). Data were presented as mean ± SEM (n = 4–5). *p < 0.05, **p < 0.01.The image A showing a bar chart of NE concentration (ng/ml). X-axis: Vehicle, Vehicle + Stress, Stress, Stress + SSD. Y-axis: NE concentration (ng/ml), scale 0 to 15. Bars show Vehicle about 5, Vehicle + Stress about 9 to 10, Stress about 5 to 6, Stress + SSD about 7. Significance marks: double asterisk and asterisk. The image B showing a bar chart of CORT concentration (ng/ml). X-axis: Vehicle, Vehicle + Stress, Stress, Stress + SSD. Y-axis: CORT concentration (ng/ml), scale 0 to 800. Bars show Vehicle about 320, Vehicle + Stress about 520, Stress about 400, Stress + SSD about 380 to 400. Significance mark: asterisk. The image C showing four micrographs labeled Vehicle, Stress, Stress + SSA, Stress + SSD, described as β2 adrenergic receptor in tumor tissues. The image D showing a study scheme with four rows: Pro 30 min before stress, SSA 30 min before stress, SSA and Pro 30 min before stress, SSA and ISO 30 min before stress. Each row connects to Restraint 5 days, then H22 cells inoculation, then Restraint 21 days, then Tumor and immunological analyses. The image E showing photos of tumor blocks arranged by group labels: SSA + Pro, SSA, Pro, SSA + ISO. The image F showing a bar chart of Tumor volume (mm superscript 3). X-axis: SSA, Pro, SSA + Pro, SSA + ISO. Y-axis: Tumor volume (mm superscript 3), scale 0 to 1000. Bars show SSA about 350, Pro about 250, SSA + Pro about 200, SSA + ISO about 750. Significance marks: double asterisk. The image G showing a bar chart of Body weight (g). X-axis: SSA, Pro, SSA + Pro, SSA + ISO. Y-axis: Body weight (g), scale 0 to 25. Bars are about 23 to 24, with SSA + ISO slightly lower. Significance mark: asterisk. The image H showing a bar chart of Gain weight (g). X-axis: SSA, Pro, SSA + Pro, SSA + ISO. Y-axis: Gain weight (g), scale 0 to 5. Bars show SSA about 3.0, Pro about 3.8, SSA + Pro about 4.3, SSA + ISO about 2.6. Significance mark: asterisk. The image I showing four micrographs labeled Ki67 with group titles SSA, Pro, SSA + Pro, SSA + ISO. Overall theme stated in the figure text and group comparisons: SSA-related groups are compared with Pro and ISO under stress for hormones, tumor volume, body weight, gain weight, β2 adrenergic receptor and Ki67.Multi-panel infographic on SSA and β-adrenergic signaling in stressed mice: hormones, tumors, weights, Ki67. SSA inhibited HCC growth in stressed mice by blocking β-adrenergic signaling. The effects of SSA and SSD on the NE (A) and Cort (B). Representative images of β2 adrenergic receptor in tumor tissues (C). The study scheme (D). Photos of tumor blocks collected from mice in each group (E). Comparison of tumor volume (F) body weight (G) and gain weight (H) among SSA, Pro, SSA + Pro, SSA + ISO groups. Representative images of Ki67 in tumor tissues (I). Data were presented as mean ± SEM (n = 4–5). *p < 0.05, **p < 0.01. SSA Changed the Immune Cell in Stressed Mice Through β-Adrenergic Signaling Immune cell populations were also analyzed in these four groups. Blood gMDSC (Figure 7A and B) and mMDSC (Figure 7A and C) frequencies were significantly higher in the SSA + ISO group than in the SSA, Pro, and SSA + Pro groups. No significant differences in blood gMDSC or mMDSC frequencies were observed among the SSA, Pro, and SSA + Pro groups (all P > 0.05). In addition, blood CD3+CD4+ (Figure 7D, P < 0.01) and CD3+CD8+ (Figure 7E, P < 0.01) T lymphocyte frequencies were significantly lower in the SSA + ISO group than in the SSA + Pro group. No significant differences in tumor CD3+CD4+ T lymphocytes (Figure 7F), CD3+CD8+ T lymphocytes (Figure 7G), mMDSC (Figure 7H), or M1 macrophages (Figure 7I) were observed among the four groups (all P > 0.05). Tumor gMDSC (Figure 7J) and M2 macrophage (Figure 7K) frequencies were significantly higher in the SSA + ISO group than in the SSA, Pro, and SSA + Pro groups. No significant differences in tumor gMDSC (Figure 7J) or M2 macrophage (Figure 7K) frequencies were observed among the SSA, Pro, and SSA + Pro groups (all P > 0.05). Figure 7SSA modulates immune cells in stressed mice via β-adrenergic signaling. Representative dot plot flow figures of blood gMDSC and mMDSC (A). Comparison of blood gMDSC (B) blood mMDSC (C) blood CD3+CD4+ T lymphocytes (D) blood CD3+CD8+ T lymphocytes (E) tumor CD3+CD4+ T lymphocytes (F) tumor CD3+CD8+ T lymphocytes (G) tumor mMDSC (H), tumor M1 macrophages (I) tumor gMDSC (J) and tumor M2 macrophages (K) among SSA, Pro, SSA + Pro, SSA + ISO groups. Data were presented as mean ± SEM (n = 4–5). *p < 0.05, **p < 0.01.A composite figure with panels A through K. A) Flow cytometry dot plots gating CD11b-positive myeloid cells, then separating granulocytic myeloid-derived suppressor cells (Ly6G-positive Ly6C-negative) and monocytic myeloid-derived suppressor cells (Ly6G-negative Ly6C-positive) across four groups. Gate percentages: SSA CD11b 56.1 percent, Ly6G 79.9 percent, Ly6C 7.95 percent; Pro CD11b 64.8 percent, Ly6G 10.6 percent; SSA plus Pro CD11b 35.7 percent, Ly6G 66.1 percent, Ly6C 14.9 percent; SSA plus ISO CD11b 84.0 percent, Ly6C 6.71 percent. B and C) Bar charts with x-axis showing four groups and y-axis showing percent of leukocyte, ranging 0 to 70 for B and 0 to 10 for C. SSA plus ISO is highest in both, with significant differences marked by asterisk (less than 0.05) and double asterisk (less than 0.01) versus SSA, Pro and SSA plus Pro. D and E) Bar charts for blood CD3-positive CD4-positive and CD3-positive CD8-positive T lymphocytes; y-axis percent of leukocyte 0 to 10. SSA plus Pro is highest; SSA plus ISO is lowest, with double asterisk significance. F and G) Tumor CD3-positive CD4-positive and CD3-positive CD8-positive T lymphocytes; y-axis percent of live cell 0 to 4 and 0 to 2 respectively; no significant differences observed. H) Tumor monocytic myeloid-derived suppressor cells; y-axis percent of live cell 0 to 0.8; no significant differences. I) Tumor M1 macrophages (CD11b-positive F4/80-positive iNOS-positive); y-axis percent of live cell 0 to 0.5; no significant differences. J and K) Tumor granulocytic myeloid-derived suppressor cells and M2 macrophages (CD11b-positive F4/80-positive CD206-positive); y-axis percent of live cell 0 to 12 and 0 to 1.5 respectively. SSA plus ISO is highest in both; significant differences marked versus SSA, Pro and SSA plus Pro.Flow cytometry compares immune cell frequencies in SSA, Pro, SSA+Pro and SSA+ISO groups. SSA modulates immune cells in stressed mice via β-adrenergic signaling. Representative dot plot flow figures of blood gMDSC and mMDSC (A). Comparison of blood gMDSC (B) blood mMDSC (C) blood CD3+CD4+ T lymphocytes (D) blood CD3+CD8+ T lymphocytes (E) tumor CD3+CD4+ T lymphocytes (F) tumor CD3+CD8+ T lymphocytes (G) tumor mMDSC (H), tumor M1 macrophages (I) tumor gMDSC (J) and tumor M2 macrophages (K) among SSA, Pro, SSA + Pro, SSA + ISO groups. Data were presented as mean ± SEM (n = 4–5). *p < 0.05, **p < 0.01. Discussion To our knowledge, this study is the first to demonstrate a novel role for SSA in inhibiting chronic stress-induced tumor growth. First, we found that SSA reversed stress-induced depressive behavior. Second, SSA significantly inhibited HCC growth and prevented the stress-induced upregulation of Ki67 in tumor tissues. Furthermore, SSA reversed the stress-induced alterations in CD3+CD4+ T lymphocytes, CD3+CD8+ T lymphocytes, gMDSC, mMDSC, and M2 macrophages. Mechanistically, SSA blocked the stress-induced increase in NE and the upregulation of β2 adrenergic receptors. Finally, we demonstrated that SSA inhibits HCC growth and reverses immune cell alterations in stressed mice by blocking β-adrenergic signaling. TCM posits that emotions represent the body’s diverse emotional responses to external environmental stimuli. The regulation of emotions in TCM is attributed to the function of “liver governing regulation and release”. When the liver’s regulatory function is disrupted and emotions become excessive, qi flow is obstructed, leading to imbalances in yin–yang and qi–blood within the organs—a condition referred to as “emotional imbalance.” This results in liver qi stagnation, which corresponds to a psychological stress response in modern medicine and contributes to depression.17–19 Radix Bupleuri is a traditional Chinese medicinal herb with functions that include harmonizing the interior and exterior, soothing the liver and relieving depression, uplifting yang to alleviate depression, and reducing fever and malaria.20 Thus, Radix Bupleuri has been used to treat depression in China for thousands of years.21 Radix Bupleuri is the primary component of Xiaoyao San, a classic antidepressant formulation originating from “Tai Ping Hui Min He Ji Ju Fang” during the Song Dynasty, which remains in use in modern China.8,9 Saikosaponins are bioactive compounds isolated from Radix Bupleuri, and modern medical research has demonstrated their antidepressant properties.22 Wang et al demonstrated that SSA improved depression-like behavior in a rat model of cerebral ischemia, presumably by increasing brain-derived neurotrophic factor expression.15 SSD has also been reported to exhibit antidepressant effects.23,24 Our results showed that SSA reversed stress-induced depressive behavior. However, SSD at a dosage of 0.5 mg/kg only partially alleviated chronic stress-induced depressive behavior. The limited antidepressant efficacy of SSD in our study may be attributable to differences in animal models and drug dosages. Liu et al investigated the antidepressant effects of SSD by oral administration at 1.50 mg/kg and 0.75 mg/kg in a rat model of chronic unpredictable mild stress, whereas Su et al administered SSD at 1 mg/kg by gavage in a lipopolysaccharide-induced depression-like mouse model.23,24 The tumor immune microenvironment comprises not only anti-tumor immune cells, such as T lymphocytes, but also immunosuppressive cells, including TAN, TAM, and MDSC.3 MDSCs are the predominant cell population in the tumor immune microenvironment and function to suppress anti-tumor immunity and promote tumor progression.25 MDSCs promote tumor progression by inducing M2 macrophage polarization, inhibiting cytotoxic T lymphocyte function, and recruiting regulatory T cells (Treg) to tumor tissues through overexpression of programmed cell death ligand 1 (PD-L1), PD-L2, cytokines, chemokines, and other factors.3,26 TAM, also referred to as M2 macrophages, promote tumor progression by inhibiting natural killer (NK) cell and T cell function, or by secreting cytokines that act directly on tumor cells.27,28 Previous studies, including our own, have demonstrated that chronic stress promotes tumor progression by remodeling the tumor immune microenvironment through recruitment of MDSC and TAN, and by altering the M1/M2 macrophage balance.7,13,29 SSA has previously been reported to possess immunomodulatory effects.14 Zhao et al showed that SSA enhanced antitumor immunity by increasing lymphocyte infiltration into tumors and shifting the Th1/Th2 balance toward Th1.30 We found that SSA altered the tumor immune microenvironment in chronically stressed mice, as evidenced by decreased IL-6 levels and reduced frequencies of gMDSCs and M2 macrophages in tumor tissues. Chronic stress induces hormonal disturbances via neuroendocrine pathways, which in turn contribute to immune abnormalities, tumorigenesis, and tumor progression.5,31 Activation of the sympathetic nervous system leads to the release of NE and Epi, representing a classic stress response. Adrenergic receptors, particularly β-adrenergic receptors, are closely associated with cancer incidence and prognosis.32–34 On one hand, NE and Epi activate β-adrenergic receptors—particularly the β2 subtype—on tumor cells, promoting tumor cell proliferation, survival, migration, and epithelial–mesenchymal transition through regulation of gene expression and cellular metabolism.35 On the other hand, NE and Epi modulate the tumor immune microenvironment via β-adrenergic receptors, thereby promoting tumor progression. For instance, Zhao et al found that chronic stress increased lung metastasis of breast cancer by promoting tumor-derived exosome secretion and enhancing neutrophil recruitment via β-adrenergic receptor activation.36 Yang et al also reported that β-adrenergic receptor blockade alleviated the immunosuppressive tumor microenvironment by promoting naive CD8+ T cell priming and effector T cell egress from lymph nodes, reducing immunosuppressive cell populations, and increasing tumor infiltration by B cells and NK cells.37 Therefore, β-adrenergic receptor blockers may represent potential therapeutic agents for cancer treatment.38 Radix Bupleuri has been shown to regulate central monoaminergic neurotransmitter systems, including NE, 5-hydroxytryptamine, and 5-hydroxyindoleacetic acid, in hippocampal and cortical tissues.39 Our study demonstrated that SSA blocked the stress-induced increase in serum NE and reduced beta-2 adrenergic receptor expression in stressed mice. Mechanistically, we found that SSA inhibited hepatocellular carcinoma growth in stressed mice by blocking β-adrenergic signaling. TCM is a valuable component of China’s cultural heritage. Previous studies have demonstrated that TCM can be used to treat chronic stress-induced tumor growth. For example, Jiang et al reported that Tong-Xie-Yao-Fang retarded colorectal tumor growth in stressed mice by promoting dendritic cell maturation, increasing CD4+ T cell percentages, and shifting the Th1/Th2 balance toward Th1 cells.40 Another study reported that quercetin inhibited breast cancer progression in chronically psychologically stressed mice by blocking the β2-adrenergic receptor pathway.41 Baicalin was shown to directly bind to β2-adrenergic receptors and reduce chronic stress-induced breast cancer metastasis.42 Chaihu Shugan San reduced psychological stress-induced susceptibility to ovarian cancer by inhibiting ubiquitination and degradation of Toll-like receptor 2 in macrophages and enhancing macrophage phagocytic function.43 Our results indicate that SSA inhibits hepatocellular carcinoma growth by modulating the tumor immune microenvironment through β-adrenergic signaling blockade in chronically stressed mice. CX3CL1 and its receptor CX3CR1 are primarily expressed in various cell types, including monocytes, macrophages, MDSCs, dendritic cells, T cells, and NK cells.44 The CX3CL1-CX3CR1 axis plays a pro-tumorigenic role by recruiting TAM and MDSC to tumor tissue.44,45 Tumor cells, TAM, TAN, and MDSC are the main sources of CXCL16 in tumor microenvironment.46 CXCR6, the receptor of CXCL16, is expressed on both tumor cells and immune cells. The CXCL16–CXCR6 axis also exerts pro-tumorigenic effects by promoting oncogenic signaling, epithelial–mesenchymal transition, and metastatic migration of tumor cells, as well as by recruiting TAM, TAN, and MDSC to tumor tissue.46,47 We found that SSA reduced CX3CL1 and CXCL16 levels in stressed mice. However, further studies are needed to elucidate the roles of CX3CL1 and CXCL16 in SSA-mediated inhibition of tumor progression in chronically stressed mice. Numerous studies have demonstrated that chronic stress promotes tumor progression through activation of β-adrenergic signaling. Administration of ISO mimics the pro-tumorigenic effects of chronic stress.34,48,49 One limitation of this study is that in the β-adrenergic signaling mechanistic experiments, groups treated with ISO alone and non-stressed controls were not included. While the comparisons among the SSA, Pro, SSA+Pro, and SSA+ISO groups demonstrated that SSA inhibits tumor growth and modulates the immune microenvironment by blocking β-adrenergic signaling, inclusion of ISO alone and non-stressed control groups would have further strengthened the mechanistic conclusions. In a separate study (under review), we have found that ISO promotes HCC progression and increases MDSC proportions, similar to the effects of chronic stress. Another limitation is that the mechanism was primarily validated using pharmacological modulation of β-adrenergic signaling. Although this approach effectively demonstrates the involvement of this pathway, genetic approaches, such as β-adrenergic receptor knockout models, would provide more definitive evidence. Future studies employing such genetic models are warranted to further validate the mechanistic role of β-adrenergic signaling in SSA’s antitumor effects. Finally, the relatively small sample size (n = 4–5 per group) in some experiments may affect the robustness of statistical comparisons. Future studies with larger cohorts are warranted to validate these findings. Conclusion Our results suggest that SSA reduces NE levels and downregulates beta-2 adrenergic receptor expression, alters the tumor immune microenvironment, and slows hepatocellular carcinoma growth in a chronically stressed mouse model in vivo. The underlying mechanism may involve enhancement of the anti-tumor immune response through blockade of the β-adrenergic signaling pathway. This study also provides new insights into the importance of positive psychological neuroendocrine immune intervention in the treatment of hepatocellular carcinoma. However, further research is needed to elucidate the mechanism by which SSA reduces NE secretion.